PDA

View Full Version : How to use Coaxial Indicator



Mosey
10-21-2009, 07:07 PM
I purchased a coaxial indicator and would like to know how to set it up.
The directions are in Chin-english, and they don't make sense.
The dial doesn't seem to move more than about 25 thou. Is there a lock screw that releases it?
(what do you expect for a cheapo?)

lazlo
10-21-2009, 07:14 PM
You're talking about a Blake clone, right?

That screw that holds the pivot arm should have a pair of brass washers on either side of it. I'd loosen it, remove, clean and re-install the washers.

The general idea is that you attach the stop bar (that keeps it from spinning), and select one of the probes that reaches the hole or cylinder you want to indicate.
Make sure the spindle is clear (!), and run it at about 500 - 700 RPM. Move the table feeds until you minimize the runout.

J. R. Williams
10-21-2009, 07:28 PM
Do not try to insert a rotating probe into a hole. I use my Blake at around 100 rpm and that is after rotating the spindle by hand to be sure things free. Then move the X and Y axis to get the indicator to the minimum point. I use a corner chair (magnetic double edge finder) and set my X and Y axis to zero from the corner of the vise which is fine for many jobs.

JRW

Pherdie
10-21-2009, 07:29 PM
Most of the import units are shipped with a split protective plastic ring/collar around the unit to limit indicator travel. This must be removed prior to use.

Fred

Bill Pace
10-21-2009, 08:11 PM
and run it at about 500 - 700 RPM. Move the table feeds until you minimize the runout.

Yeow!! Geez, I cant imagine running it that fast! I'm like JRWilliams and run mine at 100 - or even less, all it needs is a gentle sweep around the perimeter. Would think it would be dificult to follow the needle and make the X & Y adjustments at that much rpm.

lakeside53
10-21-2009, 08:20 PM
likewise.... and slower. I start very slow (VFD), but rarely go over 60 rpm even when it nicely centered. .

ahidley
10-21-2009, 09:19 PM
Its the plastic collar like stated above. Awhile ago I posted a link to a site that had a video of how to use one. If you search here for my name and indicvator and video I bet you'll find it...

wierdscience
10-21-2009, 09:27 PM
Be sure and oil the thing between uses,they will get sticky.3-1 or sewing machine oil works for mine.

Mosey
10-21-2009, 10:39 PM
ahidley,
You certainly hit the nail on the head! Removed little green plastic sleeve and it works nicely. thanks!
I would like to know how to set it up with minimal jockeying around. How do you initialy position the arms in relation to the hole you want to center on?

McNeillMachine
10-21-2009, 10:55 PM
I dont' often speak w/ such pronouncements, but allow me...

lower the indicator such that it is just above the hole/boss you're indicating. spin it by hand. eyeball it in the x direction one side, then the other. adjust. repeat. then y. adjust. repeat. lower it into the hole/over the boss. Rotate by hand, comparing value for x left/right. adjust halfway. same with Y. maybe do it a couple of times. engage the spindle and run it at 500 rpms and tweak it such that the needle stops vibrating. If it takes you any longer than 2 or 3 minutes, you're doing something wrong.

Disclaimer: I only know what I've taught myself. I could be way off accepted practice.

-Phil

hornluv
10-21-2009, 11:58 PM
I get it close with an edge finder. Eye ball the center on one axis, find the edge, move to what should be the center from that edge and find the edge on the other axis. Now move back to what should be center from there and repeat the process. That usually gets you well within .010" and I switch to the Co-ax. From there I just treat it like dialing in a four jaw chuck. Work one axis, then work the other, turning by hand. Then I turn it on and make sure. It took a lot longer to type this than it does to get a part centered.

John Stevenson
10-22-2009, 04:12 AM
If it's like most home shop use of one, follow the 8 steps below.

[1] Unpack co-ax and gloat
[2] Try to read Chinglish instructions
[3] Post on 5 forums that you have this in the hopes that someone will give the secret away how to use it.
[4] Try to decipher the 28 ways you have been told as to which way is best.
[5] Attempt to try all 28 to check.
[6] wind table al the way down so the co-ax will fit, setup, remove and wind back up to cutter hight.
[7] Do this 23 times then visit Accident and Emergency with co-ax elbow disease.
[8] Get home and put co-ax on a shelf until needed for pictures for helping next user out.

.

JCHannum
10-22-2009, 07:40 AM
While I have little good to say about the Chicom Co-ax's, the Blake is a good unit. The Chicoms are variable, good ones are OK, bad ones do not last too long.

Sir John does not like them, objecting to their length. This is only a minor problem if a bit of thought is applied. Keep in mind that the stylus is removeable and most mills have a moveable quill.

Most of the cranking up and down Sir J refers can be avoided by diligent use of these two facts, the work piece needs to be lower than the overall length of the Co-ax less the stylus to facilitate removal once center has been located. This length can, in most cases be accomodated by the quill movement. The stylus can be removed with the Co-ax in place. If a DRO is in use, it can be zeroed and the table shifted to accomodate removal. It need not be as onerous as SJ describes.

They do have theri drawbacks, but are very useful in many cases.

METALARCHER
10-22-2009, 08:05 AM
Dang! Yesterday I didn't even know what they were. Now I want one!
The LMS special isn't helping my condition either.

I tell ya it's a sickness.:D

gld
10-22-2009, 08:57 AM
I use one once 3 years ago. Got everything all set up, took the indicator out , put in my tooling, spindle would not reach work. Have not used it sense. New still in the box.

lazlo
10-22-2009, 09:33 AM
Yeow!! Geez, I cant imagine running it that fast! I'm like JRWilliams and run mine at 100 - or even less, all it needs is a gentle sweep around the perimeter.

Shrug. Works for me (on a real Blake). The official Blake instructions say not to go over 800 RPM.

I run my edge-finder at 750 rpm. Instructions say not to exceed 1,000 RPM.

How fast do you run your edge finder?

hwingo
10-22-2009, 09:49 AM
Most of the import units are shipped with a split protective plastic ring/collar around the unit to limit indicator travel. This must be removed prior to use.

Fred

This was the case with my unit. The protective sheath had to be removed before being used.

Harold

rkepler
10-22-2009, 09:50 AM
Disclaimer: I only know what I've taught myself. I could be way off accepted practice.

If you're doing it wrong that will make at least 2 of us. The only difference is that since I have a VFD I rough in the indicator on a really low spindle speed, 60 rpm or so. Everything after that is done in the same way.

Carld
10-22-2009, 10:28 AM
:eek: my my, so many ways. I tend to agree with Jim H but I have gotten to where I use the co-ax by hand rather than have the mill spin it. I have found I get a more accurate reading that way. Most holes are so rough the indicator point chatters as it runs over the surface. I can understand why some say they aren't accurate and it's not always the quality of the co-ax that causes the inaccuracy, it's how it's used.

With most indicators if you move the plunger against a flat smooth surface (such as the table on the mill) at a set height the needle will move back and forth as the indicator moves back and forth over the surface. An expensive indicator will have less slop there than the cheap ones but the slop is not as much a problem when moving in the same direction all the time. The slop comes into play when it is running over a rough surface.

I guess one of my problems is seeing where errors can come from and trying to eliminate or control them. For that reason I always seek why something happened and how to control or remove it. A dial indicator with a clock type movement is rife with slop so all you can do is understand what is happening and control it as much as possible and that is why I use the co-ax by hand rather than powering it.

Try this, bore a hole and leave the bottom a rough bore finish and polish an area at the top of the bore. Then without moving anything put your co-ax in the spindle and run it at 100 rpm in both parts of the bore and see what the needle does. You can't trust a fast spinning co-ax.

JCHannum
10-22-2009, 10:38 AM
Lower speeds tend to make reading and adjusting simpler. My mill goes down to 370 RPM, and that is what I usually use. I do turn the spindle by hand to rough it in, but running the mill makes final adjustment quicker.

McNeill's procedure is as good as any. Once the concept of operation is understood, application is rather intuitive.

Bill Pace
10-22-2009, 11:02 AM
Lower speeds tend to make reading and adjusting simpler.

McNeill's procedure is as good as any. Once the concept of operation is understood, application is rather intuitive.

I agree, McNeill's procedure is the way I do it also - except the rpm, having a VFD I take it down to 50-100 rpm. And, as you say, Jim, once you do it a couple times and get visuallised what its doing, I think its rather a simple and efficient way to get center.

Laz -- again, having the VFD, I also edge find at slower speeds - probably around 150-200ish-- just enough to get the 'kick'

mochinist
10-22-2009, 12:22 PM
You can't trust a fast spinning co-ax.I like double check expensive or tight tolerance parts anyways, I do have a blake and use it at around 200 rpm's on the manual and 500 rpm's on the cnc mills. I've gotten out my indicol holder and a tenth indicator on numerous occasions to double check the hole zero location on a tight tolerance part and the blake is always within .0005", manual or cnc.


I will say at that rpm especially on a aluminum part, it is helpful to put a dab of some oil on the surface where the co-ax indicator point will be touching, otherwise it will screw up the surface

Peter N
10-22-2009, 12:31 PM
I have one of the cheaper LDL co-ax indicators and have found it to be more than precise enough, and very useful.
I also use it the same was as Carl does, with a spanner on the drawbar and turn it by hand, then zero in one axis first, then the other.
Very quick and simple, and it needs less headroom than changing from a collet/collet chuck/drill chuck.

Peter

Glenn Wegman
10-22-2009, 12:46 PM
I dont' often speak w/ such pronouncements, but allow me...

lower the indicator such that it is just above the hole/boss you're indicating. spin it by hand. eyeball it in the x direction one side, then the other. adjust. repeat. then y. adjust. repeat. lower it into the hole/over the boss. Rotate by hand, comparing value for x left/right. adjust halfway. same with Y. maybe do it a couple of times. engage the spindle and run it at 500 rpms and tweak it such that the needle stops vibrating. If it takes you any longer than 2 or 3 minutes, you're doing something wrong.

Disclaimer: I only know what I've taught myself. I could be way off accepted practice.

-Phil

Not to criticize Phil's post but...

By the time you do that, a DTI on an Indicol (clamps ON the spindle, not IN it)is as quik or quicker, and there is no doubt about the accuracy! Along with that, the DTI can be used for numerous other operations around the shop. If you have neither a DTI or a CoAx, it may be a matter of preference, but if you already have a DTI why spend the money on a CoAx?

There may be the odd occasion where a CoAx would be a better choice over a DTI, but not very often!

Peter N
10-22-2009, 12:52 PM
Not to criticize Phil's post but...

By the time you do that, a DTI on an Indicol (clamps ON the spindle, not IN it)is as quik or quicker, and there is no doubt about the accuracy! Along with that, the DTI can be used for numerous other operations around the shop. If you have neither a DTI or a CoAx, it may be a matter of preference, but if you already have a DTI why spend the money on a CoAx?

There may be the odd occasion where a CoAx would be a better choice over a DTI, but not very often!

I think you're missing the point entirely.
With a co-ax indicator, the dial is facing straight at you 100% of the time. No mirrors, no straining your neck round the back or the sides, no parallax errors from viewing it at odd angles.
That's what makes it so simple to use, without even mentioning the extended probes so that the indicator can be down the bottom of a long bore whilst the dial is above the bore in plain sight.

Peter

jdunmyer
10-22-2009, 01:21 PM
I have one of the Chinese co-ax indicators, but have gotten away from using it. Instead, I use a TDI clamped in an Indicol and now can center a part pretty quickly:

Spin the spindle by hand with the indicator point above the part until it's "close". Put the TDI point in contact with the part. Adjust the 'X' axis so the indicator reads the same on both the Left and Right sides; I'll usually tweak the 'Y' axis a bit at the same time. Once I'm happy with the 'X', I'll make sure the 'Y' reading is the same. Note that during this procedure, I turn the spindle only 180 degrees, keeping the face of the TDI in view at all times. When I'm done, I might crank the spindle so the face is to the rear and check it with a mirror, but if the part is round, the reading is always the same as at the front.

Here's a hint for using a TDI: Do NOT try to use the Zero reading on the dial. Always adjust the dial so that pressing 'more' on the point makes the reading go 'up' or increase. My TDIs are bi-directional, and it used to be a problem figuring out which way to move the table until I learned this trick.

lazlo
10-22-2009, 01:30 PM
With a co-ax indicator, the dial is facing straight at you 100% of the time. No mirrors, no straining your neck round the back or the sides, no parallax errors from viewing it at odd angles.

Exactly. Plus you don't have to mess with the getting the DTI up/down on the Zero-Set beam. With the various tips, a Blake can indicate an 8" hole/cylinder. You need to make a beam to do that with a DTI.

Not a huge advantage either way. Different strokes...

gwilson
10-22-2009, 01:32 PM
I actually have a Blake. I don't think any of these replies have been very complete. I think the Blake type co-ax is wonderfully handy to have.

The first thing I advise you to do,is first screw on the long straight handle that you hold in 1 hand to keep the Blake from rotating. I run my unit SLOW.

Next,and VERY important,when getting ready to center something like a hole,put on the shortest leg you can,and run the leg round and round close over the hole. Adjust the mill's table until you can see that the hole is pretty well visually centered below the probe you selected. IF YOU PUT THE PROBE INTO THE HOLE WHEN IT IS TOO FAR OFF CENTER,THE PROBE CAN REACH ITS LIMIT OF TRAVEL AND BREAK THE SHAFT OFF OF THE BLAKE!!!!!!!
This last info is VERY important to remember. I saw a broken off Blake in a professional machine shop.

Running the mill slow is very adequate to indicate the hole onto center by manipulating the X and Y axes until the indicator's needle stays in 1 place. You can zero the needle's position by rotating the bezel if you want to. The important thing is to get the needle to where it moves as little as possible,with just the smallest tremor. Running the mill faster just makes centering the hole(or whatever) more difficult.

I also got the needle tip for centering punch marks,which is very handy to have. This needle tip is spring loaded,which it needs to be as the needle tip goes in and out as it is used to center the punch mark. DO NOT ATTEMPT TO USE A HOME MADE,NON SPRING LOADED NEEDLE TIP. It will break your Blake.

I have heard bad things about the Chinese Blake clones. I hope yours works o.k..

Glenn Wegman
10-22-2009, 01:48 PM
With the various tips, a Blake can indicate an 8" hole/cylinder. You need to make a beam to do that with a DTI.

Not a huge advantage either way. Different strokes...


Ever check the accuracy on that CoAx with the arm at a 45° angle to the axis? Or the wrap up on a long arm?

A properly setup DTI does not suffer from that with diameter changes.

DTI's also come with dovetails on the upper surface!!

http://img.photobucket.com/albums/v647/Fighter1/dti.jpg

There is no need for a mirror if mounted vertically anyway as you only need to sweep three sides of a bore if it it round. Left on X axis, 180° right on X axis, and toward you on Y axis. If you are not sure and want that warm feeling, a final check on the far side.

I also have a Blake.............that lives in it's box!

John Stevenson
10-22-2009, 02:58 PM
Don't even start the spindle.

Use one of these and you'll never go back to a co-ax or DTI

http://homepage.ntlworld.com/stevenson.engineers/lsteve/files/3dtester.jpg

Watch this video.

http://www.youtube.com/watch?v=RTzxebz9xkQ&feature=channel

.

lazlo
10-22-2009, 03:21 PM
John, I just got mine (a Haff & Schneider), and it's set-up on the mill -- it's great for picking up edges, like the Haimer video shows, but it seems like a pain to pick up circular features, unless I'm missing something obvious?

So to pick up a hole you drop down in the hole, wind one axis until it hits the ball, zero the DRO, wind to the opposite side, then trigger the DRO's diameter function. Then repeat for the other axis.

Seems a lot less work to use the Blake for holes and cylinders.

mochinist
10-22-2009, 05:35 PM
I centered up a hole with my blake in less time than it took to watch that video:D


I am intrigued though, who sells them in the US? I just looked at MSC and they only have an SPI brand name one, and it is quite expensive for an SPI tool.

Mosey
10-22-2009, 06:57 PM
Well thanks to all of you who responded.
I got the split plastic ring off, and it went right to work. It is best I think to pull your punches by roughly centering the feeler to the subject piece you want to center. I run it at 100 rpm (my slowest speed) and move the table and it just reaches apoint where it almost stops the needle oscillations. Then move the table in the other direction, and voila! Centered to within a thou.
By the way, I paid $40 plus shipping. Wish I had a Blake though. For my level of accuracy this is fine.

gwilson
10-22-2009, 10:30 PM
I think my Blake centers to much less than .001",but haven't used it lately,and can't remember the accuracy. Usually the item being centered isn't smooth enough to let the Blake do its maximum accuracy. I've used it several times to center a table saw blade so I can bore the hole to 3/4".

hwingo
10-23-2009, 01:07 AM
Well thanks to all of you who responded. I got the split plastic ring off, and it went right to work. I run it at 100 rpm (my slowest speed) and move the table and it just reaches apoint where it almost stops the needle oscillations. Then move the table in the other direction, and voila! Centered to within a thou.


Mosey:
You did the same thing as I in the beginning by failing to take the protective ring off the shaft. As Pherdie (Fred) first mentioned, and as I later mentioned, removal of the protective covering allows the indicator to function. For my level of work, it's fast and quite simple to operate. Glad our suggestion to remove the protective covering helped.

Harold

ulav8r
10-23-2009, 11:43 AM
My Blake indicated about .0007 run out without being in a hole. I carefully cleaned it and reduced that slightly. I then scotchbrited? the high spots but it still shows about 4 tenths run out. Someday I may send it back for repair but it will do as is if I am careful to note where the high spots are when I set it up.

JCHannum
10-24-2009, 06:09 PM
The Blake is faster on centering bores. The 3D tasters look interesting, but are pricey and offer no real advantage that I can see unless the machine is also equipped with a DRO. Mine is not, and probably never will be, so the Blake is for me.

Accuracy of reading with a Blake is not affected by sine error or such, you are merely looking for minimal movement of the indicator. Neither the Indicol or Blake are necessities, but the Blake is much simpler to use that an Indicol & DTI for me for round ID & OD features. I very very rarely ise an Indicol, thaogh I have both.

JCHannum
10-24-2009, 10:55 PM
I did not try the Blake for it's gee whizz factor. I don't waste my money on something that does not prove useful. I have had at one time or another many different tools and instruments. I give them a try, and if I do not find them to be worth having, pass them on. The Blake is a keeper, for its purpose, I have found nothing better.

lazlo
10-25-2009, 12:55 PM
The Blake is faster on centering bores. The 3D tasters look interesting, but are pricey and offer no real advantage that I can see unless the machine is also equipped with a DRO.

I've been playing around with my 3D Taster this weekend, and I'm coming to the same conclusion. It's great for indicating-in square parts. So I can indicate the fixed jaw on the vise, the side of the parallel or workpiece, and then indicate the top of the workpiece or parallel (the Z) as I'm tightening up the vise to see if the part is lifting.

You can obviously do that with a Zero-Set or Indicol holder, but you have to change setups for each axis.

I'll post some pictures later today...

But the 3D Taster is PITA for holes or cylinders. I'm staying with the Blake for that.

By the way 'Tiff, I think all the 3D Tasters have a 20mm shank, which will be a problem for someone that only has R8 (Imperial) collets or endmill holders. My Haff & Schneider looks identical to the Haimer (one German company is probably the OEM for the other), and the Chinese version is clearly copied from the German tool.

Glenn Wegman
10-25-2009, 01:25 PM
You can obviously do that with a Zero-Set or Indicol holder, but you have to change setups for each axis.
.

The IndiCol clamps to the spindle so you do not have to remove or install it in the spindle or remove or install cutting tools as you do with the Taster.

Isn't that considered a "setup"? :)

lazlo
10-25-2009, 01:41 PM
The IndiCol clamps to the spindle so you do not have to remove or install it in the spindle or remove or install cutting tools as you do with the Taster.

Isn't that considered a "setup"? :)

Glenn, point taken :) I have an Indicol, and for reasons I can't explain (habit?) I've never used it much.

I'm about to have my 3 hour weekly allotment in the shop (2 young kids :rolleyes: ), so I'll give it another shot.
My Excello has a substantially larger quill than a Bridgeport, so I'm actually curious if I can open the clamp screws that wide...

Mosey
10-25-2009, 01:52 PM
Well, now I know how to make this Chinese indicator work. Forget running it under power for now, as my machine only goes down to 150 rpm, which is too fast to see the dial hand zipping back and firth.
Start with the straight feeler straight up and down and move the table until you are eyeballed at the center. Than push the feeler out until it is past the circumference you are centering, and 1/2 of the range of the dial travel (the little hand in the middle that goes to about 6 units). Push the feeler so it clears the edge of the circular object to be centered. Drop the quill so the feeler is just below the edge of the object. Turn the quill until the feeler is roughly facing one of the quadrants of the object. Read the dial. Turn the quill 180 degrees and read again. Move the table until the reading is 1/2 the difference between the 2 readings. Turn the quill 90 degrees and read again. Turn the quill 180 degrees again. Move the table to split the difference in half again. You are now centered or very close to it.
If you want you can now turn on the machine and see how close you are to centered. Probably within 2 thousands.
Very slowly move the table to bring it to exactly centered in both axis.
I have now appreciated why one would want to pay $180 to $220 for a Blake instead of $40 to $80 for a ripoff. The ripoff does the job, but it is loose and shaky...not suitable for real precision work. I will be looking for a used Blake for the future.

Mosey
10-25-2009, 02:06 PM
Have you seen the Haimer video of the setup device? And their other wonderful devices? Wow, their stuff is incredible. If I were a pro I would have one of each.

lazlo
10-25-2009, 02:29 PM
Well, now I know how to make this Chinese indicator work. Forget running it under power for now, as my machine only goes down to 150 rpm, which is too fast to see the dial hand zipping back and firth.

You need to rough-in the centering of the indicator by hand first, to get it reasonably close. Then, when you run the spindle, you tweak the leadscrew to minimize the wobbling or "vibration" of the needle.

But doing the whole process by hand is fine too, just a bit slower.

JCHannum
10-25-2009, 03:13 PM
My mill's lowest RPM is 372RPM, and have also used it at the next higher ~700 RPM range. Once rough centered, you are looking for zero, or minimal movement of the hand. Rocking the handwheel back & forth will soon reveal when that state has been reached.

Mosey
10-25-2009, 05:08 PM
The information here is extremely valuable to beginners and other learners like me. Thanks to all of you.
Each machinist I know has their own way of doing things that works best for them, but this site gives a range of opinions and methods, with the best usually rising to the top.
The manufacturers could learn a lot from viewing what is here.

tattoomike68
10-25-2009, 05:28 PM
Yeow!! Geez, I cant imagine running it that fast! I'm like JRWilliams and run mine at 100 - or even less, all it needs is a gentle sweep around the perimeter. Would think it would be dificult to follow the needle and make the X & Y adjustments at that much rpm.


yea I turn the spindle by hand with the mill in neutral just to save the poor unit. at best use back gear and have the speed at its lowest.

most of the time I never turn the spindle on at all to be safe, I used real high quality units and see no need to run them at high speeds at all.

really if you cant indicate a bore in two or 3 laps of the Coax then you are not very good, try selling vacuum cleaners for a living. (no dig at you Bill Pace whatsoever.)

lazlo
11-05-2009, 07:50 AM
The Co-Ax indicator is not better than an Indicol with DTI or 3D taster, nor is the Indicol and DTI or 3D taster better than the Co-Ax. All three are good instruments, each with different advantages and disadvantages for a given application.

To add to what Jim says, the actual measuring instrument in both devices is the same: an ordinary dial indicator.

In the case of the Blake, the probe is is cantilevered on a smooth coaxial ring, such that lateral force on the probe is translated to an axial offset to the dial indicator.
In the case of the 3D Taster, the dial indicator has some kind of gimbal that translates force in one of three axis' into an axial offset to the dial indicator.

In other words, the 3D Taster can only measure one axis at a time. You can feel the mechanical selection mechanism "lock" into the corresponding axis when you push on it...

So although I'm finding the 3D Taster is a great time-saver for setting up square items in the mill (you hit each axis and zero the DRO), it's a pain for holes and cylinders. The Blake, a Zero-Set, or an Indicol is more suited, IMHO.

Glenn Wegman
11-05-2009, 08:49 AM
OK, so I just did a little check on the CoAx for error.

I set it up with the shortest arm, 1.5", and set the arm at approx 40° angle to the spindle axis as if indicating a 3" bore. With that setup I moved the probe over and touched a 123 block and zero'd the handwheel dial. I turned the handwheel until I had a .010" needle deflection on the CoAx and the handwheel dial showed .019" movement. So it is near a 2:1 movement ratio at that setup and the shortest arm. If you used a 3" probe on a large bore it would be WAY more error than that.

I agree 100% that a CoAx is a relative reading indicator, but just want to point out that the error can be huge in that with a 3" probe, you you could be a couple thousandths from center and have no visible needle deflection.

Is that a problem for home shop? Most likely not, but just want to make others aware of the inherant error and cosine error if someone does want to use one for fairly close tolerance work.

lazlo
11-05-2009, 09:22 AM
So it is near a 2:1 movement ratio at that setup and the shortest arm. If you used a 3" probe on a large bore it would be WAY more error than that.

I agree 100% that a CoAx is a relative reading indicator, but just want to point out that the error can be huge in that with a 3" probe, you you could be a couple thousandths from center and have no visible needle deflection.

Glenn, your measurement sounds reasonable: i.e., the cantilever mechanism on the Blake has a 2:1 ratio, but doesn't that relate to sensitivity, not error? In other words, on a Blake, I wind the axis in to a reasonable preload, and then adjust the axis' to minimize the needle wobble.

Blake's product literature indicates a centering accuracy of 2 tenths. The dial indicator has 5 tenths graduations, so minimizing the needle wobble with a reasonable preload, that seems achievable.

JCHannum
11-05-2009, 09:39 AM
One of the simplest methods of indicating a center pip on a lathe is to use a wiggler ala the Starrett No. 65 Center Test Indicator. These are a couple of mine;

http://i320.photobucket.com/albums/nn351/jchannum/P1070201.jpg

These indicators are quite accurate, and have no graduations at all. The Co-Ax indicator is no different, the dial indicator need not be graduated for it to function within its limits. This is a write up by Long Island Indicator on the Blake Co-Ax describing its use and accuracy.

http://longislandindicator.com/p5.html

As to the use of a center punch for locating center, the old hands did it regularly and with great accuracy. Marking out is an art that all should master. Use of a prick punch and magnifying glass will produce a center pip that is quite accurately centered and can be used as a reference point for subsequent operations.

Glenn Wegman
11-05-2009, 09:40 AM
lazlo,

I'm referring to the combination of probe length/cosine error that one can encounter. Try my above test with varying probe lengths and relative probe angles.:)

How can the error not vary with either?

Does Blake specify the probe length and relative angle that they claim will give you .0002"? They would make far more sense if it had just graduations and no numbers on the face!

My Blake came out of the box with a "built it" .0005 "hiccup" needle deflection. I was impressed! But I didn't but it to replace a DTI either!:)

JCHannum
11-05-2009, 09:51 AM
The LII site gives the Blake accuracy statement, it is with the furnished short probes, longer probes will lower accuracy.

It must be remembered, the Blake is measuring the hole and is no more accurate than the hole. If it is a simple drilled hole, surface imperfections will affect the reading and accuracy. If the hole is not accurate to 0.0002", the Co-Ax will not be capable of "improving" that accuracy.

Glenn Wegman
11-05-2009, 09:56 AM
I just did the same test as above, but this time I set the short probe vertical (in line with the spindle axis) and with .010" needle deflection I needed .024" handwheel dial movement.

lazlo
11-05-2009, 10:00 AM
I'm referring to the combination of probe length/cosine error that one can encounter. Try my above test with varying probe lengths and relative probe angles.:)

Right, you're measuring the cosine of the lateral force (just as the 3D Taster is doing) but why does that matter?

The needle wobble is minimized when the Blake is centered.



Does Blake specify the probe length and relative angle that they claim will give you .0002"?

Blake doesn't specify.

# Designed for centering a work piece of fixture.
# Operates at any angle.
# Use vertical or horizontal.
# Use internal or external.
# While spindle rotates, the centering indicator always faces the operator.
Includes:
# Restraining Rod.
# 4" Straight feeler.
# 4" Curved feeler
Specifications:
Length . . . . . . . . . . . . . . . . . . . . . . . . 6-7/8"
Shank . . . . . . . . . . 3/8" diameter x 7/8" long
Weight . . . . . . . . . . . . . . . . . . . 9-1/2oz. net
Feeler ball . . . . . . . . . . . . . . . . 1/8" diameter
Measuring range . . . . . . . . . . 250 axis offset
Capacity . . . internal: 5/32 to 4-1/4 diameter
. . . . . . . . . . external: zero to 4-1/4 diameter
Speed . . . . it is recommended that operating
. . . speed not exceed 800 rpm approximately
Locating . . . . . . .accuracy to within 0.0002"
. . . . . . . . . . . . . . . . . . . . . . . . of true center
Graduation . . . . . . . . . . . . . . . . . . . 0.0005"

Glenn Wegman
11-05-2009, 10:21 AM
Right, you're measuring the cosine of the lateral force (just as the 3D Taster is doing) but why does that matter?

The needle wobble is minimized when the Blake is centered.
"

The point is....

I'm just pointing out that the probe can move without any needle indication due to reduced sensitivity through cosine and probe length error.

Under certain circumstances, you could be .002" or more from the bore center and have no visible needle deflection. It could be no problem, or it could be a real problem, depending on the accuracy needed for the task.

The Taster is constant, and a DTI is constant with the probe properly positioned regardless of bore size. :)

JCHannum
11-05-2009, 10:46 AM
Don't forget that the Co-Ax is to be used with the spindle rotating. It is indicating the sweep of the stylus, not a point measurement. If it is away from center on one side, it will be twice that distance off on the other side.

On a nice clean flat surface, a wiggler can be lined up to a scribed line or the intersection of two lines with reasonable accuracy. That is not always possible, castings come to mind, sometimes only a center pip is the only point of reference. In these cases, the Co-Ax is a better choice.

The Co-Ax is not always the best choice, but is a good tool to have in hand for when it is needed. It is relatively economical, quick to set up and accurate.

lazlo
11-05-2009, 10:49 AM
The needle wobble is minimized when the Blake is centered.
...

Don't forget that the Co-Ax is to be used with the spindle rotating. It is indicating the sweep of the stylus, not a point measurement. If it is away from center on one side, it will be twice that distance off on the other side.

Exactly my point :)

JCHannum
11-05-2009, 11:15 AM
I was prompted to get my Blake Co-Ax out and refer to the instruction sheet. It states; "Dial Graduation is divided such that each division represents 0.005 inch of AXIS OFFSET (deviation from true location) when feeler is tracing a diameter of two inches. (Varies slightly with feeler angle.)"

It does not state, but I would assume, that it refers to the standard probes for use for diameters up to 4-1/4".

Glenn; who is the manufacturer of your Co-Ax?

dp
11-05-2009, 11:34 AM
This thread can't be considered complete until one has also seen this one:

http://bbs.homeshopmachinist.net/showthread.php?t=30323

Alistair Hosie
11-05-2009, 11:42 AM
could someone explain ,or better show me what such an indicator looks like please?Alistair otherwise known as Grandpa :D

JCHannum
11-05-2009, 12:01 PM
Alistair; here is an assortment of BlakeCo-Ax indicators on eBay.

http://business.shop.ebay.com/i.html?_nkw=Blake+indicator&_sacat=12576&_trksid=p3286.m270.l1313&_odkw=blake+Co-axial+indicator&_osacat=12576

I still do not buy the argument of severe cosine error since it is not measuring anything.

Glenn Wegman
11-05-2009, 01:04 PM
Jim,

The cosine error/extended probe length desensitizes the needle movement on the indicator in relation to the probe movement.

When using a CoAx, you are looking for no needle movement to indicate thet you are centered.

By changing the ratio, the needle can appear stationary when the probe still has movement. Therefore you can be off center and the probe can still be moving with no evident needle deflection. Although not a huge amount, it is still there and could be a problem if you are not aware of it, such as centering over a bore on a gear to enlarge it.

The smaller the bore, the more sensitive the needle will be.

lazlo
11-05-2009, 01:09 PM
could someone explain ,or better show me what such an indicator looks like please?

http://i164.photobucket.com/albums/u15/rtgeorge_album/Servo%20Plate/Blake.jpg

JCHannum
11-05-2009, 01:51 PM
Jim,

The cosine error/extended probe length desensitizes the needle movement on the indicator in relation to the probe movement.

When using a CoAx, you are looking for no needle movement to indicate thet you are centered.

By changing the ratio, the needle can appear stationary when the probe still has movement. Therefore you can be off center and the probe can still be moving with no evident needle deflection. Although not a huge amount, it is still there and could be a problem if you are not aware of it, such as centering over a bore on a gear to enlarge it.

The smaller the bore, the more sensitive the needle will be.

I understand the reduction of sensitivity that might be caused by use of a longer stylus, but don't think it is quite the same as cosine error. The instructions do recognize that the offset varies with feeler angle but center will still be center. Again, the dynamic aspect of the instrument's use removes many of the errors inherent in static methods. It is dependant on the diameter of the part being indicated, out of round or roughness will probably have more of an effect than any other in the accuracy. Measurement of a bore with a keyway for instance.

dp
11-05-2009, 02:00 PM
Here's a couple pictures I just took of mine, and a short video.

http://thevirtualbarandgrill.com/machinery/coaxindicator/

Caveats: This is a setup for photos only - normally I would clamp the rotor t the table upside down from the way it is shown but this provides better photos of the process.

Glenn Wegman
11-05-2009, 02:12 PM
Jim,

I should clarify here. By "probe length/cosine error" I am referring to both, and not meaning to imply that they are one and the same. Both have an effect on the accuracy of the CoAx.

Alistair Hosie
11-05-2009, 02:26 PM
OK thanks to you guys now I know what it is.Alistair

John Stevenson
11-08-2009, 07:04 AM
http://i200.photobucket.com/albums/aa294/oldtiffie/measuring/Co-axialindicator1.jpg

.[/quote]

Anyone noticed on the Co-ax's that the part that holds the stylus can only tilt one way because of the milled angle ?
OK with that but now look on the curved stylus, the flat to hold it makes the stylus point to the centre of the circle, not to OD ????????

Decided to use mine the other day when setting a large motor housing up on the mill and I noticed this as I was setting it up with the small curved stylus.

It was a moot point anyway as the housing was tall enough not to allow the Co-ax in even when the bed was right down.

You will just have to polish it up Tiffie and put it back in the box with the rest of the unused tools.

.

JCHannum
11-08-2009, 07:15 AM
Anyone noticed on the Co-ax's that the part that holds the stylus can only tilt one way because of the milled angle ?
OK with that but now look on the curved stylus, the flat to hold it makes the stylus point to the centre of the circle, not to OD ????????.

I don't see your point John, the curved stylus is used for centering on the OD of a round part, not the bore.

BTW, there is no law against modifying or making your own stylus to serve special applications. (Within reason of course.)

John Stevenson
11-08-2009, 07:22 AM
I don't see your point John, the curved stylus is used for centering on the OD of a round part, not the bore.

BTW, there is no law against modifying or making your own stylus to serve special applications. (Within reason of course.)

OK didn't think of that, most of my work is with bores but ironic that no one has picked up on this before? 8 styli in my box, 6 bent ones all external ?

John Stevenson
11-08-2009, 08:23 AM
I often wonder if we don't get carried away by all that's out there to offer ? What I call the magpie effect and I should know because I have one :D

http://www.stevenson-engineers.co.uk/files/magpie1.jpg

I went to use the Co-ax on Friday because I need to pick this bore up and also to refresh myself on how easy / hard it was to use.
Moot point as it wouldn't fit but it was quite a tall housing, 8" - 9" or so.

I then went on to set up as a normally do , there was already a collet and tool in the spindle, 10mm as it happens so i just eversed it, nipped it up,not tight and held a blue felt tip on the shank as it was revolving.

I then ran it to one side until it touched the bore and wiped the blue off, zero the DRO and refresh the blue and do the other side, half the distance and lock the axis.
Do the same for the other axis and jobs a good un.

Sometimes if the boring head is already in and I don't want to swap it back for a setup tool I will turn the head backwards so it swipes the side to get a reading then do the same on the other side.

It's not because of lack of equipment, I have boxes of it springing out my ears but is it all needed ? Did Grandad have all this and he built the industrialised world.

Take centring a shaft for a keyway, something I do loads of. I have dial gauges, those V indicators, slip gauges, DRO's, comparators and haemorrhoids.

One of the best tools we all have is the MKI eyeball. Look at the shaft below.

http://www.stevenson-engineers.co.uk/files/centreshaft.jpg

This has been setup and very accurately brought to centre under the cutter to the best of my ability using a DTI and the DRO on the machine and then lightly touched on the work as shown by the indentation on the left.

Now the table has been moved slightly sideways in X and then purposely moved 0.001" to the side using the DRO [ new Sino system so quite decent ] and a new indent made.

Look at the centre pip on the right, this is only 1 thou off centre but it stands out like my dick [ possibly because it's about the same size ? ]

So do we need all this engineering bling when we can get by with less ?

.

oldtiffie
11-08-2009, 08:59 AM
Thanks John.

That magpie is a classic case of all black and white (no grey -male) or shades of grey (more aggressive) female too.

The IRA will getcha for wearing that green coat - but it'd be worse if Ian Paisley came pounding on your door!!!.

Perhaps you are Irish after all - one of them little green men you see at the Pub!! - a Leprechaun no less.

http://en.wikipedia.org/wiki/Leprechaun

(Mechanical/shop questions:
Q1.
What would you have if you had two small balls in your hand?

A1.
The undivided attention of a Leprechaun)

Q2.
What is a ball-bearing rat-catcher?

A2.
The shop Tom cat.

Pick up my coat on your way - and I will see you there.

lazlo
11-08-2009, 09:04 AM
Anyone noticed on the Co-ax's that the part that holds the stylus can only tilt one way because of the milled angle?

OK with that but now look on the curved stylus, the flat to hold it makes the stylus point to the centre of the circle, not to OD ????????

There's a Phillips screw that holds the probe socket (with the milled angle) on the cantelever beam. Unscrew it, flip the probe socket around, and voilà -- you've just switched from bore indicating to cylinder indicating ;)

http://longislandindicator.com/Resources/blakeco-axindica.jpeg

hwingo
11-08-2009, 09:44 AM
Hi Guys,

I have been following this thread because I have a Co-Ax indicator as that shown by Tiffie.

I am somewhat confused regarding accuracy of the Co-Ax and use of the curved probe vs the straight probe.

1. Is it the general consensus that the Co-Ax is basically accurate for indicating a bore when compared to the other instruments mentioned by Tiffie?

2. Is the curved probe designed exclusively for OD and the straight probe designed for ID or can the curved probe be used as pointed out in the previous post?

Harold

lazlo
11-08-2009, 09:47 AM
Is it the general consensus that the Co-Ax is basically accurate for indicating a bore when compared to the other instruments mentioned by Tiffie?

It's very accurate. Blake's specs indicate centering within 2 tenths. We're still arguing whether that accuracy decreases due to the length (cosine angle) of the probes, since it an indirect reading setup that's measuring the axial bore offset...


Is the curved probe designed exclusively for OD and the straight probe designed for ID

No, you can flip the probes around. See the picture above, from Long Island Indicator. I do it all the time.

John Stevenson
11-08-2009, 09:59 AM
There's a Phillips screw that holds the probe socket (with the milled angle) on the cantelever beam. Unscrew it, flip the probe socket around, and voilà -- you've just switched from bore indicating to cylinder indicating ;)

http://longislandindicator.com/Resources/blakeco-axindica.jpeg

Think again, all you are doing is the same as revolving the co-az 180 degrees.
In your picture try to push that probe out to pick a 4" bore up, t will foul the angle.

Only way to do it is two flats but why has no one else picked this up ?

lazlo
11-08-2009, 10:04 AM
Think again, all you are doing is the same as revolving the co-az 180 degrees.

No, the mill cutout on the probe socket makes it one-way. I.e, it drives the plunger ring if pushed inward (for centering a hole) or pushed outward, for centering a cylinder.

If you have the probe socket set for pushing inward (bores) and you push it outward, it just folds open, and won't nudge the indicator. That's why the setscrew flat is only on one side.

Play around with the Blake in your hand and you'll see what I mean.

hwingo
11-08-2009, 10:06 AM
It's very accurate. Blake's specs indicate centering within 2 tenths. We're still arguing whether that accuracy decreases due to the length (cosine angle) of the probes, since it an indirect reading setup that's measuring the axial bore offset...



No, you can flip the probes around. See the picture above, from Long Island Indicator. I do it all the time.

Thanks Lazlo. I had in fact flipped mine around for indicating a bore. Admittedly, flipping the probe around was intuitive rather than knowing for certain.

Also, pointing out that the discussion is based on length (cosine angle) is of great help (although I still have much to learn regarding cosine:o ).

I appreciate your response.

Harold

hwingo
11-08-2009, 10:11 AM
Humm, I need to go back and see how I had my Co-Ax set up because I, for certain, used it when centering the bore under my cutter.

Harold

EDIT: I obviously used the straight arm because it's still in the Co-Ax indicator.

dp
11-08-2009, 01:25 PM
There's a Phillips screw that holds the probe socket (with the milled angle) on the cantelever beam. Unscrew it, flip the probe socket around, and voilà -- you've just switched from bore indicating to cylinder indicating ;)

http://longislandindicator.com/Resources/blakeco-axindica.jpeg

You're making a funny, right?

dp
11-08-2009, 01:51 PM
Anyone noticed on the Co-ax's that the part that holds the stylus can only tilt one way because of the milled angle ?
OK with that but now look on the curved stylus, the flat to hold it makes the stylus point to the centre of the circle, not to OD ????????

Just have to be smarter than the problem, as a friend of mine says. :D

http://thevirtualbarandgrill.com/machinery/coaxindicator/ci-bore.jpg

dp
11-08-2009, 02:18 PM
I may give it some use but I would still check it with an Indicol (more of that in later post/s). But I suspect that it won't get any regular sunburn. I was not impressed with the lesser accuracy and sensitivity as the lengths of the styli increased.

It depends on the setup, I think. In the case of my example, a shorter probe would be less accurate. The absolute error increases with distance.

http://thevirtualbarandgrill.com/machinery/coaxindicator/

And when using it differentially in this manner the total indicated error is twice the actual error.

It is a matter of learning to know the numbers on the dial are simply there for the ride and have no particular meaning. Except the 0.0005" number. If that said 0.005" you would likely have a useless tool for the purpose.

BobWarfield
11-08-2009, 02:24 PM
I decided figuring out the accuracy of one of these Co-Axial indicators would be a good geometry exercise for my G-Wizard Machinist's Calculator. Anyone is welcome to try out G-Wizard, BTW. You will find information on how to do that here:

http://www.cnccookbook.com/CCGWizard.html

Back to our calculation. The CoAx works by swinging its arm across a pivot. The end above the pivot pushes up a ring that the indicator reads off of. The lever on mine is about 1/2" long. So an important question to understand is how far must that lever move in order to move the indicator a particular distance, such as 0.001"?

Here is the geometry:

http://www.cnccookbook.com/img/Tooling/CoAxGeometry.jpg

And here is G-Wizard calculating that the angle of the swing is 0.1146 degrees to move the indicator 0.001":

http://www.cnccookbook.com/img/Tooling/CoaxAngle.jpg

Given that knowledge, its easy to use the same calculator to figure out for a given probe length, how much it deflects to generate the same angle (0.1146 degrees).

Here is our result:

For a probe length of 2", here is the probe deflection when the indicator measures the given deflection:

0.001" of indicator deflection = 0.004" of probe deflection
0.0005" of indicator deflection = 0.002" of probe deflection
0.0001" of indicator deflection = 0.0004" of probe deflection

So, how closely can you read your indicator? Remember, you aren't trying to take a precise measurement, you're just trying to see whether you've "stopped the indicator" moving.

The Blake shown by Lazlo is calibrated at 0.0005" increments. I think one could easily see whether that needle was moving an amount to show a 0.0001" deflection, which would yield an accuracy of 0.0004" on the probe.

Heck, you might even see half a tenth looking closely and hit the spec they claim.

My guess is the cheap clones are half as accurate, as they only have 0.001" indicators in them.

I wrote this up in a little more detail with more probe lengths and so on in my blog here:

http://www.cnccookbook.com/index.htm

Look for the article called, "How Accurate is Your Co-Axial Indicator?", posted 11/8/09 if you're coming on this thread late.

My conclusion is just what I'd said before: The Co-Axes are a really fast and easy way to get within a thou. If I need closer, I break out my Zero Set with an Indicol tenths indicator:

http://www.cnccookbook.com/img/Tooling/WidgitZeroSet.jpg

That one was shopmade by CNCZone's Widgitmaster, BTW. He sells them periodically. Really a nice piece!

Cheers,

BW

dp
11-08-2009, 03:33 PM
Back to our calculation. The CoAx works by swinging its arm across a pivot. The end above the pivot pushes up a ring that the indicator reads off of. The lever on mine is about 1/2" long. So an important question to understand is how far must that lever move in order to move the indicator a particular distance, such as 0.001"?

Here is the geometry:


In practice this is probably close enough but is not an accurate description. The pivot point is not fixed, for one thing - it moves at an angle to the probe tip, and that geometry change changes everything. Just not for small degrees of motion. The probe tip does not describe a circle centered on the quill center when in use, and does not necessarily move perpendicular to the surface being measured.

BobWarfield
11-08-2009, 05:56 PM
In practice this is probably close enough but is not an accurate description. The pivot point is not fixed, for one thing - it moves at an angle to the probe tip, and that geometry change changes everything. Just not for small degrees of motion. The probe tip does not describe a circle centered on the quill center when in use, and does not necessarily move perpendicular to the surface being measured.

And yet, do those differences materially affect what's really going on?

Or, as is so often the case, are we back to counting how many angels fit on the head of a pin?

Cheers,

BW

JCHannum
11-08-2009, 06:01 PM
Thanks Jim.

The Taster (mine) reads directly very accurately to a calibrated 0.01mm (0.0004") which far exceeds the accuracy of the Co-ax even at its smallest/shortest stylus.

I used the example of just using the lead-screw do show that the traditional way (no DRO's) works - and it works very well. It would be better or easier with a DRO - of course.

That the Taster was not on the centre at "Y" when doing "X" is irrelevant as the hole/bore is just as symmetrical anywhere as it is at centre.
Personally, if I were centre-ing over a cylinder (internal or external) I prefer to use an Indicol as it is easy, quick, cheap and takes up a lot less "Z"/"head" space - which is important on a smaller mill like my Sieg X3.

I have some pics to edit and up-load to show the Co-ax, Taster and Indicol on my Sieg X3. I will post them shortly.

The "centre-ing" is a secondary feature of the Taster as its prime forte' is in "blue-printing" as it can indicate surfaces/features in "X","Y" ("Z" too) very accurately - and that is what I bought it for - as well as functioning as excellent edge-finder as it puts the machine/mill spindle axis right on the edge -directly - and well within the 0.01mm/~0.0004" dial calibrations -I don't know of any other reasonably priced and readily available similarly featured product with such a range of accuracies.

The Blake's (I cannot speak for the Chicom copies.) stated accuracy is to within 0.0002" of true center, more than twice the accuracy of the 3D taster.

I asked that you describe the use of a 3D taster in centering a 2" bore to prove the point that a Blake is faster and more accurate than a 3D taster. There is no way you can argue that it is not. This is what the Blake was designed to do, and it does it very well.

When used in conjunction with a DRO, regardless of the accuracy of the 3D taster, the results will be no more accurate than what the DRO is capable of plus any error inherent in the setup. Since locating center requires at minimum six distinct DRO measurements, there are six potential sources of error when centering by this method.

The Blake Co-Ax costs a bit over $200.00, the cheapest 3D tasters appear to start at about twice that, adding a thousand or so dollars for a DRO makes the Blake a bargain for its intended purpose.

The various arguments about cosine error are meaningless. The Blake achieves stated accuracy with the supplied feeler at a diameter of two inches, reversing the curved probe adds about 1-1/2" of diameter to that. Using a longer feeler induces cosine error of course, but the same holds true with any other instrument of this type.

lazlo
11-08-2009, 06:11 PM
The Blake's (I cannot speak for the Chicom copies.) stated accuracy is to within 0.0002" of true center, more than twice the accuracy of the 3D taster.

The Haff & Schneider 3D Taster is spec'd with a measured (hand-signed) accuracy of .005 mm, or 2 tenths.


The Blake Co-Ax costs a bit over $200.00, the cheapest 3D tasters appear to start at about twice that, adding a thousand or so dollars for a DRO makes the Blake a bargain for its intended purpose.

If you want to compare apples to apples, the Haimer/Haff & Schneider retails for $850. Clearly not feasible for a HSM'er.

The Chinese Blake clones run ~ $75, and the Chinese 3D Tasters are around $150. The Tasters are really intended for CNC setup, so if you have a home-shop CNC, they seem ideal for zero'ing your axis, but a Blake, or Zero-Set, or Indicol is really more practical and price efficient for a manual machine.

clutch
11-08-2009, 06:14 PM
A test indicator is more precise but depending on your needs it isn't better. Time and ease matters.

For the tolerances on many of the parts we run on our vmc's, a Coax or for that matter the MHC clone is just fine.

Clutch

S_J_H
11-08-2009, 07:42 PM
I'm baffled why some of you guys run the co-ax so slow. I run mine around 300-500rpm. I find it faster to use.
When comparing it with my best indicator, a Brown&Sharp T.I. with .00005" graduations the co-ax is not in the same league. But very rarely do I need to be that accurate.

I still think a co-ax would make for a great project. I'm going to make one and improve upon my import.


Chinese 3D Tasters are around $150
Robert, do you have a link to anybody selling them that cheap? I have to wonder just how poorly they are made compared to the real thing.

Steve

dp
11-08-2009, 07:47 PM
I'm baffled why some of you guys run the co-ax so slow. I run mine around 300-500rpm.

I run mine quite fast as well but for a video it's a blur so I slow it to under 50 rpm.

oldtiffie
11-08-2009, 07:49 PM
The Blake's (I cannot speak for the Chicom copies.) stated accuracy is to within 0.0002" of true center, more than twice the accuracy of the 3D taster.

I asked that you describe the use of a 3D taster in centering a 2" bore to prove the point that a Blake is faster and more accurate than a 3D taster. There is no way you can argue that it is not. This is what the Blake was designed to do, and it does it very well.

When used in conjunction with a DRO, regardless of the accuracy of the 3D taster, the results will be no more accurate than what the DRO is capable of plus any error inherent in the setup. Since locating center requires at minimum six distinct DRO measurements, there are six potential sources of error when centering by this method.

The Blake Co-Ax costs a bit over $200.00, the cheapest 3D tasters appear to start at about twice that, adding a thousand or so dollars for a DRO makes the Blake a bargain for its intended purpose.

The various arguments about cosine error are meaningless. The Blake achieves stated accuracy with the supplied feeler at a diameter of two inches, reversing the curved probe adds about 1-1/2" of diameter to that. Using a longer feeler induces cosine error of course, but the same holds true with any other instrument of this type.


Jim.

Just for info - put your co-ax in your mill and mount the shortest straight stylus into it - vertically.

Mount a say vise or angle plate on the mill table. Bring the co-ax stylus up to the angle plate/vise and zero the co-ax dial.

Now bring a good DTI up to the angle plate/vise and zero the DTI. Re-set the zero on the co-ax if needed. Now move the mill so that the DTI is off-set by 0.001". Now read off the co-ax indicator. Is it reading 0.001" - or not.

Now repeat the exercise with the stylus at 45 degrees to the vertical.

Now repeat the exercise with all styli (short, medium and long) at the vertical and at 45 degrees.

I will be very surprised if most do not register a 0.001" off-set on the co-ax indicator - perhaps 0.0005" as it is eccentricity - "run-out" - we are after.

I will just as surprised if the co-ax indicator indicates less as the length of the styli increases - and the "sensitivity" with it.

If this is the case then I'd be less than confident that the "minimum deflection" sought using the longer styli was meaningful at all in terms of knowing what the "least swing/oscillation" represented as regards degree of accuracy of the centre-ing.

Likewise, unless the co-ax indication is not consistently correct through-out its range then the 0.0005" on the co-ax indicator is either not correct of is not achievable - or both. If this is so,then the co-ax should come with a conversion chart to convert indicated off-set in units of 0.0005" to "actual".

If I am centre-ing I not only want to know that it has a degree of "swing" but what that represents in terms of the accuracy of the concentricity between the job and the mill quill spindle.

My concern here is mainly that some here may assume or rely on that "0.0005" accuracy through-out the co-ax operating range when it may not be.

If I am correct then some here may have spent money on an item that really only runs a bad second to an Indicol or a good 0.0005"/0.01mm Test Dial Indicator.

I will post the pics promised in an earlier post later in the day - I think that they will be "interesting".

dp
11-08-2009, 08:08 PM
I will just as surprised if the co-ax indicator indicates less as the length of the styli increases - and the "sensitivity" with it.

In the manner I use it, it is safe to say that removing all the tick marks and numbers from the dial of the coaxial indicator would do nothing to hurt the functionality of the device. The numbers are meaningless and could as easily be pictures of Disney characters for all the value they have. :)

oldtiffie
11-08-2009, 08:14 PM
Seems so Dennis.

Goofy??

http://www.google.com.au/search?hl=en&source=hp&q=disney+goofy+images&meta=&aq=1m&oq=goofy+%2B+image

Carld
11-08-2009, 08:16 PM
dp, I feel the same way. All you need on the dial is a "0" with a line. The only thing a co-ax is really good for is to center something under the spindle and I have found it does as well with the short feeler as with the long feeler. I have found the smoothness of the surface the feeler is running has a lot to do with how accurately you will center the work.

All I ever do is use "0" at the center of the extremes of the needle movement and then work down from there to as little wiggle as I can get.

The biggest advantage of the co-ax is the dial always faces you while your centering the work.

Tools are only as good as the operator can make them work.

lazlo
11-08-2009, 08:24 PM
Robert, do you have a link to anybody selling them that cheap?

800Watt carries them occasionally.

Glenn Wegman
11-08-2009, 08:27 PM
Blake clearly states that it is accurate to .0002" measuring a 2" bore using the shortest probe. Any bore larger or smaller than that or a probe longer than that is subject to error.

Any devise with an articulating arm is subject to error period. Like it or not, it's there! How much error and whether it affects your particular job is irrelevant here, or up to your discression.

An Interapid DTI is only accurate with the probe at a 12° angle. There is a reason for that. It's called "Cosine Error" and they recognize and state that! All articulated arm devises will only be accurate with the probe or arm at one particular spot in it's range of movement.

As JCHannum and I have stated, a CoAx will get you centered to within a thousandth. If you need better than that, it's not the tool to use, so get out a DTI.

A CoAx is a handy devise and I use mine regularly, but if I want to be sure of where I am, I use a DTI.

lazlo
11-08-2009, 08:27 PM
In the manner I use it, it is safe to say that removing all the tick marks and numbers from the dial of the coaxial indicator would do nothing to hurt the functionality of the device. The numbers are meaningless

What Dennis said. The Blake is just a relative run-out indicator. You tweak the axis' until the needle stops wobbling.

Is the horse dead yet? :D

lazlo
11-08-2009, 08:35 PM
Blake clearly states that it is accurate to .0002" measuring a 2" bore using the shortest probe. Any bore larger or smaller than that or a probe longer than that is subject to error.

Glenn, where are you reading that? The instructions that came with my Blake just say:


"Locating accuracy: To within .0002 in. of true center. "

No qualifications are stated.

Same deal on their web page:

http://www.blakemanufacturing.com/pages/aboutus.html

I'll call them in the morning and ask tech support. You're making me wonder if some of my motor plates are off-center :) I'll have to try a couple of experiments...

JCHannum
11-08-2009, 08:37 PM
I agree with dp, the graduations are uneccessary.

Stop trying to compare the Co-Ax to a DTI, they are two different instruments, used for two different purposes. The Co-Ax dial reads offset, not a point. It is designed to be used dynamically to locate center. You are not looking for any number, you are looking for minimal deflection of the pointer.

Does a too long feeler or a too steep angle of the feeler induce cosine error in the Co-Ax, of course it does. Does a too long feeler or too steep of an angle of the feeler induce cosine error in a DTI, yes again. Using any instrument outside the range of its design parameters will result in erroneous readings. The Co-Ax will actually tend to be less affected as it is not used for absolute measurement, but averages the reading obtained through the sweep of the feeler.

JCHannum
11-08-2009, 08:44 PM
Glenn, where are you reading that? The instructions that came with my Blake just say:


"Locating accuracy: To within .0002 in. of true center. "

No qualifications are stated.

Same deal on their web page:

http://www.blakemanufacturing.com/pages/aboutus.html


I'll call them in the morning and ask tech support. You're making me wonder if some of my motor plates are off-center :) I'll have to try a couple of experiments...

The instruction sheet states 0.0002" in a 2" bore. It does not mention feeler length, but I think you can assume the standard short feeler.

I see no reason to doubt that given perfect conditions. Those conditions having to do not with the Blake, but with the bore being indicated and the spindle the Blake is in. If there is wobble in the spindle or the bore is not perfectly round, the pointer will not be dead still and minimal deflection will have to be settled for.

This is another advantage of the Blake over the static devices, it will give the best picture of where center should be. Given an egg shaped bore, a static instrument will only see the points actually contacted. Depending on the orientation and degree of the eggedness, a static indicator could be off quite a bit.

BTW, I just checked the Blake link and they offer this table of correction factors for feeler length and bore diameter.

http://www.blakemanufacturing.com/pages/coaxvalues.html

lazlo
11-08-2009, 08:57 PM
The instruction sheet states 0.0002" in a 2" bore. It does not mention feeler length, but I think you can assume the standard short feeler.

Jim, you mentioned eons ago that you had a PDF of the Blake instructions -- could you email them to me? I have a single-page sheet that was folded-up in my Blake box, and it doesn't mention the 2" bore.


I just checked the Blake link and they offer this table of correction factors for feeler length and bore diameter.

http://www.blakemanufacturing.com/pages/coaxvalues.html

So that table is saying that with a 2" feeler, the dial graduations are 5 tenths. With a 4" feeler the graduations are a thou, and with a 6" feeler the graduations are 1 1/2 thou. But you minimizing the wobble in either case, so how much accuracy do you think you're losing on the longer feelers?

JCHannum
11-08-2009, 09:07 PM
I don't remember a PDF, but I have CRS. I do have the little sheet. On the third page, just above the CARE instructions, it mentions Dial Graduations and the 2" bore.

Referencing the table, any loss of accuracy appears to be a non-problem as far as I am concerned.

oldtiffie
11-08-2009, 10:15 PM
The instruction sheet states 0.0002" in a 2" bore. It does not mention feeler length, but I think you can assume the standard short feeler.

I see no reason to doubt that given perfect conditions. Those conditions having to do not with the Blake, but with the bore being indicated and the spindle the Blake is in. If there is wobble in the spindle or the bore is not perfectly round, the pointer will not be dead still and minimal deflection will have to be settled for.

This is another advantage of the Blake over the static devices, it will give the best picture of where center should be. Given an egg shaped bore, a static instrument will only see the points actually contacted. Depending on the orientation and degree of the eggedness, a static indicator could be off quite a bit.

BTW, I just checked the Blake link and they offer this table of correction factors for feeler length and bore diameter.

http://www.blakemanufacturing.com/pages/coaxvalues.html

Thanks Jim.

The table from the link at Blake that you provided let the cat out of the bag as regards what the value of the scale graduations are at varying probe lengths and diameters. The accuracy of the 2" feeler at 2" diameter is spot on and the rest of the 2" feeler readings are about "half a tenth" over the scale 0.0005" calibrations. The 4" and 6" scale actual movements have error or "scale" factors of about 2X for the 4" feeler and up to 3X for the 6" feeler.

It is pretty much what I expected - right from the start - but getting that info was like drawing a tooth.

I don't know if Blake includes that correction sheet in the package wit a new item or not. From the comments thus far it seems not - but it should!!!


I see no reason to doubt that given perfect conditions. Those conditions having to do not with the Blake, but with the bore being indicated and the spindle the Blake is in. If there is wobble in the spindle or the bore is not perfectly round, the pointer will not be dead still and minimal deflection will have to be settled for.

I agree as regards the "wobble" with an out-of-round bore but provided the bore is round, and the spindle bearings are OK, any "wobble" of the co-ax will not matter at all as the pointer will still follow a true circular path that is concentric with the spindle axis. No different to any test indicator in or on a cylinder and held in any holder in drill chuck or a collet.


This is another advantage of the Blake over the static devices, it will give the best picture of where center should be. Given an egg shaped bore, a static instrument will only see the points actually contacted. Depending on the orientation and degree of the eggedness, a static indicator could be off quite a bit.

Nope - don't agree there either.

The co-ax will "see" two highs and two lows on an "egg" or "oval" shaped cylinder and if everything was perfect the minimum needle fluctuation would be the difference between the radii of the high and low points on the axis of the egg/oval. It may happen that if the differences and the fluctuations as well as the mill spindle were too high the co-ax feeler may not have time to "recover" and a reduced "error" as well as "hammering" of the feeler on the cylinder under test may well occur.

A co-ax may/will indicate the extent of the "highs" and "lows" but not where they are.

A good dial indicator swung - by hand - with the "direct reading" dial indicator in the spindle chuck/collet will not only accurately pick up the extent of the irregularity - ie "out of round" but will also locate/position all "errors" and irregularities.

The "Indicol" and its variations are looking better all the time as "half a thou" on an indicol indicator (within its "co-sine error" parameter) is "half a thou" at any position.

dp
11-08-2009, 11:48 PM
Glenn, where are you reading that? The instructions that came with my Blake just say:


"Locating accuracy: To within .0002 in. of true center. "

If anyone finds a hole I've made in anything in my shop that is that round I'll buy them a beer and have one my own damn self.

However - when I checked the alignment of my tail stock with my coax indicator it was withing a needle's width in all quadrants with a 2" probe when I finished aligning it. Not bad for a 9" Grizzly. Hell, I'll bet PapaGriz will want it back for his museum and give me a free G0602 for my trouble :).

lazlo
11-08-2009, 11:58 PM
After everyone's done dissing the Blake, can we argue about the Centricator? :D

No cosine error...

oldtiffie
11-09-2009, 12:09 AM
Dennis.

Will that stop anybody "grizzling" - or not - or don't they need any excuse - 'coz they like "grizzling"?

Is it too much of a long bow to draw to suggest that as everything at "Grizzly" is "Chinese" that "grizzling" is a "Chinese" activity?

Ipso facto, can I expect a lot of "grizzling" about the "good(ness??)" (or lack of it?) with your Chinese co-ax in a Chinese chuck on your Chinese lathe?

(Back-ground noise: "grizzle, grizzle, mumble, mumble, sniffle, snivel - f**k "Tiffie" - "good idea" etc.etc.).

dp
11-09-2009, 12:22 AM
Dennis.

Will that stop anybody "grizzling" - or not - or don't they need any excuse - 'coz they like "grizzling"?

Is it too much of a long bow to draw to suggest that as everything at "Grizzly" is "Chinese" that "grizzling" is a "Chinese" activity?

Ipso facto, can I expect a lot of "grizzling" about the "good(ness??)" (or lack of it?) with your Chinese co-ax in a Chinese chuck on your Chinese lathe?

(Back-ground noise: "grizzle, grizzle, mumble, mumble, sniffle, snivel - f**k "Tiffie" - "good idea" etc.etc.).

No, actually I like my fake Blake and I even like my Grizzly lathe but I do wish I'd had more money when I bought it - I'd have bought a larger one. I've never really been bothered much by the Made in China label. I don't like junk but I don't care where it's made. It's still junk. I have a lot of made in China tools that I like quite a bit. Hell, I even like my Snap-On tools I bought in the 1960's.

JCHannum
11-09-2009, 01:06 AM
Thanks Jim.

The table from the link at Blake that you provided let the cat out of the bag as regards what the value of the scale graduations are at varying probe lengths and diameters. The accuracy of the 2" feeler at 2" diameter is spot on and the rest of the 2" feeler readings are about "half a tenth" over the scale 0.0005" calibrations. The 4" and 6" scale actual movements have error or "scale" factors of about 2X for the 4" feeler and up to 3X for the 6" feeler.

It is pretty much what I expected - right from the start - but getting that info was like drawing a tooth.

I don't know if Blake includes that correction sheet in the package wit a new item or not. From the comments thus far it seems not - but it should!!!



I agree as regards the "wobble" with an out-of-round bore but provided the bore is round, and the spindle bearings are OK, any "wobble" of the co-ax will not matter at all as the pointer will still follow a true circular path that is concentric with the spindle axis. No different to any test indicator in or on a cylinder and held in any holder in drill chuck or a collet.



Nope - don't agree there either.

The co-ax will "see" two highs and two lows on an "egg" or "oval" shaped cylinder and if everything was perfect the minimum needle fluctuation would be the difference between the radii of the high and low points on the axis of the egg/oval. It may happen that if the differences and the fluctuations as well as the mill spindle were too high the co-ax feeler may not have time to "recover" and a reduced "error" as well as "hammering" of the feeler on the cylinder under test may well occur.

A co-ax may/will indicate the extent of the "highs" and "lows" but not where they are.

A good dial indicator swung - by hand - with the "direct reading" dial indicator in the spindle chuck/collet will not only accurately pick up the extent of the irregularity - ie "out of round" but will also locate/position all "errors" and irregularities.

The "Indicol" and its variations are looking better all the time as "half a thou" on an indicol indicator (within its "co-sine error" parameter) is "half a thou" at any position.

There was no cat and no bag. Blake has simply published the correction factors for the instrument. It was not like drawing a tooth, it was right there on page two of Blake's web page. Why should they include them with their instruments? Does any manufacturer include a similar sheet with their DTI?

You must remember that the Co-Ax is used under power and reflects dynamic changes in the spindle that a static DTI cannot. Wobble and runout of a spindle under power will be different than at rest.

The rest of my statements are correct, the Co-Ax will give a picture of the center of an irregular bore that a DTI cannot. The Co-Ax does not care where the highs and lows are, but will indicate the best average of them. You cannot do that with a DTI.

oldtiffie
11-09-2009, 01:18 AM
As I am here to assist where possible (and while I'm on a roll to stick more up the "co-ax" aficionados too - mustn't miss out on that) I thought I'd contribute a bit more as regards indication and concentric alignments.

Here are some of the pics I promised in earlier posts.

The mill is my Sieg X3 mini mill-drill (Chinese of course - and VERY good too it is!!) instead of my (Chinese - again) larger HF-45 square column dove-tail mill/s. This is to give some of the folks with "smaller" mills a bit of ago - for a change.

I will use links instead of images as there is a limit of four images per page and some here have slow/er internet connections.

So here we go - and not a co-ax insight!!!

It is all about being as basic and as effective and accurate as I can with frequently used tools that are readily available and cheap - no exotic Tasters or co-ax indicators at all - but with the job done well never the less.

Here is the basic stuff - a common 0.01mm (~0.0004") "direct reading" dial indicator on a cheap flexible holder in a drill chuck (collets not needed) in my X3 mill. The set-up is very rigid.

Centre-ing a bit of brass rod in the collet adaptor (from my lathe). The indicator holder has a ball joint - just like a common "wiggler". The indicator has dove-tails front, top and back which are clamped in/by the indicator holder - very flexible and versatile!!

http://i200.photobucket.com/albums/aa294/oldtiffie/measuring/Dialindicator1.jpg

http://i200.photobucket.com/albums/aa294/oldtiffie/measuring/Dialindicator2.jpg

Inside or outside a tube in a "front-face mounted/bolted" 3-jaw chuck that belongs to my lathe and that I can bolt directly to my mill table or my 6" "Vertex" rotary table:
http://i200.photobucket.com/albums/aa294/oldtiffie/measuring/Dialindicator3.jpg

http://i200.photobucket.com/albums/aa294/oldtiffie/measuring/Dialindicator4.jpg

And for use on the centre bore etc. on a rotary table. No need for a co-ax or a Taster:
http://i200.photobucket.com/albums/aa294/oldtiffie/measuring/Dialindicator5.jpg

http://i200.photobucket.com/albums/aa294/oldtiffie/measuring/Dialindicator6.jpg

http://i200.photobucket.com/albums/aa294/oldtiffie/measuring/Dialindicator7.jpg

http://i200.photobucket.com/albums/aa294/oldtiffie/measuring/Dialindicator8.jpg

I have more to go - later - which I hope will be of interest.

JCHannum
11-09-2009, 02:00 AM
No one ever said that a DTI was not suitable, so don't feel you have the need to stick anything anywhere.

If only one instrument is in the budget, or starting a shop, the purchase of a good DTI, Interapid quality at least, is the best initial investment you can make. Do not stint on quality here and do get the full kit with all attachments. This will be the shop's primary precision instrument and will accomplish all the centering and set up functions needed.

The Co-Ax is not a requisite, but it is an adjunct that will provide a faster and simpler means of centering a bore. Its accuracy will equal that attainable with the DTI and if the job is a frequent occurence, the ease and speed it affords will be appreciated.

dp
11-09-2009, 02:09 AM
I agree with dp, the graduations are uneccessary.

Here's why. In the second half of this video I've placed a large bore on the mill table (since John S said it couldn't be done :) ) and have centered it by reducing the indicator movement to zero. Then I turn the Y axis feed screw to show the increase in needle movement as the bore moves off center. I then move it back to center then through center and finally back.

At no point do I give a rip what the numbers are because what I'm after is movement so small as to be lost in the accuracy of the indicator. If after moving the X and Y feeds I've reduced all movement possible then what is left is bore error or tram error. When I've got it to that point, I'm done.

http://www.youtube.com/watch?v=ZfioLDhBNBQ

oldtiffie
11-09-2009, 02:41 AM
In my previous post I mentioned the "wiggler" which has been around since for-ever. It is one the most useful, versatile, cheap and under-rated tools in the shop. Here is mine:
http://i200.photobucket.com/albums/aa294/oldtiffie/measuring/Wiggler1.jpg

It has a ball-joint that is "loaded" to suit. I use mine extensively on the mill and the drill etc. - lathe too on occasion. A drill chuck is all that is needed but collets will do too.

The left probe is a "nail" (pointy) that is used at about 1,000 RPM. I centre it by eye by adjusting it under speed with the back of my thumb (nail). I use it to pick up lines (edges too!!) as well as centre-punched marks etc. Very useful!!

The next two are a 0.100" and 0.200"diameter balls that are used for edge-finding - these work very well too. They work well to quickly locate or pick up a hole or a cylinder as I do with my dial indicators.

The last one on the right is used to fasten a test dial indicator to. Works very well too.

I showed a cylinder in my previous post. It was one of these:
http://i200.photobucket.com/albums/aa294/oldtiffie/measuring/Cylindermastersquares1.jpg

They are my "master cylinder squares". They are made from 3" diam x 1/4" wall thickness precision-drawn tube. They are very accurate for round and straightness. I faced them off at both ends very accurately on my lathe so that the cylinder and the ends are very accurately square to each other. I bolt them (vertically - through the hollow centre) to my mill table to give me a very accurate vertical reference and locating line or plane. I clamp them together or apart as needs be. I used them very successfully to "re-square" my angle plates and my tilting angle plates.

oldtiffie
11-09-2009, 02:54 AM
Here's why. In the second half of this video I've placed a large bore on the mill table (since John S said it couldn't be done :) ) and have centered it by reducing the indicator movement to zero. Then I turn the Y axis feed screw to show the increase in needle movement as the bore moves off center. I then move it back to center then through center and finally back.

At no point do I give a rip what the numbers are because what I'm after is movement so small as to be lost in the accuracy of the indicator. If after moving the X and Y feeds I've reduced all movement possible then what is left is bore error or tram error. When I've got it to that point, I'm done.

http://www.youtube.com/watch?v=ZfioLDhBNBQ

That was/is a very well produced instructive video Dennis -well done - I really did appreciate and enjoy it.

oldtiffie
11-09-2009, 03:04 AM
No one ever said that a DTI was not suitable, so don't feel you have the need to stick anything anywhere.

If only one instrument is in the budget, or starting a shop, the purchase of a good DTI, Interapid quality at least, is the best initial investment you can make. Do not stint on quality here and do get the full kit with all attachments. This will be the shop's primary precision instrument and will accomplish all the centering and set up functions needed.

The Co-Ax is not a requisite, but it is an adjunct that will provide a faster and simpler means of centering a bore. Its accuracy will equal that attainable with the DTI and if the job is a frequent occurrence, the ease and speed it affords will be appreciated.

Thanks Jim for an excellent post.

It tells every one that there are many tools and many solutions to suit similar circumstances.

If some here were as objective as that - giving credit as well as the "pros and cons" - with reasons - then a much better and more informative result will be the outcome as this thread is now becoming.

If some who are new to HSM-ing learn that the most expensive is not always needed - or the best either - and that a better outcome is achieved as regards needs, value and cost, then we will all be better off.

John Stevenson
11-09-2009, 03:13 AM
Here's why. In the second half of this video I've placed a large bore on the mill table (since John S said it couldn't be done :) )



Didn't say it couldn't be done in my case Dennis but that your rotor is 2" high ? my housing was about 8" or 9" high and with the table right down there wasn't enough room for the co-ax.

It was a hight problem and not a large bore problem.

I can't see at this point who's arguing with who ?
They do the job well if you like them, a taster finds edges very quickly which a co-ax won't.
A taster can find the centre of a bore but it requires does more work and a DRO to be really useful.

The cosine error is a red herring because if it exists it's consistent all the way round and so cancels out.

I'll bet any practising machinist is laughing out loud at this post, 3 pages [ on my screen setup ] to to explain something that can be done by various means in 30 seconds and I'm not reading all the posts. Pages of maths won't get the PRACTICAL job done any quicker.

.

JCHannum
11-09-2009, 08:01 AM
Such is the nature of the internet. The discussion had died a natural death until disinterred a few days back in an attempt to prove the Co-Ax was no good because of the cosine error non-problem. The only other rather vocal dissenter was a certain person who said the Co-Ax was useless because it would not fit in some applications.

Much of the remainder centered around the use and application of the Co-Ax. Some readers even learned some pointers on its use. For instance, one respondent was amazed to learn of the common modification of flatting the other side of the curved feeler to facilitate its use in large bores.

John Stevenson
11-09-2009, 08:21 AM
The only other rather vocal dissenter was a certain person who said the Co-Ax was useless because it would not fit in some applications.

Well if it won't fit then it is useless :D


one respondent was amazed to learn of the common modification of flatting the other side of the curved feeler to facilitate its use in large bores.

Note to factory "Put two flats on " :D

.

dp
11-09-2009, 10:28 AM
Didn't say it couldn't be done in my case Dennis but that your rotor is 2" high ? my housing was about 8" or 9" high and with the table right down there wasn't enough room for the co-ax.

Just having fun, John. It was the direction of the probe curve I was having fun with. You'd made the point that the flat and the curve of the probe required the probe always point toward the center. And Lazlo offered that the problem could be solved by turning the probe carrier 180º.

My probes have just one flat but can easily be turned around as shown.

lazlo
11-09-2009, 10:32 AM
I got a couple of PM's asking what a Centricator is.
It's the Deckel version of the Blake. Typical over-the-top German engineering :)

http://www.michael-deckel.de/documents/uk_501_CIII.jpg

Glenn Wegman
11-09-2009, 10:53 AM
Not like a CoAx, as it has basically no error if used properly, similar to using a DTI!

Germens understand basic problems!:D

lazlo
11-09-2009, 11:16 AM
Not like a CoAx, as it has basically no error if used properly, similar to using a DTI!

Germens understand basic problems!:D

How about this one? Is this more like a Blake? :D

http://www.michael-deckel.de/documents/uk_461_big_centricator_cocos.jpg

The only difference is the deflection mechanism. The CIII Centricator (the first one I posted) has an offset beam deflector, and the CO Centricator (here) has a cantilever like the Blake.

I don't think you'll see a big difference in centering accuracy :)

Glenn Wegman
11-09-2009, 11:28 AM
I don't think you'll see a big difference in centering accuracy :)

I don't disagree with that!

My point is that there is error with an articulating arm regardless of how big or small.

John Stevenson
11-09-2009, 06:17 PM
I got a couple of PM's asking what a Centricator is.
It's the Deckel version of the Blake. Typical over-the-top German engineering :)

http://www.michael-deckel.de/documents/uk_501_CIII.jpg


And the Russian version.

http://www.stevenson-engineers.co.uk/files/ruskiecoax1.jpg

I bought two Russian items some while ago, one was this "Centrescope" and the other was an optical centre finder that I'll cover in a later post.
Both these are serious kit and were on ISO50 tapers, I have redone the mountings and put then on 16mm diameter so they will fit a 16mm collet as all the machines they are destined for have 16mm collets, they are too big for the Sieg KX mills [ as is a Co-ax ]

Only two styli provided but because of the build up that is all that is needed as you will see in the next post as I need the full 4 photo.
limit.

One point about this Russian design is that it can do both inside and outside diameters without changing anything unless you want to, more later.

.

John Stevenson
11-09-2009, 06:17 PM
This is it out of it's box.

http://www.stevenson-engineers.co.uk/files/ruskiecoax2.jpg



Note that knurled knob, far right in line with the gauge, I'll come to it later.
The dovetail slide can side in and out all the way and is locked by a gib strip and the side know. It can also reverse and so give a far greater range than an other piece of tooling we have seen so far.

http://www.stevenson-engineers.co.uk/files/ruskiecoax3.jpg

This is it setup in the Bridgy, staged shot as the bearing [ scrap ] is clamped in the vise.
If that knurled knob is set for internal or external the gauge has full deflection which is 5 revolutions of 50 thou per rev so you only need to be with a 1/4" of a circle centre and in theory one turn by hand will have you dead on.

I did three turns to set this up but could have done it in two because I had to double check as believe it or not that bearing was 3 thou oval because of clamping it !!

http://www.stevenson-engineers.co.uk/files/ruskiecoax4.jpg

The dovetail has been reversed here showing just how big it can go, internal or external without relying on skimpy styli.

Remember that knurled knob <g>

http://www.stevenson-engineers.co.uk/files/ruskiecoax5.jpg
Well it runs thru the tool and bears on the lever that connects to the gauge.

It like a bolt in a gun but spring loaded with three positions, held back so that it does internal with full deflection, all the way in so it loads the gauge and does external as it releases.

And finally it has a mid position where it loads the gauge in the middle and in this position you can measure either way, internal or external but you only have +0.125" and -0.125" travel.

Neat bit of kit what ?

.

John Stevenson
11-09-2009, 06:18 PM
And now to the centring microscope, also Russian but with one feature not seen in the west, it doesn't need an outside light source because it glows in the dark <g>

I am under the impression that the Russians have good optics, Evan may know better, we have a far amount of Russian gear over here, even during the cold war we still dealt with them and we never regarded then as an outright enemy as the US did. Most of it is very high quality unless electronics are concerned.

http://www.stevenson-engineers.co.uk/files/ruskiecentrescope1.jpg

This is it in it's box, again was setup for ISO 50 taper hence the unused support.

I put a piece of alloy block in the vise and centre drilled it as lightly as i could with a small centre drill, then fitted the scope but I had to drop the table a couple of inches to get it in.

http://www.stevenson-engineers.co.uk/files/ruskiecentrescope2.jpg

You can't see the centre pop as it's too small. Tried for a phot thru the lens and got a passable one which surprised me as I've never tried this before and don't have any decent camera gear or skills, just a Panasonic point and shoot.

http://www.stevenson-engineers.co.uk/files/ruskiecentrescope3.jpg

It's not off centre, it's the camera that's off centre to the lens. It's just possible to see outer rings marked in 0.01mm grads and the centre line running from 8 o'clock to 2 o'clock. The other line is there from 11 o'clock to 5 o'clock but nor showing. That is a tiny centre drill mark.

However it wasn't on centre and I had to tweak the table to get it to line up.

http://www.stevenson-engineers.co.uk/files/ruskiecentrescope4.jpg

They are in mm so it's 0.00098" out in the X and 0.0013779" out in the Y which doesn't surprise me as the trams is off AGAIN on this POS Bridgeport with the licorice head and dropping the table 2" has shown this up.

I'll leave Tiffie to work the rest of the maths out.

Goodnight.

.

lazlo
11-09-2009, 06:46 PM
And the Russian version.

One point about this Russian design is that it can do both inside and outside diameters without changing anything unless you want to, more later.

Very cool! That definitely looks like something you could make in the shop.

I have a 50X centering scope that came off a K&T boring mill. It's pretty slick -- there's a single horizontal crosshair. You align one axis, rotate the reticle 90°, then do the opposite axis.

Definitely not fast, but it's probably the most accurate way to pick-up random surface features.

oldtiffie
11-09-2009, 07:52 PM
I got a couple of PM's asking what a Centricator is.
It's the Deckel version of the Blake. Typical over-the-top German engineering :)

http://www.michael-deckel.de/documents/uk_501_CIII.jpg



Originally Posted by Glenn Wegman
Not like a CoAx, as it has basically no error if used properly, similar to using a DTI!

Germens understand basic problems!

Well, let's have a look at those statements Lazlo.

First of all the German model is a direct-reading instrument as it has all of the attributes of the shortest (2") "Blake" straight probe at centre/vertical with none of the errors due to the Blake longer straight ("internal") feelers (bending) or all of its "bent" "external" probes (bending plus "twisting/torque).

Avoidable bending is bad enough but combining twisting with it makes a bad design/situation a lot worse.

The German design has pretty well all bases covered and is far superior to the Blake.

The German model seems to be like Goldilocks porridge: "Just right".

If, as you say, the "German" model is over-engineered (which I don't think it is - at all) and as/if that German model is the bench-mark, then the USA-designed Blake is inferior to the German-designed and manufactured product.

The German model seems to have been made by "Deckel" - a manufacturing company with a long history of excellence in design and manufacturing of quality machines and product. It is "way up there" with other European companies similar to Schaublin and the like - and many of them.

I have not had a "Blake" co-ax indicator in my hand or on my machine lately, but from from what I remember of what it was like when I did use one, it is no better than the Chinese one I bought from Little Machine Shop (USA) recently - and the Chinese one does well enough - within its limitations.

lazlo
11-09-2009, 08:01 PM
I have not had a "Blake" co-ax indicator in my hand or on my machine lately, but from from what I remember of what it was like when I did use one, it is no better than the Chinese one I bought from Little Machine Shop

Your memory is not good :) I have both, and the real Blake is, not surprisingly, vastly better build quality, especially the dial indicator.

Also note that the Centricator is not direct reading. Just like the Blake, it has an offset beam that drives an indicator. You use it to minimize needle wobble (relative runout), just like a Blake.

oldtiffie
11-09-2009, 08:23 PM
John,

many thanks for showing that super-bloody nice Russian stuff.

I am well aware of some of the superb (and lately a lot, but by no means all - some is crap -Russian stuff).

That really good, versatile and very well made protractor is a case in point. I bought a (presumably Chinese clone) from a supplier in the USA and it required a lot of "re-work" to make it right. I bought another directly from China (Hong Kong) and that one restored my faith (faith - me? - yep I wondered about that too) and it was all that I could ask for - delivered in 4 days instead 3 weeks from the USA and at half the cost and triple the service.

If ever Russia and Eastern Europe get their act/s together, they will be another "China".

I could see a need for a co-axial microscope so I bought a kit (I have yet to make it from the parts supplied) from Hemingway Kits (UK) - a great firm to deal with - here:
http://www.hemingwaykits.com/

http://www.hemingwaykits.com/acatalog/Centering_Microscope.html

http://www.hemingwaykits.com/acatalog/info_HK_171X.html

I also bought one of their "Quick Step Mills" - a great tool which I will post details of on one of the "tail-stock" threads - shortly.

http://www.hemingwaykits.com/acatalog/The_Quick_Step_Mill__.html

oldtiffie
11-10-2009, 08:31 AM
John Stevenson made a remark that may have "passed" (by?) some - but it was most important.

He said that had "head-space" problems on his BP mill - ie not enough space between the mill spindle and the table to fit the job and a co-axial into that space.

Quite a few here have BP's.

Some others like me have smaller machines - like my Sieg X3's (not too small most times) and my HF-45 (not all that small at all).

I would JUST fit the set-up (with a co-ax indicator) onto my HF-45 but not onto my X3's.

I had no problem using the compact test dial indicator (TDI).

Here is the co-ax (Blake clone):
http://i200.photobucket.com/albums/aa294/oldtiffie/measuring/Co-axialindicator1.jpg

http://i200.photobucket.com/albums/aa294/oldtiffie/measuring/Co-axialindicator2.jpg

http://i200.photobucket.com/albums/aa294/oldtiffie/measuring/Co-axialindicator3.jpg

and the TDI:
http://i200.photobucket.com/albums/aa294/oldtiffie/measuring/Dialindicator3.jpg

I am the first to say that I may have been "stretching it" a bit using a Jacob's chuck instead of a collet holder - but its a normal set-up.

I suggest that anyone seriously considering buying/getting a co-axial indicator also consider the "head-space" (aka "will it fit on my mill") situation.

lazlo
11-10-2009, 09:36 AM
How about this one? Is this more like a Blake? :D

http://www.michael-deckel.de/documents/uk_461_big_centricator_cocos.jpg

The only difference is the deflection mechanism. The CIII Centricator (the first one I posted) has an offset beam deflector, and the CO Centricator (here) has a cantilever like the Blake.

Just for grins, I looked up the price of the Deckel Centricator. The smaller one without the dovetail slide is £1,454 or $2,436. The Centricator CIII with the dovetail is £3,918 GBP or $6,538. For that price, one of the Mitutoyo calendar models should deliver and install it :)

http://www.measureshop.biz/en/measuring-instruments/dial-bore-and-depth-gauges/centring-devices/centricator-centring-device.html#shop

Glenn Wegman
11-10-2009, 10:08 AM
[QUOTE=lazlo]Just for grins, I looked up the price of the Deckel Centricator. The smaller one without the dovetail slide is £1,454 or $2,436. The Centricator CIII with the dovetail is £3,918 GBP or $6,538. For that price, one of the Mitutoyo calendar models should deliver and install it :)

Do you think the difference in price reflects a difference in accuracy? :)

They must have seen the Moore price sheet!

BobWarfield
11-10-2009, 11:08 AM
I suggest that anyone seriously considering buying/getting a co-axial indicator also consider the "head-space" (aka "will it fit on my mill") situation.

Wow, Tiff, sorry about the small mill. I think they have pills nowadays to fix that sort of thing <G>.

I also don't think I'd be using that Jacobs chuck if I cared so much about the accuracy as you seem to, but to each his own (there, the pin is pulled and the 'nade sits on the table!).

Here's my IH not breaking a sweat with the Co-Ax and a longish probe:

http://www.cnccookbook.com/img/MillStuff/CNC/ZAxisMod/P1010262.JPG

I'm probably at about 1/3 of the Z-travel. Yup, coulda used a shorter probe, but the bore came out very nicely and the job was all over in just a few minutes.

Cheers,

BW

JCHannum
11-10-2009, 12:10 PM
If you are going to use every workholding device in the shop to prove a point, why not throw in a cinder block or two.

If price and size present a problem, make your own. They are really quite simple to construct. Plans are in one of the Lautard books and were in HSM several years ago. They have been reprinted in one of the HSM collections.

This is one I made when I had my Benchmaster mill at a cost of $10.00 for the flea market indicator.

http://i320.photobucket.com/albums/nn351/jchannum/P1070407.jpg

It used an AGD1 indicator. Miniature indicators are available from Federal and a few other manufacturers that would make up into an even smaller instrument.

lazlo
11-10-2009, 12:33 PM
If you are going to use every workholding device in the shop to prove a point, why not throw in a cinder block or two.

LOL!

I pointed out that 'Tiff was going to have that issue on page five :)

http://bbs.homeshopmachinist.net/showpost.php?p=480119&postcount=43


By the way 'Tiff, I think all the 3D Tasters have a 20mm shank, which will be a problem for someone that only has R8 (Imperial) collets or endmill holders.

Carld
11-10-2009, 12:36 PM
It's interesting that the original question was "how to use a co-ax indicator" and has in 17 pages turned into a discourse on the failings, limitations and few, if any, posts dealt with "how to use a co-ax indicator.

It's hard to tell if his question was answered and I guess he has long since ignored the thread.

Peter N
11-10-2009, 01:41 PM
I also don't think I'd be using that Jacobs chuck if I cared so much about the accuracy as you seem to, but to each his own

You beat me to it Bob:D :D
How any posts on the need for accuracy was it Tiffie? You'd also gain a few inches of headroom by using a collet directly in the spindle, and that would solve your other problem.

Peter

dp
11-10-2009, 02:06 PM
It's interesting that the original question was "how to use a co-ax indicator" and has in 17 pages turned into a discourse on the failings, limitations and few, if any, posts dealt with "how to use a co-ax indicator.

It's hard to tell if his question was answered and I guess he has long since ignored the thread.

Well, I did post a video that showed the storing ring and how to reverse the probe for checking large bores :)

oldtiffie
11-10-2009, 05:58 PM
Originally Posted by oldtiffie
I suggest that anyone seriously considering buying/getting a co-axial indicator also consider the "head-space" (aka "will it fit on my mill") situation.


Wow, Tiff, sorry about the small mill. I think they have pills nowadays to fix that sort of thing <G>.

I also don't think I'd be using that Jacobs chuck if I cared so much about the accuracy as you seem to, but to each his own (there, the pin is pulled and the 'nade sits on the table!).

Here's my IH not breaking a sweat with the Co-Ax and a longish probe:

http://www.cnccookbook.com/img/MillStuff/CNC/ZAxisMod/P1010262.JPG

I'm probably at about 1/3 of the Z-travel. Yup, coulda used a shorter probe, but the bore came out very nicely and the job was all over in just a few minutes.

Cheers,

BW

Thanks Bob.

I made a mistake and gave my X3 mill a Viagra pill to make it "grow" but sadly all that it fu**ed was itself.

If I have time today, I will duplicate the "job in a 3-jaw chuck on a rotary table with a co-ax indicator in the chuck" on my HF-45 (same as the one in your post) and will up-load and post the pics.

It does not matter at all whether the indicator - any indicator - is held in the mill spindle with a drill chuck, a collet, a "wiggler" arm or just held onto the knurled surface of a chuck with "Blue tack". I could even compound my "sin's" and put a "large-ish" shim under one of the Jacob's chuck jaws and still get a similar or the same result.

All that is needed is that the "feeler" describe a circle that is concentric with the mill spindle.

There is no mention or need anywhere for the co-ax to be adjusted/checked for centre which is odd if the sort of accuracy that using a collet seems to infer.

When was it so important that collets had to be used to hold an indicator to tram a mill spindle to the table when the procedure and requirement is exactly the same as using a co-ax - or ANY indicator in a mill spindle.

All of this pre-supposes that the spindle bearings are OK - in ALL cases.

I suggest that some here go back to basic mechanics and geometry and have another look at how it really is instead of just repeating (regurgitating??) stuff that "people who (seem to??) know stuff" keep on saying - and their apostles and disciples (ventriloquists dolls??/"dummies"?) use as something between a monastic chant and a mantra.

Never mind whether you do or don't like me (it doesn't matter to me either) - just concentrate on and address the issue/s.

oldtiffie
11-10-2009, 07:07 PM
If you are going to use every workholding device in the shop to prove a point, why not throw in a cinder block or two.

If price and size present a problem, make your own. They are really quite simple to construct. Plans are in one of the Lautard books and were in HSM several years ago. They have been reprinted in one of the HSM collections.

This is one I made when I had my Benchmaster mill at a cost of $10.00 for the flea market indicator.

http://i320.photobucket.com/albums/nn351/jchannum/P1070407.jpg

It used an AGD1 indicator. Miniature indicators are available from Federal and a few other manufacturers that would make up into an even smaller instrument.

Jim.

The point I was making is that "head space" ("Z") is an issue as John Stevenson had with his BP as he advised/replied to dp (Dennis).

My X3 may not be an issue for many here but John's BP mill will be.

My HF-45 will be an issue for many here as well. As I advised Bob W, I will duplicate the set-up on my HF-45 (with a 40" ruler for "scale") to see how it goes. I will include the co-ax indicator as well as my normal shop indicator. I will also check/advise the difference between the space use/saving between the Jacob's chuck and the ER-32 collets.



Originally Posted by JCHannum
If you are going to use every workholding device in the shop to prove a point, why not throw in a cinder block or two.



Originally Posted by Lazlo
LOL!

I pointed out that 'Tiff was going to have that issue on page five

http://bbs.homeshopmachinist.net/sho...9&postcount=43


Originally Posted by Lazlo
By the way 'Tiff, I think all the 3D Tasters have a 20mm shank, which will be a problem for someone that only has R8 (Imperial) collets or endmill holders.

Bad luck about the limitations of an R8 collet system. An R8>ER-32 collet adaptor will soon sort that out. All of my mills have MT3 tapers and ER-32 collets (20mm ~0.800" - maximum).

I think the "cinder" problem will be resolved pretty damned quick if Bob W let's the handle go on that grenade. Lots of "blood and gore" in the cinder blocks I make from it though.

But don't worry Bob, as I am pretty sure that our Navy cooks cut/diced/minced the meat with a hand grenade. $hit of a way to get a metal filling in my teeth though - but still better than a Navy Dentist though as they were worse than the cooks - and it meant a trip to a Navy Doctor who was worse than the Dentist.

Err Bob. DON'T try to put the pin back in after you have released the handle.

lazlo
11-10-2009, 07:21 PM
Bad luck about the limitations of an R8 collet system. An R8>ER-32 collet adaptor will soon sort that out.

I wouldn't call it a limitation -- its just that most machine tapers including R8, Morse, Jarno, NT... are Imperial and ER is Metric.

In any event, that's what I'm using to mount the 3D Taster on my Excello: an R8-ER40 collet chuck, but the damn thing adds about 3" of overhang.

I need to play around with an ER-16, maybe an ER-20, and see if I can get it to fit inside the R8 spindle, so you don't lose headspace.

BobWarfield
11-10-2009, 07:34 PM
It does not matter at all whether the indicator - any indicator - is held in the mill spindle with a drill chuck, a collet, a "wiggler" arm or just held onto the knurled surface of a chuck with "Blue tack". I could even compound my "sin's" and put a "large-ish" shim under one of the Jacob's chuck jaws and still get a similar or the same result.

All that is needed is that the "feeler" describe a circle that is concentric with the mill spindle.

There is no mention or need anywhere for the co-ax to be adjusted/checked for centre which is odd if the sort of accuracy that using a collet seems to infer.


Why Tiffie, that's excellent! I see you did finally understand and explain perfectly clearly why all this prattle about the various errors was irrelevant to the actual use of the tool.

If only we could have come to that before there were so many posts, so many pictures, and so many cul-de-sacs in the discussion.

Now if you can avoid stacking the cinder blocks on the mill, perhaps there is enough headroom on the RF-45 to do a setup or two with your coax.

Lazlo, I think I'd look for an endmill holder a touch smaller than your Taster's metric shank and take your lathe's boring tool to it. Softly, softly, catch a monkey and you should be able to get there quite accurately. A 3/4" doesn't need much doing to get to 20mm.

Cheers,

BW

oldtiffie
11-10-2009, 07:46 PM
Originally Posted oldtiffie
Bad luck about the limitations of an R8 collet system. An R8>ER-32 collet adaptor will soon sort that out.


Originally Posted by Lazlo]

I wouldn't call it a limitation -- its just that most machine tapers including R8, Morse, Jarno, NT... are Imperial and ER is Metric.

In any event, that's what I'm using to mount the 3D Taster on my Excello: an R8-ER40 collet chuck, but the damn thing adds about 3" of overhang.

I need to play around with an ER-16, maybe an ER-20, and see if I can get it to fit inside the R8 spindle, so you don't lose headspace.

Lazlo.


The world is neither encompassed,defined nor restricted to or by the high or low tidal marks of the continental shelf or the "waters" of the USA or continental North America.

Don't let the (so-called) "World Series" American Football and Base-ball series confuse (or convince) you.

Maybe you should "get out" more.

There "others" that are "out there" and I'd think that most tapers for the mills used by HSM-ers outside the USA will be 2MT or 3MT. That most or many tapers in the USA HSM shops may be R8 is not a real issue. We can't help what Hardinge does with its proprietary designs and product restrictions.

ER collets are not just "metric" - they are universal as "inch" ER collet sets will include any "metric" stuff within the 1mm (~0.04") range of each collet. The same applies for nominal "metric" collets and their ability to cope with "inch" tools etc.

I have addressed the issue of the "inch" being a metric standard (as enacted by the US Congress) - so pretty well everything is metric. I would bet that there is a metric/ISO/DIN standard/s for the common R8 taper - and its collet system as well.

I think you have made my point about limiting head-space too.

Thanks.

John Stevenson
11-10-2009, 07:59 PM
In any event, that's what I'm using to mount the 3D Taster on my Excello: an R8-ER40 collet chuck, but the damn thing adds about 3" of overhang.

I need to play around with an ER-16, maybe an ER-20, and see if I can get it to fit inside the R8 spindle, so you don't lose headspace.

I have a 20mm R8 collet here, it's not deep but it's enough to hold the Taster.
.

oldtiffie
11-10-2009, 08:02 PM
Originally Posted by oldtiffie
It does not matter at all whether the indicator - any indicator - is held in the mill spindle with a drill chuck, a collet, a "wiggler" arm or just held onto the knurled surface of a chuck with "Blue tack". I could even compound my "sin's" and put a "large-ish" shim under one of the Jacob's chuck jaws and still get a similar or the same result.

All that is needed is that the "feeler" describe a circle that is concentric with the mill spindle.

There is no mention or need anywhere for the co-ax to be adjusted/checked for centre which is odd if the sort of accuracy that using a collet seems to infer.


Why Tiffie, that's excellent! I see you did finally understand and explain perfectly clearly why all this prattle about the various errors was irrelevant to the actual use of the tool.

If only we could have come to that before there were so many posts, so many pictures, and so many cul-de-sacs in the discussion.

Now if you can avoid stacking the cinder blocks on the mill, perhaps there is enough headroom on the RF-45 to do a setup or two with your coax.

Lazlo, I think I'd look for an endmill holder a touch smaller than your Taster's metric shank and take your lathe's boring tool to it. Softly, softly, catch a monkey and you should be able to get there quite accurately. A 3/4" doesn't need much doing to get to 20mm.

Cheers,

BW

Thanks Bob.

I just hope the self-induced fog of desperation "lifts" so that some others here can see things in a better light.

I'd have expanded your "cul-de-sacs" to include dark and blind alleys - and perhaps even into orbit (or beyond).

My main concern was - and still is - that some may not only believe but rely on the co-ax indicator calibrations being 0.0005" ("half a thou") in all circumstances (all feelers at all diameters - inside and out) given that so many here gave it such a big unqualified "(w)rap" and that the manufacturer did not say otherwise on the literature that came with a new item.

Its worse if the tool was bought even in good condition without documentation.

The same applies to my clone and so the "real" and "clone" are equally at fault in that regard.

John Stevenson
11-10-2009, 08:17 PM
Note to Blake and the Chinese, just put one mark and a zero on the dial. That way no one will get confused.

My thread gauge on one lathe stands vertical and has 4 marks, the other also has 4 marks but leans at an angle of 22.98376452746597654 degrees, [ approx ]Do I need to take into account the cosine error when closing the half nuts. ?

lazlo
11-10-2009, 08:25 PM
Maybe you should "get out" more.

There "others" that are "out there" and I'd think that most tapers for the mills used by HSM-ers outside the USA will be 2MT or 3MT.

Which, as I said, is an Imperial standard.


ER collets are not just "metric" - they are universal as "inch" ER collet sets will include any "metric" stuff within the 1mm (~0.04") range of each collet.

ER collets are not universal -- they only meet their runout spec at the nominal dimension, which is why there are Inch ER collets as well as Metric ER collets.
My point being that it's not a limitation of an R8 collet that it doesn't fit Metric tools.

John, Bob -- a quicker solution would be to bore an R8 emergency collet for 20mm. Added to my infinitely-long todo list...

John Stevenson
11-10-2009, 08:28 PM
My point being that it's not a limitation of an R8 collet that it doesn't fit Metric tools.
Don't quite understand that last statement, most of my work is done with metric R8 collets, I do have imperials but most of my tooling is metric shanked.

These people have them up to 1" in imperial and 24mm in metric.

http://www.rdgtools.co.uk/acatalog/R8_COLLET_S_.html

Not sure how they do this , extended nose or very shallow hole ? My 20mm is a shallow bore, OK for holding things and very light cuts but definitely not for hogging.

JCHannum
11-10-2009, 09:11 PM
My main concern was - and still is - that some may not only believe but rely on the co-ax indicator calibrations being 0.0005" ("half a thou") in all circumstances (all feelers at all diameters - inside and out) given that so many here gave it such a big unqualified "(w)rap" and that the manufacturer did not say otherwise on the literature that came with a new item.

Its worse if the tool was bought even in good condition without documentation.

The same applies to my clone and so the "real" and "clone" are equally at fault in that regard.

I cannot understand why you continue to dwell on this. Your statements are incorrect and very misleading. The graduations on the Blake dial are not 0.0005". They represent 0.0005" Axis Offset. This information is clearly marked on the face of the dial. In addition, the instructions very clearly state the conditions for the 0.0005" dimension, ie. at a 2" diameter bore. The table of error compensation for other feelers and diameters is readily accessable on the website.

Any DTI or lever type measuring instrument in the world is subject to the exact same cosine error you keep flouting as being such a major fault of the Blake. Many come with assorted lengths of feelers, but none that I know of publish a table giving error compensation figures for use with the various lengths.

lazlo
11-10-2009, 09:25 PM
Don't quite understand that last statement, most of my work is done with metric R8 collets, I do have imperials but most of my tooling is metric shanked.

Oh nice! I didn't know they made R8 collets in Metric. I just checked the usual suspects over here, and Enco carries Chinese R8 Metric collets (marked "NEW!") for $18 each. Imperial R8 from the same vendor are $5 :( I only need the 20mm though, so as long as the runout isn't obnoxious...

dp
11-10-2009, 09:39 PM
Note to Blake and the Chinese, just put one mark and a zero on the dial. That way no one will get confused.

I think I still prefer using Disney characters instead of numbers - that way it doesn't look so empty.

I'm surprised anyone would actually try to measure something with such an instrument even if they knew and applied all the conditions needed to do so. As an inexpensive gage that remains stationary while it's probe spins away it seems well suited for the purpose of centering a thing under the quill. It's that stationary part that provides the only advantage over a DTI that I'm aware of.

oldtiffie
11-10-2009, 10:31 PM
Originally Posted by John Stevenson
Don't quite understand that last statement, most of my work is done with metric R8 collets, I do have imperials but most of my tooling is metric shanked.

Oh nice! I didn't know they made R8 collets in Metric. I just checked the usual suspects over here, and Enco carries Chinese R8 Metric collets (marked "NEW!") for $18 each. Imperial R8 from the same vendor are $5 :( I only need the 20mm though, so as long as the runout isn't obnoxious...

Lazlo.

Seems you are on a pretty big/steep learning curve here.

The run-out of the collet or what-ever you hold the "Taster" with/in does not matter in the least as long as that run-out at the "ball" on the feeler/probe does not exceed the "X"-"Y" adjustments at the top of the Taster body.

Use a good test dial indicator (0.01mm/0.0005" - or "better") and adjust "X" and "Y" TIR to a minimum/zero - you are then ready to roll as regards centre-ing the mill/machine spindle axis over the centre of the ball when the dial is "centered/zero-ed".

No need to centre or adjust it at all if you use the same place on the dial (need not be zero) at each measuring "station" as absolute/relative distances/difference will be the same as read on your DRO or lead-screw dial. As the"feeler arm" is spring-loaded" there is no need to be concerned about "back-lash" in the Taster (as with most indicators).

Off-sets/movements on the indicator are "true" as any "Co-sine" or other "errors" are compensated for.

Once you "get on top of" that Taster you will be very well satisfied as it is a precision tool that has a lot of use. Its easy to "split" the 0.01mm/0.0005" graduations into "halves and quarters) for real/better accuracy.

In normal use or set-up, it does/may take up a fair bit of "Z" space and/as most will opt for a 20mm collet.

If head-room was a problem I would simply get a 3MT/R8 blank (machinable end) from Little Machine Shop or CDCO Tools or "other/s" and drill and bore out the 20mm hole in my lathe head-stock. A set/grub screw will suffice to hold the Taster in it. Its cheaper than collets and frees up a lot of/more head room. (Not unlike an end mill holder really).

Go for it.

http://i200.photobucket.com/albums/aa294/oldtiffie/measuring/Taster-3D-1.jpg

Edit:

pic of my Taster added.

lazlo
11-10-2009, 10:52 PM
Lazlo.

Seems you are on a pretty big/steep learning curve here.

The run-out of the collet or what-ever you hold the "Taster" with/in does not matter in the least as long as that run-out at the "ball" on the feeler/probe does not exceed the "X"-"Y" adjustments at the top of the Taster body.

Of course it does Tiff -- the 3D Taster isn't spinning like the Coax, so any runout on the collet holding the taster is going to translate directly into measuring error. That's why the instructions go into great detail on calibrating the tester with the four set screws at the top.

You didn't read the instructions yet, did you :rolleyes:

dp
11-10-2009, 11:01 PM
Question for Tiffie - on your Blake clone and mine there is a short black hinged probe. I've not figured out what that is used for so around here it's going to outlast all the other probes. Any idea how it's used?

lazlo
11-10-2009, 11:03 PM
Question for Tiffie - on your Blake clone and mine there is a short black hinged probe.

It's for picking-up a center punch mark.

oldtiffie
11-10-2009, 11:07 PM
Wrong again Lazlo.

In the first instance, I not only read the instructions - and understood what they were intended to achieve - and/but I used them - as per instructions.

In the second instance:
If the stem and the collets were as accurate as you seem to infer then there would be no need for the "X" and "Y" adjusters - would there?

But there is - isn't there?

Now why would that be do you think?

Here's what I think:

The purpose of the "X" and "Y" adjustment/s is to use the dial indicator on the ball to cancel out any errors between the axis of the mill spindle and the ball at the end of the probe/arm. The axis of both the spindle and the ball are co-incident and any "errors" are "in orbit" around that joint/common axis.

QED

Or as the saying goes here: "Up the old red rooster!!"

lazlo
11-10-2009, 11:17 PM
The purpose of the "X" and "Y" adjustment/s is to use the dial indicator on the ball to cancel out any errors between the axis of the mill spindle and the ball at the end of the probe/arm.

Exactly, but that's not going to zero-out the collet error:


The run-out of the collet or what-ever you hold the "Taster" with/in does not matter in the least

Yes, the run out matters a lot. Because every time you stick a collet in the spindle, the point of max runout is going to be at a different position, and unless you're planning to painstakingly re-calibrate the taster each time you stick it in the collet, the collet error is going to translate directly into indicating error.

Collet error, of course, doesn't matter on the Blake, because the probe is orbiting. But it's a huge issue with the 3D Taster, unless you're proposing to indicate the edge with the dial facing one way, flip it around 180° and indicate the edge again?

By the way, it's called a "3D Taster" because Taster in German means "caliper".

oldtiffie
11-10-2009, 11:22 PM
Originally Posted by dp
Question for Tiffie - on your Blake clone and mine there is a short black hinged probe.


It's for picking-up a center punch mark.

That's so Lazlo.

It is also spring-loaded.

The instruction sheet (this one is in bloody "Chinglish" too!!) says/suggests putting the probe into the centre-mark/punch and depressing it by 1/16". You are then ready to roll. It infers, but does not say, that the probe should be as near "centered/vertical" as you can get it - and away you go!! - as before - to centre it.

I meant to remark on that feature early but it got lost in the noise and black powder (cordite??) smoke.

It is pretty unique as regards that capability. It does work really well.

See - I did say something "nice" about the co-ax.

It didn't hurt my "pride" (pig-headedness???) much at all - but I noticed that my ar$e was bloody sore after it. Do you you think there is/was a "connection" some-how??

I have been advised to be a lot less "picky" with my ar$e and a lot more picky with my "friends". I can't see the need really. Hmmm. I'm ex-Navy - perhaps there is a "link" there too - so I'm told.

Keep probing!!

oldtiffie
11-10-2009, 11:50 PM
Originally Posted by oldtiffie
The purpose of the "X" and "Y" adjustment/s is to use the dial indicator on the ball to cancel out any errors between the axis of the mill spindle and the ball at the end of the probe/arm.

Exactly, but that's not going to zero-out the collet error:


Originally Posted by Tiffie
The run-out of the collet or what-ever you hold the "Taster" with/in does not matter in the least

Yes, the run out matters a lot. Because every time you stick a collet in the spindle, the point of max runout is going to be at a different position, and unless you're planning to painstakingly re-calibrate the taster each time you stick it in the collet, the collet error is going to translate directly into indicating error.

Collet error, of course, doesn't matter on the Blake, because the probe is orbiting. But it's a huge issue with the 3D Taster, unless you're proposing to indicate the edge with the dial facing one way, flip it around 180° and indicate the edge again?

By the way, it's called a "3D Taster" because Taster in German means "caliper".

Lazlo.

You really are a trial at times.

Perhaps instead of "flipping" "Tasters" 180 degrees, you might like to flip "burgers" at "Maccas".

If having to check and perhaps (re)adjust a tool each time it is mounted is such a tedium, then I suggest that you "get real" or "chuck it in".

I suggest that instead of "thinking (and saying) so" that you do as the instructions say (ie "do so") and then "see so" and then you will (hopefully) and "know so" then come back and "say so" to enligten me at least.

Now just make sure that in your haste to have a "taste" that you don't break an arm or a leg (or have me do it) and then need a caliper. (Yep I "Googled" it too. For any futher assistance - just ask Evan nicely. I'm sure he'd be only too happy to oblige you).

I've checked my extensive library of "Noddie" books (my "Bible" and ultimate reference - they are my version of "Machinery's Handbook") and there is no mention of perfect tools, machines - or collets either.

If you really do want (lust after?) such things then behave yourself and put it on you Christmas stocking list.

Tell me about it on 26th. December 2009.

dp
11-10-2009, 11:57 PM
It's for picking-up a center punch mark.

So I just tried that and I discovered it resembles a monkey and a football at work. I'm doing something very wrong, I'm sure.

dp
11-10-2009, 11:59 PM
The mystery is ended - from the Blake site: http://www.blakemanufacturing.com/pages/repairs.html


The Blake CO-AX INDICATOR is a centering device. It requires a reading of both sides of the bore or boss, in each axis, to arrive at the center. It's sole function is to find center. It is NEVER to be used as a measuring instrument.

Popcorn, anyone?

oldtiffie
11-11-2009, 12:07 AM
Thanks Dennis.

I'd have thought (past hoping??) that this too would have been included in the literature that the manufacturer put in with the tool - all manufacturers - "Blake" and clone-makers alike.

Its a fine line between not telling the whole truth - deliberately or not - on the one side and dishonesty on the other.

There should not be any need to chase all over the web to get this stuff.

lazlo
11-11-2009, 12:17 AM
The run-out of the collet or what-ever you hold the "Taster" with/in does not matter in the least
...

If having to check and perhaps (re)adjust a tool each time it is mounted is such a tedium, then I suggest that you "get real" or "chuck it in".

So then the accuracy of the collet does matter, right? :)
Point being, that the accuracy of the collet holding the 3D Taster matters a lot, the accuracy of the collet holding the Blake does not.

Which is why I said that I needed to find an accurate 20mm collet...


The Blake CO-AX INDICATOR is a centering device. It requires a reading of both sides of the bore or boss, in each axis, to arrive at the center. It's sole function is to find center. It is NEVER to be used as a measuring instrument.

Yeah, maybe if Jim, Glenn, Blake, you and I repeat it a couple of hundred more times he'll believe it :)


I'd have thought (past hoping??) that this too would have been included in the literature that the manufacturer put in with the tool - all manufacturers - "Blake" and clone-makers alike.

Well, Blake does, and it's been said many times since page 3, but isn't it glaringly obvious by the design of the mechanism, that it shows the relative runout of the hole, and not an absolute measurement?

dp
11-11-2009, 12:18 AM
Thanks Dennis.

I'd have thought (past hoping??) that this too would have been included in the literature that the manufacturer put in with the tool - all manufacturers - "Blake" and clone-makers alike.

Its a fine line between not telling the whole truth - deliberately or not - on the one side and dishonesty on the other.

There should not be any need to chase all over the web to get this stuff.

The Blake site is pretty small. If you enter "site:www.blakemanufacturing.com" into google you'll get all 6 pages back, including this gem:

http://www.blakemanufacturing.com/pages/diagram.html

I get the feeling the store is off line and that Blake makes and repairs only this one item. A rebuild costs only $40.00 USD which really isn't bad.

I'd still like to know how that center finder probe works.

oldtiffie
11-11-2009, 01:21 AM
...


So then the accuracy of the collet does matter, right? :)
Point being, that the accuracy of the collet holding the 3D Taster matters a lot, the accuracy of the collet holding the Blake does not.

Which is why I said that I needed to find an accurate 20mm collet...



Yeah, maybe if Jim, Glenn, Blake, you and I repeat it a couple of hundred more times he'll believe it :)



Well, Blake does, and it's been said many times since page 3, but isn't it glaringly obvious by the design of the mechanism, that it shows the relative runout of the hole, and not an absolute measurement?

Lazlo.

Once again you are not only being disingenuous but you are selectively "cherry-picking" stuff that seems to suit your purposes and/or distorting it and/or using it out of the context in which it was written.

I should not have to tell or remind you that when quoting (I think you have called it "citing") then keep (to) the context or leave it out. Its an "all or bugger all" situation.

I have never had an issue with the Blake co-ax as a centreing device - only issue or concern was with with the accuracy - or lack of it - in terms of whether the scale reading is 0.0005" per increment of needle movement - or not. No more and no less.

Instead of getting under my feet (and my skin shortly) I suggest that you try the "Taster" in anything that will hold it - anything - collet drill chuck or an adaptor that you make to take the Taster and the other end to fit into a chuck or collet. With the sole proviso that the run-out at the Taster ball does not exceed the Taster "X" and "Y" adjustment range/s.

Now - once and for all - the concentricity does not matter if it is within the capacity of adjustment of the Taster or the limits of axial movement of the co-ax plunger.

Now you, "Fred down the road" and Uncle Tom Cobley and all can include it in your hymn books and chant/sing that sad refrain as long and as often as you like.


Numbers of people and/or posts don't matter -as it won't change the facts.

I doubt that looking in chicken bones or entrails will help much either. Try my voodoo doll - it itches a bit!!

oldtiffie
11-11-2009, 01:35 AM
Wow, Tiff, sorry about the small mill. I think they have pills nowadays to fix that sort of thing <G>.

I also don't think I'd be using that Jacobs chuck if I cared so much about the accuracy as you seem to, but to each his own (there, the pin is pulled and the 'nade sits on the table!).

Here's my IH not breaking a sweat with the Co-Ax and a longish probe:

http://www.cnccookbook.com/img/MillStuff/CNC/ZAxisMod/P1010262.JPG

I'm probably at about 1/3 of the Z-travel. Yup, coulda used a shorter probe, but the bore came out very nicely and the job was all over in just a few minutes.

Cheers,

BW

Thanks Bob.

As I said,that was a very good post with quite valid and pertinent points as regards fitting my set-ups on my X3 mills onto my HF-45 mill.

I said that I'd do that set up on my HF-45 - and here they are - with a 40"/1 metre rule for scale. I have included the longer internal and external probes/feelers as well.

http://i200.photobucket.com/albums/aa294/oldtiffie/measuring/Co-axialindicator4.jpg

http://i200.photobucket.com/albums/aa294/oldtiffie/measuring/Co-axialindicator5.jpg

http://i200.photobucket.com/albums/aa294/oldtiffie/measuring/Co-axialindicator6.jpg

and as a link - I've exceeded my 4 images per post otherwise:
http://i200.photobucket.com/albums/aa294/oldtiffie/measuring/Co-axialindicator7.jpg

Glenn Wegman
11-11-2009, 07:34 AM
The mystery is ended - from the Blake site: http://www.blakemanufacturing.com/pages/repairs.html
Popcorn, anyone?



...
Yeah, maybe if Jim, Glenn, Blake, you and I repeat it a couple of hundred more times he'll believe it :)


lazlo,

I posted this on page 7, glad to see you guys are making progress! :)


lazlo,
Does Blake specify the probe length and relative angle that they claim will give you .0002"? They would make far more sense if it had just graduations and no numbers on the face!

JCHannum
11-11-2009, 07:37 AM
Thanks Dennis.

I'd have thought (past hoping??) that this too would have been included in the literature that the manufacturer put in with the tool - all manufacturers - "Blake" and clone-makers alike.

Its a fine line between not telling the whole truth - deliberately or not - on the one side and dishonesty on the other.

There should not be any need to chase all over the web to get this stuff.

Why do you feel there is any dishonesty in any of Blake's claim. It is a Co-Axial Indicator nothing more or less. Blake advertising specifically states it "Provides a simplified method for centering workpiece or fixture on horizontal or vertical machines, external or internal diameters..." It cannot get much clearer than that. The instruction sheet clearly and succinctly details application and use. Again clear as a bell.

The manufacture does assume some level of intelligence in the end user and does not include a list of applications it is not intended for. These misapplications would include, but not be limited to, use as a DTI, voltage meter, hammer, or pry bar.

Robert's point taken on the concentricity of the collet or toolholder holding the 3D taster is correct. If the bore is offset or angular, the axis of the taster will be off. Similarly, any lack of concentricity in the machine spindle will have the same result.

oldtiffie
11-11-2009, 08:26 AM
Why do you feel there is any dishonesty in any of Blake's claim. It is a Co-Axial Indicator nothing more or less. Blake advertising specifically states it "Provides a simplified method for centering workpiece or fixture on horizontal or vertical machines, external or internal diameters..." It cannot get much clearer than that. The instruction sheet clearly and succinctly details application and use. Again clear as a bell.

The manufacture does assume some level of intelligence in the end user and does not include a list of applications it is not intended for. These misapplications would include, but not be limited to, use as a DTI, voltage meter, hammer, or pry bar.

Robert's point taken on the concentricity of the collet or toolholder holding the 3D taster is correct. If the bore is offset or angular, the axis of the taster will be off. Similarly, any lack of concentricity in the machine spindle will have the same result.

Jim.

This is what Blake's web site says - pretty well as you say:
http://www.blakemanufacturing.com/pages/aboutus.html

Blake also published a table of correction factors to four decimal places that was posted here - by you? - but I can't see it at the moment in this thread and I can't find it on Blake's web site either. Can you re-post it please?

With a scale marked/calibrated to 0.0005" and the correction scale referred to, the end-user can, on the face of it, be confident of such an outcome - can he not? But can or will he get it? I think not and it is agreed that he will probably not.

Based on your argument or assumption of an indicated but not necessarily accurate centreing, it would or may need to be verified with a good dial indicator - which operates in a static condition.

I know the merits and limitations of the Blake or any other similar co-ax indicator and will make my assumptions accordingly. Which is both fair enough and as it should be. I would hope that other actual and potential "Blake" co-ax indicators are, or can be, similarly well informed.


Robert's point taken on the concentricity of the collet or toolholder holding the 3D taster is correct. If the bore is offset or angular, the axis of the taster will be off. Similarly, any lack of concentricity in the machine spindle will have the same result

Taking that statement both literally and at face value to its limits, there is a good case to say that any mill/spindle/collet chuck/collet combination that has any detectable radial or angular run-out cannot and will not be able to use a Taster to its optimum performance. It could be similarly said that there is no need for the Taster "X" and "Y" adjustments at all or if fitted any use or need for them.

I doubt that is so.

I will post the "manuals" - such as they are - for my "clone" Taster and co-axial indicators - hopefully tomorrow or the day after. Can you please post the manual for (your) "genuine" "Blake" co-axial indicator please. I expect that some here - for what-ever reasons - will not have a copy of the manual for their Taster or "Blake" or clone co-axial indicators.

Carld
11-11-2009, 09:31 AM
Here ya go Tiff.

http://www.blakemanufacturing.com/pages/coaxvalues.html

Glenn Wegman
11-11-2009, 09:50 AM
Also note that the capacity is listed to 4 1/4" bore max.

http://www.blakemanufacturing.com/pages/aboutus.html

JCHannum
11-11-2009, 10:07 AM
With a scale marked/calibrated to 0.0005" and the correction scale referred to, the end-user can, on the face of it, be confident of such an outcome - can he not? But can or will he get it? I think not and it is agreed that he will probably not.

Based on your argument or assumption of an indicated but not necessarily accurate centreing, it would or may need to be verified with a good dial indicator - which operates in a static condition.

I know the merits and limitations of the Blake or any other similar co-ax indicator and will make my assumptions accordingly. Which is both fair enough and as it should be. I would hope that other actual and potential "Blake" co-ax indicators are, or can be, similarly well informed.


Taking that statement both literally and at face value to its limits, there is a good case to say that any mill/spindle/collet chuck/collet combination that has any detectable radial or angular run-out cannot and will not be able to use a Taster to its optimum performance. It could be similarly said that there is no need for the Taster "X" and "Y" adjustments at all or if fitted any use or need for them.

I doubt that is so.

I will post the "manuals" - such as they are - for my "clone" Taster and co-axial indicators - hopefully tomorrow or the day after. Can you please post the manual for (your) "genuine" "Blake" co-axial indicator please. I expect that some here - for what-ever reasons - will not have a copy of the manual for their Taster or "Blake" or clone co-axial indicators.

Tiff;
I have posted the table of correction factors and the instructions are clear enough that there is little need to waste more bandwidth and time reproducing them here. The link is right at the bottom of the page you have posted. Similarly don't waste any more bandwidth and time with endless copies of taster manuals, you have done that to death.

You continue to state the Blake dial is graduated to 0.0005", while it is clearly graduated to 0.0005" Axis Offset. Blake states an accuracy of 0.0002" of true center, and I see no reason to question that, nor do I see anybody other than yourself unable to absorb the fact that the Blake will produce that accuracy.

I suggest you reread the final paragraph of the Blake link you posted, it fully explains the advantages of the Blake over static indicators.

No amount of zeroing of a 3D taster will overcome an eccentric collet or toolholder. Yes, you can zero it to the existing point, but remove it, turn the spindle 90 degrees, reinsert the taster and it will be off the amount of the eccentricity. It cannot be any other way.

lazlo
11-11-2009, 10:13 AM
Blake also published a table of correction factors to four decimal places that was posted here

With a scale marked/calibrated to 0.0005" and the correction scale referred to, the end-user can, on the face of it, be confident of such an outcome - can he not?

That table is just doing the trig for you based on the probe length. The Blake uses a 5 tenths dial indicator to measure the plunger offset, so with a 2" probe, the graduations line up to 5 tenths, but with a 4" probe, the graduations are effectively a thou.

But you don't really care what the graduations are, you're just minimizing the wobble (the total runout).

Or, going back 11 pages:


It's very accurate. Blake's specs indicate centering within 2 tenths. We're still arguing whether that accuracy decreases due to the length (cosine angle) of the probes, since its an indirect reading setup that's measuring the axial bore offset...

lazlo
11-11-2009, 10:17 AM
No amount of zeroing of a 3D taster will overcome an eccentric collet or toolholder. Yes, you can zero it to the existing point, but remove it, turn the spindle 90 degrees, reinsert the taster and it will be off the amount of the eccentricity. It cannot be any other way.

Tiff, I think I threw you off when I responded to John's mention of Metric R8 collets -- that I needed to find an accurate collet, and not a cheap Chinese collet.
I was talking about needing an accurate collet for my 3D Taster. If you recall, I was annoyed at the amount of headspace that the ER40 collet chuck takes.

As I've said, you don't need an accurate collet for the Blake, because it's rotating.

You often like to refer back to the original post :) I posted a description in the first reply to this thread describing how to use a coaxial indicator, and John replied that a 3D Taster was much better. I replied that my experience was that the 3D Taster was great for setting-up square edges, if you have a DRO, but that a Blake, or Zero-Set, or Indicol was much easier and quicker for round features.

Somehow that spun into a 15 page criticism of the coaxial indicator, which ironically, was the OP's question.

JCHannum
11-11-2009, 10:31 AM
Somehow that spun into a 15 page criticism of the coaxial indicator, which ironically, was the OP's question.

You can go to the options section of the User CP and select 40 posts per page, reducing that to only six pages.

rode2rouen
11-11-2009, 10:38 AM
You can go to the options section of the User CP and select 40 posts per page, reducing that to only six pages.


Now that's helpful!
I'm showing 21 pages!!


Rex

Carld
11-11-2009, 11:06 AM
This has been an entertaining thread to nowhere. The question of how to use a co-ax indicator was answered long ago. Then it went on to describe the inadequacy and errors a co-ax has.

Well, a co-ax indicator is a tool to locate center of a post, dowel, bore, etc. I have even used it to check if the work was flat in the vise or on the table which is not what it is designed for.

I just got back from the shop where I bored a piece of DOM and faced it. I tested the collet with a dial indicator and dowel rod and it had less than .0005" runout. I laid the DOM tube on the mill table and oiled the bore and used the short, medium and long probes to test the centering of the work. I first centered it with the short probe and indicated it in very near perfect. I tried the medium and it showed the same variation on the needle. I tried the long probe and found a little less than .0005" variation of the needle and I think that is because the longer probe increases the sensitivity of the indicator and produces a more accurate centering of the work.

Now here is something that will stick in the craw of some, my co-ax is a $100 Chinese version I bought about 10 years ago.

The needle was solid during the readings, it did not flutter at all on the smooth oiled surface as it would on a dry surface. I have used a co-ax indicator for about 15 years since it was first shown to me and I have found them to be very accurate if used CORRECTLY for what they were intended.

I have always felt confident I was within .0005" of center when I used them in a smooth bore with oil on the surface. If it was a dry rough bore it was never accurate to better than maybe .001"-.002".

The co-ax is only as good as the surface it is used on and the ability of the user. How you use it is up to you but if you follow the instructions and are a reasonable person you can expect accurate readings.

mochinist
11-11-2009, 11:25 AM
If John would have came in and said that the co-ax was the best invention ever, then tiffy would have loved the thing. truth :cool:

Peter N
11-11-2009, 12:37 PM
Somehow that spun into a 15 page criticism of the coaxial indicator...<snip>

That's Intractable Antipodean Obduration :D

dp
11-11-2009, 01:11 PM
The co-ax is only as good as the surface it is used on and the ability of the user.

You've just described my Harley and all my indicators :)

oldtiffie
11-11-2009, 04:07 PM
Here ya go Tiff.

http://www.blakemanufacturing.com/pages/coaxvalues.html


Thanks Carl -appreciated.

There it was - right at the bottom of my screen!! - it came up fine when I enlarged the screen - F11

I apologise to Blake, Jim H and others in that regard.

Mea culpa, mea culpa, mea maxima culpa

http://en.wikipedia.org/wiki/Mea_culpa

Just HAD to get another Wikipedia quote in!!!

(Just??) as you'd expect???!!

oldtiffie
11-11-2009, 04:19 PM
Originally Posted by Carld
The co-ax is only as good as the surface it is used on and the ability of the user.

[QUOTE=dp]You've just described my Harley and all my indicators :)

And you Dennis have just described my 1943 WLA 750cc Harley-D (ex Army - still in its original Army livery!!!).

Rough, ready and just needed a good kick in the guts to get her ticking and kicking back!! Some girl friends were much the same - but the H-D was uglier (not by much).

Now, my 500cc Vincent HRD "Comet" was an entirely different bike - beautiful machine. I had o settle for the 500 as I couldn't get a 1,000 "Black Shadow/Lightning" at the time (~ 1955) for love nor money - they were like rocking horse $hit - and with good reason. Just as well as I bloody near killed myself often enough on my 500.

oldtiffie
11-11-2009, 09:52 PM
I have up-loaded the manuals for my "Chinese" clones of the "Blake" co-axial indicator plus a screen-shot of the Blake web page for the adjustment table for the various feeler arms and diameters.

I have also uploaded the manual for my clone "3-D Taster".

These are typical "Chinglish" as regards text etc. plus the copies that came with my (new) tools were pretty poor as regards quality - so I scanned them at a lot higher resolutions than I otherwise might have.

The details are there - its just a matter of persevering with them to make sense of them.

Co-axial indicator:
http://i200.photobucket.com/albums/aa294/oldtiffie/measuring/Co-axindicatormanual1.jpg

http://i200.photobucket.com/albums/aa294/oldtiffie/measuring/Co-axindicatormanual2.jpg

http://i200.photobucket.com/albums/aa294/oldtiffie/measuring/Co-axindicator-Blake-adjustmentfact.jpg

"3-D Taster":
http://i200.photobucket.com/albums/aa294/oldtiffie/measuring/3DTastermanual1.jpg

http://i200.photobucket.com/albums/aa294/oldtiffie/measuring/3DTastermanual2.jpg

I have only included them as links as the files may be a fair size.

I hope it helps.

dp
11-11-2009, 11:46 PM
I realize this topic has been beat to death but I still did not have an adequate answer for using the center point probe with the Fake Blake. I decided to have another go at it so I put the Fake Blake in my Chinese mill, used a Chinese one-shot centerpunch to mark a piece of 1018 I had. This I set under the Fake Blake with the center point probe in place. I lowered it to nearly touching the work and started the spindle turning.

TMidget might prefer I call it a quill. Anyway, I observed the swing and minimized the probe swing as best I could, shifted the table to best eye-ball center, and lowered the probe into the center point. I compressed the probe (it is spring loaded) about 1/8". I started the quill spinning and using the x-y wheels I was able to bring the needle deviation to under half a mark. It is very sensitive as you get to 0/0 offset and it is easy to overshoot.

I stopped the spindle (Sorry, TM - old habit) and replaced the Fake Blake with my wiggler w/pointer and damn, it was dead on.

But in my search for more information I blundered onto the Little Machine Shop store and found this video:

http://littlemachineshop.com/video/Coaxial_Centering_Indicator.wmv

It doesn't show the center point tool in use but it's a pretty decent video.

So the one good thing that has come from this thread is I finally shimmed my mill (same kind as in the video) and also centered my tailstock ram using only this indicator and of course my favorite made in China Volkswagon brake rotor. Oh, and the center point tool works with it but is far slower than a wiggler.

oldtiffie
11-12-2009, 12:19 AM
Good for you Dennis. I knew the "pointer" worked - 'coz I tried it. Its in the "manual" that I posted.

Just a thought on square column mills and tramming.

It is quite possible to get a seemingly "good" tram by tilting the head left-right (for "X") and by shimming high/low at the tilting head for "Y".

That takes no account or nor does it identify any "out of square" error between the vertical column dove-tail and (in the "Z" plane) at the "X"-"Z" and "Y"-"Z" planes.

This has to be rectified first if necessary.

Example:
If the column is tilted "left" by say 1 degree and the head tilted right by one degree a test tram will show it to be OK when it is/may not be.

If you were to have a very small point of light centred on and parallel to the mill-head axis and when you raise/lower the milling-head on the column the spot of light on the table - instead of staying stationary on the table would move left/right (in "X") an amount equal to the sine of the angular off-set (away from 90 degrees).

The error is unlikely to always be all "X" or all "Y" but a vector sum/off-set composed from (X" + "Y") so both "X" and "Y" will/may need attention - by shimming under the base of the column. If you are unlucky it can take quite while (chasing your tail??) to get right.

This is the cause of a lot "out of line" errors apparent in other wise seemingly correctly trammed and positioned works.

This can be very difficult to explain to people who either cannot understand it -or those that can but who don't want to.

Keep up the good work with the co-ax indicator.

dp
11-12-2009, 12:37 AM
Good for you Dennis. I knew the "pointer" worked - 'coz I tried it. Its in the "manual" that I posted.

Just a thought on square column mills and tramming.

It is quite possible to get a seemingly "good" tram by tilting the head left-right (for "X") and by shimming high/low at the tilting head for "Y".

That takes no account or nor does it identify any "out of square" error between the vertical column dove-tail and (in the "Z" plane) at the "X"-"Z" and "Y"-"Z" planes.

I recall that you did quite a good write-up about that two or three years ago. The particular construction of the mill does require the head be aligned to the column and the column aligned to the table. The head can be off from the column on three axes and there are precious few tools built in for correcting that. None, in fact.

I was thinking a couple nights ago that if one were explore a mill of this construction with lines drawn from all the points of error one could quickly find themselves looking at a parallelogram.

To explain the point - if the column is off perpendicular to the table on the X axis and the head is rotated back to vertical, The mill will be in tram, but the center point of the quill will move along the X axis as the head is raised and lowered. It is the same with the Y axis. This is where a fine laser beam projecting from the quill would be a great alignment testing tool.

I should dig out the way-back machine and see if I can find that piece you wrote.

Edit: Interesting thing - I didn't see any of your post below what I quoted. It didn't arrive here with the rest of the post, so I pretty much repeated what you'd said :). Computers will be the death of us.

oldtiffie
11-12-2009, 12:56 AM
Thanks Dennis.

Use a good 2-3-6 block bolted/clamped to the table, put an indicator base in your quill or on the mill head with the indicator feeler on the 2-4-6 block, zero the indicator and run the head up and down on the column. It should be pretty close to zero. If not, loosen the bolts at the base of the column and shim to suit. Do for both "X" and "Y".

When OK - tram the mill head - use the co-ax at maximum sweep - it should at least get it pretty right. Its your call as tho whether you (re)"sweep" with a good dial indicator or not.

I reckon that as your co-ax is as good as it is, I'd use it tram your mill.

Use the correction table from Blake that I posted to re-assess the new value of actual deflection per the indicator dial ("0.0005") calibration/indexing. I reckon it will not only work at all but will work well. I was very impressed with that video of yours using the co-ax to tram the VW brake disk - very - as I had never tried it and until your video I was more than just a bit skeptical of it - but not now!!

If for no other reason, I just had to include the co-ax just to stay "on topic".

dockrat
11-12-2009, 01:35 AM
This is where a fine laser beam projecting from the quill would be a great alignment testing tool.

Here you go. One of these:

http://i253.photobucket.com/albums/hh67/Dockrat1/IMGP1224Medium.jpg

oldtiffie
11-12-2009, 03:53 AM
That's a good unit/tool and a very good suggestion Ernie as it has an "X"-"Y" adjuster (just like the "Taster") to centre the beam to a minimum spot and a minimum "throw" and so align it with the lathe axis.

As I recall, and without going out to the shop to get the unit, I think it has an optimum focal length of about 100mm (~ 4") - but as 4" of travel of the head on the vertical column dove-tail guide will soon show if there is a serious alignment problem in that regard - or not.

If needed and after that, Dennis can use the best of whatever (other?) tools he has at his disposal to do the job/s.

dp
11-16-2009, 12:15 AM
Here's great vid from Glacern that shows all the various edge and center finders at work in one place. http://www.glacern.com/index.php?page=video_index&video=1&video_name=crash_course_mill_05

Except the DTI, oddly.

hwingo
11-16-2009, 02:06 AM
Nice find! Thanks for sharing.

Harold

Mosey
12-12-2009, 07:58 PM
23 pages and still going!! Boy, am I pleased to see that my question invoked/provoked such full responses! Thank you.

John Stevenson
12-12-2009, 08:05 PM
23 pages and still going!! Boy, am I pleased to see that my question invoked/provoked such full responses! Thank you.

Well it wasn't, it died a month ago :rolleyes:

.