PDA

View Full Version : Threads pitches are off center



SSC
01-18-2012, 11:48 AM
Hey guys, I am glad I finally found this place. Looks like tons of great info.


I have run into a snag on single point threading on my G4003G lathe.

I was wondering what could be causing my threads to look more like NNNNN than VVVVVVV? Does that make sense? the thread pitches are almost flat on one side. I will try to post a picture later.

Black_Moons
01-18-2012, 11:52 AM
Wrong shape of tool? Wrong infeed? Poor alignment of tool?

Protip: When they say 29.5 degrees, they mean 29.5 degrees off the cross slide axis, Not off the carriage axis (ie, the compound should be closer in angle to straight in/out of the thread, And not aligned to the length of the work/carriage feed)

My lathe, and may others, are marked where 29.5 degrees on the scale is actualy 60.5 degrees (compaired to other lathes) and that does not work well at all for threading. Solution is to use a protractor and/or put another mark on the lathe 90 degrees from the origional, if your angle scale does not extend far enough to put it to 60.5 degrees.

SSC
01-18-2012, 11:56 AM
forgot to link to the tool i use.

Threading Toolr (http://www.use-enco.com/CGI/INSRIT?PMAKA=422-3041+)
http://www.use-enco.com/ProductImages/0012513-11.jpg

SSC
01-18-2012, 12:03 PM
I checked my angle with a protractor we are right at 29.5.

since the threading tool has such a small angle to check with a fish tail, on quick jobs i normally set it up off of a facing cut and make sure the holder is flat against the end of the work. did that make sense? I saw that on an MIT video (http://www.machinistblog.com/mit-techtv-machine-shop-videos/)

jimmstruk
01-18-2012, 12:07 PM
That picture of toolholder is NOT a threading tool!!

jimmstruk
01-18-2012, 12:13 PM
My bad, tool catalog says it is. Sorry JIM

EVguru
01-18-2012, 12:18 PM
I checked my angle with a protractor we are right at 29.5

With referrence to what?

Try a radial infeed (just the cross-slide). You don't HAVE to feed in at 30 or 29.5 degrees, you just often get better results doing it that way becuase you're cutting only/mostly on the leading edge.

You can also leave the top-slide set parallel to the spindle axis and advance the topslide 50% of the radial infeed. That puts most of the cut on the leading edge, but saves messing with the top-slide.

Oldbrock
01-18-2012, 12:18 PM
You need to do some reading of machine tool operation books, that tool you show will not cut 60 degree threads the way it would be normally set up. You need a fishtail to set up a threading tool so that the centerline of the tool is square to the work not the side. Peter

SSC
01-18-2012, 12:19 PM
Now i will say that my protractor is not a precision tool so it could also be off. Is there a way to test the angle my compound is set to? It could be my hash marks are off.

EVguru
01-18-2012, 12:23 PM
You need to do some reading of machine tool operation books, that tool you show will not cut 60 degree threads the way it would be normally set up. You need a fishtail to set up a threading tool so that the centerline of the tool is square to the work not the side. Peter

If you have a hand ground tool you might need a fishtail, but a jig ground tool (or insert) should just be able to be set up perpendicular to the work.

SSC
01-18-2012, 12:24 PM
You need to do some reading of machine tool operation books, that tool you show will not cut 60 degree threads the way it would be normally set up. You need a fishtail to set up a threading tool so that the centerline of the tool is square to the work not the side. Peter


I am sure i just didn't explain it right. I am not setting it to the side of the work. typically all of my threading is near the end of a work piece, so I make a light facing cut to the end of the work. then I loosen the tool post and press the length of tool holder to make perfect contact across the front work. doing this I have tested it and majority of the time it is within a .001 or less across the length of the tool holder.

SSC
01-18-2012, 12:26 PM
With referrence to what?

Try a radial infeed (just the cross-slide). You don't HAVE to feed in at 30 or 29.5 degrees, you just often get better results doing it that way becuase you're cutting only/mostly on the leading edge.

I placed the protractor along the side of the compound and and the other side along the back of the cross slide.

SSC
01-18-2012, 12:27 PM
Okay I did a plunge still thread and it was a much much cleaner thread. does that mean that my compound could be off and when it reads 29.5 it really isn't 29.5?

Black_Moons
01-18-2012, 12:42 PM
You need to do some reading of machine tool operation books, that tool you show will not cut 60 degree threads the way it would be normally set up. You need a fishtail to set up a threading tool so that the centerline of the tool is square to the work not the side. Peter

Nah, those inserts are really funky, they actualy have a thread 'point' on them thats not outward from the center of the triangle, but on an angle.

http://image.made-in-china.com/2f0j00SvDaZYsFkKcl/Threading-Insert-4RD-.jpg
(example image.. athough those are for some weird round thread?!?)

http://static.traderscity.com/board/userpix30/7041-threading-insert-cemented-carbide-1.jpg Another.

SSC: anywhere from 0 to 30 degrees will work. Closest to 30 degrees is best, hence 29.5 is picked as 'closest you can get without going over', Try 25 degrees or whatever you can 'guarentee' is less then 30 degrees given whatever accuracy you think you have.

If it looks awful, your likey at 60+ degrees instead of 29 because your compound is marked realative to the carriage angle and not the cross feed angle.

LKeithR
01-18-2012, 01:05 PM
Is there a way to test the angle my compound is set to? It could be my hash marks are off.
To accurately set the compound angle you can use a dial indicator. Just to be clear that the angle is in the right place imagine that the centre line of the compound is parallel the the centre line of the cross-feed. In other words, both handles point back at your stomach. Now push the compound handle 29.5 degrees to your right and align as close as you can with the protractor hash marks.

To check your angle mount a dial indicator on the bed and align it so that it reads off the side of the compound. Zero the dial on the indicator and also set the cross-feed handle to zero--being sure to take out the backlash in the process. Now move the cross-feed exactly one inch towards or away from you--either way works as long as you have enough room to travel a full inch and you're moving the compound the same direction you used to remove the lash. Adjust the angle of the compound till your dial indicator reads exactly .566" when you have moved the compound one inch and you'll be dead nuts on 29.5 degrees. Of course the compound doesn't need to be that accurate--working from the protractor is fine--but you can do this as a one-time test and then put matchmarks on the compound and cross-feed so you can quickly find the same spot the next time...

J Harp
01-18-2012, 01:12 PM
Suggestion, try chucking up something with a known good thread. Then set your threading tool so it fits into the groove as close as you can get it. It doesn't matter whether you are feeding in with the compound or the cross slide, the center line of the threading tool has to be at 90 deg. to the work to produce a good thread. With a small lathe you will probably get a cleaner thread by feeding with the compound at slightly less than 30 deg.

LKeithR
01-18-2012, 01:15 PM
SSC: anywhere from 0 to 30 degrees will work. Closest to 30 degrees is best, hence 29.5 is picked as 'closest you can get without going over', Try 25 degrees or whatever you can 'guarentee' is less then 30 degrees given whatever accuracy you think you have.



We use 29.5 degrees so that when you feed the tool in with the compound there is a half degree clearance on the back face of the tool. For whatever depth of thread you require the tool ends up cutting the full face of the thread on the front side but leaves a series of tiny steps on the back side. The trick to getting a clean looking finish is to make your final .001" - .002" cut by moving the cross-feed in. This allows the tool to take a final full face cut on both faces of the thread...

John Garner
01-18-2012, 01:21 PM
SSC --

You need to set the cutting edge of the tool relative to the workpiece -- typically done by centering the tip in the V of a centergage held against the cylindrical surface of the workpiece -- instead of setting the shank of the toolholder parallel to a faced surface of the workpiece.

John

Black_Moons
01-18-2012, 01:43 PM
We use 29.5 degrees so that when you feed the tool in with the compound there is a half degree clearance on the back face of the tool. For whatever depth of thread you require the tool ends up cutting the full face of the thread on the front side but leaves a series of tiny steps on the back side. The trick to getting a clean looking finish is to make your final .001" - .002" cut by moving the cross-feed in. This allows the tool to take a final full face cut on both faces of the thread...

If you have 'tiny steps' on the backside, thats a sign your using 30.5 degrees and not 29.5 degrees.

The whole point is to be criticaly under 30 degrees so you don't get little steps on the back side. Using cross feed for the final pass can be useful to help clean a thread as both sides get even cutting pressure and hence similar finish/etc, but it should'nt be needed.

TGriffin
01-18-2012, 01:46 PM
SSC,

Forget about the angle of the compound rest, it isn't important. Just set it at 30 degrees and forget about it. The whole 29 1/2 degree, 30 1/2 degree thing that the text books promote just creates a lot of confusion and doesn't really do anything. Only two settings are important; the angle of the tool relative to the work and the height of the tool. Since a threading tool is a form tool and you are copying its shape, it must be set up so the center line of the tool is perpendicular to the work and it is exactly on center vertically (line it up with the point of your center). Feed the tool in using the compound and when the pitch diameter is a few thousandths over size, switch to feeding with the cross slide to take it to the finished diameter. That will nicely clean up the thread and make it much easier to hit the final pitch diameter because you can read the diameter directly off the cross slide dial instead of trying to interpret the reading on the compound dial. It also helps to take a few free cuts without moving the tool in to compensate for any deflection of the part before measuring the pitch diameter.

Tom

baldysm
01-18-2012, 01:54 PM
If you using a Chinese lathe, there have been reports of the graduations on the compound being off. Like having more than a single 20 degrees between 0 and 90.

I know the textbooks always teach you to set the compound at 29.5 degrees, and feed in on the compound, returning the cross slide to 0 before taking the next cut.

Haven't done things that way in years. The compound angle is irrelevant to the way I thread. I feed straight in. Never had a problem, but it is a little harder on the tool.

JCHannum
01-18-2012, 02:00 PM
since the threading tool has such a small angle to check with a fish tail, on quick jobs i normally set it up off of a facing cut and make sure the holder is flat against the end of the work. did that make sense? I saw that on an MIT video (http://www.machinistblog.com/mit-techtv-machine-shop-videos/)

I think this is your problem, the holder should be set at 30* to the facing cut to position the insert properly.

dp
01-18-2012, 02:11 PM
If your compound when set to 0 moves parallel to the face of the chuck then you rotate it CCW to 29.5. If it is parallel to the ways at 0 then you rotate it CW 60.5.

Various lathes come with one or the other method for 0 alignment.

marinesteam
01-18-2012, 02:30 PM
If your compound when set to 0 moves parallel to the face of the chuck then you rotate it CCW to 29.5. If it is parallel to the ways at 0 then you rotate it CW 60.5.

Various lathes come with one or the other method for 0 alignment.

My 4003G compound is marked as zero when the compound slide move parallel to the ways. Threading is done at 60.5 as the previous post noted.


marinesteam

lynnl
01-18-2012, 02:39 PM
I was wondering what could be causing my threads to look more like NNNNN than VVVVVVV? Does that make sense? the thread pitches are almost flat on one side. I will try to post a picture later.

You didn't say which side of the thread is flat, but if you're getting NNNN shape, flat on one side, then obviously one flank of your cutting tool is oriented perpendicular (or nearly so) to the work piece. Think about. Each flank of the 60 deg cutter has to be at a 30 degree angle. Make that happen.

Like TGriffin said, don't get wrapped around the axle over the 30 vs 29.5 angle of the compound. There's no significant difference, and 30 degrees is easier to find.

willmac
01-18-2012, 02:45 PM
Unless you have validated the accuracy of the angular graduations on your compound, you should not rely on them. Even if you have, it is generally preferable to check with a protractor.

Stepside
01-18-2012, 02:58 PM
My Enterprise lathe has the witness mark hidden under the compound. It takes a flashlight to see the setting. So I used a piece of 1/4 inch thick Aluminum plate to make a tool. One side of the plate is 29.5 degrees off the other side. Think of a 30-60 drafting triangle with the tip cut off, only it is a 29.5-60.5 triangle. One side is placed against the face of the chuck and the other side locates the compound. It hangs on a screw next to the rest of the lathe tooling. One could use a 30-60 triangle and it would work fine except for the hot chips and plastic.

I think the 29.5-30 degree and the crossfeed only-compound feed issues are like the Ford-Chevy-Dodge issue. Find what works for you and be consistant.

SSC
01-18-2012, 03:06 PM
My 4003G compound is marked as zero when the compound slide move parallel to the ways. Threading is done at 60.5 as the previous post noted.


marinesteam


thanks guys. this did the trick!!!

This makes me feel kinda dumb. Everyday I am given the chance to be humbled.

Black_Moons
01-18-2012, 03:37 PM
thanks guys. this did the trick!!!

This makes me feel kinda dumb. Everyday I am given the chance to be humbled.

Haha don't worry about it. My lathe is the exact same way, and I was wondering why my threads looked so bad too. Then I looked carefuly and realised "hey wait a second, 29.5 degrees makes no sense this is feeding across not in... Hey maybe its the OTHER 29.5 degrees!"

Stepside: Please don't tell people that 29.5 vs 30 is 'not an issue', its very clearly a mechanical issue.

'29 vs 29.5' or even '28 vs 29.5' is 'not an issue'.

The whole point of 29.5 is just to be *critialy* less then 30 degrees. Anything less then 30 degrees works great. Anything more does not. 'Close to but not exceeding 30 degrees' does arguabley work 'better' then 0 degrees.

'0 verus less then 30 degrees (aka 29.5)' degrees usage would be a 'ford verus dodge' issue, Some people do it one way, other people do it the other way.
Less then 30 degrees verus more then 30 degrees is a 'doing it right, or doing it wrong' issue.

If you look at how it cuts, anything from -30 degrees to +30 degrees will produce the exact same thread. Anything under -30 or over +30 will produce a 'steped' thread, As close to 30 degrees as you can get will reduce the cutting force. Hence: 29.5 degrees.
To be exact, its 30 degrees minus whatever you think your accuracy is. Hence you can't tell people 30 degrees without giving them a tollerance of -? +0

Boucher
01-18-2012, 05:01 PM
Where are you in central Texas? There are several members here scattered around the area.

lynnl
01-18-2012, 05:43 PM
Anything under -30 or over +30 will produce a 'steped' thread, As close to 30 degrees as you can get will reduce the cutting force.

How does it produce a stepped thread? The compound is still being fed in a straight line. Granted, if the compound is skewed around at, say 31 deg the thread will be cut with an included angle of more than 60, because that flank on the headstock side will be at 31 degrees, rather than 30.

But assuming you take some spring passes at the end, it will clean up to whatever angle is ground on the cutter tip, which should be 60.

Tony Ennis
01-18-2012, 07:00 PM
mrpete222 (tubalcain) has some great videos (http://www.youtube.com/watch?v=9y0MmvscBzg) on youtube.

J. Randall
01-18-2012, 11:34 PM
If you have 'tiny steps' on the backside, thats a sign your using 30.5 degrees and not 29.5 degrees.

The whole point is to be criticaly under 30 degrees so you don't get little steps on the back side. Using cross feed for the final pass can be useful to help clean a thread as both sides get even cutting pressure and hence similar finish/etc, but it should'nt be needed.

BM, I have always been under the belief that the steps are there, even when set accurately at 29 1/2 degrees. I think it would take pretty good magnification to see them on a fairly fine thread, but you can sometimes see them with the naked eye on a real coarse thread. That's the reason for the final cleanup cut straight in, to clean the backside up.
James

Jim Hubbell
01-19-2012, 12:33 AM
To get an idea of what is going on, chuck a piece (scrap) with an 8 TPI (large)thread and set it turning in back-gear (slow). Set up your tooling as if to cut the same thread. Now do some " air cutting ". As the piece is turning adjust the single point to follow and watch the point enter the existing thread. Using the compound, the point should go all the way in without cutting. If it is run in to the point of starting to cut, it will cut on the left-hand side of the tool. This setup will show how the various angles mentioned in the previous posts affect the cut and therefore the finished thread.
Try this and it should make things more clear.
Good luck

oldtiffie
01-19-2012, 12:54 AM
Hey guys, I am glad I finally found this place. Looks like tons of great info.


I have run into a snag on single point threading on my G4003G lathe.

I was wondering what could be causing my threads to look more like NNNNN than VVVVVVV? Does that make sense? the thread pitches are almost flat on one side. I will try to post a picture later.


SSC.

To clear up any misunderstandings, can you please do a thread, leave the tool close to full depth (so as not to damage the insert) and take a pic or two from right above the tool. Show the job as well. Then post the pics here.

EVguru
01-19-2012, 06:02 AM
Some people might find this picture of some infeed methods useful.
http://d2n4wb9orp1vta.cloudfront.net/resources/images/cdn/cms/Pic298.jpg

With 60 degree threads the 30 degree method is often referred to as 'flank infeed' and the angle of less than 30 degrees as 'modified flank infeed'.

Smaller less rigid lathes may 'need' to use the flank/modified flank infeed method to reduce the cutting forces and prevent chatter etc. If you can get good results by simple radial feed, why not do so? It's much simpler! You can read your thread depth straight off the dial.

Often problems only occour on later passes as the depth of cut increases and reducing the infeed can help. Initial passes can be much more agressive and for that reason CNC software often has an 'equal volume of chip per pass' algorithm built in.

Both flank and modified flank infeed methods advance the cutting tool in both radially and axially. Using the cross slide to feed in radially, whilst using the top slide (set in it's usual position parallel to the spindle axis) to feed in axially simulates feeding in using the top slide set at an angle, but without the bother of actually changing the angle of the top slide and the thread depth can be read directly from the cross slide dial. You could do the trig to work out what ratio of feeds to apply to generate a 30 degree (or less than for modified flank) infeed, but in practice 50% feed on the top slide works well.