PDA

View Full Version : Cutting a 57 tooth gear



duckman
03-15-2016, 06:49 PM
Person at work said that we can't cut a 57 tooth gear, asked why and did not get an answer, so what's the answer.

dave_r
03-15-2016, 07:58 PM
You aren't authorized to make one? Nobody where you work knows how to do it? :p

Mcgyver
03-15-2016, 08:41 PM
make gears? I thought you had to buy gears!

likely because its not part of the plate set you have for the dividing head.....you'd need a 57 hole plate, each move is 40 holes on a 57 hole plate....either that or differential indexing.

dp
03-15-2016, 08:42 PM
Person at work said that we can't cut a 57 tooth gear, asked why and did not get an answer, so what's the answer.

What tools are available for those gears you can make?

Paul Alciatore
03-15-2016, 09:04 PM
You CAN make a 57 hole plate if you have a dividing head or RT and a lathe.

duckman
03-15-2016, 09:09 PM
These gears are on the outside of large tie bar nuts for molding machines. I've just taken over for there hackerchinist, the welder came and made a comment that I've already surpassed the other guy, in 2 weeks I've not had to have anything welded to fix a woops , back to the gear we have, a dividing head but don't know what plates we have, the head is 40-1 , differential is out no gears . We make screws for injection and extruders machines, tie bars, nozzles, and repair other screw ups, I just finished shortening a carbide lined barrel , used a CBN insert ($100.00 bucks a pop) it didn't work that good, the customer had tried to do the work themselves but they faced from the outside in and when they got to the carbide layer they fractured it, so I set up the Dumore with a HF diamond wheel and will miracles never cease I ground off almost 1/8" carbide and steel at .001"/.0015" per pass, yeah I know don't grind steel with a diamond wheel well it worked and the wheel is still usable .

mars-red
03-15-2016, 11:31 PM
If you're only making one, and you don't need to reduce the small amount of error any further, then you don't need plates for anything as long as the handwheel on the dividing head has decent graduations. I've made oddball tooth counts that way before, it works fine as long as you calculate the correct absolute angular positions for each tooth, so that there is no compounding error.

Video Man
03-16-2016, 12:01 AM
My machining program says if you have a 90:1 rotary table and the common set of plates, one turn plus 22 holes
on the 38 plate would give 57 divisions exactly....

Optics Curmudgeon
03-16-2016, 12:55 AM
The two common standards for dividing heads are Brown & Sharpe and Cincinnati, both are 40:1 and use plate sets that include a 57 hole circle. If yours is one of those (or equivalent) you're all set.

ahidley
03-16-2016, 02:13 AM
If your making a 57 tooth gear you need 58 divisions to return back to zero with 57 cuts

Paul Alciatore
03-16-2016, 03:30 AM
Well, yes that is true. Although at first glance it looks like it would be, 57 is not a prime number. 3 x 19 = 57.

So the 38 hole plate provides the factor of 19 (2 x 19 = 38) and the 90:1 worm provides the factor of 3. If you have a 60:1 worm, it would also provide that factor of 3 so a 38 hole plate would work there too: you would use one turn plus two holes.

Unfortunately a 40:1 worm would not provide that factor of 3. Then the 57 hole plate would be needed. And 40:1 worms are the most common ones in indexing heads. So this is probably why you were told that you/your shop can not make a 57 tooth gear.

But I still say you CAN make one that will be just as accurate as the OEM plates that came with the indexing head.




My machining program says if you have a 90:1 rotary table and the common set of plates, one turn plus 22 holes
on the 38 plate would give 57 divisions exactly....

Forrest Addy
03-16-2016, 03:43 AM
Edited to repair an error pointed out below in Post 13:

General response for gear noobs confronted with gears of wierd tooth counts. Gears having tooth counts involving prime numbers pose problems when dividing head hole circles or index gearing are unavailable. It's a math thing where prime numbers are those having no factors other than themselves and 1. Here's a string of prime number examples 3, 5, 7, 11, 13, 17, 19, 23, 27, 31, 37, 41, 43, 47, 51, 53, 59, 61 ... Find a math book and there will likely be a table of prime numbers. You may have noticed that many of the smaller prime numbers are also preferred numbers used as thread pitches, 3, 5, 7, 11, 13 (Unified thread.)

Prime number gearing can be a real hassle for the gear cutter because they are hard to index for cutting and require some complicated work-arounds to achieve. The other is prime number gearing meshed in compounded trains produce long decimals that are particularly handy when calculating oddball ratios that have to be accurate to six significant figures for example.

Here's a start for those interested in prime numbers

https://www.youtube.com/results?search_query=prime+numbers

Personally I gave up dividing head hole circles for gadgets like the DivisionMaster

http://divisionmaster.co.uk/divisionmaster.html

and the Tiny Controls Programmable Stand Alone Indexing Table Controller

http://www.tinycontrols.com/standalone-indexing-table-controller.html

You can set either of these gadgets up to index ANY number of divisions - factor products or primes - to 10,000 or so. Cut the tooth space, traverse back, poke the button, watch the thing while it indexes to the next increment in a second to two, hit the feed and cut the next tooth space, repeat while trying o stay awake. They don't just index. They can be programmed to feed a rotary table to angles, power feed at variable RPM, etc. The tiny Controls indexer is about 1/4 the cost of the DivisionMaster and while it has more features it does not have an on-board (puny) stepper driver.

The OP problem is 57 teeth and how to index for it. As was pointed out in post 13, 3 and 19 are factors of 57. The next question is: can you index for 57 teeth using the 19 hole circle in a dividing head's hole circle plate? 38 any better?

Got an up to date DRO? Don't need no stinkin' CNC. These days of DRO's with bolt circle functions make drilling new hole circle plates duck soup. Make a few new plates and leave them un-drilled for later projects but drill one with 57 holes.

No DRO with bolt circle function but do have a good printer? CAD up a hole circle drilling template and print it out 1 for 1 scale. Stick it on a blank hole circle plate and drill the pattern by eye. Before you panic, remember the ratio is 40 to 1 and a 0.010 positioning error (sloppy work!!) on a 2" radius in any one hole translates to 0.010/40 on the same radius on the dividing head spindle or 0.00025". You can hear indexing errors on this scale in lightly loaded trains but they aren't enough to cause problems in low speed applications.

I like grinding a suitable sized two flute end mill with a drill point; a center drill and short, stiff hole circle drill in one package. Procede with project.

Richard P Wilson
03-16-2016, 03:55 AM
Well, yes that is true. Although at first glance it looks like it would be, 57 is not a prime number. 3 x 19 = 57.

So the 38 hole plate provides the factor of 19 (2 x 19 = 38) and the 90:1 worm provides the factor of 3. If you have a 60:1 worm, it would also provide that factor of 3 so a 38 hole plate would work there too: you would use one turn plus two holes.

Unfortunately a 40:1 worm would not provide that factor of 3. Then the 57 hole plate would be needed. And 40:1 worms are the most common ones in indexing heads. So this is probably why you were told that you/your shop can not make a 57 tooth gear.

But I still say you CAN make one that will be just as accurate as the OEM plates that came with the indexing head.

Can you tell us how? (This is a serious enquiry, I'm not trying to start an argument)

(Forrest, with respect, 57 isn't a prime number. As another poster pointed out, its divisible by 19, which is a prime number. Neither, for that matter, is 27 a prime number, its divisible by 9 and 3.)

Mcgyver
03-16-2016, 08:17 AM
Can you tell us how?


http://theoreticalmachinist.com/BoltCircleCalc

easy with cnc, tedious otherwise, but you basically drill a plate out using a set of bolt hole circle coordinates. There are other ways involvng turning a circle to a precise diameter that will fit 57 'whatever's of a know size

Stepside
03-16-2016, 08:44 AM
If it is 40 spaces on a 57 hole plate and it is a problem that comes up quite often, why not have someone with a CNC mill make a plate. Maybe even build a custom sector plate that is non-adjustable.

Forrest Addy
03-16-2016, 09:43 AM
Can you tell us how? (This is a serious enquiry, I'm not trying to start an argument)

(Forrest, with respect, 57 isn't a prime number. As another poster pointed out, its divisible by 19, which is a prime number. Neither, for that matter, is 27 a prime number, its divisible by 9 and 3.)

Arrgh!. Shoot me. My whole discussion is fatally flawed because of one error. But wait can you find a 19 tooth change gear? 19 is a common hole circle. Oh well, shoot me anyway but see the new material in Post 12 afterwards which I fixed. (Post 12.)

Richard P Wilson
03-16-2016, 10:32 AM
So shoot me. My whole discussion was fatally flawed because of one error? But wait 'til you find a 19 tooth change gear. Oops! 19 is a common hole circle. Oh well, shoot me anyway.

No, that certainly wasn't the intention. I just tried as gently as possible to point out that 57 and 27 were not prime numbers. No I haven't seen a 19T change gear, but in the UK we have 19 TPI pipe threads, just normally use 38T gears to generate them because they are a more convenient size. Wouldn't be difficult to do a 19T if needed though would it?

loose nut
03-16-2016, 10:39 AM
If you have a rotary table then you can download and use the excel spreadsheet at this link. Just input 57 (or an other number) in the blue box and it will calculate the Degs-mins setting you need for 57 the divisions. No muss, no fuss.

https://sites.google.com/site/bluewatermes/indexing-converter

Forrest Addy
03-16-2016, 11:04 AM
No, that certainly wasn't the intention. I just tried as gently as possible to point out that 57 and 27 were not prime numbers. No I haven't seen a 19T change gear, but in the UK we have 19 TPI pipe threads, just normally use 38T gears to generate them because they are a more convenient size. Wouldn't be difficult to do a 19T if needed though would it?

Prime numbers and British pipe. Know them both pretty well on the shop level.

No sweat. Richard. No offense taken; you were quite correct in pointing my error out. I'm supposed to be this all-wise guru but then I screw up the numbers and I'm back to aw-$hit. I gave myself a well deserved dope slap on that one. Excuse me, I gotta get back to rolling my rock uphill.

Loosenut: ever dial 57 divisions on a rotary table? Accurately? Like one arc minute? Takes incredible focus and keen eyes observing precautions for parallax error, consistent drag, phase of the moon, etc. It can be done but the perp is a wreck by the time the last tooth is cut..

RichR
03-16-2016, 11:20 AM
... the ratio is 40 to 1 and a 0.010 positioning error (sloppy work!!) on a 2" radius in any one hole translates to 0.010/40 on the same radius on the dividing head spindle or 0.00025".

I've seen it mentioned in other threads on this forum that more accurate plates can be generated from previous plates, but I never
really got it. That one concisely worded fragment made it click into place, thank you.

JCHannum
03-16-2016, 11:48 AM
I would not recommend using a rotary table to cut a 57 tooth gear, but I did use one (a low end Palmgren) to make the dividing plates for my Phil Duclos dividing head. In his write up, he described several methods of accomplishing this. Any of them will produce a plate of sufficient accuracy for the job at hand, that old 1/40 thing.

mars-red
03-16-2016, 11:51 AM
Loosenut: ever dial 57 divisions on a rotary table? Accurately? Like one arc minute? Takes incredible focus and keen eyes observing precautions for parallax error, consistent drag, phase of the moon, etc. It can be done but the perp is a wreck by the time the last tooth is cut..

It all depends on what constitutes "accurately". There are 21600 arc minutes in one revolution (360 degrees). As an example for a 2" diameter gear, one arc minute amounts to just under 3 ten thousandths of an inch of an arc length. A 40:1 ratio, apparently common in dividing heads, gives you 540 arc minutes per turn of the input handwheel. With 60 graduations along the handwheel, that makes 9 arc minutes per graduation. It's pretty easy to eyeball halfway between the graduations, making it pretty easy to get down to 4.5 arc minutes of resolution. For our 2" diameter gear that makes for less than 1.5 thousandths of an inch of error along the surface, and that's a non-compounding error as long as you have calculated your positions absolutely, and not relatively (meaning, there is also no cumulative error in your values you are indexing to). I don't do a lot of gear cutting, but that seems like it would be sufficiently accurate for many purposes.

I have an old shop made dividing head with a 120:1 ratio, and 90 generously spaced graduations on the handwheel, so each graduation is 2 arc minutes. At that point I'm sure the inaccuracies of the shop made dividing head components are giving more error than me having to use approximate positions on the handwheel.

Anyway, as others have stated several times, just use the dividing head to make an appropriate plate if the small amount of error from manual indexing is a problem.

Paul Alciatore
03-16-2016, 02:52 PM
Finally someone asked. I have posted this idea several times before but no one ever seems to pay attention. Forrest hit on the basic idea when he said just make a crude hole circle that is accurate to perhaps 0.010". What he said about the error becoming less when that circle is used is completely true.

How to make a hole circle with any desired number of holes on ANY indexing head or rotary table. And make it with just as much accuracy as your indexing head or rotary table is capable of? Here goes.

A worm gear will act as an accuracy amplifier. A 40:1 worm will reduce the error in your crude circle by a factor of 40 so an error of 0.010" in that crude circle will be reduced to 0.010" / 40 = 0.00025" in a SECOND copy of that circle. So you just have to make a crude one and use to make a second plate on your indexing device that will be at least 40 times more accurate. And the trick works a second time as well. A third generation plate with that number of holes will, in theory, have an error of only 6 millionths of an inch. Of course, neither your nor my equipment is capable of that level of work so you will be down to whatever level your equipment and skill dictate.

If you will use your shop made plate for indexing on the worm gear driven indexer or RT, then a second generation plate will be sufficiently accurate even if your original was far worse than the 0.010" error that Forrest suggested. And in fact, you can draw a circle freehand (it doesn't even need to be round) and divide it freehand into your desired number of holes and drill it in your drill press for your first circle. A second generation plate, made with that really crude, hand drawn one may not be quite good enough, but a third generation plate will be so accurate that you and I can NOT measure any angular problems in it. It would be down to the basic level of accuracy of you and your equipment.

So just make a second generation plate for use on an indexer or RT or make a third generation one if it will be used for direct indexing. If you use due care, that third generation plate WILL be just as accurate as ANY factory made one.

This technique does not need any CAD or CAM work. It does not even need any particular care in making the first generation plate. And it will work for ANY number of holes. ANY NUMBER WHAT SO EVER. Well, any whole number anyway.

I suspect that this is basically what was done to make the factory plates in the days before CAM was even a glint in a machinist's eye. Once they had an accurate worm gear, they were home free for dividing by any number of divisions.




Can you tell us how? (This is a serious enquiry, I'm not trying to start an argument)

(Forrest, with respect, 57 isn't a prime number. As another poster pointed out, its divisible by 19, which is a prime number. Neither, for that matter, is 27 a prime number, its divisible by 9 and 3.)

Baz
03-16-2016, 03:06 PM
You are only allowed to make a 57 tooth gear if it is for a tin opener for opening Heinz products :eek:
In Paul's earlier post he mentioned a problem with a 40:1 dividing head not providing a factor of 3 for a standard plate as opposed to the 90:1. However it might be worth noting that most plates are fixed with 3 screws. So you can index the plate round to 3 places effectively increasing the count of all you holes by 3. Yes incredibly tedious but it's there.
You can also make a bridge piece that has 2 pins to fit in adjacent holes and provides a hole in it's middle for the detent - doubling all hole counts.

Richard P Wilson
03-16-2016, 04:20 PM
Using Paul's suggestion of starting crude but after a couple of iterations, getting the right accuracy, how about this as a crude start.
Turn up a piece of decent ply or MDF to a diameter of 181.43mm. Then mark off round the edge at 10mm intervals, which should give you 57 intervals. Use that to drill the first plate, then make 2nd plate, should be somewhere near.

duckman
03-16-2016, 04:56 PM
I checked at work today and we only have 3 plates A B & C so it looks like I'll just have to make one , I printed the indexing converter for rotary tables , just need to make a blank.

John Stevenson
03-16-2016, 05:21 PM
The two common standards for dividing heads are Brown & Sharpe and Cincinnati, both are 40:1 and use plate sets that include a 57 hole circle. If yours is one of those (or equivalent) you're all set.

Partly true.
The B&S set of plates, 3 in total stop at 49 holes. This is the standard set copies by most of the other makers.

The Artful Bodger
03-16-2016, 06:42 PM
. However it might be worth noting that most plates are fixed with 3 screws. So you can index the plate round to 3 places effectively increasing the count of all you holes by 3. Yes incredibly tedious but it's there.

Surely not so tedious, cut first 19 teeth (leaving two spaces between each) then advance the plate one screw hole, cut another 19 teeth then advance again to cut the final 19 teeth.

smithdoor
03-16-2016, 06:44 PM
Use compound indexing simple been around for over 100 years
http://bbs.homeshopmachinist.net/attachment.php?attachmentid=715&d=1455468834
http://bbs.homeshopmachinist.net/attachment.php?attachmentid=713&d=1455148088

Dave

loose nut
03-16-2016, 07:50 PM
I would not recommend using a rotary table to cut a 57 tooth gear, but I did use one (a low end Palmgren) to make the dividing plates for my Phil Duclos dividing head. In his write up, he described several methods of accomplishing this. Any of them will produce a plate of sufficient accuracy for the job at hand, that old 1/40 thing.


It all depends on what constitutes "accurately". There are 21600 arc minutes in one revolution (360 degrees). As an example for a 2" diameter gear, one arc minute amounts to just under 3 ten thousandths of an inch of an arc length. A 40:1 ratio, apparently common in dividing heads, gives you 540 arc minutes per turn of the input handwheel. With 60 graduations along the handwheel, that makes 9 arc minutes per graduation. It's pretty easy to eyeball halfway between the graduations, making it pretty easy to get down to 4.5 arc minutes of resolution. For our 2" diameter gear that makes for less than 1.5 thousandths of an inch of error along the surface, and that's a non-compounding error as long as you have calculated your positions absolutely, and not relatively (meaning, there is also no cumulative error in your values you are indexing to). I don't do a lot of gear cutting, but that seems like it would be sufficiently accurate for many purposes.

I have an old shop made dividing head with a 120:1 ratio, and 90 generously spaced graduations on the handwheel, so each graduation is 2 arc minutes. At that point I'm sure the inaccuracies of the shop made dividing head components are giving more error than me having to use approximate positions on the handwheel.

Anyway, as others have stated several times, just use the dividing head to make an appropriate plate if the small amount of error from manual indexing is a problem.

On top of what Mars said how accurate do most guys get the cutter height. There are many places where errors can creep in. The center to center error is probably better then using an old dividing head with worn parts anyway. I have used this to machine gears and it works just fine for regular type gears. If you need high end ground gears, for NASA, then that's a different matter.

bob308
03-16-2016, 07:58 PM
do you have the old gear? is there enough of it to use as a indexing plate?

Paul Alciatore
03-17-2016, 01:28 AM
That will probably work, but it will be very confusing as well as tedious. You may need to make three or five gear blanks for every one you produce.

Edit: OK, I read your second post on this and I guess it would be less confusing than I first thought. But still a bit of a PITA to rotate the plate twice.




You are only allowed to make a 57 tooth gear if it is for a tin opener for opening Heinz products :eek:
In Paul's earlier post he mentioned a problem with a 40:1 dividing head not providing a factor of 3 for a standard plate as opposed to the 90:1. However it might be worth noting that most plates are fixed with 3 screws. So you can index the plate round to 3 places effectively increasing the count of all you holes by 3. Yes incredibly tedious but it's there.
You can also make a bridge piece that has 2 pins to fit in adjacent holes and provides a hole in it's middle for the detent - doubling all hole counts.

Paul Alciatore
03-17-2016, 01:31 AM
That will definitely work. Just remember you can use a second generation plate on an indexer or a RT, but if you use the plate for direct indexing, then you should take the trouble to make a third generation.




Using Paul's suggestion of starting crude but after a couple of iterations, getting the right accuracy, how about this as a crude start.
Turn up a piece of decent ply or MDF to a diameter of 181.43mm. Then mark off round the edge at 10mm intervals, which should give you 57 intervals. Use that to drill the first plate, then make 2nd plate, should be somewhere near.

oldtiffie
03-17-2016, 07:05 AM
Drill wander, both along and across the pitch circle may be more a problem then most other things so the selection of the chuck/collet as well as the drill bit its self may well need to be considered before the job is started rather than wonder(ing) about any post-job defect/s.

Why use the dividing/indexing plate/s at all when you can use Marv Klotz' excellent facility to generate the angular dividing head table motion directly using the calibrated "as supplied "Degrees:Minutes:Seconds" (aka D:M:S and d:m:s)" on the vernier-calibrated (as supplied) hand-wheel?.

Marv's utility really does make it so much easier.

Yep you can make a quite (very!!) accurate 127 tooth indexing plate and/or spur gear as well.

I am very satisfied with my several 6" "Asian" "Vertex" rotary tables.

But my real preference is for my "DivisionMaster" attachment which makes so many difficulties with rotary tables and methods redundant.

John Stevenson is truly the guru and subject matter expert here with regard to the "DivisionMaster" as he is very talented indeed.

Magicniner
03-17-2016, 07:50 AM
Use compound indexing simple been around for over 100 years

Dave

Those instructions are as clear as mud, obviously put together by, and of use to, someone who already knows how ;-)

- Nick

smithdoor
03-17-2016, 09:40 AM
It one the simplest ways of indexing odd numbers.
My father and my self had to find this out on a 51 tooth gear back in the 1974 Work great. FYI my father only new about this from a class he took back in 1940 and never used. I can send you a copy of the full page that may me clearer on how it works, just email me.

Dave


Those instructions are as clear as mud, obviously put together by, and of use to, someone who already knows how ;-)

- Nick

Richard P Wilson
03-17-2016, 10:05 AM
It one the simplest ways of indexing odd numbers.
My father and my self had to find this out on a 51 tooth gear back in the 1974 Work great. FYI my father only new about this from a class he took back in 1940 and never used. I can send you a copy of the full page that may me clearer on how it works, just email me.

Dave
Pardon my ignorance, but what is a Smith Index Pin? It looks interesting, but how do I find out more, and can it be done on a standard (Vertex) dividing head?

smithdoor
03-17-2016, 10:29 AM
Yes it works with all dividing heads
I think that the Vertex is a copy of the BS dividing head
The Smith Index is drawings for my Enco dividing head. I few have ask about the name simple had to name it some for a book.

Dave


Pardon my ignorance, but what is a Smith Index Pin? It looks interesting, but how do I find out more, and can it be done on a standard (Vertex) dividing head?

Richard P Wilson
03-17-2016, 10:44 AM
Yes it works with all dividing heads
I think that the Vertex is a copy of the BS dividing head
The Smith Index is drawings for my Enco dividing head. I few have ask about the name simple had to name it some for a book.

Dave

I'm still confused. It refers to a Smith Indexer pointer, an Smith Indexer needle and a Smith Indexer pin, while saying in underlined text 'do not pull out indexer needle' The only one of these I understand my Vertex (yes it is a copy of a BS head) to have is the indexer needle which engages with the holes in the plate. So what are Smith Indexer needles, pointers and pins?

smithdoor
03-17-2016, 10:49 AM
It is all made in the shop there are nothing buy but you indexing head
Just use the name smith for the parts

Dave


I'm still confused. It refers to a Smith Indexer pointer, an Smith Indexer needle and a Smith Indexer pin, while saying in underlined text 'do not pull out indexer needle' The only one of these I understand my Vertex (yes it is a copy of a BS head) to have is the indexer needle which engages with the holes in the plate. So what are Smith Indexer needles, pointers and pins?

lynnl
03-17-2016, 12:58 PM
But my real preference is for my "DivisionMaster" attachment which makes so many difficulties with rotary tables and methods redundant.


Why the hell would anybody ever want to make difficulties REDUNDANT?
Even one difficulty is one too many for most purposes.

Magicniner
03-17-2016, 01:25 PM
I few have ask about the name simple had to name it some for a book.

Come again?

Baz
03-17-2016, 01:34 PM
Pardon my ignorance, but what is a Smith Index Pin? It looks interesting, but how do I find out more, and can it be done on a standard (Vertex) dividing head?
There was a previous thread on compound indexing (http://http://bbs.homeshopmachinist.net/threads/69590-Compound-Indexer-FYI). Persevere, somewhere in the middle of it I provided a clear English explanation. It also appears Smithdoor invented some names for items for which he did not know the standard name and now freely uses those invented names which can cause confusion. Note that almost no dividing heads are equipped for compound indexing as professional machine shops just go with differential indexing as they need the other features a universal head provides.

Richard P Wilson
03-17-2016, 02:23 PM
OK, thanks for that Baz. I was familiar with the basic principle of compound indexing, having made a device many years ago which enabled me to easily index my lathe spindle by degrees. I got thoroughly confused by smithdoor's explanation, but now I've read your explanation, I understand the principles of what is needed to do compound dividing on a standard dividing head, although I have no idea at present how to achieve it.
Now, does anyone have a standard dividing head which they have modified to enable indexing on the back as well as the front of the plate, and if so, are they prepared to divulge details?

dp
03-17-2016, 02:41 PM
Wouldn't this be accomplished by drilling three holes around the perimeter of a plate, or gnawing three spaces in it with a gear cutter or other tool?

smithdoor
03-17-2016, 03:35 PM
This was list in the Machinery Hand Book as one way to divide a gear. The last book I have shown the compound indexing chart for BS heads. I do not know when stop making this type of dividing head but I have few made in to the 40's It still higher precession than CNC indexers today precession of 20 sec . This head can divide to 14 sec. There are few I have that could index down to lest than 2 sec. They use a 100 hole center indexing plate and outer one of 101 indexing plate that can divide down to less than 3 sec. Here is a photo off the internet
http://s95.photobucket.com/user/lca078/media/dividinghead.jpg.html?t=1165801868
http://s95.photobucket.com/user/lca078/media/dividinghead.jpg.html?t=1165801868
Dave


There was a previous thread on compound indexing (http://http://bbs.homeshopmachinist.net/threads/69590-Compound-Indexer-FYI). Persevere, somewhere in the middle of it I provided a clear English explanation. It also appears Smithdoor invented some names for items for which he did not know the standard name and now freely uses those invented names which can cause confusion. Note that almost no dividing heads are equipped for compound indexing as professional machine shops just go with differential indexing as they need the other features a universal head provides.

Paul Alciatore
03-17-2016, 03:49 PM
Just how you should "drill" the holes in a plate is a whole different subject. If you just use a jobbers length, HSS drill and start drilling, then you are probably destroying most of the accuracy that the process of making that plate is capable of.

For drilling holes for indexing, at a bare minimum, I would take several steps.

1. Use an appropriate spotting drill to accurately start the hole. I would ensure that the web width of this spotting drill was SMALLER than the web width of the drill that will follow it and that the tip angle is greater. This will ensure that the next drill will start on the same center.

2. The spotting drill would be used to make a dimple only. That means that the ends of the cutting edges would never touch the work. This ensures that the next drill will be located by it's center point at the center of the dimple, not by the edges of the hole.

3. A second drill would be a screw machine length drill, preferably a solid carbide one. It would be a few thousandths smaller than my final size.

4. Finally, I would ream the hole to the final size.

While making this plate I would have all movements on my mill locked down tight. I would loosen the locks on the RT or indexer for movement from one hole to the next and lock them before drilling. And yes, I know this means that to make one hole circle, you must go around it at least three times. Four if you use a countersink to clean up the edges: I would countersink before reaming to avoid a fifth clean up step.

There is a case for using a center drill for the first step. It will have a small center hole so the second drill can start on that. And, if the OD is big enough, it can provide the beveled edge so the countersink is not needed. But this creates a conflict with the OD being large enough and the small, center drill segment being smaller than the web of the second drill.




Drill wander, both along and across the pitch circle may be more a problem then most other things so the selection of the chuck/collet as well as the drill bit its self may well need to be considered before the job is started rather than wonder(ing) about any post-job defect/s.

...<snip>...

Richard P Wilson
03-17-2016, 06:27 PM
OK, it doesn't look like a simple adaptation of a Vertex/BS type dividing head.
Back to doing a rough plate and a couple of iterations to get an accurate one.

Forrest Addy
03-17-2016, 06:34 PM
Just how you should "drill" the holes in a plate is a whole different subject. If you just use a jobbers length, HSS drill and start drilling, then you are probably destroying most of the accuracy that the process of making that plate is capable of.

For drilling holes for indexing, at a bare minimum, I would take several steps.

You could free hand a drill point on a two flute endmill of suitable size and combine the center/spot drill, pre-drill. and ream operation in one swell foop using a single tool. If the endmilll is the right brand, the flutes flare a little in the washout to the radius leading to the body diameter. Control the depth to kick-out just right and you can champher the hole slightly eliminating the need to debur all those holes.

Work smart, use your tooling to save steps.

Mcgyver
03-17-2016, 07:49 PM
Just how you should "drill" the holes in a plate is a whole different subject. If you just use a jobbers length, HSS drill and start drilling, then you are probably destroying most of the accuracy that the process of making that plate is capable of.

For drilling holes for indexing, at a bare minimum, I would take several steps.

1. Use an appropriate spotting drill to accurately start the hole. I would ensure that the web width of this spotting drill was SMALLER than the web width of the drill that will follow it and that the tip angle is greater. This will ensure that the next drill will start on the same center.

2. The spotting drill would be used to make a dimple only. That means that the ends of the cutting edges would never touch the work. This ensures that the next drill will be located by it's center point at the center of the dimple, not by the edges of the hole.

3. A second drill would be a screw machine length drill, preferably a solid carbide one. It would be a few thousandths smaller than my final size.

4. Finally, I would ream the hole to the final size.

While making this plate I would have all movements on my mill locked down tight. I would loosen the locks on the RT or indexer for movement from one hole to the next and lock them before drilling. And yes, I know this means that to make one hole circle, you must go around it at least three times. Four if you use a countersink to clean up the edges: I would countersink before reaming to avoid a fifth clean up step.

There is a case for using a center drill for the first step. It will have a small center hole so the second drill can start on that. And, if the OD is big enough, it can provide the beveled edge so the countersink is not needed. But this creates a conflict with the OD being large enough and the small, center drill segment being smaller than the web of the second drill.

imo you guys are way over thinking it. spot drill, drill, ream. Ream in the drill press for a faster job. Because of the 40:1 ratio, if you're out as much as 4 thou on a hole location (very unlikely if using a spot drill) the division will be out .0001" ....not an amount worth considering in a milling op with a form cutter.

oldtiffie
03-17-2016, 09:15 PM
A lot if not all of the "drill wander" might be eliminated if the "drill" were a good (new) centre drill in a good collet and the pin/plunger end accurately concentric with the plunger and the plunger end turned to a very accurately concentric "ball/sphere" (or similar)

The plunger "ball" would locate very accurately with the plunder "vee" with no back-lash or clearances at all.

No need for more than just the centre drill - no other drills at all.

ahidley
03-18-2016, 04:51 PM
Geeze six pages of comments and nobody said "just buy a harbor freight 8" rotary table and your good to go!"
Several years back I posted on here and offeed a free chart, either excell or a pdf, which showed divisions 1-60 and which hole plate, that was included with the rotary table, to use.
Well now I'm in theshop looking at it and it says:
57 divisions--- 6.31579 degrees--- 1.2631579 turns on a 72:1 worm--- whole turns 1--- part turns 0.26316--- use division plate 19 set indexer for 5 holes.
Please note this table is made in India and it has a 72:1 worm.
I don't know why manufacturers don't include a chart. But if anybody wants a copy send me your email addy and I'll attach a copy.
Happy rotating

John Stevenson
03-18-2016, 05:42 PM
Will work on a 60 turn table at 1 turn and 1/19 on a 19 plate which is standard.
Will work on a 72 turn table at 1 turn and 5/19 on a 19 plate.
Will work on a 90 turn table at 1 turn and 11/19 on a 19 plate.


Just won't work on a 40 turn table without a 57 plate.

Forrest Addy
03-18-2016, 06:18 PM
....

Just won't work on a 40 turn table without a 57 plate.

And THAT's the inescapable fact.

Rich Carlstedt
03-18-2016, 07:42 PM
" I don't know why manufacturers don't include a chart."

They do, but probably not when it comes from China or India !

My "NEWS" ( Yuasa ) 12" 80:1 not only has the charts, but lots of extra plates that include the 57 capability.

( One turn + 23/57 )

Sorry, but the old adage is you get what you pay for...... applies to RT's

Rich

Rich Carlstedt
03-18-2016, 08:18 PM
My above comment is not to be a snide remark, or a put down, but to point out the realities of machining.
If you expect to hold tenths on a lathe, then get a Hardinge or 10 EE ........ BUT LISTEN UP
The caliber of a real machinist is to maximize the potential of his machine and equipment.
Having good equipment means it is easier ( and faster)
Having poorer equipment means more difficult solutions are required ... and that sometimes separates the men from the boys..
Note I said "sometimes" . Sometimes it's possible and sometimes, it just cannot be done (using regular procedures)

There are guys who come up with fantastic solutions to complex problems and I believe that is why most of us here, participate in this board.
To either help or discover ( for our selves) a way of solving a problem.

When an old timer says "Don't waste your time buying that lathe" ,it may not be a criticism , but reflect on what the poster is thinking of doing.
Us oldtimer's laugh ( sorry) when we hear or see a newbie wanting to have a lathe ( w/chucks) and expects to hold tenths for a purchase of a few hundred dollars. Or he buys a lathe and the cross-slide has .040 backlash and he thinks it should be zero !
The concept of owning machine tools has TWO facets that should be covered .
Many think of their tools like their cars and want them put back to factory condition ( hey, no harm there ) but journeyman machinists look at them as tools to accomplish the work at hand , and not a collectible. We do not damage the machine or destroy its components ( hammer machinist) but do expect to modify it for better performance. I have met guys who would not install DRO"S because it meant tapping into a factory casting .....wow.
So rotary tables fit into the equation, the cheaper ones are not as accurate, nor as complete as are more expensive ones.

Just some thoughts
Rich

John Stevenson
03-18-2016, 09:13 PM
I think Rich's above post sums it all up and is one of the better ones, if not the best one I have read so far this year.

Well done.

Forrest Addy
03-18-2016, 11:57 PM
Hear, hear, Mr Carlstedt. First lathe I ever ran less than 30 years old was the one I bought in 1971. Very good machine' cos the price of a new car but it still wasn't perfect and required a little tweaking. Some operations have to be managed in a certain way if I am to hold very tight tolerances. Same with every machine I ever ran. Sloppy machine tools are not an obstacle, merely a challenge. If I know of an error I can work around it, compensate. If a machine has error or other limitation, I have the full time I'm roughing off the stock to learn of it, its magnitude, how its distributed, how it manifests itself.

Kinda like Zen or the kind words from an ancient Shaolin monk to a promising novice. "Patience Grasshopper, the machine parts off work leaving the spindle end of the cut concave. What does that teach you?"

smithdoor
03-19-2016, 09:33 AM
Great
I did not look on my rotary table charts

Dave


Will work on a 60 turn table at 1 turn and 1/19 on a 19 plate which is standard.
Will work on a 72 turn table at 1 turn and 5/19 on a 19 plate.
Will work on a 90 turn table at 1 turn and 11/19 on a 19 plate.


Just won't work on a 40 turn table without a 57 plate.

gambler
03-19-2016, 09:36 AM
so much info. some of you guys need to write some articles.:)

Baz
03-19-2016, 09:44 AM
Why expect a mfr to provide charts? They are in lots of handbooks for 40:1 and if you can't work it out anyway you should be spending some time in an adult maths education class instead of faffing around deciding which rotary table to get. It only involves multiplication and division which is the most basic educational achievement. For the under 30's there is probably an 'app' if their little brains can work out how to turn the drill on.

smithdoor
03-19-2016, 01:47 PM
My self over years I found needing charts for
20, 40, 60, 72 and 90 tooth worm gears
I even and time owned one that had 120 tooth worm gear.
There may be app for dividing chart

Dave


Why expect a mfr to provide charts? They are in lots of handbooks for 40:1 and if you can't work it out anyway you should be spending some time in an adult maths education class instead of faffing around deciding which rotary table to get. It only involves multiplication and division which is the most basic educational achievement. For the under 30's there is probably an 'app' if their little brains can work out how to turn the drill on.

Paul Alciatore
03-19-2016, 10:15 PM
OK, the word "Smith" refers to some kind of shop made device that can be used with any indexing head.

BUT, EXACTLY WHAT IS THAT SHOP MADE DEVICE???????????????????? The above posts are less clear then mud. Photos? Drawings? ?????




It is all made in the shop there are nothing buy but you indexing head
Just use the name smith for the parts

Dave

Richard P Wilson
03-20-2016, 07:09 AM
OK, the word "Smith" refers to some kind of shop made device that can be used with any indexing head.

BUT, EXACTLY WHAT IS THAT SHOP MADE DEVICE???????????????????? The above posts are less clear then mud. Photos? Drawings? ?????

Paul
I tried to find out what this was earlier in the thread, but I don't think hes going to tell us, even if if it exists, which I doubt. And even if if he did tell us, would we understand the description?

JCHannum
03-20-2016, 08:24 AM
Not all of us are great communicators, that does not mean that our knowledge lacks value. As far as I can tell, Dave is describing a variation of compound indexing. It is covered in Machinery's Handbook and the description in there is enough to send you screaming from the room.

Basically, he is using two circles on the same plate and making two moves to achieve the desired number of teeth. It is a known method and has been in use for many years.

John Stevenson
03-20-2016, 09:12 AM
It's a little know fact that mental institutions have a high proportion of failed compound indexing devotee's

Mcgyver
03-20-2016, 10:12 AM
when are they going to bring more Jello?

Jimmer12
03-20-2016, 10:15 AM
Wire cut the gear and call it a day 😊

Baz
03-20-2016, 10:33 AM
Lets give Smithdoor some credit for bringing this up as it has almost no exposure on the machining forums so far.

Most dividing head plates are held by 3 screws (K&T by a ring which might make this easier) so you have to make a set-up that holds it on a bush that allows it (the plate) to rotate freely while also fitting the normal detent plunger mechanism.
Next devise another system for a detent to hold the plate by using another set of holes to stop it rotating.
A long flat spring with a pin as often used on watchmakers lathes for indexing the spindle would do but Smithdoor set a block onto the dividing head frame (top left in his diagrams) that overlaps the outer rim of the plate and allows a pin to engage through both but only to the outer holes.
An improvement would be to allow 2 plates to be fitted back to back to make more hole sets available.

My machinery's handbook has a formula that doesn't allow for primes - if it works at all I'm not sure. Does anyone have a copy with a table they could post up? I have not found any on the web so far.
The table given by Smithdoor earlier in the thread only seems to work for some of the values - those that have a 1 in the right most column "times a round" and only if that is taken as 0 not 1. I know I can calculate options but values would be nice for checking so I guess this is a time when I wish manufactures provided tables.

mklotz
03-20-2016, 10:57 AM
I did a little research on compound indexing since the concept was new to me.
Apparently, one leaves the hole plate loose on the shaft so it can be rotated.
A separate indexing pin, NOT the one on the crank handle, is used to keep the
plate from rotating. With the hole plate secured by this extra pin, the crank
is rotated some number of holes as would be done in normal, non-compound
indexing. Then, the extra pin is released and the hole plate is rotated in the
designated direction for the required number of holes. When this is done, the
crank, which is fixed to the hole plate by its indexing pin, also rotates.
Thus the net rotation of the worm is the (algebraic) sum of the two separate
rotations (crank and hole plate) and the workpiece is rotated 1/40 of that
amount (assuming a standard B&S gear ratio).

It should be patently obvious that the hole plates can't be swapped while this
tedious process is being done so all the rotations must be accomplished via the
holes that are available on the single plate that is mounted. For example, we
can't mix 1/15 steps (15 hole circle on plate #1) with 1/49 steps (49 hole
circle on plate #3).

Now, you'll note that, in the MH (Machinery's Handbook) table, all the hole
plate movements are specified as holes on either the 20 (plate #1), 33 (plate
#2) or 49 (plate #3) hole circles. IOW, they confine themselves to solutions
that only use the outer hole circle on the plate. I'm not certain, but I
think this is because it's mechanically difficult to mount the separate
indexing pin (the one that indexes the plate) so it can reach the other hole
circles. Rich Kuzmack has addressed a method for indexing on all the circles -
HSM Jan/Feb 1998 pg. 54.

Some of the solutions in the MH table specify a negative (i.e. opposite to
crank rotation direction) indexing of the hole plate. For instance, it
indicates that, for 87 divisions, one moves the crank 23 holes on the 29 hole
circle and then rotate the hole plate 11 holes on the 33 hole plate IN THE
OPPOSITE DIRECTION. Considering the effects of backlash, solutions that
require negative rotations are not to be preferred. For both simplicity and
accuracy, we would like to always be moving everything in the same direction.
Rich has produced a table with all positive crank plate movements in his 1-2/88
article in HSM.

To test my understanding, I wrote a computer program to find all the compound
indexing solutions with error less than some input quantity. Running it for
this problem, I found that there were acceptable solutions using any of the
three plates. The MH solution is the most accurate if one requires indexing
on the outer circle of holes. If that condition is relaxed, there is a
slightly more accurate solution.

Partial listing of program output:

Number of divisions = 127
Approximation accuracy = 0.005000 %
Required increment = 40/127 = 0.31496063

PLATE #1 (15,16,17,18,19,20 holes)

0 & 1/16 +1 * (15/17) = 0.94485294 [ 3] (-3.0637E-003 %) **
1 & 1/16 -1 * ( 2/17) = 0.94485294 [ 3] (-3.0637E-003 %) **
0 & 15/17 +1 * ( 1/16) = 0.94485294 [ 3] (-3.0637E-003 %) **
0 & 5/16 +1 * (18/19) = 1.25986842 [ 4] (+2.0559E-003 %) **
0 & 18/19 +1 * ( 5/16) = 1.25986842 [ 4] (+2.0559E-003 %) **
1 & 17/19 +1 * (10/16) = 2.51973684 [ 8] (+2.0559E-003 %) **
3 & 8/18 +1 * (13/20) = 4.09444444 [13] (-1.0684E-003 %) **
3 & 17/18 +1 * ( 3/20) = 4.09444444 [13] (-1.0684E-003 %) **
4 & 8/18 -1 * ( 7/20) = 4.09444444 [13] (-1.0684E-003 %) **
4 & 17/18 -1 * (17/20) = 4.09444444 [13] (-1.0684E-003 %) **
4 & 3/20 -1 * ( 1/18) = 4.09444444 [13] (-1.0684E-003 %) **
4 & 13/20 -1 * (10/18) = 4.09444444 [13] (-1.0684E-003 %) **
3 & 15/17 +1 * (16/19) = 4.72445820 [15] (+1.0320E-003 %) **
3 & 16/19 +1 * (15/17) = 4.72445820 [15] (+1.0320E-003 %) **
4 & 14/15 +1 * ( 8/19) = 5.35438596 [17] (+1.0320E-003 %) **
5 & 14/15 -1 * (11/19) = 5.35438596 [17] (+1.0320E-003 %) **
5 & 13/15 +1 * ( 2/17) = 5.98431373 [19] (+1.0320E-003 %) **
5 & 13/19 +1 * ( 6/20) = 5.98421053 [19] (-6.9252E-004 %) **
6 & 13/19 -1 * (14/20) = 5.98421053 [19] (-6.9252E-004 %) **
5 & 6/20 +1 * (13/19) = 5.98421053 [19] (-6.9252E-004 %) **
6 & 6/20 -1 * ( 6/19) = 5.98421053 [19] (-6.9252E-004 %) **

PLATE #2 (21,23,27,29,31,33 holes)

2 & 10/21 +1 * ( 1/23) = 2.51966874 [ 8] (-6.4700E-004 %) **
3 & 10/21 -1 * (22/23) = 2.51966874 [ 8] (-6.4700E-004 %) **
2 & 27/29 +1 * (28/33) = 3.77951933 [12] (-2.1769E-004 %) **
3 & 27/29 -1 * ( 5/33) = 3.77951933 [12] (-2.1769E-004 %) **
2 & 28/33 +1 * (27/29) = 3.77951933 [12] (-2.1769E-004 %) **
3 & 28/33 -1 * ( 2/29) = 3.77951933 [12] (-2.1769E-004 %) **
5 & 14/23 +1 * ( 2/33) = 5.66930171 [18] (+1.8299E-004 %) **
6 & 14/23 -1 * (31/33) = 5.66930171 [18] (+1.8299E-004 %) **
5 & 2/33 +1 * (14/23) = 5.66930171 [18] (+1.8299E-004 %) **
6 & 2/33 -1 * ( 9/23) = 5.66930171 [18] (+1.8299E-004 %) **
6 & 4/29 +1 * ( 5/31) = 6.29922136 [20] (+1.3904E-004 %) **
7 & 4/29 -1 * (26/31) = 6.29922136 [20] (+1.3904E-004 %) **
6 & 5/31 +1 * ( 4/29) = 6.29922136 [20] (+1.3904E-004 %) **
7 & 5/31 -1 * (25/29) = 6.29922136 [20] (+1.3904E-004 %) **

PLATE #3 (37,39,41,43,47,49)

The Machinery's Handbook recommendation:
2 & 23/39 +1 * (12/49) = 2.83464155 [ 9] (-1.4536E-004 %)

3 & 29/41 +1 * (33/47) = 4.40944473 [14] (-9.2668E-005 %) **

MH's preoccupation with the outer row of holes leads to downright errors in
their table. For 51, divisions, MH shows:

8 & 41/47 -1 * (12/49) = 8.62744247 [11] (-9.8686E-005 %)

Rich converts this to a more usable positive hole plate rotation in his HSM
article:

7 & 41/47 +1 * (37/49) = 8.62744247 [11] (-9.8686E-005 %)

but there are numerous ZERO error solutions on plate #1, e.g.,

1 & 5/15 +1 * ( 4/17) = 1.56862745 [ 2] (+0.0000E+000 %) **
1 & 4/17 +1 * ( 6/18) = 1.56862745 [ 2] (+0.0000E+000 %) **
2 & 10/15 +1 * ( 8/17) = 3.13725490 [ 4] (+0.0000E+000 %) **

Sorry for the extra long post but I wanted to put Rich's excellent work in
context and make everyone aware that even MH has to be used with forbearance.


smithdoor
03-20-2016, 12:32 PM
Just try cutting a HELICAL GEARS 157 tooth with a BS 2 dividing head
Ether you make a very large indexing plate or compound indexing ether works
On gear cutting machines it do with cams and gearing simple for large runs. Not so good for just one gear.
I have found in the past with BS and Cincinnati dividing head both use Compound indexing Helical gear cutting. Cincinnati later use very large indexing plates.

Dave



Not all of us are great communicators, that does not mean that our knowledge lacks value. As far as I can tell, Dave is describing a variation of compound indexing. It is covered in Machinery's Handbook and the description in there is enough to send you screaming from the room.

Basically, he is using two circles on the same plate and making two moves to achieve the desired number of teeth. It is a known method and has been in use for many years.

John Stevenson
03-20-2016, 01:44 PM
Just use the Cincinnati high number plate B and use 40 of the 157 tooth holes on the B plate.

TGTool
03-20-2016, 01:51 PM
Just use the Cincinnati high number plate B and use 40 of the 157 tooth holes on the B plate.

Ah, yes. Thanks, I hadn't thought of that strategy. :)

smithdoor
03-20-2016, 03:05 PM
Just look at fig 4-b this book is from 1966
http://vintagemachinery.org/pubs/detail.aspx?id=5381

Baz
03-20-2016, 04:47 PM
Thanks Marv
Managed to decode this "1 & 5/15 +1 * ( 4/17) = 1.56862745 [ 2] (+0.0000E+000 %) **"
1 full turn
5/15 of a turn using the 15 hole plate
+1* so in the same direction
4/17 or a turn by moving the plate using the 17 hole circle
gives 1.56... of a turn of the worm which is a bigger increment than expected so
[2] it cuts alternate teeth
which also sort of explains the 'times around ' column in Smithdoor's table.

Paul Alciatore
03-20-2016, 05:24 PM
Oh, so it is differential indexing? That is simply combining two fractions that add or subtract to the desired number in a mathematically precise manner. It is not quite correct to say that it can be done on any indexer. You need the second arm and pin to allow the plate to be moved on one circle while the normal arm is moved on another. As to how easy it is, I would certainly want a dry run if I was going to use it. You know, "Measure twice, cut once".

In indexing, there are many, many ways of achieving the desired result. By far the best solution in today's world is the use of an indexer or RT that is fitted with a stepping motor and the necessary software for dividing by any number of divisions. The only reservations that I have about that solution is the fact that I know that the original worm, plus the stepper motor with it's number of steps (usually 200), and the ratio of any gearing or other drive between them will still have a definite number of steps. And there WILL be some divisions that are not going to come out with an exact solution for that overall number of steps of the device. SO, the software will have to make some kind of rounding or adjustment when doing division by some numbers.

With software that I didn't write, I must wonder just how these rounding errors are handled. Are they rounded to the nearest step or are they just truncated to the whole number with any decimal, including 0.999999 being just dropped? And what do they do when they return to zero after a full revolution. If you need to make a second pass, do they really start at the same exact step or do they possibly miss or add a step or three? There would always be a small doubt in my mind about this. And do the manufacturers actually say what this number of total steps is? So I can know when the software is rounding and when it is not. I would want to know.




Not all of us are great communicators, that does not mean that our knowledge lacks value. As far as I can tell, Dave is describing a variation of compound indexing. It is covered in Machinery's Handbook and the description in there is enough to send you screaming from the room.

Basically, he is using two circles on the same plate and making two moves to achieve the desired number of teeth. It is a known method and has been in use for many years.

JCHannum
03-20-2016, 06:13 PM
It is not differential indexing, it is compound indexing, two different methods. It does require making a couple of parts to adapt the dividing head. It is an accepted method that has been in use for years. It is not a slam dunk, but it is certainly do-able.

mklotz
03-20-2016, 06:40 PM
Thanks Marv
Managed to decode this "1 & 5/15 +1 * ( 4/17) = 1.56862745 [ 2] (+0.0000E+000 %) **"
1 full turn
5/15 of a turn using the 15 hole plate
+1* so in the same direction
4/17 or a turn by moving the plate using the 17 hole circle
gives 1.56... of a turn of the worm which is a bigger increment than expected so
[2] it cuts alternate teeth
which also sort of explains the 'times around ' column in Smithdoor's table.

Yes, exactly, and the number in parentheses at the end is the error percentage. The perhaps overly concise notation allowed me to get a lot of solutions onto a relatively small file. When I did this work, many years ago, that seemed important. Today I'm not sure why I thought that. At any rate, I thought my treatise would help some folks to understand what this thread is on about.

Paul Alciatore
03-20-2016, 08:23 PM
OOps, sorry. I mix them up.




It is not differential indexing, it is compound indexing, two different methods. It does require making a couple of parts to adapt the dividing head. It is an accepted method that has been in use for years. It is not a slam dunk, but it is certainly do-able.

John Stevenson
03-20-2016, 08:26 PM
You can be excused Paul.
It has been a long day and a long post Is the paint dry yet ?

Toolguy
03-20-2016, 08:32 PM
It seems to me in this era, the simple thing would be to make an indexing plate with the desired number of holes using a CNC mill. This would be simple to make and simple to use. No counting of revolutions and parts (or losing count), each hole is the next tooth.

Richard P Wilson
03-21-2016, 04:18 AM
It is not differential indexing, it is compound indexing, two different methods. It does require making a couple of parts to adapt the dividing head. It is an accepted method that has been in use for years. It is not a slam dunk, but it is certainly do-able.

I understand the principles of this but we keep being told things like 'it needs a couple of parts making' and 'its do-able', but no one has yet shown us how in practical terms. If this works, and its as simple as we are told, it would be useful to folk like me who don't have access to CNC equipment.

JCHannum
03-21-2016, 10:23 AM
I mentioned Machinery's Handbook as a source of information on the method. It is a good place to start. I don't have the current version, but I do know a detailed description is included in earlier copies. My shop copy is the Eleventh Edition.

I really don't know much more than that. I have no need for a 57 tooth gear and if the need were to arise, there are other methods I would avail myself of.

Mcgyver
03-21-2016, 11:21 AM
I understand the principles of this but we keep being told things like 'it needs a couple of parts making' and 'its do-able', but no one has yet shown us how in practical terms. If this works, and its as simple as we are told, it would be useful to folk like me who don't have access to CNC equipment.

I've not done it, but it involves indexing the plate location (which is normally fixed) as well as the regular movement of the pin. The couple of parts are a retractable pin that allow the plate itself to be indexed.

this shows it

https://www.youtube.com/watch?v=S1zovk1KODs

Richard P Wilson
03-22-2016, 04:17 AM
Thats useful - at least gives me some ideas.

Forrest Addy
03-22-2016, 04:53 AM
Where's the ca-do spirit? Those on this forum who cavil at making a 57 hole plate if a clear need arises should leave his tools for the next guy and go into real estate, fashion design, politics or something where objective competence in a physical world is not a Darwinian attribute.

I can think of eleven ways to make an acceptable 57 hole indexing plate using basic home shop tools and/or a modest home computer.

There are among us a cadre of defeatist weenies. You know who you are and don't try to weasel out by ad-honinum counter-attack. Want to do better? Acknowledge your shame and reform your faint-hearted ways. This trade is not for weenies. A certain amount of enthusiasm, arousal to the challenge, curiosity, objective humility, etc is necessary besides the the technical skills and theoretical knowledge to make a good all around machinist. Don't get me wrong; it takes all kinds, There is a nice in this trade for nearly everybody except the timid, lazy, mendacious, and indifferent. Character DOES count. To paraphrase: skill without character is like a sow with a gold ring in her snout

Richard P Wilson
03-22-2016, 07:21 AM
[QUOTE=Forrest Addy;1039295]Where's the ca-do spirit? Those on this forum who cavil at making a 57 hole plate if a clear need arises should leave his tools for the next guy and go into real estate, fashion design, politics or something where objective competence in a physical world is not a Darwinian attribute.

I can think of eleven ways to make an acceptable 57 hole indexing plate using basic home shop tools and/or a modest home computer.

And these 11 ways are? Any of them don't need a home computer?

Jim2
03-22-2016, 09:18 AM
Personally, I favor the method that Baz mentioned earlier in the thread where you take your 19-hole plate and index it 3 times. My plates are held to the dividing head with a 3-hole bolt circle, so this would work for me w/o any mods. I haven't tried it, but I'm pretty sure it would work?

Jim

Carm
03-22-2016, 09:20 AM
And these 11 ways are? Any of them don't need a home computer?

Several have been mentioned.
You likely know a few off the top of your head that weren't.
Band saw teeth, Pi tapes ( or typical graduations) with stylus pick up.
Known diameters butted around calculated circumference, or the circumference evenly ticked off.
Radius bar w/pivot & shotpin holes with chord bar & drill bushing. Step sequence.
Nothing new under the sun but it could be in shadow.

Carm
03-22-2016, 09:21 AM
Personally, I favor the method that Baz mentioned earlier in the thread where you take your 19-hole plate and index it 3 times. My plates are held to the dividing head with a 3-hole bolt circle, so this would work for me w/o any mods. I haven't tried it, but I'm pretty sure it would work?

Jim

Post #24. Aye.

Carm
03-22-2016, 09:25 AM
Post #24. Aye.

Mmmm, wait a minuet. That depends.

smithdoor
03-22-2016, 10:22 AM
I agree on the other side is how to keep the cost down on making one replacement gear
1) It takes time to make a indexing plate you only use one time ever.
2) The differential methods also works the head you need larger mill and it weight is great now need a host too.
3) Find the a index with the 57 holes is like finding hens teeth and then if do find gold is cheaper to buy.
4) If you have gear cutting machine this also take time to set up for just one gear.
5) how about the new CNC indexers not cheap for just a very few gears you may make and not accurate as you may think it is

This NOT how I came by compound indexing my father new some we only had a head with the standard dividing place and no money to buy any and need money today. Back in 70's the CNC (NC) just fantasy. FYI we had steak after cashing the check. To this day I remember the number of teeth of that gear 53 teeth and did for half the price of any one else we later found out.
FYI my father was not a machinist he took a few class in 1939 just in case that skill was need. He was working a brick layer back in 1939. To this I still have one of his books for time he did join the Army back 1942.

Dave

differential dividing head weight
http://www.travers.com/precision-dividing-head/p/65-850-095/

Here a some more info
http://www.slideshare.net/JavariaChiragh/indexing-or-dividinghead-12005688


[QUOTE=Forrest Addy;1039295]Where's the ca-do spirit? Those on this forum who cavil at making a 57 hole plate if a clear need arises should leave his tools for the next guy and go into real estate, fashion design, politics or something where objective competence in a physical world is not a Darwinian attribute.

I can think of eleven ways to make an acceptable 57 hole indexing plate using basic home shop tools and/or a modest home computer.

And these 11 ways are? Any of them don't need a home computer?

Forrest Addy
03-22-2016, 11:19 AM
And these 11 ways are? Any of them don't need a home computer?

Non-computer ways:

Rectangular coordinates from a calculated table using a

height gage,

dial settings on a mill table

DRO if one is available

measuring rods as on a jig borer.

Radial layout via

stepping off chords on a scored circle Note: striding off 57 division by chord setting on dividers is chancy. Error naturally accumulates.

stride 24 divisions along the circle one way and 23 the other with the last space being the confirming chord.

divide the circle into 8 equal sectors to ensure the "straddles" - distance of adjacent holes either size of each radial are correct. The sector lines ensure spacing error does not accumulate as successive chords are laid down by each stride of the dividers.

That's seven ways to layout a 57 hole circle without a computer, four for layout prior to drill and three for direct positioning.

Remember earlier remarks how incremental error on the hole circle produces 1/40 the error on the work piece. 0.010 error results in 0.00025 error along a circumference on the work equal to the diameter of the plate's hole circle.

If one had an "optical" center punch

http://www.ebay.com/itm/PEC-On-Mark-Optical-Center-Punch-OP-270-SUPER-DEAL-NEW-in-BOX-/141915234756?hash=item210acea1c4:g:IuoAAOSwQiRUqF2 m there are far cheaper imports just as accurate)

CEP of the punch mark from the layout intersection would be 0.002".

I've never made a hole circle plate per se but back in the day I made a cooler tube sheet of copper nickel alloy in half a shift. Making an index plate with one 57 hole circle aint hard, just daunting. Machinists can made any machined part from metal and most rigid materials that are not metal. We are the go to people for making unobtainable parts. If you need something bad enough and its is not available then by default you have the time to make it

JCHannum
03-22-2016, 12:21 PM
You need nothing more than pencil, paper and a rule to layout equal divisions it is simple geometry.
http://lumberjocks.com/GnarlyErik/blog/33282

Toolguy
03-22-2016, 12:21 PM
It seems that it wouldn't be that hard to dial it off with a rotary table. The angle between holes is 6 degrees, 18 minutes, 56.84 seconds. One could round off to 6 degrees, 18 minutes, 57 seconds. This x 57 would end up at 360.0025 degrees.

Also, it may be possible on this board that someone could prepare a blank and send it to someone else who has a CNC mill to drill the 57 hole pattern and return it to the originator.

chucketn
03-22-2016, 12:31 PM
Does anyone really need a 57 tooth gear? Or hole plate? I thought this thread came about as a question why someone couldn't make a 57 hole plate or a 57 tooth gear... I have built an Arduino based Rotary Table controller and can easily set up for any # of divisions. I can also make a straight hobb, and cut gears, just need to know DP/Module and PA.

Chuck

smithdoor
03-22-2016, 12:50 PM
No
But if some one hands you a old gear and say please make today the dollars goes off in your head
Manufacture will make odd gear to stop some from make. This so the so Manufacture can charge a lot for a replacement.
I was in Manufacture for almost 40 years and I would do side jobs in repair and replacement of gears. It was good money.
How ever need one and needs today will pay a lot to have one today.

Dave



Does anyone really need a 57 tooth gear? Or hole plate? I thought this thread came about as a question why someone couldn't make a 57 hole plate or a 57 tooth gear... I have built an Arduino based Rotary Table controller and can easily set up for any # of divisions. I can also make a straight hobb, and cut gears, just need to know DP/Module and PA.

Chuck

Richard P Wilson
03-22-2016, 01:01 PM
Does anyone really need a 57 tooth gear? Or hole plate? I thought this thread came about as a question why someone couldn't make a 57 hole plate or a 57 tooth gear... I have built an Arduino based Rotary Table controller and can easily set up for any # of divisions. I can also make a straight hobb, and cut gears, just need to know DP/Module and PA.

Chuck

Maybe you are right, but I thought the OP (Duckman) actually did need to cut a 57T gear for a job at work, and thats why he asked the question. Like a lot of these long threads, the original reason for the question tends to get forgotten.

Carm
03-22-2016, 01:06 PM
(snip)It seems that it wouldn't be that hard to dial it off with a rotary table. The angle between holes is 6 degrees, 18 minutes, 56.84 seconds. One could round off to 6 degrees, 18 minutes, 57 seconds. This x 57 would end up at 360.0025 degrees.



"Wouldn't be that hard" is a relative thing. Even if time/money aren't in the equation and someone is methodical, there are a lot of steps there. The x57=360.0025 is a best case.

Carm
03-22-2016, 01:07 PM
Maybe you are right, but I thought the OP (Duckman) actually did need to cut a 57T gear for a job at work, and thats why he asked the question. Like a lot of these long threads, the original reason for the question tends to get forgotten.

But the digressions can be informative and sometimes entertaining.

John Stevenson
03-22-2016, 03:24 PM
Does anyone really need a 57 tooth gear? Or hole plate? I thought this thread came about as a question why someone couldn't make a 57 hole plate or a 57 tooth gear... I have built an Arduino based Rotary Table controller and can easily set up for any # of divisions. I can also make a straight hobb, and cut gears, just need to know DP/Module and PA.

Chuck

Pretty certain that 57 is one of the Colchester gears for doing DP or MOD, could be mistaken.

Working on the know fact that any plate will improve when a new one is made by an error of 40:1 I made a new one for a guy who wanted some spurious numbers out of plastic on the laser.

http://www.stevenson-engineers.co.uk/files/division%20plate%20plastic.jpg

Think it took about 3 - 4 minutes to cut. Main thing is that it got him going for a bit of scrap perspex and a stamp.

Mcgyver
03-22-2016, 04:55 PM
Maybe you are right, but I thought the OP (Duckman) actually did need to cut a 57T gear for a job at work, and thats why he asked the question. Like a lot of these long threads, the original reason for the question tends to get forgotten.

yup, that's the internet.....making it past the first page semi on topic is an accomplishment.