Gear cutting with a shaper?

[Andrew TinsleyParticipant @andrewtinsley63637]

[Andrew TinsleyParticipant @andrewtinsley63637]

I will soon have a shaping machine. A small version to replace my huge one which I gave away to a deserving cause.

Now I want to try my hand at gear cutting, I have the Myford kit required to do this on my ML7, when its protracted overhaul is complete.

Meanwhile I would like to try gear cutting on the shaper. I assume that I will need a profiled cutter to replicate the gaps between the teeth! Is this at all possible? Also what is the best method of indexing the blank? Not sure if I can somehow use my rotary table, although I don't know how one would do that!

I am curious to know if a shaper can be used for this task. I am sure that I have read about this, years ago, in an ancient engineering book.

Regards,

Andrew.

[Michael GilliganParticipant @michaelgilligan61133]

Andrew,

I don't want to open-up old wounds [particularly mine] … so I will just post this link instead of referring you back to a recent thread:

**LINK**

http://neme-s.org/Shaper%20Books/Michael_Moore/shaper%20gear%20cut.pdf

It is well-worth reading.

MichaelG.

Edited By Michael Gilligan on 11/02/2017 17:56:30

[Andrew TinsleyParticipant @andrewtinsley63637]

Hello Michael,

I will take a look at the link, somewhat bothered about stepping on sore toes, please accept my apologies for not using the search engine. I really ought to get to grips with the search facility, although I seem to be having a problem, as I always seem to get other topics mixed with what I am looking for.

Thanks again,

Andrew.

[vintagengineerParticipant @vintagengineer]

Yes no problem. The hardest bit is grinding the form tool. Using a rotary table or dividing head won't present any problems as long as you have a bit of clearance after the tool clears the job.

[Gray62Participant @gray62]

All depends on what size shaper you have as to whether you can easily mount a rotary table on it. A few years back I was demonstrating on the SMEE stand at Sandown MEX how to cut internal gears on a 8" Boxford shaper. In order to mount My BS0 dividing head, I had to put an extension onto the front of the table, this was done by bolting a heavy angle plate onto the front surface of the table, the dividing head was then mounted onto this. Cutters can be readily profiled out of the same HSS blanks used for lathe tools just as you would for single point cutting on a mill. I do mine on a Quorn using profiled grinding wheels and comparing the finished profile to an existing gear.

[Michael GilliganParticipant @michaelgilligan61133]

Posted by Andrew Tinsley on 11/02/2017 18:19:24:

Hello Michael,

I will take a look at the link, somewhat bothered about stepping on sore toes, please accept my apologies for not using the search engine.

.

No problem at all, Andrew

… It was some of the content of the previous thread that pained me, not your lack of awareness.

MichaelG.

[Phil PParticipant @philp]

Posted by vintagengineer on 11/02/2017 18:25:56:

Yes no problem. The hardest bit is grinding the form tool. Using a rotary table or dividing head won't present any problems as long as you have a bit of clearance after the tool clears the job.

You only need to grind a very simple tool for this process, ie one that has the form of a straight sided rack tooth, the only thing to worry about is the pressure angle required.

The teeth are generated as the blank rotates when the table moves sideways, the indexing is to move the blank round to the next tooth after each pass of the table, the dividing head is normally linked to the movement of the table with a band of very thin spring steel around a pulley of the same diameter as the PCD of the gear being cut.

The idea is that the blank rotates and rolls the tooth space past the cutter as it does so, thus generating the gap between teeth.

Phil

Edited By Phil P on 11/02/2017 19:48:34

[Mark RandParticipant @markrand96270]

I cut two 14 tooth gears for my Hardinge HLV rebuild using the 'Base Circle' technique in Michael's link. One was 14.5° PA full form and the other was 20°PA stub form. Just needed two different ACME type tools for the correct PA. I used 2 thou shim stock instead of the wire in the link. Other than that, the setup was very similar.

 

A not very good video is here:- **LINK**

I had used involute cutters for the other nine gears in the apron gearbox, but these two could not be done that way because there wasn't enough clearance for the cutter on one and the cutters didn't have the correct PA on the other. Hence the shaper job.

 

Once I'd built the jig to index and rotate the blank as the table moved sideways the work was simple, if somewhat monotonous.

Edited By Mark Rand on 11/02/2017 20:37:55

Edited By Mark Rand on 11/02/2017 20:39:27

[Michael GilliganParticipant @michaelgilligan61133]

Thanks for the demonstration video, Mark

Mine is an Adept No.2 hand-operated shaper, but I'm determined to give this method a go [when I can get back into the workshop].

MichaelG.

[HopperParticipant @hopper]

It might be simpler to grind your own involute-shaped cutter from a piece of HSS, using the the two discs method outlined in Ivan Law's book "Gears and Gearcutting", a part of the Workshop Practice series. You should have a copy of this book if you are getting into gear cutting anyways. No need to use his twin disc forumla to make turning tools and then turn up cutter blanks etc, just use the geometry to make a sheet metal template then grind and diamond file your HSS shaper tool to match the template.

[Andrew TinsleyParticipant @andrewtinsley63637]

Hello Michael,

Thanks for giving me the link. I am in some considerable pain at the moment and cannot get my head around the gear blank rotating as the shaper moves across the work. I cannot even see how to mount the blank on my rotary table, which I think is far too big for the job in hand.

I shall take some painkillers and print out the link, that should make it more understandable!!!

Many Thanks,

Andrew.

[Michael GilliganParticipant @michaelgilligan61133]

Andrew,

I have sent you a P.M.

MichaelG.

[richardandtracyParticipant @richardandtracy]

I must admit I am impressed with the simplicity of the idea used in that link.

However, I think it needs one thing pointed out. The diameter of the indexing slot wheel should be the same as the outer diameter of the gear wheel, so that the rack effect of the lateral translation is replicated at the gear wheel to get a perfect form. If there is a diameter error, it would be better to be as shown, with the diameter of the indexing slot being smaller than the wheel. The effect of the indexing wheel diameter error is to alter the effective pressure angle with the smaller index wheel increasing the effective pressure angle & vice versa.

Regards,

Richard.

[Ady1Participant @ady1]

I use a cross slide collet chuck from a turret/large lathe on an acorn/atlas 7"

Early days yet and my gears don't look great but they do mesh and it's a stable cutting platform

I use a changegear on the LHS and the work goes on the RHS

Edited By Ady1 on 13/02/2017 14:52:54

[Michael GilliganParticipant @michaelgilligan61133]

Posted by richardandtracy on 13/02/2017 14:17:05:

I must admit I am impressed with the simplicity of the idea used in that link.

However, I think it needs one thing pointed out. < etc. >

.

Richard,

I don't quite understand your point … could you please clarify ?

As I see it, the diameter of the indexing wheel is irrelevant; it is the diameter of the 'pitch circle disc' which matters, and the dimensioning of these is explained in the text. [*]

I should mention that there are some minor drawing errors in Fig.1 the isometric view … however the portrayal of the diameter of Disc 'C' is fairly arbitrary, since the diametral pitch and the tooth count are both unstated.

If I have misunderstood, then please put me right … because I intend this to be one of my first projects when I can get back into the workshop.

Thanks

MichaelG.

.

[*]

The pitch circle discs are7/16in. thick and can be turned from any material available—mild- steel, brass, aluminium, or even wood. The diameter is got from the formula. Dia. = pitch dia., of wheel to be cut — 0.048 in. (i.e., thickness of tension wire).

Edit: that quote is copied direct from the PDF, so I would just mention that the character preceding  0.048 in. is supposed to be a minus.

Edited By Michael Gilligan on 13/02/2017 15:07:17

[richardandtracyParticipant @richardandtracy]

Don't think so.

Take it to an absurd limit to consider the effect. Assume index wheel half the diameter of the actual wheel. Move the shaper table 1mm over, causing a 1 degree angle change of the index wheel. The gear wheel perimeter rotates by 1 degree because it's linked to the index wheel, but because it's twice the diameter of the index wheel, its perimeter actually moves 2mm (remember, though, the table has only moved the axis of rotation 1mm to one side while the tip of the gear tooth has moved 2mm), meaning much more is cut off the tooth than you would have expected if they were the same diameter – thus forming a much reduced tooth size.

If they are the same size, both the axis of rotation & tooth tip will have moved 1mm. It's not possible for the shape to be the same in both cases.

I'll see if I can draw it up.

Regards,

Richard.

 

 

Edited By richardandtracy on 13/02/2017 16:04:50

[Another JohnSParticipant @anotherjohns]

MichaelG:

One thing I mis-understand from the Base-Circle article (which I read a decade+ ago when I had a Shaper) is that the PCD actually runs through (I guess) 1/2 of the "moving wire" and the actual diameter of the pulley.

IIRC, the article states that the pulley should be of PCD; but does NOT take into account the centre of bending of the wire wrapped around the pulley.

You are a (very) bright guy, and have read up on this, so you'll be able to tell me if I'm correct or not in this memory of mine.

It does look like fun, but with CNC in my workshop now, any gear cutting is focused on that machine. Please tell us how it goes (and, I like Mark's video, thanks Mark).

Opinions on the PCD and wire thickness welcome.

John.

[richardandtracyParticipant @richardandtracy]

Sorry – run out of time this evening. Been doing it on SolidWorks and got to leave it behind now. Will get a chance to look at it again tomorrow morning. Will try to get something up by lunch, but by the look of it with 4 off steps at 5mm increments on a 60mm wheel with 29 degree cutter angle and assuming 60 dia & 30 dia index wheel, it produces two very different cut shapes. Need to increase the number of steps considerably before can really show the wheel properly.

Regards,

Richard.

[Michael GilliganParticipant @michaelgilligan61133]

 

Posted by John Alexander Stewart on 13/02/2017 16:22:47:

IIRC, the article states that the pulley should be of PCD; but does NOT take into account the centre of bending of the wire wrapped around the pulley.

…

Opinions on the PCD and wire thickness welcome.

John.

.

John,

My quote, earlier [15:00:58 today] was directly copied and pasted from the linked PDF of the article.

… and this, I think, is the missing piece in your puzzle [again a direct copy & paste]:

"Fastened to the pitch circle disc by a screw, is a length of 18 s.w.g. music wire which winds round the disc and is fastened by the ends through straining screws " G" to the adjustable ends " H " of the bracket " K " fastened to the sliding carriage of the machine. "

 

In my understanding, we therefore have a disc "C" of diameter = PCD of the desired gear, minus 48 thou.

[and, of course, 18swg has a diameter of 48 thou … so I think he had it sussed]

 

MichaelG.

[Glad to have finally roused some interest in this]

Edited By Michael Gilligan on 13/02/2017 19:17:35

[Mark RandParticipant @markrand96270]

Yes:- the diameter of the pulley or drum must be the gear's pitch circle diameter minus the thickness of the wire. This puts the neutral axis of the wire at exactly the pitch circle diameter. Any other diameter will cause the gap between the teeth to be too wide, as the gear will rotate too fast or too slowly and the resulting form will also not be a true involute.

The wire should also run exactly straight and parallel to the motion of the table, otherwise it'll either get loose or too tight as the table traverses.

On mine, since I used 2 thou shim stock instead of wire, the drum was 2 thou smaller then the PCD of the pinions.

Edited By Mark Rand on 13/02/2017 20:19:14

[Brian HParticipant @brianh50089]

I'm finding this very interesting as I need to make a wormwheel of 1" inch outside diameter but I need to check what TPI and therefor DP I can cut on my Boxford. The wheel then needs a 1/4" Acme thread in the centre but I can cheat on that as I'm waiting for some Acme nuts and threaded rod to arrive and will then silver solder a nut into the wormwheel.

Brian

[BazyleParticipant @bazyle]

I just tried an experiment wrapping some Bowden cable inner round my drill pillar 7 times measuring etc. I found that rather than half the cable width compensation it behaved as if the cable was 1.8mm not the 1.54mm measured. Probably a good idea to use thin shim as suggested above

[Michael GilliganParticipant @michaelgilligan61133]

Posted by Bazyle on 13/02/2017 21:57:08:

I just tried an experiment wrapping some Bowden cable inner round my drill pillar 7 times measuring etc. I found that rather than half the cable width compensation it behaved as if the cable was 1.8mm not the 1.54mm measured. Probably a good idea to use thin shim as suggested above

.

That's interesting, Bazyle … and I agree that shim should be the way to go.

I must just mention, though: If you mean flexible [multistrand] Bowden cable … I very much doubt if it is truly circular in cross-section; so perhaps your result is not so suprising [?]

MichaelG.

[Neil WyattModerator @neilwyatt]

Make a 20 tpi silver steel worm and matching tap and you will find that it will make a very nice fitting 60-tooth worm wheel on a 1" blank, if you gash it then free-hob.

Ditto 10 tpi and 2" blank.

I made two dividing heads this way not perfect accuracy, but have served me well.

Neil

[Author]

Posts