Gear Cutting

[martyn nutlandParticipant @martynnutland79495]

Hello Everyone

Seasons greetings and best wishes for peace and health from rural France.

I have treated myself to a dividing head (not a Christmas present!) This may be a bit ambitious on my part as I'm not sure my skill levels will enable me to master it. But I'd like to.

As far as I can determine from images on the internet, it's a clone of a Brown and Sharpe O Model Semi-Universal. It's probably Chinese judging by the incomprehensibly garbled manual, although it doesn't state it's parentage either on it or the box. That said, and I'm loathe to admit this, it seems rather beautifully made.

The prime purpose of buying it, is I wanted to make Simms vernier couplings for the magneto drive on my Austin Seven projects and this is where I would. appreciate some help. I'd like to do other work in the future.

As I understand it, one buys sets of gear cutters each one of which is suitable for a gear with a specific range of teeth. Say, for the sake of the argument, any number between 36 and 48. The next cutter would then take you from 49 teeth to 60 and so on.

But how do you establlsh the angle between the teeth? I think I'm talking about diametrical pitch (?). There seems to be a value in degrees on the cutters – for example 20°. Is that it?

So would the way forward, along the 'Simms road', be to count the teeth on the gear (19) measure the angle between them (?) and choose a cutter that includes 19 in its range and is of the appropriate angle?

I have a Machinery's Handbook and basic instruments such as grinding gauge, protractor, combination set, calipers etcetera.

As always very many thanks in advance for any comments and advice with renewed good wishes.

Martyn

[martyn nutlandParticipant @martynnutland79495]

Creating a Simms Vernier Coupling

[not done it yetParticipant @notdoneityet]

The diametral pitch (DP) determines how fine or coarse the teeth will be. (Number of teeth on the gear plus two) divided by the diameter in inches equals the DP.

THE angle denoted on the gear cutters basically refers to the shape of the teeth on any gear of that pressure angle (PA). Most plder gears were 14 1/2 and later (more modern) ones 20 degrees.

The cutters generally come in a set of eight ranging from the smallest tooth count to a rack (infinite diameter🙂 ). You will need number three of the set of eight for a 19 tooth gear (I think).

[JasonBModerator @jasonb]

I don't think an off the shelf gear cutter will have the right profile for these items. You are likely to have to make up your own be it a single point tool or multiple tooth.  Looks like the tooth gap remains parallel and the teeth taper so you should be able to get away with one pass per cut.

Your 19T spacing will come from the division plates of the dividing head which will give the angular pacing of the teeth around the full circle.

Edited By JasonB on 26/12/2020 12:01:33

[Phil PParticipant @philp]

Martyn

The angle stated on the cutter is the tooth pressure angle and is more often than not either 20° or 14½°.

You will need to work out the cutter required from the finished diameter of your coupling and the number of teeth required on it, so it will be the DP (Diametral Pitch) number if using imperial cutters, or the M (Module) if if using metric.

Have you seen these by the way, it might save you some time.

https://www.vintagecarparts.co.uk/categories/vintage-car-parts-miscellaneous-vernier-coupling

Jason………..

Have a look at the above link, you can see that on the Simms outer couplings the gap is parallel and the tooth is tapered, The inner coupling that sits between them has parallel teeth.

Phil

Edited By Phil P on 26/12/2020 12:02:50

Jason……You just beat me to it with your correction 

Edited By Phil P on 26/12/2020 12:04:12

[Howard LewisParticipant @howardlewis46836]

Strictly speaking, a Simms vernier coupling should have round topped teeth, rather like a knuckle thread, but probably gear teeth will suffice, since there should be no relative movement, and the object is to provide a fine angular adjustment.

You will be constrained by the diameter of the original coupling. The number of teeth required on each part may determine the DP that you can use

The space available for the completed coupling will also be a factor in determining the DP

Dimensions of the existing components, if avaliable, should provide some guidance.

The finer the the DP the shallower will be the depth of each tooth, the more teeth will be cut onto the each part of the coupling. A 20 DP tooth will be 0.108" deep, whilst a 40 DP tooth will be 0.054" deep, leaving the space for the body of each part of the coupling.

Presumably, the Driving and Driven parts of the coupling will either have a keyway in the bore, or a tapered bore (Possibly a standard numerical taper?&nbsp.

The DP will help to choose the depth of the tooth, and so the overall length of the coupling.

Total length of coupling will be: (Overall thickness of Driving part + Overall thickness of Centre coupling + Overall length of Driven part ) This will be determined by the space between the Magneto and the the Timing Case

The counterbores on each of the three pieces will provide space for the nut retaining the Driving and Driven parts to their shafts

The finer the tooth pitch (Diametral Pitch) the finer the adjustment that you will have. and the shorter the length of the complete coupling,

SO, picking numbers out of the air, purely as an example..

The idea is to have one half with,say, 20T and the other with 19T. So that an advance of 18 degrees (1T on the 20T ) is counteracted by retarding the other half by 1T ( to give a retard of 18.947 degree, which effectively retards by 0.947 degrees, when the two parts re engage Or vice versa, to give a fine adjustment.

Since the gear teeth will not be rotating against each other, it would seem allowable to cut the teeth across the face of the components, rather like a flat bevel gear (if you can describe the parts such a thing )

Presumably the technique is to cut three sets of teeth. .In each case the teeth will have to be cut into a face that has been counterbored to a depth, at least, if not slightly deeper than the teeth.

The Driver will have 20T, The Centre coupling will have 20T on one side, and 19T on the other, so may well need to have one set cut whilst on a sacrificial arbor.

The Driven component will of the coupling will have 19T on the face.

It is possible that the complete assembly will,need to be checked for length against the space available, and the rear faces of Driver and / or Driven components be faced to reduce to the required length.

HTH

Howard

[Neil AParticipant @neila]

My understanding of the construction of the Simms Vernier Coupling is that it comprises of two serrated flanges, with a one serration difference, and a moulded hard rubber disc fitted between them. I assume that the rubber disc is to allow for slight misalignment between the two shafts, it may also cushion the drive depending on the hardness of the rubber. As has been stated, the serrations look similar to the teeth of a knuckle gear and seem to be cast rather than machined.

While machining the metal flanges and serrations, once the shape needed has been determined, would be reasonably straight forward, I think that you will still need to use a moulded rubber centre section to complete the coupling. Using metal for the centre section would result in fretting between the parts as there will be no way of lubricating them.

The link given by Phil P shows all the parts of the coupling, including the rubber disc. You may wish to purchase the rubber disc to get the profile you need. Remember, you could spend a lot of time making a part from scratch when you could purchase parts which only need the bore machining.

An interesting project, I hope it turns out well.

Neil

[Oily RagParticipant @oilyrag]

Martyn,

Simms couplings are coaxial couplings which take into account a 'slight' misalignment of the two components and also act as a torsional damper for intermittent loading, especially where an intermediate rubber 'sandwich filling' is used. The tooth profiles of these face couplings is not generally of an involute form – more akin to a spline form. The trick in the vernier coupling is the dissimilarity between the numbers of teeth which allows a finite adjustment between the two components to dial in a precise timing. This is achieved by the 'sandwich filling' having differential teeth. It is a more complex arrangement to an Oldham coupling.

For using the dividing head you will need to know the turn ratio and then look for a suitable plate. For instance a 40 turn ratio with a 21 hole plate gives 21 x 40 =840 then say divide by the number of teeth required, say 20 then 840/20 = 42 so you need to advance the plate by 42 holes = 2 turns exact. say you need 28 teeth spacing then 840/28 = 30 so you need to advance the plate by 30 holes, which equals 1 turn and (30 – 21 = 19) 19 holes.

Edited to correct a typo

I recommend Ivan Laws 'Gears and Gear Cutting' number 17 in the Workshop Practice Series for an introduction to all gear cutting techniques.

Edited By Oily Rag on 26/12/2020 20:46:28

[martyn nutlandParticipant @martynnutland79495]

Many thanks for all of that and for the encouragement.

As Lieutenant Columbo might say: 'There's still one thing bothering me'.

The tool supplier I use is RC Machines in Luxembourg who do indeed have eight piece cutter sets, just as we said. They are marked 20°, again as we said and then they have different number ranges to indicate the number of teeth each can cut. Understood.

Then there is a 'module' number. I now understand the 'module' to indicate the coarseness of the tooth; thus a low module should give a fine tooth, a high number a coarse one. However, the 'RC' cutters have numbers such as 0.5, 0.6, 0.8 etcetera.

So, following Jason B's most helpful advice of taking a Number 3 cutter from the eight piece set to achieve my Simms vernier gears, what module would I be looking for?

And, if I do other gears for different things in the future, do I need to keep buying additional eight piece sets of different modules (at 130-160 Euros a throw!)

Clarification would be, again, much appreciated.

Martyn

[Michael GilliganParticipant @michaelgilligan61133]

Posted by martyn nutland on 27/12/2020 08:31:05:

Many thanks for all of that and for the encouragement.

As Lieutenant Columbo might say: 'There's still one thing bothering me'.

[…]

Then there is a 'module' number. I now understand the 'module' to indicate the coarseness of the tooth; thus a low module should give a fine tooth, a high number a coarse one. However, the 'RC' cutters have numbers such as 0.5, 0.6, 0.8 etcetera.

[…]

.

Permit me to chip-in with one convenient benchmark : Module 0.8 is almost exactly a match for 32 DP

… with a little arithmetical gymnastics, the relationhip between M and DP becomes clear.

MichaelG.

[John HinkleyParticipant @johnhinkley26699]

Oh, well, if we're into chipping in ……..

The way I've read this thread – rightly or wrongly – is that there's a little confusion on the OP's part regarding the exprsssions Diametral Pitch and Module. Here's my 4d-worth of chips ( without salt or vinegar ).

I find this chart useful:

Note it specifically references spur gears, so whether it can be applied to the gears being sought here, I would caution against, their tooth form being anything but spur gear-like.

But I find it a useful aide memoire when cutting gears.

John

 

Edited to remove a little mud from the waters.

Edited By John Hinkley on 27/12/2020 09:52:37

[Michael Smith 15Participant @michaelsmith15]

All the Simms couplings that I have had have been lost wax castings and the teeth are left as cast ,It seems a lot of work to make one when they are readily available from several car restoration suppliers for about £30 each and £15 for the fabric coupling. Mike

[SillyOldDufferModerator @sillyoldduffer]

Never mind Lieutenant Columbo, I feel like a particularly dim Doctor Watson – at least two steps behind. With zero practical experience I may be over-thinking this!

Looking at the pictures provided by Phil's Vintage Car's link, I see the Simms Coupler is made like a Contrate gear, ie the teeth are at a right angle to the face, not around the edge.

Oily Rag explained how Simms Coupler's are adjusted, and here's words and music from 1922! Note the description refers to 'serrations', not gear teeth.

Neither of the pictures show involute teeth of the type produced by an expensive module cutter:

Gear cutters are used to make involute gears, with lots of fuss about dedendums, addendums, and the very particular curve needed to make 'n' teeth with the correct pressure angle:

Involute gears are designed to run against each other and are accurately made to a high specification to mesh without grinding. As the Simms coupler is a clever sort of permanently engaged dog clutch I don't think there's any reason to cut accurate involute teeth. I'm not even sure involutes cut face-on rather than edge-wise would mesh suitably – the blue involute gear above has broad shaped Y valleys, whereas the 1922 Simms Coupler tooth form is like a sine wave. (This may be less likely to tear the rubber.)

Serrations gripping via a rubber pad suggest the Rotary Table need not be set particularly accurately to cut 19 and 20 serrations on a pair of discs. 360/20 = 18° and 360/19 = 18.94°, which is 19° between friends for a total error of 1.14/360 degrees around the circle.

As the tooth form on the replica Simms spare part looks more like a simple U than the sine shape shown in the 1922 drawing it should be possible to make suitable serrations with a ball-end milling cutter, and to round off the tops with a file. It's not precision work. Otherwise a U-shaped form-tool.

If several couplers are needed, casting is probably the best approach. The moulds are a doddle to make with CAD/CAM and a 3D-printer.

My guess is involute gear cutters are unnecessary and possibly harmful for this job. Feel free to put me right if I'm talking rubbish again!

Dave

[Simon Williams 3Participant @simonwilliams3]

The gear elements of the coupler seem to be pretty much sussed, so turning our attention to the flexy disc…..

I've made something similar by casting it from silicon RTV (silicon sealer). I made an aluminium mould to constrain the goo from going walkabout, and coated the surfaces with 3 – in – 1 oil to stop it sticking. Then just added RTV silicon from a squirty cartridge and let it set for a couple of days.

There is an expensive non-corrosive RTV option; I used the simple builders gloop which stinks of acetic acid as it sets. Once it's solid the acid isn't a problem.

Season's Greetings all

Simon

[Neil WyattModerator @neilwyatt]

It might be possible to 3D print one in flexible filament.

Neil

[Dave HalfordParticipant @davehalford22513]

The best way to do this is make a single point cutter to match a new rubber joiner.

If the original ones are cast from monkey metal, modern alloy would last a lot longer.

A little slow typing (I blame the tea)

High modulus silicon (not bathroom) might survive as a joiner depending on the magneto load, if the engine suddenly runs like a dog you'll know why.

Edited By Dave Halford on 27/12/2020 11:37:54

Edited By Dave Halford on 27/12/2020 11:39:19

[HopperParticipant @hopper]

Brand new Simms Coupling rubbers for Austin 7s are readily available for purchase if you google it. Cheap too.

You would need to grind up a fly cutter to match the profile of the rubber "teeth" on both sides. They may be slightly different width between the 19 and 20 teeth fitted in the same diameter. . Unless they use the same width and reduce the gaps between them to suit

Then to get the "angle" between the teeth on the metal coupling, you would need to use the chart with your dividing head to first make one with 20 divisions, or teeth, and then one with 19.

You should get yourself Harold Hall's book "Dividing" to learn the basic concepts and terminology of using your dividing head and how the indexing plates etc all work. Can be confusing but simple when you see it explained..

Then get yourself Ivan Law's book "Gears and Gear Cutting" to learn the basic concepts and terminology of gear cutting for future projects that use standard involute gear cutters (which the Simms does not.).

Both books are part of the Workshop Practice series so readily available and cost only a few quid each. Best money you can spend right now.

Edited By Hopper on 27/12/2020 12:01:29

[old martParticipant @oldmart]

From the picture by S O D, the coupling would be easy to make with a rotary table and a suitable diameter ball end mill, and a rounding off cutter.

Edited By old mart on 27/12/2020 18:49:07

[Roger BestParticipant @rogerbest89007]

What a clever gismo, like a harmonic drive.

[old martParticipant @oldmart]

I can't quite get my head around the smallest ammount of adjustment achievable, could it be 0.947 of a degree?

I have just read Howards post, he had already answered my question.

Edited By old mart on 27/12/2020 19:52:00

[BazyleParticipant @bazyle]

If you get the rubber bit and examine it it might show the profile is a semicircle so you could make two at once by sandwiching two plates together and drilling radially.

[not done it yetParticipant @notdoneityet]

Posted by Bazyle on 27/12/2020 20:15:03:

If you get the rubber bit and examine it it might show the profile is a semicircle so you could make two at once by sandwiching two plates together and drilling radially.

OK if you have two cars or are making one for a friend as well. Or go into business on the internet?

[Simon Williams 3Participant @simonwilliams3]

Hedge your bets and carry a spare

[Anonymous]

Posted by old mart on 27/12/2020 18:47:37:

……the coupling would be easy to make with a rotary table and a suitable diameter ball end mill, and a rounding off cutter

+1

A dividing head isn't the best tool for the job as the work ideally needs to be horizontal. Talk of involute tooth forms and cutters is irrelevant. From the drawings/pictures of the part the tooth forms are nothing like involute. The only important parameter for the tooth form is that it matches the mating rubber coupling. The parts should be pretty simple to make; don't get bogged down thinking it's a gear, it isn't.

Andrew

[Howard LewisParticipant @howardlewis46836]

No one seems to have answered the question, obviously in the OP's mind, about the difference between DP and Module.

Effectively, they are the same, other than the units used in the calculation..

DP is for Imperial dimensioned gears. Module is for Metric dimensioned gears.

A 20 DP gear with 20T will have an OD of 1.100 inches. OD = (Tooth Count +2 ) / DP (20+2 ) / 20 = 1.100

A Module 1 gear with 20T will have an OD of 22 mm, purely by using mm instead of inches as the unit of measurement in the formula.

This formula, and many others are in Ivan law's book, "Gears and Gear Cutting", No 17 in the Workshop Practice series.

Now for the bad news,

If you buy complete sets of eight gear cutters; you will need a set for each module, or DP.

Each set should allow you cut gears from 12T to a rack;. No.1 will cut from 135T to a rack, while a No 8 will cut 12T anf 13T only.

The depth of cut (Usually marked on the gear. as Df ) will be the same for all the gears of a particular DP or Module, but will differ as the DP or Module changes. (2.16 mm for a Module 1, but increasing to 8.63 mm for a Module 4 gear, or 0.0108 inch for 20 DP gear to 0.360 inch for a 6 DP gear )

to find and compare pictures of 6DP gears, likely to be used on Traction Engines, with the 20DP changegears on a Myford ML7. It will be obvious that they are different. ) Similarly compare a 0.8 Module gear with a 2.5 Module gear

You should not mesh gears with different pressure angles, such as a 20 degree PA with one with 14.5 degrees .

The tooth form is slightly different

In the absence, (Not the case apparently ) of proper Simms couplings of the correct size, I advocated using gear cutters, across the face of the couplings, rather than the OD, to produce what Meccano called "Contrate" gears, as already mentioned..

HTH

Howard

[Author]

Posts