A New Dividing Plate for my Dividing Head

[JouniPParticipant @jounip]

[JouniPParticipant @jounip]

Hello

 

I have a Dividing head with 60:1 reduction worm and 3 plates with

13, 17, 22, 24, 45, 49

23, 29, 37, 43, 53

19, 28, 31, 41, 47, 59

holes.

 

The following divisions are possible: 2 through 60, 60 through 100 (except 63, 64, 81, 91, 99 and prime numbers) and 81 divisions between 100 and 360.

 

I need to cut wheel with 64 teeth, which I can not cut. No big deal because I can make an extra dividing plate with 16, 32 or 48 holes, and with that I can divide by 64.

 

I am too lazy to calculate is there any practical difference between 16, 32 or 48 holes and which one is best. Does someone have a opinion?

 

Also, when I make a new dividing plate, can You recommend any other number of holes to drill?

 

JP

[John Stevenson 1Participant @johnstevenson1]

Posted by JouniP on 08/07/2011 16:27:21:

 

I am too lazy to calculate is there any practical difference between 16, 32 or 48 holes and which one is best. Does someone have a opinion?

 

 

JP

 

 

 

I have an opinion but I’m also too lazy to post it

John S.

[JouniPParticipant @jounip]

English is not my mother language, as You perhaps has noticed. It’s not always easy to me find right expressions. Perhaps I may try again.

 

Is there any practical difference between 16, 32, or 48 holes in my problem. Of course I can make calculations myself, but that takes time. Perhaps someone already have sensible opinion?

 

JP

[John Stevenson 1Participant @johnstevenson1]

JP,

I did put a smiley at the end of my post.

 

As regards your numbers any would do but the higher the better for the odd chance that one large number will not break down lower.

 

As regards other number, and you may be lacking in space but 63, 127 and 25 come t mind.63 and 127 for metric translation gears and 25 because believe it or not a dividing head can’t do 125 divisions or 250 divisions without a 25 hole circle.

 

125 or 250 is very common on feed screws with 8 or 4 tpi again very common on older British machines.

 

John S.

[JouniPParticipant @jounip]

I noticed smiley and regret nuance of my post. Sorry.

 

127 (and 113) divisions would be useful but I can not find way to make new plate to make them. 25 holes is good advice.

 

 

 

JP

[alan frostParticipant @alanfrost17805]

An old trick I admit to never trying is saving your old broken bandsaw blades and turning a wooden,aluminium, or steel disc to a diameter accomodating exactly the no. of teeth required, and using the teeth with a lashed up detent as an indexer. Guy Lautards bedside reader ( number two or three ) has a more sophisticated and probably more accurate method using ball bearings (I think ) spaced round a suitably sized perimeter ). There must be better ways-such as using CNC but I’m too old for such passing fads.

 

Don’t see the need for them-for the benefit of regular readers the time is 00.12 (9th July ) and I’ve nearly half finished making my brush ring for an electric motor using a file. Started only just over a week ago-who needs CNC ?. Mind you I’m not rushing -stopped for a coffee yesterday.

 

PS If no-one comes up with a better way will look up Lautard method and post details. Rgds.

Edited By alan frost on 09/07/2011 00:25:23

[alan frostParticipant @alanfrost17805]

A ps. Don’t know how you value your time but M.E. Tools (no connection ) produce some reasonably priced dividing plates which I seem to remember have extra rings of holes with useful numbers such as 25 and 127. My memory is not what it was-perhaps someone else can confirm ?

[Richard ParsonsParticipant @richardparsons61721]

The answer to your problem can be found in an earlier thread. This thread is called “Cutting a 365 tooth gear”.>>

If you copy the title into ‘Search for keyword’ you will find it.>>

Good luck>>

Dick>>

[Anonymous]

There are several ways to achieve the required dividing plate. If we stick to the 16 holes needed for the job in hand, then the holes can be marked out by hand to a high degree of accuracy. How so? Any angle can be divided exactly in half using a compass and straight edge, no actual measurement needed. So, starting from a diamterical line just keep dividing by 2.

 

The method expounded by Lautard is a good one, either directly, or used to generate a dividing plate. I hadn’t heard about using ball bearings; in the edition I have he mentions using toolmakers buttons, but ball bearings would be cheaper and quicker.

 

Of course a CNC mill can be used, but it’s debatable as to whether it is quicker than using a manual method.

 

It would be possible to use a rotary table to generate the holes, each increment is 5.625°, so not too onerous.

 

Finally, any half decent DRO should have a bolt circle feature.

 

I’m in the process of making a dividing plate with 63 and 69 holes. While I have the facilities to use any of the above methods I’ll probably end up using the manual vertical mill and the DRO, as I think it will be quicker to set up than using the CNC mill.

 

Regards,

 

Andrew

[Anonymous]

Fired with enthusiasm I made my dividing plate yesterday afternoon. It’s in alumimium, as I had an end of bar offcut of the right size, and it will only be used a couple of times, 2 off 69 tooth gears and 2 off 63 tooth gears. The OD is 6-3/8″. All machining was done manually.

 

Regards,

 

Andrew

[alan frostParticipant @alanfrost17805]

Nice looking job , pretty quick too

[Sam StonesParticipant @samstones42903]

Andrew,

That’s very impressive.

When you told us you would probably use the DRO, was that X and Y?

If so, then what was your coordinate source.

Regards,

Sam

[John Stevenson 1Participant @johnstevenson1]

Sam, if it’s anything like my DRO you centre the work up and set for X and Y zero’s then punch in how many divisions and PCD.

 

It then spits out two co-ordinates that take you to the first hole, just move until both read zero and drill.

 

The arrow key they gives the co-ordinate out for the second hole, again move to all zero’s and drill.

So basically the co-ordinate source is in the DRO.

 

Rinse and repeat.

 

John S.

[Sam StonesParticipant @samstones42903]

What a wonderful gadget John,

 

Every home should have one.

 

But wait a minute, I’ve got exactly the same in my CAD package.

 

Sam

 

 

[John Stevenson 1Participant @johnstevenson1]

CAD package.

Dedicated computer program.

Rough and ready Excel spreadsheet.

 

Many ways to do the same job, none are wrong just that some methods are better suited to how you work.

 

I am a big believer is doing a job the way that suits you and your equipment, no good saying use this gadget when no one else has one.

 

I must admit though that when I come to use this bolt hole circle function on the DRO I have to get the book out for the exact method as I don’t use it enough for the little quirks on keystrokes to stick in what’s left of a brain.

 

I must admit though I use CNC for all the plates I make. Doing these commercially in the quantity I do makes this an ideal job and if I had to do them by hand it would be a total pain and not enjoyable.

A typical set of watch makers plates has 1,264 holes in them and I probably do 5 sets per day when they are running.

 

John S.

[Anonymous]

Thanks for the kind words. I am pleased with the way it turned out; doesn’t usually work that way for me!

 

The starting offcut was 7″ diameter and varied in thickness from 3/8″ to 1/2″. The final thickness was 5/16″. Sequence of operations was:

 

Mount the blank in the 4 jaw chuck roughly concentric and square. Machine the front face, outer diameter and drill and bore the central hole. Turn blank round in 4 jaw chuck, ensuring squareness to lathe axis. Face second side to required thickness. Mount the blank on parallels on the vertical mill; two clamps on the outside. Set the DRO so that 0,0 is the centre of the blank. We don’t need any great accuracy here (a couple of thou is fine) so use a piece of silver steel in the drill chuck, indicate off the central hole, using a cigarette paper, and use the 1/2 function on the DRO to give the centre in each axis. Use the bolt function to drill and then countersink the three mounting holes. Insert a clamp in the central hole and remove the two outer clamps. Use the bolt function on the DRO to drill the two outer circles of 69 and 63 holes. Chamfer holes by hand with a countersink; job done.

 

John S has the essence of the bolt function on the DRO. In my case you enter the centre (in this case 0,0 although it can be anywhere), the diameter of the circle, the number of holes, the starting angle (with 9 o’clock being zero) and then successive presses of the > button give you the location of each hole when the displays are zero’d. As Sam says you could get the X,Y co-ordinates of each hole from a CAD system. For dummies like me the bolt function is easier as you just mindlessly zero the displays for each hole, rather than have to set to a specific number.

 

I’ve also hogged out some riser blocks for my dividing head, so no excuses now, I need to get on and cut the gears.

 

Regards,

 

Andrew

[John Stevenson 1Participant @johnstevenson1]

Posted by Andrew Johnston on 11/07/2011 19:47:42:

 

 

the starting angle (with 9 o’clock being zero)

 

Regards,

 

Andrew

 

 

Mine must be on British summer time as zero is at 3 o’clock. <g>

 

John S.

[Anonymous]

Mine must be on British summer time as zero is at 3 o’clock.

 

John S.

 

Hmmm, didn’t think of that; certainly another solution. Since each successive hole increments anti-clockwise I rather assumed that the starting angle from 0 was also positive anti-clockwise.

 

Then again I am left-handed, and therefore sinister, so may be my mill is backwards, in true Dennis Wheatley fashion.

 

Regards,

 

Andrew

[Sam StonesParticipant @samstones42903]

In my CAD package (Keycreator), zero is also at 3 o’clock, and also counts ACW. This seems to gel with my memory of trigonometry.

The first gear wheel I made for the skeleton clock back in the 70’s was the Great Wheel. This brass wheel transfers the torque from the fusee upwards into the main gear train. At that time, I had a Myford ML7 and a home-made 40:1 dividing attachment with three dividing plates. DRO’s were a thing of the future.

John Steven’s drawing describes the 96 tooth wheel OD as 2 23/32″ (or about 69mm), and 7/32″ (5.5mm) wide. I turned the blank to size (as you would), and ground and hardened a piece of round silver steel, the end of which had the all important tooth profile. This was my fly cutter. In a Melbourne summer, there can be a lot of flies around.

I began cutting grooves between the teeth, gaining confidence tooth by tooth as the job progressed. At last I was cutting the one remaining groove, and decided as a final check to `drop’ into the first groove.

Whoops! What’s this ‘ere.

The first tooth disappeared before my very eyes, which should have been better trained for such serious work. Further investigations showed that I didn’t actually have the right number of holes in the dividing plate to `do’ 96 teeth, and had in fact chosen the wrong combination. My luck was `in’ however, because, of all the gears in the clock this, and its mating lantern pinion, were the only two gears where the tooth count didn’t really matter.

So, as with many traps for new players, the Great Wheel (and the Maintaining Wheel) ended up with only 94 teeth.

 

Regards to all,

 

Sam

[Anonymous]

Just to round off part of the thread I have now started cutting the gears using my new division plate. Two sets, each with one gear with 63 teeth and one with 69 teeth.

 

Here’s the division plate in position:

 

The larger of the two gears being cut:

 

And finally the two gears, together with the part machined differential centre:

 

For reference the larger gear is a shade under 12″ in diameter.

 

The more observant will note the green clamp in the picture of the gear cutting in progress. This is to stop the gear rotating during cutting. First time round I optimistically assumed that the dividing head spindle lock and tailstock would be sufficient, given that the cutting forces, in theory, are along the x axis. Sadly my optimism was misplaced, and on the 60th or so tooth the gear slipped by about 20 thou on the periphery. I’ve been round and recut the gear, and I’m sure that in practice it would work, but I’m not happy with it, so I’ve ordered a new casting. An expensive lesson learnt.

 

Regards,

 

Andrew

[Anthony RhodesParticipant @anthonyrhodes37318]

Posted by John Stevenson on 08/07/2011 19:33:47:

JP,

I did put a smiley at the end of my post.

 

As regards your numbers any would do but the higher the better for the odd chance that one large number will not break down lower.

 

As regards other number, and you may be lacking in space but 63, 127 and 25 come t mind.63 and 127 for metric translation gears and 25 because believe it or not a dividing head can’t do 125 divisions or 250 divisions without a 25 hole circle.

 

125 or 250 is very common on feed screws with 8 or 4 tpi again very common on older British machines.

 

John S.

 

John,

 

The numbers I want are:

127 inch-metric converion

113 pi

77 pi

71 pi

51 pi

50 pi and many others

 

Pi for DP and modular pitches.

71/113 = .62832 = .200000017 pi

77/25 x 51/50 = 3.1416 = 1.000002338 pi ~

 

63 would be OK too.

 

Anthony

[maurice (Mo) HardyParticipant @mauricemohardy58097]

I have been doing some meccano model work and considered making some odd gear sizes. I have a corel draw program and am using it to produce a dividing plate for a dividing head. It looks easy to use the corel draw tools to draw the plate and when I print it out it comes to the right measurements. I was thinking i could stick the printout to a metal plate and centre punch the holes using it as a template. any small errors would be further reduced by the 60-1 ratio on the headstock.

[Chris Pattison 1Participant @chrispattison1]

To have a plate with suitable hole pitches, I would start looking at having one made in MDF using a laser cutter. I would expect that these holes would be precisely spaced. This would then be used as a template to manufacture the plate in steel.

[John OlsenParticipant @johnolsen79199]

If you make a plate by any method that comes to hand, then use it with geared dividing head or rotary table to make another plate, the errors will be divided by the ratio of the worm. Practically this means that you could mark out the first plate by almost any means and the second plate would be pretty good. So ideas like printing out a CAD diagram and then spotting through would be fine.

I did make a conversion for my Vertex BS0 dividing head to do differential dividing, so I can do just about any number that takes my fancy. I was asked by a previous editor of ME to write up an article but then the editor changed and I think the article was binned. Along with about three others.

John

[Author]

Posts