Dividing head advice

[Robert TretheweyParticipant @roberttrethewey]

It's been pointed out to me that there are at least two glaring errors in my latest thread 4in Dividing Head – the first being that the label rivited to the outter edge of the plate says 1 division = 1 min (not 1 mm) the second is that the two chamfered holes that lock the plate in place ate 6.3mm dia not 3.3mm dia. Sorry for any confusion. I've had one reply saying that "why cant I produce the plates myself" – no reason why if I knew the combination of holes in each plate.

Edited By Robert Trethewey on 14/04/2022 23:06:47

[Neil LickfoldParticipant @neillickfold44316]

Your dividing head is unusual in that it has the Deg and Mins on the plate. I have not seen that before. With a 1 min accuracy, you can produce the plates that you need, just have to plan what you are making and then create or get made the number of holes on that plate. No point having plates that are full of hole numbers that will never be used and just sit on the shelf. Was there any stamp marks on the dividing head or parts or plates?

Some plates are drilled from each side but are not through holes, and the other divisions are on a different pitch radius on each side of the plates.

Nice to know more about your dividing head. Was it a differential unit, or does the back shaft just turns the worm gear at 1:1

[Michael GilliganParticipant @michaelgilligan61133]

Posted by Neil Lickfold on 15/04/2022 01:53:51:

.

[…] just have to plan what you are making […]

.

I think that’s the essence of Robert’s problem

He has one plate out of a set of [presumably] three or more; and wishes to put the appropriate patterns on each of the replacements.

Knowing the outside diameter of a plate, and the diameter of the indexing holes; he can probably deduce the realistically feasible maximum hole-count … but there’s a long way to go after that.

The short-cut, of course, would be to identify the device [which is where we came in].

MichaelG.

Edited By Michael Gilligan on 15/04/2022 06:29:59

[DC31kParticipant @dc31k]

Posted by Michael Gilligan on 15/04/2022 06:20:07:

The short-cut, of course, would be to identify the device [which is where we came in].

The short cut, of course of course, is to use the brains with which we are blessed and work this information out for ourselves.

We are asked to ID a dividing head. The only pictures we have of it are in a fully disassembled state. We have no data on the indexing arrangement (i.e. the max. and min. hole PCD that it will accommodate) nor the indexing pin size. We do not do Mystic Meg here, we do engineering.

If identification is our aim, we need photos of the full piece. Supplementary information that also provides a good lead is the worm ratio (you do not need any plates at all to find that), the fasteners used on it (metric UN, Whit), the spindle fitting (e.g. a metric spindle nose thread would point us in a different direction to a Myford spindle nose thread).

John Hinkley provides reasoning for a 90:1 ratio. I tend to agree with him. So we go and look at something that has the same worm ratio (e.g. the HV6 given as an example). If we find any difference between the plates on that and the plate we have, we do not give up, we identify exactly what that difference is and see if it is an equivalence rather than a difference.

Below is the info. for an HV6. It has a 20 hole and an 18 hole. Everything you can do with these, you can do with the 40 and 36 on the plate above. Note that it comes with three plates with six hole circles in each, which may not be a suitable arrangement for the one above if you cannot fit six circles in the range of the indexing arm.

—

90:1 worm ratio

Hole circles=[15,16,17,18,18,19,20,21,23,27,29,31,33,37,39,41,43,47,49]
—
All divisions possible with above plates
[1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 45, 46, 47, 48, 49, 50, 51, 54, 55, 57, 58, 60, 62, 63, 65, 66, 69, 70, 72, 74, 75, 78, 80, 81, 82, 85, 86, 87, 90, 93, 94, 95, 96, 98, 99, 100, 102, 105, 108, 110, 111, 114, 115, 117, 120, 123, 126, 129, 130, 135, 138, 141, 144, 145, 147, 150, 153, 155, 160, 162, 165, 170, 171, 174, 180, 185, 186, 189, 190, 195, 198, 200, 205, 207, 210, 215, 222, 225, 230, 234, 235, 240, 243, 245, 246, 255, 258, 261, 270, 279, 282, 285, 288, 290, 294, 297, 300, 306, 310, 315, 324, 330, 333, 342, 345, 351, 360, 369, 370, 378, 387, 390, 405, 410, 414, 423, 430, 435, 441, 450, 465, 470, 480, 486, 490, 495, 510, 522, 540, 555, 558, 570, 585, 594, 600, 615, 630, 645, 666, 675, 690, 702, 705, 720, 735, 738, 765, 774, 810, 846, 855, 870, 882, 900, 930, 945, 990, 1035, 1110, 1170, 1215, 1230, 1290, 1305, 1350, 1395, 1410, 1440, 1470, 1485, 1530, 1620, 1665, 1710, 1755, 1800, 1845, 1890, 1935, 2070, 2115, 2205, 2430, 2610, 2790, 2970, 3330, 3510, 3690, 3870, 4230, 4410]
—
Missing divisions up to 100
[28, 44, 52, 53, 56, 59, 61, 64, 67, 68, 71, 73, 76, 77, 79, 83, 84, 88, 89, 91, 92, 97]
—
Hole circles needed to produce missing divisions
[14, 22, 26, 53, 28, 59, 61, 32, 67, 34, 71, 73, 38, 77, 79, 83, 14, 44, 89, 91, 46, 97]

[Michael GilliganParticipant @michaelgilligan61133]

Posted by DC31k on 15/04/2022 07:25:47:

Posted by Michael Gilligan on 15/04/2022 06:20:07:

The short-cut, of course, would be to identify the device [which is where we came in].

The short cut, of course of course, is to use the brains with which we are blessed and work this information out for ourselves.

[…]

.

I disagree completely

… but let’s not argue about it

We have simply interpreted the task differently.

I consider the ultimate objective to be completion of a set of plates, of which there is currently only one plate available.

MichaelG.

Edited By Michael Gilligan on 15/04/2022 07:42:08

[Michael GilliganParticipant @michaelgilligan61133]

Posted by DC31k on 15/04/2022 07:25:47:

.

John Hinkley provides reasoning for a 90:1 ratio. I tend to agree with him.

.

John’s reasoning seems incontrovertible

… except that there are only 240 divisions around the plate.

He does have the right answer though !

MichaelG.

 

Edited By Michael Gilligan on 15/04/2022 08:52:23

[SillyOldDufferModerator @sillyoldduffer]

Posted by Michael Gilligan on 15/04/2022 08:26:40:

Posted by DC31k on 15/04/2022 07:25:47:

.

John Hinkley provides reasoning for a 90:1 ratio. I tend to agree with him.

.

John’s reasoning seems incontrovertible

… except that there are only 240 divisions around the plate.

He does have the right answer though !

MichaelG.

Edited By Michael Gilligan on 15/04/2022 08:52:23

Except I jumped to the conclusion it was 60:1! Given my impressive track record of mistooks, the first thing I'd do is confirm the ratio on the actual device by counting the number of handle turns needed to rotate the head once. Always good to nail facts rather than join dots!

I fear it won't be possible to identify the make and model of Robert's dividing head or to find spares for it. Many firms could have made it, some of them with tiny production runs.

As making and using division plates is a pain in the rear end, especially if the number of holes have to be calculated and drilled from first principles, I repeat my suggestion of driving the head with an Arduino and stepper motor instead. Doesn't require the operator to look anything up in a table or keep count and the head can be moved to any angle within the accuracy of the stepper and worm, not needing the differential gear fitted to posh dividing heads to produce angles ordinary hole plates can't do.

Dave

[John HinkleyParticipant @johnhinkley26699]

Michael,

You're quite right, of course. My observation of 360 divisions is wrong. What I should have said was that there are 240 divisions – 4 sections of 60 divisions each, with supplementary engravings at the 1, 2 and 3 degree points. That is 4 degrees per handle revolution. 4 degrees x 90 handle turns = 360 degrees. QED as we used to say.

That'll teach me not to "show my workings"! Maybe.

John

[Howard LewisParticipant @howardlewis46836]

IF the ratio IS 90:1, having one Division Plate, by reference to the HV6 chart, another plate could be made carrying some of the hole circles missing from the existing plate. (Ideally, using the iterative system of using a "mule" plate to reduce the errors on a second)

In this way successive plates could be made to extend the number of divisions available.

The three plates supplied for the HV6 do not cover all the possibilities between 1 and 100.

One day, I'll get round to making up one or more plates to cover some of the blanks. But even then, there will still be some,

Howard

[Michael GilliganParticipant @michaelgilligan61133]

Now … as we’re all trying to do some ‘reverse engineering’ can we please agree that the worm:wheel reduction ratio is actually 1:40

.

Cropped from a photo in Robert’s Album, in which the single-start worm is also visible

MichaelG.

[Author]

Posts