View Full Version : info on dividing head

01-16-2007, 02:36 PM
i just bought this dividing head it is a Ellis and the only number i could find on it is 44531. It also has a 6 jaw buck chuck on it i would like any info i can get plus where i can get plates

here is a link to pics


01-16-2007, 02:54 PM
Jason-- nice find. I have heard of Ellis, but that is about all I can say. Good luck finding plates. You may have to make them which has always sounded to me like a project that is an exercise in frustration. The whole premise is that they need to be precise, and yet you don't have a dividing head to do the indexing :-) I would think that if you had a DRO on a mill, you might be able to do it merely with x and y coordinates.

If you get frustrated with it, just sell it to me :-)


01-16-2007, 03:08 PM
Cool! a six jaw! Mine has a 6" adjust true. I also got only one plate with the head. His Lordship (our John Stevenson) from time to time sells new plates that would need to be finished to fit this head I believe and his has the 127 hole option plate as well I think. Look for Marypoppinsbag on ebay UK.

The documents are available on the web as PDF's and I have downloaded them. Google them up or I can send them to you if you prefer (once I find them on this PC at work).

The spindle taper is B&S 9 fyi. On mine I have to remake the brass arm that holds the pin for the holes. Other than that it seems in good shape and the runout was pretty small (how small I don't remember). I also didn't get the swivel base with mine or the tail stock but that could be made up. Another 'round tuit' to get!

01-16-2007, 03:16 PM
thanks for the replies i got everything in the link i gave you for i think cheap i also got some stuff for my south bend (a thread cutter and cross slide)and a coffee can of resharpened ball mills and key way cutters, a couple of v blocks,and measurment tools from starret not sure what they are used for but i got them and everything was a package deal.


01-16-2007, 03:17 PM
I found this just after my last posting from a Google search, but the bbs appeared to be unreachable for a bit:


pretty nice to find the manuals for something, for free, on the web.


01-16-2007, 03:26 PM
Actually, making the plates is not bad at all due to the gear train reduction of error. Just lay them out on a decent rotab and the gearing of the head will cut any error dramatically. If your really anal, you could then use the first plate to make a second far more accurate plate. :D

01-16-2007, 08:01 PM
Making the plates is easy if you have a rotary table or can borrow one . http://www.cp-tel.net/mary/mary-index.htm Go here and down load (calculates any number of Divisions on super spacer } The formula is easy to use and very,very accrate I have used it many times and on high dollar parts at work

Paul Alciatore
01-16-2007, 11:33 PM
I have to totally agree with Bad Dog. If you want to make your own, you can lay out a first try by hand or by a CAD generated pattern for all the hole circles you want. Drill them on the mill-drill or drill press by hand.

A head with 40:1 gearing will produce a circle that is 40 times as accurate or in other words, the error will be only 1/40 of the error in your hand made circle. Thus if your original is accurate to +/- 1/2 degree, the circle made from it with the head would be +/- 45 seconds. That's pretty good. If you want better, do it a second time with the circles you just made as the new pattern and you will be down to under +/- 1 1/8 seconds. I guarantee your head is not that good so the error from the head will be the limit at this stage. Probably 20 or 30 seconds. But you could just use the first one you made on the head and your work would be down to the accuracy of the head anyway. And it doesn't get any better than that.

Ian B
01-17-2007, 03:51 AM
The rotary table does indeed work fine - I just made the 125 tooth gearwheel using one. The link that Lane gives is good in that it divides the circle up into deg, min & secs. I did this with Excel, but with one difference;

Like most rotary tables, mine has a 90 tooth wheel. The handwheel is divided into 4 degrees (one turn) and subdivided in minutes. With the spreadsheet, I subtracted multiples of 4 degrees from the settings - so, if the 3rd tooth was at 18 deg 10 mins, the printout that I was following showed 2 deg, 10 mins - as this was the next reading to go for on the handwheel. Saved a lot of confusion.

If you use a rotab, always approach each setting from the same direction - minimise the effects of backlash on the worm & wheel.


01-17-2007, 10:05 AM
I guess I posted what I did under the presumption that he did not have a rotary table, but certainly they could be made.

The link I posted, however, seemed to have a note at the bottom that indicated that the site owner had some extra indexing plates and that he could be contacted about purchasing them....not a bad idea if they could be had cheap. I guess I believe in making tooling where it makes financial sense or where you can make something that can't be purchased (because it isn't made or you can make one more suitable).

Either way, good luck with it Jason. Let me know if I can help somehow (sorry, I don't have a rotary table yet :-)


01-17-2007, 08:33 PM
"The Shop Wisdom of Philip Duclos" (book from V.P.) has an article on building a small dividing head and he tells how to make the plates.