Rotary Table CNCing

So for Xmas I treated myself to a rotary table... I didnt want to spend a fortune, but I wanted one that was a bit above the usual nasty chinese and indian ones. So I went over to Chronos and had a look at what they had and decided on this new design Soba 4", a copy of the 4" Vertex. OK its not a Vertex, but is a cut above their usual quality. It has a MT2 central taper so can mount a chuck, or use a centre...

Attachment 1357

One reason I chose it is that its a good candidate (IMHO) for CNCing...

Removiing the handle, reveals a 10mm spindle

Attachment 1359

and removing the vernier collar reveals a usefull 15mm deep by 18mm dia place to clamp a stepper mount to.

Attachment 1360

The plan is simple. Machine a circular collar 14mm thick 34mm OD, 18mmID with two or three 4mm tapped holes for locking grubscrews, to fit over the bearing. Place a 10x18x5.5 thrust washer on the 10mm spindle, drill out a L050 coupling and lock to shaft while using some puller arrangement to pull shaft outwards and rotating table to push shaft out - this will preload the thrust bearing. A 67mm length of 40mm OD, 3mm wall tube forms the basis of the stepper mount. This will locate into a 40mm dia, 8mm deep bored hole in a 10mm piece of aluminium 56mm square drilled out for a NEMA23 motor and counterbored 38.1mm. the tube is retained in the plate by 2 or 4 grubscrews through the sides of the plate. The general idea is shown below:

Attachment 1361

Nice one Irving, problem I had was fitting a thrust bearing on the inside to take the trust in reverse so to speak. There is not a lot of room inside the unit but I managed to do it although I had to do some machining inside the block to make room for the thrust race.
Your diagram shows only one thrust race so OK in one direction only.

Peter

Use an Oldham coupling, Lovejoy couplings don't take misalignment into account.

I have done one of the 4" tilting units, the cheap and nasty ones just to see if a type17 motor will run it, yes it does but only for positioning, not enough power to overcome cutting forces if you want to machine say an arc.

.

Yes I pulled the worm out and internally there is a cast iron to cast iron bearing. It would be possible to fit a thrust bearing in there but the problem is finding a small enough OD diameter with an ID of 10mm. Possible a needle thrust if i could find something small enough, but the smallest is 24mm OD, or 18mm as a ball thrust washer. Another option could be an oiliite washer maybe?

It really needs to be <15mm OD i.e. about the same dia as the worm gear, I think, because of the offset of the spindle. The alternative would be to bore out the base of the cavity just above the wormwheel to a diameter of about 20mm to allow the eccentric to rotate the thrust washer into when disengaging the drive. Not impossible to do, but tricky, and means stripping the rest of the table down.

Attachment 1369
Attachment 1368
Attachment 1370

Quote:

---

Originally Posted by John S

Use an Oldham coupling, Lovejoy couplings don't take misalignment into account.

I have done one of the 4" tilting units, the cheap and nasty ones just to see if a type17 motor will run it, yes it does but only for positioning, not enough power to overcome cutting forces if you want to machine say an arc.

.

---

John,

Where is the misalignment? using my proposed tube mounting approach everything is concentric. And I have several LC couplings going spare.

I am using a type 23 1Nm unit which I have here, but I could get a 1.85Nm type 23 from Zapp if needed. I measured the torque needed to turn the table and the 1Nm motor should be able to give about 5rpm for positioning and 1/2 of that for cutting, assuming alloy only (20N cutting forces as per other discussions on torque). My main interest is not cutting arcs but engraving handwheels and cutting gears...

Quote:

---

Where is the misalignment? using my proposed tube mounting approach everything is concentric.

---

Sounds like famous last words to me:heehee:
Bet you end up using a flexible coupler of some sort :naughty:

Irving 1Nm is more than enough,I used a type 17 at 0.36Nm just to see if it works and it does.
I use 1.8Nm on the ones I do commercially but that's only because I get the best deal on this size of motor from China.

.

Irving,
You reminded me, I had to fit a sintered bush inside the offset bush as the thrust I used inside was a needle and so did not provide for any side thrust as ball units do. It is not easy to assemble due to the offset arrangement but at least it works as long as I dont try to reduce backlash to much.

Peter

Quote:

---

Originally Posted by BillTodd

Sounds like famous last words to me:heehee:
Bet you end up using a flexible coupler of some sort :naughty:

---

Maybe Bill, but i work on the KISS principle... I'll try it with what I've got to hand... and if and when...

Quote:

---

Originally Posted by ptjw7uk

Irving,
You reminded me, I had to fit a sintered bush inside the offset bush as the thrust I used inside was a needle and so did not provide for any side thrust as ball units do. It is not easy to assemble due to the offset arrangement but at least it works as long as I dont try to reduce backlash to much.

Peter

---

Peter, can you explain further. I was looking at sintered bushes with a view to taking a .375" ID .6" OD bush, opening out to 10mmID and turning it down to 15mm OD. It will have less friction than cast-iron to cast-iron. But as I said above, the torque requirement at the moment seems to be OK with a 1Nm motor so maybe I'll leave that till later if needed. The end-float I can set now with a 1.5thou feeler gauge and the motion feels smooth. The worm is a 2mm pitch approx so, if my calcs are right, this represents an additional 5 or 6 minutes of backlash on top of the 15minutes already present and this can be taken up by the software

As I said I managed to machine to the inside of the offset to accept a needle thrust bearing and then realised that there would be no support at that end so I inserted a sintered bush and reamed to size on the outside a normal ball thrust could be used as there is more room.
There were several momenst during the whole thing where I thought I had taken on more than I could do with my limited skill set. I also ended up with several bearing revisions as the whole thing was designed from bad idea to bad idea, if I had to do it again I would just buy a Arc euro(John S) finished one!

Peter

Never one to be phased by my incompetence or lack of skills (just a learning curve right?) if I decide to fit a thrust bearing at the lower end of the worm spindle I reckon there are two options:

1/ mill 5.5mm of the bearing housing and fit a 10x18x5.5 ball thrust bearing, then bore out the table body to 20mm dia at the right depth for about 6 - 7mm. Assembly would be put the worm in, drop the thrust bearing over the spindle, push to one side so that the offset bearing can be inserted over the spindle and into the table body. See pic...


Attachment 1372

2/ manufacture a new offset bearing, 5mm shorter than the current one and bored out 8mm not 10mm. turn down the worm shaft to 8mm and use 8x15x5 hrust bearings each end. Or maybe reuse the existing offset bearing but line with a 8mm ID, 10mm OD oilite bearing. This won't require boring out of the table body.

Irving, your option 1 sounds very similar to my escapades only unforseen problem I had was that with the inner bearing fitted the offset wouldnt(couldnt) turn sufficently to engage the drive. I cured it by using a carbide burr in my drill press to make enough clearance for the bearing.

Peter

How did you do that? I've added a drawing to show how I propose to do it..

Irv, why add the thrust bearing inside the housing?

Would it not be easier to add a pair of deep groove ball bearings in your motor mount to take the thrust?

Quote:

---

Or maybe reuse the existing offset bearing but line with a 8mm ID, 10mm OD oilite bearing. This won't require boring out of the table body.

---

A much better idea IMHO

The thrust bearings are for the worm drive so as to limit back lash. My reason to fit an internal one was that I wanted to rotate in both directions and so wanted no movement of the screw. If you are cutting gears I think you could get away only having the outside one as you could always go in the same direction.

peter

Quote:

---

Originally Posted by BillTodd

Irv, why add the thrust bearing inside the housing?

Would it not be easier to add a pair of deep groove ball bearings in your motor mount to take the thrust?

---

hmmm... thats an idea.. makes the mount more complex, but would be easier to machine... i'll have a think...

Ok, heres my thoughts.... the left hand diagram shows an approach using two 10x18x5.5 thurst bearings and a miniature 10x15x4 bearing. the top bearing cap screws on to contol endfloat. The right hand version shows the same idea using two 10x26x8 deep groove bearings to give both axial and radial support.

In both cases there is a central piece that acts as a shaft extender that goes over the existing shaft and is locked to it using a grubscrew or two. The only issue with this design is cutting the threads which are 25 x 1.5 (LH diag) or 27 x 1.5 (RH diag) for the end cap. I don't have a way to do that right now and the tap and die is £50!

Comments?

Hi Irving,
Just one comment, the inner thrust bearing is to take the thrust when the table rotates in reverse which was the reason I did it in the first place, otherwise in reverse the worm part will wear on the offset part and worm drives excert a lot of force and I foresore a lot of wear.

Peter

I agree Peter, but my thinking is by controlling the thrust externally the worm doesn't come into contact with the base of the offset part - its held off it by a few thou so that wear can't occur.

Well the machining of the stepper mount has taken a back seat for a while due to the weather - it being too cold to work out in the garage (and I've had a cold too). So I decided to buy a PIC programmer with the original intention of using Kwackers software and a 16F452 PIC but then I decided I wanted the stepper driver to be built in, not a seperate board and, hey ho, I like writing code. So I knocked up a quick breadboard using a 16F877 I had lying around and wrote a fledgling set of routines to read the keyboard, drive the display, create stepper pulses, etc. At the moment this is driving my L297/298 stepper board, but I have some IRF520 logic level MOSFETS and I plan to drive the stepper directly in Unipolar mode with a chopper current limiter. I have a ex-laptop 24v 65W (2A+) power supply that is perfect for the job. The PIC and display take less than 60mA so a straight 7805 regulator on a small heatsink will provide the 5v rail.

Attachment 1734

Quote:

---

Originally Posted by irving2008

I was looking at sintered bushes with a view to taking a .375" ID .6" OD bush, opening out to 10mmID and turning it down to 15mm OD.

---

Sintered bushes come oversize and go to size when you drive them home. I did try machining one once and it was distinctly odd, they are a bugger to chuck and smear rather than cut :eek: