Rotary Table Alignment

[Jmccrack]

Hey guys. All the conventional wisdom says to align a rotary table you put a dial or a coaxial indicator or a rod in the spinal and swept the center hole of the rotary table. Than rotate the spindle and 0,0 your table. OK..........After you do that put the dial on the rotary table top and sweep the spindle preferably the inside of the spindle. I would be interested in what you get. I have an 8” Taiwanese Vertex table. When I do what I suggested above i am .003” out. I suggest the best way to align a rotary table is to put the indicator on the table and sweep the spindle. Thoughts??

 

[rlo1]

I always thought that if you center the rotary table with with a center finder to the mill spindle you are good.

 

[Jmccrack]

So did I and maybe some of the more expensive ones are. But just for giggles put a magnetic dial on the Rotary table and run the indicator around your spindle or better yet inside your spindle,you might be surprised

 

[BaronJ]

If you centre your RT properly, why should I be surprised by dial indicator readings taken against the spindle bore !

 

[Jmccrack]

On some or a lot of tables you will find the center of the RT centered off the spindle will be different than if you center the RT off the table. Check it out

 

[BaronJ]

If the mill head is out of tram I can understand that there may be a difference.

 

[Jmccrack]

Hey guys check out this video from Joe Pi. I think most of us know him from his you tube videos Go to minute 12 and watch from there.

 

[BaronJ]

Whilst I agree that he shows in the video that there is a slight difference, he doesn't make any mention of why. There are two points that will cause a difference, the tram of the mill and the parallelism of the table surface to the bed.

 

[clockworkcheval]

A couple of years ago I purchased a Vertex HV-6 150 mm diameter Rotary Table. After some adverse results I checked the whole thing out. I found that the MT2 bore diameter at the top of the rotary table was concentric within 0,02 mm with the outer diameter of the rotary table - what is good enough for me. But the other end of the bore was 0,4 mm out if center, what is definitely not good enough for me. So the whole bore sits slanted in the RT. Mounting an ER25 collet arbor in the MT2 bore only increases the eccentricity, making the whole RT worthless as a carrier for cutting horological gears. The RT is now slumbering in a corner of my workshop waiting for me to muster the courage to fully rework the bore. I think I must do a straight re-bore and insert a push-fit cilinder with a nice concentric MT bore.
I also found a typo in one of the Dividing Mechanism tables. To divide the circle into 13 equal parts with Plate C on row 39 the handle should be turned 6 36/39. In 36 is a typo, which took care of another gear. I now always calculate the number of handle turns by hand.

 

[BaronJ]

Hi clockworkcheval

A couple of years ago I purchased a Vertex HV-6 150 mm diameter Rotary Table. After some adverse results I checked the whole thing out. I found that the MT2 bore diameter at the top of the rotary table was concentric within 0,02 mm with the outer diameter of the rotary table - what is good enough for me. But the other end of the bore was 0,4 mm out if center, what is definitely not good enough for me. So the whole bore sits slanted in the RT. Mounting an ER25 collet arbor in the MT2 bore only increases the eccentricity, making the whole RT worthless as a carrier for cutting horological gears. The RT is now slumbering in a corner of my workshop waiting for me to muster the courage to fully rework the bore. I think I must do a straight re-bore and insert a push-fit cilinder with a nice concentric MT bore.
I also found a typo in one of the Dividing Mechanism tables. To divide the circle into 13 equal parts with Plate C on row 39 the handle should be turned 6 36/39. In 36 is a typo, which took care of another gear. I now always calculate the number of handle turns by hand.

I also own a Vertex HV6-6 RT. I've not found a problem with mine, however I've not yet used the dividing plates, so I've not come across any errors in the dividing tables. This is actually the second one I've had, the first one came with a bent shaft and was replaced by the supplier.

I completely stripped mine down and cleaned it checking concentricity of the base and table as I went. I didn't find anything that worried me other than the taper in the centre was not deep enough to allow the Myford chuck arbour to seat properly.

A few turns with a MT2 reamer sorted that one quite easily. If your taper in the bore is out then machine reaming it using the mill should correct it, though you will have to take great care that you don't over do it and don’t make the taper too deep.

 

[fcheslop]

I had the same problem as clockworkcheval on the 4inch version and it cleaned up easily using a taper reamer
The dividing paper work also has the same error so always do the maths first just to check .It saves getting those gears with the half tooth not always guaranteed though
The paper work with the table is probably the same as the 6inch
For the money and the amount of use my little one has had its been well worth it. Iv cut around 30 gears using it and more than one duffer. I find the 4inch version easier to get onto the machine and set up than a 6inch I used to have
cheers

 

[Jmccrack]

Have you fellas dialed in both ways? From the spindle to the bore then from the table to the spindle? On the subject of dividing plates I never use mine anymore. I put together the Rotary Controller from “World of Ward” and have not looked back. Check out his web site. The unit is a snap to build and costs very little. I cut gears and splines and have never been stumped.

 

[petertha]

There are two points that will cause a difference, the tram of the mill and the parallelism of the table surface to the bed.

I agree. The spindle axis first has to be perpendicular to the RT axis before one can commence the RT test, meaning the mill head must first be trammed . If the spindle is pointed L/R or In/Out (or worse yet some amount of both) you will get different DTI readings purely from that discrepancy alone because the ball would be traversing around a 3D path on a cylinder. If the spindle is perpendicular but the centers are non-coincident (ie. displaced) then the DTI is now only traversing in a 2D plane. The DTI should pick this up & alignment achieved. Crude sketches hopefully illustrate.

Now where it gets interesting is if the DTI ball is traversing an inside taper. For example if the DTI is mounted to an arbor in the RT & table rotated & the DTI ball is traversing inside the tapered seat of the spindle, which is normally what we do. Again, assuming the spindle is trammed, I think you can still zero the axis together but the DTI deflection amount cannot be relied on because its being exaggerated by the taper. But you should be able to move the RT until the DTI settles to some stationary value indicating the axis are co-linear. We dont care about how many thou on either side, just that the needle isnt moving. Sorry harder to draw this scenario without getting out the CAD tools but hopefully it makes sense.

Now if you flipped the DTI test around so it was attached to arbor in the spindle & measuring the inside taper of RT, maybe the only difference between DTI-up and DTI-down tests is you are just picking up the relative angle of say an MT3 taper in the RT vs the R8 taper of the spindle? (which is meaningless). Otherwise I cant see that either test would make a difference if aligned.

 

[Cogsy]

There are two points that will cause a difference, the tram of the mill and the parallelism of the table surface to the bed.

I think there's a third (which I also think is what Joe was suggesting when he mentioned the RT bearings), the centre hole of the RT may not be 100% central to the axis of rotation. This should be fairly east to check, just by setting up an indicator in a fixed position to indicate the centre hole and rotating the RT through 1 revolution.

 

[tornitore45]

Is a matter of geometry. With the indicator in the spindle you find the center of the RT hole. The indicator placed on the RT revolves around the RT center of revolution which depends from the bearing concentricity and may or may not coincide with the hole center.

 

[BaronJ]

Hi Guys,

Petertha, Your post describes exactly my view. Thanks for the drawings, well illustrated.

Cogsy, You have picked up on the one thing I didn't mention ! Simply because if the table is turned and the bearing spindle is co-axial with the edge they should be in line. It didn't initially occur to me that a bit of swarf under the surface of the plate when turned would cause bearing misalignment.

Tornitore45, Agreed.

Thanks for all the comments.

 

[Richard Carlstedt]

First, The Table OD and ID must run true. 3 of my tables run true ( 2 -8" and a 12" ) However, the 6" unit was off, so I chucked it up (table only ! ) in the lathe with a DTI indicating the WAY surface, then I re-bored the center hole concentric with the ways. You could also bore the hole in the mill by rotating the table with a small cutter ( Boring head could make error).
Once you have a good center-bore , whether straight or taper (ie M.T.) , then just chuck a piece of stock in your collet and bring it down into the center-bore AND then fasten the RT to your mill table - Done !
No DTI needed and it is super fast. In other words, let your spindle C/L locate the RT C/L
Rich

 

[Bruce R.]

I never indicate my rotary table, I made a stepped steel dowel, one end fits tightly into the hole in the center of my rotary table and the other end fits the half inch collet in my mill. I mount the table loosely with the dowel in the collet, then bring the quill down and fudge the table around until the dowel seats in the center hole, then I tighten the table bolts. Once the bolts are tight, I’ll run the quill up and down to make sure nothing moved. I’ve been using this procedure for the past 30 years and never had any problems with turning out good parts.

 

[BaronJ]

Hi Bruce, Guys,

One of the things that I discovered was that with my mill (BF20L) it is very difficult to get an accurate tram of the head. I actually set the table up on the lathe faceplate and very lightly reamed the bore to square it up. The bearing and bearing surface have obviously been machined at one setting and are dead square to each other. Once the RT is on the mill its no problem to use the bore to set the table in line under the quill.

This is the device I use. It allows for a slight misalignment in tram.

 

[Bruce R.]

I don’t know what you mean by “tram”. But I’m gonna go ahead and assume that you mean the vertical axis of the head of your mill is off. Mine is adjustable in 2 directions and I use a 12” parallel and a holder for my indicator that is bent on 90 degrees in my spindle. I adjust the indicator on the holder so that it swings in a roughly 10” circle, with the parallel on the mill table at 12 o’clock and 6 o’clock I bring the indicator into contact with one end of the parallel, get a reading, and twist the spindle by hand to swing the indicator to the other end of the parallel. Adjust the mill head until you get the same reading on both ends of the parallel. Then turn your parallel to 9 o’clock and 3 o’clock and repeat the process. Sometimes you have to go back and forth between the 2 directions until it’s perfectly vertical, you also need to watch when you tighten the mill head fasteners as it could move and throw the reading off. An easy way to tell if the head is off is to use a flycutter, it should cut all the way around in a circle, if it only hits at one spot then the head is off kilter.