Run out on a rotary table

Advert

New Replies

Report Bug

Help/FAQs

Run out on a rotary table

This topic has 63 replies, 21 voices, and was last updated 21 June 2017 at 11:02 by larry Phelan.

Viewing 25 posts - 1 through 25 (of 64 total)

1 2 3 →

Author

Posts
25 March 2017 at 18:25 #290515
Iain Downs
Participant
@iaindowns78295
I bought a rotary table a year or so ago (the price seemed rather good so I jumped). I've not used it much, but a friend asked me to make some pentagonal rod, which seemed like a damned good excuse.

Although I've made something with 5 sides it didn't look even, and it seemed that the rod was off-centre.

I checked the run out on the table with an 8 mm test bar (taken from a scanner in an MT2 collect. I got around .14 run out over 50mm. The actual milling was done nearer 100mm which would put it between .25 and .3 out.

The table came from Warco and my question is, 'can I fix it'?

I've a badly photocopied exploded parts list which is a little hard to read, but I thought I would seek the practical advice before I took it apart and buggered it up further…

Iain
Advert

25 March 2017 at 18:25 #8629
Iain Downs
Participant
@iaindowns78295
How to fix it?
25 March 2017 at 18:34 #290517
Chris Evans 6
Participant
@chrisevans6
You could mill a register for setting up purposes in the top of the morse taper using a carbide cutter. It would not help with the sort of set up shown running off the morse bore though. Can you mount a chuck onto a plate to set up independent of the table bore ?
25 March 2017 at 19:03 #290527
SillyOldDuffer
Moderator
@sillyoldduffer
Just a thought, but is it possible that the rod is bending under tool pressure? The forces applied during milling are quite high. It might be worth trying a tailstock.

Dave
25 March 2017 at 19:19 #290528
KWIL
Participant
@kwil
I would check the runout on the morse taper insert first (how do I know?)
25 March 2017 at 19:20 #290529
MW
Participant
@mw27036
Posted by SillyOldDuffer on 25/03/2017 19:03:43:

Just a thought, but is it possible that the rod is bending under tool pressure? The forces applied during milling are quite high. It might be worth trying a tailstock.

Dave

Or even make a steel or aluminium fabrication that aligns both your rotary table and tailstock that you can quickly clamp to your milling table without the need for tedious setup time, all you have to do then is make sure the plate is relatively square to the head. It could save thee a lot of time.

Michael W
25 March 2017 at 19:23 #290530
MW
Participant
@mw27036
Just a thought that are we taking about how parallel the probe is to the bar, or the rotary runout of the bar on the table. At 100mm you're bound to need a tailstock to keep things steady.
25 March 2017 at 20:01 #290540
larry Phelan
Participant
@larryphelan54019
Maybe your table came from the same place as mine did ! Sounds like the run out is about the same.

I,m glad I came across your post,since I was considering buying a table from Warco,good price alright but good quality——–? I think we are both in the same boat. My hexagon was like your pentagon,right number of sides,but not too much in common. What did Warco have to say about it?

Why do we have to put up with such crap ?
25 March 2017 at 20:10 #290542
Iain Downs
Participant
@iaindowns78295
I'm measuring the runout with a dial indicator on a ground bar in a collet.

Thanks all

There's no stress on the test bar so there is definately run out.

I was initially going to mount my 4 inch 4 jaw on the table, bit I pretty much gave up on that. I'd have to make a plate to mount the chuck on and then mount the plate, shim it or something and still manage to get the 4 jaw centred.

The milling actually looks to have worked quite well with the pentagon being fairly parallel – just not centred or aligned to the bar, so I'm not overly concerned about the need for a tailstock.

I want to do some gear milling with the table and this is just a bit too imprecise for that, I think.

If I take the thing apart can i tune it?

Iain
25 March 2017 at 20:17 #290545
MW
Participant
@mw27036
Posted by larry Phelan on 25/03/2017 20:01:29:

Why do we have to put up with such crap ?

Because that's engineering for you.
25 March 2017 at 20:20 #290546
Michael Gilligan
Participant
@michaelgilligan61133
Posted by larry Phelan on 25/03/2017 20:01:29:

Why do we have to put up with such crap ?

.
1. To paraphrase Mr Stevenson [of this parish] … it's "fit for purse". JS has already mentioned that a proper "fit for purpose" RT might easily cost you £3,200
2. To [allegedly] quote John Ruskin: "There is hardly anything in the world that some man cannot make a little worse and sell a little cheaper, and the people who consider price only are this man's lawful prey. "
They are saying much the same thing.

Bear in mind that the 'intrinsic value of the materials' probably only represents a small percentage of the total value of a Rotary Table … most of the added-value is in the making/testing/certifying, and that is where corners are generally cut.

I'm sure you knew this already … but it's worth reminding ourselves occasionally.

MichaelG.

Edited By Michael Gilligan on 25/03/2017 20:22:42
25 March 2017 at 21:16 #290562
not done it yet
Participant
@notdoneityet
Sorry, but I see it as a real shortcoming of the methodology. Rigidity is not considered. Blame is on the rotary table but no consideration of the possibility this run out is due to MT, the extension to a 4 jawa self centering chuck (with its own run out) and an unsupported bar being milled at a further extension from the rotary table, and no consideration as to the concentricity of the workpiece.

Try again, is my advice, but do find where the imprecision occurs – before blaming the rotary table!
25 March 2017 at 22:54 #290574
SillyOldDuffer
Moderator
@sillyoldduffer
Posted by Iain Downs on 25/03/2017 20:10:45:.

…

There's no stress on the test bar so there is definately run out.

…

Iain

Hi Ian,

We're slightly at cross-purposes. I'm sure you've measured the run-out correctly.

What I and others are suggesting is that the run-out isn't what's causing your pentagon to go adrift. Try this experiment. With the DTI in position push the end of the rod down with your finger. I think you'll be able to move it a lot more than 0.14mm!

It doesn't help that your metal rod is held by a small chuck plugged into the rotary table's centre hole rather than being bolted to it. There's a lot of leverage in the set-up and it won't take much to move it. A milling cutter applies a lot of force to the work and it's important to hold the job firmly.

Dave
26 March 2017 at 11:17 #290614
larry Phelan
Participant
@larryphelan54019
Good morning all,

Just reading the latest post from that member with the bum rotary table. WELCOME TO THE CLUB !! I too had hopes of doing some gear cutting at some stage,but I hate to think what they might turn out like,since even the chart supplied with my table is full of errors.

You can get fairly good results by using the 4 jaw chuck,it,s just a bit slow to set up and you do need to make a mount for it,but that,s no big deal,even I could manage to make one,and that,s saying something ! I had already made one,a copy of the lathe spindle,just a simple disc with a threaded nose on it and it works quite well.

I agree it,s very disappointing to find that the table is of limited use,but it may be possible to correct it,I hope to have a go at mine soon,thanks to Neil. I saw the one being offered by Warco and was tempted,but I think I might just hold fire. I hope you get a better response than I did [at least they still sell them ]

What is the point of selling stuff like this? Perhaps if they want to sell through M E W,they should have to meet proper standards. The word gets around fairly fast in these circles and if professionals wont accept such junk,why should we? I know they will say that these things are made to a price,but if they dont work,the price is irrelevant. Just some food for thought. This is how junk boxes evolve !.

Edited By JasonB on 26/03/2017 14:55:05
27 March 2017 at 02:39 #290726
John Stevenson 1
Participant
@johnstevenson1
Buy cheap, buy twice.
27 March 2017 at 08:13 #290735
Alan Waddington 2
Participant
@alanwaddington2
Posted by John Stevenson on 27/03/2017 02:39:50:

Buy cheap, buy twice.

"Chinese made rotary table in runout shock horror"

not exactly a surprising headline to be fair……..
27 March 2017 at 10:08 #290754
KWIL
Participant
@kwil
The MT taper is an inserted piece, it presses out. That s where the error was on mine. Swapped the whole RT for another which was significantly better.
27 March 2017 at 18:21 #290834
Iain Downs
Participant
@iaindowns78295
OH dear.

I thought I'd replied earlier on some things, but clearly didn't click 'send'.

My main point was that any stress on the bar would be consistent. That is it would apply to each of the facets as it was milled, making the bar pentagonally symetrical, even if not necessarily of the dimensions I expected. By carefully measuring and creeping up on the correct dimensions, I should be able to get a good pentagon (I didn't, but that's part of the learning experience – it was a struggle to get my mike in there so I would take the piece out, check it and put it back. Then attempt to alight it ). Albeit possibly slightly tapered.

I wrote to Warco this morning and I will let you know what the response is. If any.

I'l also check the run out on the table. If it's just the taper that's dodgy, I could try mounting something on the table proper.

Thanks as always.

Iain
27 March 2017 at 20:48 #290850
SillyOldDuffer
Moderator
@sillyoldduffer
Posted by Iain Downs on 27/03/2017 18:21:53:

…

My main point was that any stress on the bar would be consistent. That is it would apply to each of the facets as it was milled, making the bar pentagonally symetrical, even if not necessarily of the dimensions I expected. By carefully measuring and creeping up on the correct dimensions, I should be able to get a good pentagon

…

Iain

Sorry to dissent Iain, but I don't think that the stress on that rod can be consistent. Cutting the flats weakens the rod and changes its shape. Therefore the stresses will be different each time you cut a new flat. And any bending will be worse than expected because there's plenty of leverage; that's why I suggested supporting it with a tailstock.

If the problem was being caused by run-out alone, then your "carefully measuring and creeping up on the correct dimensions" would work wouldn't it? I think that it doesn't because the rod moves differently under every cut.

In conclusion, as your method is on the right track and your Table isn't awful, it's more likely to be the rod.

I may be wrong. No improvement after firmly supporting the rod with a tailstock would put me straight back in my box!

Dave
27 March 2017 at 21:22 #290856
Iain Downs
Participant
@iaindowns78295
Happy to have dissent, Dave. It's how I learn!

To cut to the quick – I don't have a tailstock (apart from on my lathe), so that's out!

What I should be able to do (if I'm sort of right and you're sort of wrong) is to mill a perfectly symmetrical pentagon which is no on the axis of the rod. If it ends up with significant (more than .1mm) asymmetry, I must blame the pressure of the cutter.

This is one of those jobs you probably need to do twice. Once to make mistakes and once to get it right.

However, there IS run out on the morse taper regardless of what my tool may be pressurising and I need to do something about it before I try and cut a gear (which I would like to sometime, just to round out my education). A tailstock may be part of the solution for that.

Cheers

Iain
27 March 2017 at 21:44 #290866
Michael Gilligan
Participant
@michaelgilligan61133
Posted by SillyOldDuffer on 27/03/2017 20:48:03:

… Cutting the flats weakens the rod and changes its shape. Therefore the stresses will be different each time you cut a new flat. And any bending will be worse than expected because there's plenty of leverage; that's why I suggested supporting it with a tailstock.

[…]

I may be wrong. No improvement after firmly supporting the rod with a tailstock would put me straight back in my box!

.

Dave,

May I try using your logic to make an alternative suggestion ?

If the problem originates with variations in the bending of the bar, as it is being machined; then surely the resulting pentagon would also taper along its length.

Measuring the cross-section of the machined part should be a simple exercise, and be very informative.

MichaelG.
27 March 2017 at 22:02 #290870
SillyOldDuffer
Moderator
@sillyoldduffer
Iain and Michael,

I feel the urge to experiment tomorrow. I'll stick a bendy bar in my rotary table and see what happens when I cut it. I have a tailstock and can do supported and unsupported.

I agree with Michael's taper comment; it happens when I try to turn a long thin unsupported rod in my lathe. I would certainly expect to see similar in Iain's case. Do you have the wherewithal to measure what you have Iain?

Don't bet the farm on predicting tomorrow's results!

Dave
28 March 2017 at 07:30 #290880
Iain Downs
Participant
@iaindowns78295
Sadly, the day job will stop me doing any significant work on this until the weekend .

If I get a chance I will re-measure the rod along it's length. There's only 20mm cut so I'm not sure how definitive it will be and, as I said, I could have done the work better this first time.

Iain
28 March 2017 at 10:26 #290899
Neil Lickfold
Participant
@neillickfold44316
When using Rotary tables, the more rigid and shorter the moment the better. So getting a regular 4jaw and mounting it directly onto the rotary table will make a very big difference. You can also mount a 3 jaw onto the rotary table, but will be limited to the repeatability of that chuck for the next work piece. Generally 4 jaw chucks have a greater holding ability when it comes to milling. A smaller chuck for small items and a larger chuck for larger items. Quite often it is easier to hold the rotary table by clamping or drilling and tapping into the bottom face it it is not already done so,(Take care if it does not come tapped on the underside of the base) and then use the T nuts and some bolts or cap screws to secure it to the machine table. Sub plates used to be very common many years ago, but seem to be a forgotten item these days. They can also be used to expand the clamp area of the machine tool as well. Neil
28 March 2017 at 11:14 #290902
Iain Downs
Participant
@iaindowns78295
So. Since I have a long day working from home, I took a brief break and got my mike and the pentagon out.

The image below shows the results.

There is some evidence here for Daves assertion in that the end of the rod is, in general 30 odd microns (1.5 though in old money) wider than the inner point 13mm in from the end.

Curiously this is not consistent with a thinner end in two orientations.

None of this relates to the run out which would simply result in the pentagon being off centre with respect to the axis of the bar. That would take quite a bit more measuring which I'm not going to do in this case.

Having said that an error of 1.5 though is a pretty happy day for me, so I'm not complaining. However if I was cutting a gear where the gear depth was off by 1.5 though, but the axis was off by 5 or 10 thou (.25mm) I would be happy enough with the depth error, but less so with the runout error!

Tailstock from Warco is £32 at the moment.

Iain
Author

Posts