UKC

I have a CNC cutting apparatus which has axis X,Y,Z. This is machining a part on a rotary axis Q with one degree of freedom. (Think milling machine strapped onto the side of a lathe)

Now if the part is not centred on Q then there is an offset and depending on what the orientation of Q is the offset is in either the Z or Y axis. (ie. it wobbles when rotated).

Is there a proper term for describing this? I'm looking for a more suitable term than "the part holder is wonky, leading to an offset in the z-direction for some values of Q, a y offset for some other values of Q and a little of both for all other values of Q" which is a mouthful and doesn't sound very scientific!

Many thanks

(A chemist attempting to do CNC machining!)

In reply to Reach>Talent:

Speaking as neither: Eccentric chucking/workholding?

More importantly, what are you making?
jk

In reply to Reach>Talent:
A part holding arrangement is always wonky - your problem is that it is more wonky than you are prepared to accept. You should expect to specify a tolerance for the centration of your part - you may find the wiki page on tolerancing useful? http://en.wikipedia.org/wiki/Geometric_dimensioning_and_tolerancing

Is the part holder itself a machined part, or is it a chuck built in to the lathe part of the tool? If you have had it machined then you need to specify tolerancing on the centration and radius of the chuck component etc. If it's a supplied chuck, you need to investigate why the part is not properly centred in the chuck - is it the quality of machining of the blank you are using?

Are you using the Q axis as a lathe - i.e. high RPMs, or as a low speed sample positioner? If it's the later, perhaps you can use some plastic shims (e.g. see RS) to bodge it?

Can you just re-machine the blank in a lathe mode so that it is properly centrated?

In reply to Reach>Talent:

Hmm, what a question to pose on here! you may well be better to try posting on http://www.mycncuk.com/ There are a couple of chaps on there who really do know everything, and if there's something they don't know then it's not worth knowing. One chap you should hope to reply to a thread is called JAZZCNC, an extremely well respected member of the community.

In reply to Reach>Talent:
I'm not an engineer or machinist, but how about; 'Instability of the part holder Q means that tolerances in the Z and Y axes cannot be specified accurately or achieved reliably'?

Maybe that's too simplistic...

In reply to Reach>Talent:

If the part is not centered, can you not use a four jaw chuck and dial it up? As others have said... much easier to get concentric than try and compensate in the machining code for this.

If it is round and not concentric, then (depending on the reason) you either need to remachine the part or get it in a collet chuck (or four jaw) if you need more concentricity from the chuck.

In reply to Reach>Talent:

I'm a turner to trade (unfortunetely)but i haven't been CNC machining for 7 years,Is it a rotary milling tool on a turrent sitting at 90 degrees to the part located in the chuck ?like you get on the Mazak quickturns ?or is it a vertical m/c centre like a Mori Seiki ? When you say orientation that usually means clockwise or anti in machining terms is that what you mean ? What is gripping the part...a chuck

In reply to Reach>Talent:

Oh, you're after a symbolic/science term rather than a way of phrasing it.

In reply to Reach>Talent:
The part isn't sitting true or it's running out would be the usual way to describe it.If it's a three-jaw you could tap it true with a copper mallet then clock it up using a mag base and clock to fix the run out.(as stated by others above)

Obviously tapping the compomnent you are about to cut depends on the hardness of it unless you have a set of soft jaws,is it soft jaws ? if so perhaps they need bored out.

God engineering is so boring !ahem.

In reply to all:

The kit in question is somewhat specialised and falls into the "if I told you I'd have to kill you" area of information, hence the vagueness

The part is clamped into a holder on the rotary axis which is generally used as a positioning tool (ie. discrete steps rather than high speed rotation).

At the moment I am having to code an offset when machining from certain orientations which is not a huge pain, however I have to document this in a report and using suitable terminology is

In reply to Reach>Talent:




I'd probably go with 'the machine surfaces (assuming mostly round) are non-concentric with the part of the component held in the fourth rotary axis' or something to that effect.

In reply to Reach>Talent:

The holder rotates in discreet steps and it sounds like you need/want to have smaller or no steps. In terms of stepper motors you would want a higher stepping resolution to achieve that.

In reply to richyfenn:
It isn't the step size as this is a very posh stepper, it is a slight offset on the rotary axis.

Grab a tin and stick a finger in the middle of each end, holding it so it is lying on its side. If you rotate it with your fingers as pivots then watch from above it doesn't move side to side. Now do the same but with your fingers only 1cm from the edge; as it rotates it moves in an eccentric fashion. At two points in the rotation it is offset in the Z axis, 90 degrees either side it is offset in the Y axis and in between it is offset in both.

Basically someone made the jig fractionally too thin so the workpiece sits fractionally below the centre of rotation of the stepper shaft.

Problem is I need to write a very precise report which is both technically correct and also very hard for a layperson to misunderstand by a mix of guesswork and Google

In reply to crayefish:

Looking like a good start, cheers! I have a feeling a few 3D models are required as well as I am really struggling to be precise.

In reply to Reach>Talent:

I think the word you are looking for is runout, or total runout.

In reply to Redacted:

I don't think tapping is an option, it is fairly delicate. Put it this way, if I was feeling a bit frivolous I could probably write my initials on a pinhead with it.

:-o

In reply to A Fish Shop:

Top marks!

In reply to Reach>Talent:



Ah, so not intentionally held off center. On a lathe that would commonly be referred to as run-out.

Can't you shim or grind the fixture to eliminate the buggering about measuring and applying offsets?

jk

In reply to jkarran:

The error is small and variable depending on the size of the material I am working on, it is something like 20 microns on a curved surface so a bit small for the traditional "strip of beer can".

In reply to Reach>Talent:
Fair enough, sounds like a fiddly job and you've probably got a workable solution anyway. Assuming you want it properly axial and on center the best solution where you can is to cut the blank true and to finished size/shape the chuck/collet/jig before doing the fine details. Nice and simple for round shapes where oversized stock is available.

jk