Shimming Techniques

Could Aluminium kitchen foil be used as a shim?

Don't be surprised if you break it down to chase for zero and find you get different readings, I would leave well alone but that's your decision.

I'd leave it as is and see how things go on the next engine, can't get any worse than the one you have a already made.

As you know which way it leans you can make use of that when flycutting to give a light finish cut with the trailing edge.

.

Hopefully, I am I missing an in-joke

MichaelG.

Yes, but kitchen foil is distinctly soft and likely to squish as metals go. Depends on the pressure: a lot of weight on a small area will cause instant damage, but the same weight distributed over a large area might be OK. Harder metals like brass and steel are less likely to deform.

The 10V turned out so well on the machine with more of a list to the column that any subsequent engines should be even better with the column in it's current position.

If you really want to start going to 0.0005" then buy a 12" length of 0.0005" feeler gauge stock or to make it easier to handle get 0.0015" and take out a 1 thou shim and replace with the 1 1/2

Edited By JasonB on 21/01/2021 10:32:58

leave well alone, as others say you might well make it worse byt trying to change it

Duncan, you have to remember my motto is "if it ain't broke, fiddle with it until it is"...

OK, thanks very much all. I'll say I'll leave it at that, but the reality is that I probably won't.

But before you tear it all apart again, try my suggestion of tightening up the two bolts on the "uphill" side. You might be pleasantly surprised. Past experience indicates you can get that last half a thou that way. Metal is more flexible than you might think on these small machines.

Edited By Hopper on 21/01/2021 11:50:24

Ok thanks guys - I'll look again tonight. It's actually very easy to swap shims - loosed all four bolts, then push the top of the column over to release the appropriate shims and add/remove as necessary, then re-tighten and re-check.

I don't have to remove the column or anything like that.

Great progress Dr_GMJN !

I have been involved in a number of structures that where grouted in, similar to using a liquid shim as discussed.

The general rule was to use a jacking screws to align the items preloaded against the hold-down screws, then flood the joint and tighten to the specified torque.

The large area of the plates gave huge stiffness so it doesn't matter much what grout you use but big stuff tends to be cementacious as its dirt cheap. However epoxy is a good vibration absorber, so reduces chatter and is cheap in small quantities.

If your wedge-shaped gaps are still a problem I am sure something could be done along these lines.

OK so this one’s sorted.

I got some 0.0005” feeler gauge as suggested.

Firstly, having looked at where I was measuring, I realised I was making things far more complicated by having the four datum points on the bed 90 degrees round from the bolt pattern I’m adjusting, ie the bolts are at 10, 2, 4 and 8 o’clock, the points on the bed were 12, 3, 6 and 9 o’clock. I changed the measuring points to correspond, and made the x-spacing equal to the bolt spacing in that axis. This had two benefits:

1) The x-errors give pretty much the exact shim size needed under the left or right bolt pairs.

2) It’s far easier to visualise the effects of what any shim will do in terms of how they vary corner/datum heights.

I have no idea why I was using the previous method - possibly something I’d seen online, and the fact it enables full, unobstructed rotation of the gauge within the width of the bed (which is irrelevant anyway).

Secondly, after getting all readings to within 0.001” by shimming, I used Hopper’s tension method for fine adjustment. This was trial and error stuff. However, rather than tighten a bolt further, I did the opposite and loosened it to tweak the readings to zero. I then undid the relevant bolt and slipped shims under, in 0.0005” increments, fully tightening and re-checking. After a couple of tries, It was all perfect.

I checked in three head height positions, three bed positions in x, and also adjusted the length of the gauge arm radius from about 3”, all the way out so that the gauge touched the extremities of the bed corners in one arc. The worst difference I got out of all that was less than 0.001”, with the majority of readings less than 0.0005”. The mid height, mid bed test was spot-on over 5” in x.

All the shims now have much greater area than previously, and the perpendicularity of the axis is better than it was. So that’s that. Just got to work through the backlash issue now.



Anyway, thanks for all the advice. I knew it was do-able, and it really frustrated me not being able to work through such a seemingly trivial task.

Jolly good, just shows if at first you don't succeed...........

although a friend of mine finishes that off with get a bigger hammer

Thanks Roger. In the end it didn’t need Epoxy, but it’s good to explore all options - and learn a few things along the way.

Thanks Duncan. I think that a rather harsh experience In the very distant past concerning A-levels has made me reluctant to give up on anything I consider worthwhile!

If at first you don't succeed.. don't try skydiving!

My experience with backlash in the X and Y axes was that there was quite a lot in the lead screw to table connection. The leadscrew nut had some wear compensation with a split nut but there was none around the handwheel area. The parts were held together with spring pins through the handwheels into the leadscrew shaft. It may be a good idea to put your DTI on the table and indicate on the end of the leadscrew. See what reading you get from reversing the direction of table travel.

If you have a DRO backlash in X and Y is less problematic but backlash can still reduce ridgidity of the system if you do not have the movements locked down. Climb milling is something some people recommend against because of the issues with the cutter driving the table in a direction that is free space due to backlash. Get rid of backlash and you do not have this worry.

Martin C

Thanks Martin - it’s the severe backlash in z that’s the issue. I have a cheap DRO system which has worked very well, and backlash had never been an issue on the x/y table.

Ive another entire thread on the z-axis backlash issues, which can basically lead to the head dropping and cutters grabbing when feeding in z. Obviously when doing operations where the z-axis can be locked it’s a non issue. However, for counterboring or spot facing it can easily lead to a scrapped part.

So I've been playing with the ARC shell mill today for the first time - squaring up some key steel (EN8 I think) for a rear toolpost project. It went through it like butter, producing a shower of blue chips in the process. Very satisfying:







But still...milling the ends (with the bar mounted horizontally) still produced vibrations, and a "something not right" noise, just like I've had with every tool and pretty much every material except aluminium right from day one. This is with climb or conventional milling, pretty much irrespective of depth of cut. Also over several gib adjustments, and recently confirmed to happen with no shims under the column (ie full contact) earlier in this thread.

So I ended up putting the bar vertically and it was fine (it's a short bar).

I guess my machine just doesn't do side milling and that's it...sounds stupid, but I'm at a loss as to where to go with it? The only thing I've not cheked/adjusted in spindle bearings, but there's no play in them, and runout is negligible.

Did you ever get round to checking the spindle bearing backlash?