[keen]

Ive just checked the gibs after over a years use since the last adjustment - I guess around 300-400 hours running.

Very pleased with the wear situation - gibs only needed 1/2 - 1 turn. I use the 'step reverse' method where you set the step to fine (mine is left on metric 0.01mm ( or about 0.0005" )) and with a dial indicator click an axis one way - then change direction. I allow a max of 'one lost click' before movement begins in the other direction. My Z and X had no lost movement so I began to tighten the gib - a half turn began to introduce enough drag to begin the 'lost movement click' The Y had one lost click at fully out (less slideway wear here) and no lost click in the middle of the Y travel - again I tightened up less than a turn before lost motion increased. I double checked the actual slideways clearance with a dial indicator set to measure between the sliding surfaces and then applying a heavy racking load...and I got - apart from spring that returns, the permant shift or clearance was not more than 0.001" Very pleased!

I exposed and cleaned the ways first - they looked in good shape. I always operate the oil plunger at the start of each day - plus I manually coat the slideways and screws from time to time in case the central system does not evenly lubricate. This may be in part why I have little wear - but results in quite a bit of lube/tramp oil floating on the coolant in the tank. I just thought of an easy way to remove it today - probably in common use but in case others have not heard of it.....

Spread a clean rag out on the surface of the coolant - leave for a short while, pull the rag out in a way that catchs the oil and quickly dump it before too much tramp oil escapes. Repeat.

[Freddy Bastard]

Nice!

That would be very usefull if Tormach did some videos on how to finetune our machines.

The lost clic method, is it documented somewhere?

Thank's for your feedback.

[keen]

Hi Freddy - If you are asking me....the official method from memory is more involved...I would need to find it and reread it to be sure.....anyway my method is quick and simple....

[David Bord]

Keen, very good news for you! I too would love to hear more about the process. I think I understand how you are doing it.

Using the step/jog feature set to approx .0005 you are checking to see if the gibs are sticking or the ball screws/bearings are compressing by checking for lost clicks/jog commands.

The one question I have, is how do you account for backlash with that method?

David

[keen]

Hi David - interesting question! backlash has always been a problem with conventional machine tools. But with ground ball screws set with preload on the nut and thrust bearing, backlash in theory can be a thing of the past. You must have the correct machine design, correct machine finish and assembly in the key areas, correct ball nut and thrust preload set, and the correct gib adjustment. Then it is possible to have zero backlash!

In practise this is very acheivable on a high end CNC machine - but not so likely on a low end machine. When I first got my Tormach I spent some time trying to get the maximum accuracy out of it. The end result is what really matters and I believe machining a male circle (see pic below) is a very good way to measure the end result because you can mic the part and also rotate the part in a lathe with a dial indicator and see exactly what the end result errors are.

I copy below:

Tormach PCNC Accuracy/Roundness Test 30.6.08

25 Dia 4140 6.00 cutter end milled down to 24.00 Dia.
Micrometer 23.98 – 24.005 mm
Dial Indicator in Lathe total run out 0.025mm

In answer to your question I am saying you should be aiming for zero backlash anyway - the aim is minimum screw motion backlash and minimum gib/ slideway clearance. The lost click method is a simple way to set to that minimum.

This will vary from Tormach to Tormach as there are many factors involved, thankfully most factors can be adjusted or at least improved.

[Freddy Bastard]

This thread is becoming interesting...

Did you have to play with the ball screw preload?

[keen]

Hi Freddy. Luckily my ball screws were not bad as delivered. I guess as they wear I will eventually need to strip and do that ...so far they have not worn noticably. I did need to tweak the thrust bearings preload, stepper couplings grip and the gibs when I first got the machine.

[keen]

This is what I did to the stepper couplings eventually - see pics. I occasionaly was getting 'creep'.

[307startup]

With as few hours as you have on your machine, and the turcite lining on the male way surfaces, you really shouldn't be seeing any wear.

What you're more likely finding are that the machines surfaces are mating differently than before due to gravity and settling of the cast iron.

Just as all materials have some elasticity, even stress-relieved cast iron will reseat itself as it ages...look at engine blocks...yes the bores will show wear, but the bore alignment will shift all over the place compared to fresh off the assembly line.

This is why "aged" blocks will be re line-honed prior to rebuilding, with often very little boring done to the cylinders themselves, usually just enough of a stoning to bring the cross-hatch back to life.

[keen]

Hi 307startup.

Yes I agree with you in principle as far as the X/Y are concerned. Sure a high end machine should have little wear If they are well designed, correct material choice, well proportioned, finished, assembled, and lubed.. then yes the Turcite etc should mean little wear - but many low end mills do wear rapidly because one of these factors are just not right. Many low end mill manufacturers just do not do the R&D to really refine the design. they cut the R&D corner and "out the door" leaving the price conscious owner to deal with the fall out.... 'Archilles heels' abound in low end machines. I am sure a the Mill/Drill based CNC's would flog out real fast.

Good work Tormach for getting all those factors right. As far as the Z is concerned, the potential for higher racking loads (and Turcite is not always the best in this situation, I am not convinced it is even on the Tormach Z slideway....I can't see it.....that may be a good thing) - Anyway I believe the design is sound and Tormachs achievement is even greater on the Z!

[bevinp]

Hi Keen,
Does your lost step method work reliably on the Z axis? And in both Up and Down directions?

And do you suffer any head creeping down after turning power off, or do you have the Z axis brake?

Bevin

[keen]

Hi Bevin. Yes the 'lost click method' works well on the Z. Actually the Z has less of an issue with traditional backlash if it is designed and set up to operate via gravity.

By that I mean if the Z ways are of suitable design, proportion, finish, adjustment and lub etc they can allow the head to 'sit on the screw' under gravity. I think this is an ideal low friction/wear situation if it can be acheived with accurate alignment and no 'dropping' when the steppers are unpowered.

Yes I have the Z stepper brake model - ocassionaly the brake does not kick in and the head begins to descend.