DMI Router Lost z position

[RaiderX]

Ok this is a new one to me with exception of some 1960's machines I have worked with.

I was machining on our dmi router today. new program so I checked depth of cut to z position several times (verify my zero setting) all was good. ran the part out took about 3 hours. then went back into production parts these are proven programs with org statement in program (like g10 on fanuc). first few parts were flat pattern and showed a little evidence of not breaking thru, I shrugged it off. next part had z sensitive features (esp. a wheel cut). scrapped the part and was headed for the second. I checked z setting, tlo etc.

I have been doing this long enough to know 99% of error is the guy in front of the control so I approached this as so. not the case here as far as I can tell.

finally set up an indicator to tool at set spot. zero referenced machine and z changed by .025. I suspect .0254 but did not check that close.

What gives?  the reason I suspect .0254 and it is a long shot but does this machine run metric encoder that reads .1mm? IDK. sure is suspicious.

We work in composites so it is a nasty dusty environment if that gives any one something to go on. optical encoders?

This is unexceptable in my books but need a root cause and solution. management wants me to remake the fixture, etc, etc. that is not going to solve a machine losing position.

 

input would be much appreciated.

Doug

[Leon82]

I have heard that the carbon fiber dust in not good for the electronics also. An encoder can slip a tooth and the belts can wear

Is there a value in the external work offset?

 

 

[RaiderX]

thanks for the reply. we do not cut carbon? if that is what your saying. mostly fiberglass and composites there of. I am in aware of any external offset setting?

thanks for the reply and do educate me. I have 30+ years experience on various (mostly fanuc) machines but this one is very new to me.

I am leaning heavily towards a regular cleaning of resolver, encoder?? IDK I am new to the machine and materiel being cut.

this is for sure a humbling trade.

thanks again. Doug

[Leon82]

I misread i thought you said carbon fiber.

Does it have glass feedback scales? Maybe there is a contaminant on the z?

[gcode]

Is it possible that the Z axis is being overloaded during a Z axis plunge cut

which causes the Z encoder to slip

I've seen the wafer created by a  plunge cut breaking though get stuck between the tool and fixture

The wafer can't be cut cause it's spinning with the tool.

On a strong machine, the tool blows up, on a weak machine, the encoder can slip

 

 

[RaiderX]

Thanks Gcode, I make sure and baby this machine the head is a very week point. lots of light cuts really fast has been my approach (18k spindle).

[gcode]

The best solution is to relieve the fixture at plunge points so the wafer has somewhere to go

[Leon82]

You may want to verify any programs with the g10 offset.

If it slipped your zero may have changed.

In extreme cases the machine won't home out and a grid shift is required. Some machines have a coupling to absorb shock in the event of collision that slips