Backlash you'd accept to interpolate a good circle

[precisionmetal]

jools,

I have a 2005 super mini mill (the "super" part really only comes from the 15k spindle). I've tested interpolating a ~50mm hole into a 20mm thick aluminum plate using a 12mm end mill, and as long as the feed rate is kept down, mine will hold less than 10um TIR on the bore gauge.

PM

 

[DMF_TomB]

average cnc thats been used over a decade if circle milled round within .001" most would think thats not bad. obviously some might say they hold .00001" tolerance easily on all their machines but I would say thays not typical or normal average
.
obviously depends on feed rate AND where in slide travel. often middle of travel is looser than ends of travel.
.
use test indicator and 6.0000" gage block move in one direction zero indicator and DRO against gage block against stop, backup remove gage block move to stop til indicator zero read DRO see if reads 6.0000" often its off cause DRO calibration off, then overtravel and reverse til indicator zero and zero DRO and move over 6" put gage block back, go toward it over travel and backup til indicator zero and read DRO does it read 6.0000" ? hard to describe got to overtravel and reverse (cause of backlash, always measure going same direction) like done on manual machines
.
just saying there's backlash AND DRO calibration. some cnc you check DRO calibration AND backlash compensation yearly, backlash compensation often you set for average as backlash can vary based on where you are in slide travel
.
i would never assume DRO is calibrated that is 6.0000" on X is same as 6.0000" on Y, BOTH need calibration checked, usually its a electronic compensation following a procedure. I have seen DRO off .002" per 6.0000" before, thats why DRO calibration is checked.
.
once operated a cnc where if DRO compensation was active a light would blink all the time, if power turned off then on, the compensation needed to be turned on again, i often found it off cause somebody turned machine off and was not aware about turning DRO compensation back on. obviously different cnc can operate differently. i only discovered it cause 6.0000 programmed movement was not measuring 6.0000 of parts and wanted to know why and "read the manuals of the cnc", once DRO was calibrated it ran much closer like 90% of error was eliminated
.
be aware machine AND part change size based on temperature, temperature can vary in different spots. also many parts will go out of round when unchucked or unclamped. many parts are finished at lighter chucking or clamp pressure. and many parts just distort out of round after some time has gone by hours to days often.
.
a person who says their cnc holds .00001" easy, usually will mention chucking pressure distortion and temp/size variations and parts warping after some time has gone by. not even mentioning grid shift calibration that is machine changes size as it warms up and part distance from zero return position changes with temp which can vary literally every 10 minutes talking about .0001" tolerances. cold coolant not held withing +/-1F degree temp can cause problems as well as coolant evaporation having a cooling effect. many run a "warmup" program (some with or without coolant running it depends)so cnc is at average operating temperatures( temp at 500rpm and 3000rpm can be different), obviously if part is buried under 100lbs of very hot chips the part is not at room temperature usually, there is hot and cold distortion and easily one side of part can be hotter or colder than other side of part (more a problem with bigger parts)

 

[jools]

WOOOHOOOHOOO!!
Congratulations jools!
If your experience with a Haas Minimill is anything like mine has been, you'll be very happy indeed.
I've built all kinds of weird and wonderful things with mine and it continues to perform like a champ even after twenty years.

Once you've got it set up and running you'll wonder how you ever built the stuff you did without it.

Cheers

Marcus
Implant Mechanix • Design & Innovation > HOME
Vancouver Wire EDM -- Wire EDM Machining

Thanks man I hope it is just like yours and can't wait to start building professionally

Signed the contracts today after wrangling with the finance guys, got another provider to do 10% and $420 a month, saved myself over £3000 over the course of the loan

Just waiting for delivery date now

Jools

 

[jools]

jools,

I have a 2005 super mini mill (the "super" part really only comes from the 15k spindle). I've tested interpolating a ~50mm hole into a 20mm thick aluminum plate using a 12mm end mill, and as long as the feed rate is kept down, mine will hold less than 10um TIR on the bore gauge.

PM

That's really good. Time is on my side when building as it's low volume stuff. I think it's going to be perfect to be honest as she's fresh out the crate. If not I'll call Haas in to fix her

Jools

 

[jools]

DauFug?
OPs Q was "Backlash you'd accept to interpolate a good circle".
And was talking ref flexi machines and the like.
My answer is 10 microns. If a VMC can't interpolate within that limit, it needs a tune-up.
And just for you....some documentation...

Nice ball bar trace. I'd be happy with that