Hi all, so, on a new 5 axis (AC swivel table) machine with HH530 i have a problem with adjusting the kinematics of the machine. Machine has been set up, geometry, measurements,...made a test cut peace, everything is ok. Then tried to run kinematic measurement cycle 451, cycle run ok, control gives an information about A and C axis, so confirm and do another test cut... test cut after cycle 451 is off by +0,05 Peace is a simple square block, machined with plane rotation and data shift A90 & C0-C90-C180-C270... measured peace is off in X and Y direction for +0,05 both axis. It seems to me that the problem is with the transformation in kinematic tables but I can not figure it out. I tried changing the values of the Z axis transformation for swivel table, e.g. value -100.000 changed by -0,1....measurement result is then -0,05, so tried -0,05..then error comes back to +0,05...again -0,025, error is the same, +0,05....crazy One other thing, i succeed getting a good result by changing the offset of 3d probe for kinematic and work peace measurement. To be clear, I manually input the length offset for 3d probe that is bigger the calibrated value for +0,015. Then cycle 451 run, measured, confirmed,.. then change back the offset to original (reduce by -0,015), measure zero offset, cut, measure the work peace and result are in +/-0,002. for this reason i tried changing the transformation values and see if any result, but unfortunately no. can somebody show me what i'm doing wrong or show me in the direction of the real cause for this error? cheers
Follow along with the video below to see how to install our site as a web app on your home screen.
Note: This feature may not be available in some browsers.
Largest Manufacturing Technology Community on the Web
Stay Connected:
Heidenhain 530 Kinematic adjustement problem
- Thread starter barefr
- Start date
- Replies 8
- Views 5,297
cameraman
Diamond
- Joined
- Nov 24, 2014
Probably most people are not familiar with that machine. So it's basically a gantry style 5 axis machine that can swing an 800 mm part; cause otherwise "microcut mcg5x" sounds like desktop 5 axis machine (although common sense would say that HH 530 should be on a pretty industrially scaled machine).
MICROCUT 5X Machining Center MCG-5X - YouTube
And I'm just asking a stupid question here... @barefr Your measurements and values are metric I assume and in mms?
So for example your last zero'd offset cut left you "quote +/- 0.002 unquote" so that's +/- 2 microns (right)?
Most of the PM forum members are in the USA and they commonly think in "Tenths"... (ten thousandths of an inch). So your final result is +/- 80 millionths [of an inch], (after your zero'd offset cuts?); but other wise you are off by +50 micron (0.05 mm) in X and Y or roughly 0.002" (two thousandths of an inch) due to kinematic calibration/tool probing/(dynamic work offset) weirdness ?
Just double checking, as you guys in Europe have been "Down" with the metric system for the better part of a 100 years. In the USA machine controls come specifically configured for decimal inches, and if you are a metric "Enthusiast" (or formerly European engineer or physicist) in the USA, you'd better make sure your new machine comes "pre-figured" / parameters at/in the control for metric work. (LOL).
Nice looking machine BTW.
Last edited:
LockNut
Stainless
- Joined
- Jan 6, 2007
- Location
- Bergen County
I think you might have your answer already. This just happened to me. We got a machine back from IMTS to our showroom. The people that received it from the factory to do the IMTS demo screwed up the kinematics by 1MM by making a mistake setting up the spindle probe calibration, running the Cycle 451 and saving the results to the machine with the 1 MM error. It took me quite a while to find out what actually happened because the previous people denied everything. The calibration of the probe length is paramount. It will make or break a good machine and chasing down the problem could be difficult if you are not intimate with HH controls, which I was not.
Every time the kinematics gets run, the control saves the results in a text file. Important!! This shows all measurements, kinematic and volumetric errors and lets you save a history over time. You can also run Cycle 450, which only save the kinematic values stored to the machine, also important.
Paul
cameraman
Diamond
- Joined
- Nov 24, 2014
Although its not my thread...
WOW...
Thanks for that.
Over the years there are a few threads (here) where really pretty/very experienced machinists get a new machine and get kinda tangled up with the Kinematic cycles (not their fault), but in a lot of cases the finger gets pointed elsewhere to seemingly inherent problems with a machine (like supposed spindle growth) where for such a class of machine there should not really be any thermal growth of that (magnitude) or suspected thermal growth on a particular axis (like a Y axis!?). And yet you rarely see the resolution or fix posted, or "Discovery" as to what the problem really is. So it kinda leaves you wondering if indeed there is a design problem with a particular machine/model, or leaving minor doubt in the back your mind regarding a particular machine rather than being a deeper mistake/error with a particular file deeper in the control.
Anyway THANKS again for posting that (@Locknut)!
Cheers.
thx paul for posting...
I'm crazy, tried 1000 options, non of them worked..
re calibrating the probe, cycle 451, cutting, measuring...the same
adjusting the kinematic table after cutting with kinematic compensation, result ok, in microns,...cycle 451 again, error comes back....
only what works, cycle 451 with + 0,017 mm in probe length offset,
zero set with probe -0,017, cutting, measuring, result accuracy in microns...
paul, what was the solution in your case, only probe calibration?
LockNut
Stainless
- Joined
- Jan 6, 2007
- Location
- Bergen County
It seems for some reason, your probe is not being calibrated properly. If you can consistently make a manual adjustment to your probe length to get it to work, GREAT! But I would not count on that consistently. I would try to find out why you probe is not being calibrated right.
My solution was correcting the mistake made by another person in a hurry to set the machine up for an upcoming IMTS demo. The problem with demos is that they don't have to be accurate, just pretty. When I checked the TCHPR451.txt file created every time the kinematics was run, I discovered a 1MM error in the Z axis and an even larger error in volumetric accuracy resulting from that 1MM error. I have no idea how the 1MM error came about but I discovered it when I tracked the history.
What is your procedure for calibrating your probe? I set a Z offset to the face of the spindle with a known, accurate height standard and a known tool artifact used to calibrate a tool setter (say, 125MM). Activate that Z Offset, then calibrate your probe using the same height standard. There will always be some error, hopefully in the microns.
One thing more. NEVER just run kinematics and save the first time. ALWAYS run kinematics at least 3 times and look at the TCHPR451.txt file and when 3 runs give you a consistent error, then save the kinematics to the machine parameters.
Paul
My solution was correcting the mistake made by another person in a hurry to set the machine up for an upcoming IMTS demo. The problem with demos is that they don't have to be accurate, just pretty. When I checked the TCHPR451.txt file created every time the kinematics was run, I discovered a 1MM error in the Z axis and an even larger error in volumetric accuracy resulting from that 1MM error. I have no idea how the 1MM error came about but I discovered it when I tracked the history.
What is your procedure for calibrating your probe? I set a Z offset to the face of the spindle with a known, accurate height standard and a known tool artifact used to calibrate a tool setter (say, 125MM). Activate that Z Offset, then calibrate your probe using the same height standard. There will always be some error, hopefully in the microns.
One thing more. NEVER just run kinematics and save the first time. ALWAYS run kinematics at least 3 times and look at the TCHPR451.txt file and when 3 runs give you a consistent error, then save the kinematics to the machine parameters.
Paul
probe calibration just like described...
known caliber, set offset 0, probe calibration on same point...always the same value...
checking, set 0 on a peace by cutting, check 0 with probe, ok.
i'm was running kinematic cycle few times in a row,to see repeat ability, always a small correction in a and c, few microns,nothing that worries me...
i always also check the .txt file of kinematics and see error and compensation values, similar small values
also saving the kinematics with cycle 450, the ones with the good result.
i even tried a reversed logic, set the kinematic, same probe length, measure the peace, cutting and adjustment..at one point it was successful in the point that now i get 0,002-0,005mm error in X and Y and get 0,04 error in Z. So corrected a tool length,cutting, adjusting the kinematic compensation, measuring, 0,001-0,003mm error on all axes...great....until I run kinematic cycle again...
so basely, running in circles for few days now.....
i even restored the machine original backup from the factory, and start the set up from beginning, but no....always the same...
known caliber, set offset 0, probe calibration on same point...always the same value...
checking, set 0 on a peace by cutting, check 0 with probe, ok.
i'm was running kinematic cycle few times in a row,to see repeat ability, always a small correction in a and c, few microns,nothing that worries me...
i always also check the .txt file of kinematics and see error and compensation values, similar small values
also saving the kinematics with cycle 450, the ones with the good result.
i even tried a reversed logic, set the kinematic, same probe length, measure the peace, cutting and adjustment..at one point it was successful in the point that now i get 0,002-0,005mm error in X and Y and get 0,04 error in Z. So corrected a tool length,cutting, adjusting the kinematic compensation, measuring, 0,001-0,003mm error on all axes...great....until I run kinematic cycle again...
so basely, running in circles for few days now.....
i even restored the machine original backup from the factory, and start the set up from beginning, but no....always the same...
