What's new
What's new

Mazak Glass Scale Correction

tylerfavor

Plastic
Joined
Oct 27, 2020
Anyone know if a parameter exists to correct x axis scale?

I have a SVC 2000L running Mazatrol 640m. the X axis moves only 199.9836" when commanded to move 200.0" and only 99.9926" when commanded to move 100.0". it repeats this perfectly. any way to correct this?
 
Anyone know if a parameter exists to correct x axis scale?

I have a SVC 2000L running Mazatrol 640m. the X axis moves only 199.9836" when commanded to move 200.0" and only 99.9926" when commanded to move 100.0". it repeats this perfectly. any way to correct this?
1. How do you know that 200 is in fact 200. Maybe 199.9836 is correct ??
2. Did you consider the impact of temperature expansion ??
 
1. How do you know that 200 is in fact 200. Maybe 199.9836 is correct ??
2. Did you consider the impact of temperature expansion ??

We used a FARO tracker with a tracker ball in the spindle. The tracker has a .0005 tolerance over the distance we were using it. we simply set the trackers origin at our work offset zero, used the controller to move the spindle to X- 200.0", and compared the distance the tracker ball moved with the 200.0" it was supposed to move.

We repeated this process several times with repeating results. We also confirmed these numbers with a more recently calibrated FARO tracker.
 
There are parameters on the control that use the scale to set the axis pitch error. I would highly recommend contacting Mazak service to do this with you, not only is it in a locked portion of the control but there is a very specific way the pitch error table is set. If you get it wrong it will start stacking errors everywhere.
 
You've probably already done this, since you're only asking about X, but how does the Y axis check at a similar distance?

If Y checks good but X isn't, I'd assume it was the machine. But if they were both bad, I'd be questioning the tool.


I have no experience measuring something that long nor with your Faro or machine.

Edit: I guess you can't do a similar distance in Y. Does that machine really only have 20" in Y? ouch.
 
Isn't the SVC-L a linear motor in X? AFAIK they don't have any wear parts, and the motor is the scale.

Obviously this is a huge can of worms, and if you need to ask us how to comp it, you really should be asking Mazak. Relative to the cost of the machine (and the time you have spent diagnosing it), having a tech out there for a day is pretty cheap.

If this is a sign of issues with the linear motor, perhaps you can act early to circumvent a failure. Replacing that motor will probably cost as much as the machine is worth. Again - even if you have to fly somebody out from Kentucky, seems it would be money well spent.
 
Check the level and flatness of the base. If the floor has sagged in the middle or at the ends, it may be travelling on a really gentle radius.
Check the travel 10” above the table and see if it changes - this will help prove/disprove the radius idea.
 
Last edited:
My DMG has a linear absolute and incremental encoder,I don't know how else you measure with a linear motor.

Guess I would check scale mounts, and then use a different measuring instrument. Without a ballscrew to introduce errors and backlash, I don't know what else to think. Unless someone has already messed with the offsets

Linear way machine right? Because the only other thing would be play in the y ways, but I would not think that repeatable
 
Thank you all for your help. It sounds like i need to just eat the down time and wait for a Tech. All local techs are booked out several weeks, and mazak cant help us yet due to a registration issue with this machine. I was hoping that we could adjust on our own and not have a machine down several weeks. lol.
 
You've probably already done this, since you're only asking about X, but how does the Y axis check at a similar distance?

If Y checks good but X isn't, I'd assume it was the machine. But if they were both bad, I'd be questioning the tool.


I have no experience measuring something that long nor with your Faro or machine.

Edit: I guess you can't do a similar distance in Y. Does that machine really only have 20" in Y? ouch.

Y seems perfect at 20".

Yes, only 20". this guy only runs machined wing spars. most these are 180" to 200" long with .005" hole location tolerances from one end to the other, so this is definitely a hold up. also way to many holes and features to simply comp for the error in cam.
 
This is not DYI.
One needs a tech and more importantly one that understands long table travels.
If a relativity new machine I'd demand it be done for free.
When put in place what was it certified and signed off at?
Foundation under the base is good and stable? No trains, semis or large forklifts that make a mess of things?
Bob
 
We used a FARO tracker with a tracker ball in the spindle. The tracker has a .0005 tolerance over the distance we were using it. we simply set the trackers origin at our work offset zero, used the controller to move the spindle to X- 200.0", and compared the distance the tracker ball moved with the 200.0" it was supposed to move.

We repeated this process several times with repeating results. We also confirmed these numbers with a more recently calibrated FARO tracker.

According to Faro specifications (http://www.dirdim.com/pdfs/DDI_FARO_Laser_Tracker_ION.pdf) the tracker accuracy is 49 microns (0.002 inch). It doesn't seems to me that laser tracker is the correct tool to calibrate CNC machine. The laser interferometer, with material temperature and enviroment compensation sensors, assures accuracy of 0.5 to 1 ppm (part per million). In your case - 0.0001-0.0002 inch.
 
This is not DYI.
One needs a tech and more importantly one that understands long table travels.
If a relativity new machine I'd demand it be done for free.
When put in place what was it certified and signed off at?
Foundation under the base is good and stable? No trains, semis or large forklifts that make a mess of things?
Bob

We do have several very large fixtures that require a fork lift to load. Trains near by but limited to slow speeds. the only thing that noticeably shakes the building are helicopters (Black hawks)that takeoff and land a few hundred yards away.
 
According to Faro specifications (http://www.dirdim.com/pdfs/DDI_FARO_Laser_Tracker_ION.pdf) the tracker accuracy is 49 microns (0.002 inch). It doesn't seems to me that laser tracker is the correct tool to calibrate CNC machine. The laser interferometer, with material temperature and enviroment compensation sensors, assures accuracy of 0.5 to 1 ppm (part per million). In your case - 0.0001-0.0002 inch.

I realize the tracker isnt ideal, but its the best we have for the whole 200". I assume Mazak will have a way to correctly calibrate?
 
hi tylerfavor
according to your error over the 200 inches, you could use , say a 20 inch gauge(block), and would be reading an error of .0016 over that distance, that might tell you about how well the tracker is actually reading. my 2 cents
 
According to Faro specifications (http://www.dirdim.com/pdfs/DDI_FARO_Laser_Tracker_ION.pdf) the tracker accuracy is 49 microns (0.002 inch). It doesn't seems to me that laser tracker is the correct tool to calibrate CNC machine. The laser interferometer, with material temperature and enviroment compensation sensors, assures accuracy of 0.5 to 1 ppm (part per million). In your case - 0.0001-0.0002 inch.

I realize the tracker isnt ideal, but its the best we have for the whole 200". I assume Mazak will have a way to correctly calibrate?

I don't know whose idea it was to put the tracker in there, but it sounds like someone is looking to create a problem.

Personally, I would have more faith in the accuracy of the scale than the accuracy of the tracker.

I.e. the error that you're reading is more likely to be coming from the tracker than from the machine.

Interferometer (Renishaw ML10, XL-80 are commonly used) is the correct tool to calibrate a machine tool. This was almost certainly performed at the factory, and there should be a report somewhere in the machine documentation that you can look at to see if there was any or how much compensation applied to that axis.

You can get a third party to come in and do it again if required. Environmental conditions are critical - ambient temperature stability, humidity, air turbulence, ground stability as has been mentioned.
 








 
Back
Top