We have a Centroid M400 on out millport Runs 7 days a week for the last 10 years. Never missed a beat. Well it did go down a few years back, The CMOS battery dies. One call to Centroid and we were back up in 15 minutes.
Very interesting vid! You've got me looking at the MCV-500 now. I didn't understand your comment yesterday about needing the laser, but I do now since this vid and the comments caused me to go do a bit of research on the laser. :) Now to find one.
Hi Alan, Thanks for sharing the video on your progress. Can you elaborate on Metrol tool seater and short range sensor that you used in your project, please? Thank you.
I use the metrol PTP5M1CB for homing switches. And in the video i used a linear scale from mitutoyo. Since the video i have upgraded to a laser interferometer for the mapping.
When taking additional passes and fine tuning do I need to go back into motor and turn the screw comp back off each time I am taking a new set of measurements? I have been turning it off when I take my .5" readings, enter new data, and turn back on. Am I just chasing my tail doing that way?
@mt6865 I have not done a comp in a long time but if I remember correctly, I would do the first comp off and tge fine tuning with the comp on and small adjusting the numbers plus or minus until it was dialed in, if you shut it off each time you can't see the difference your making.
@@AMHKnives Thanks, that is exactly what I ended up doing. Ran it once, turned on the comp and started adjusting the corrections little by little. Was able to get the first axis dialed into within .00015" across the 18" of travel. Should have done this 2 years ago.
Am new to cnc,still learning,your video was very helpful, how do you make the g code for the acorn,have never made a simple non part g code. To run on my acorn
You could use the intercom software in cnc 12 to try some simple moves or just some simple x+ or - moves to run the axis. To make all my parts i use fusion 360 for my cad cam and g code for thd acorn.
Did you calibrate your inches/rev and backlash first? Does this not also affect the screw compensation or do the screw compensation table override these values? Kind of seem redundant in spots... Great Videos. Thanks!
You do revs per inch first and backlash. Ballscrew comp is only for the error in the screw itself. And it uses the table you make to do small changes in the pulses to take out the error.
@@AMHKnives Thank you. Calibration is on my list of todo for the evening. I'll follow those steps in order. Trying to get my "circles" to be a bit more circular.
Alan, you choose .500 per step, curious if mapping half that would be more accurate if you wanted to get crazy- how many spots for the map are available ?
.5 is the standard setting but you can change it to small increments if you wanted to it would just take way longer to map and put in the information. I don't know how many spots are available, the paperwork does not say, you would have to email centroid cnc for an answer.
No need to go finer! Remember screw comp is compensating for screw pitch error! When trying to wrap your head around this...think in the terms of pitch not absolute positional error. Screw Pitch error is fairly consistent due to the manufacturing process equipment. The Centroid CNC12 software interpolates the measured error between the correction points. In other words the pitch error compensation is spread out over the .5" distance. for example: from the beginning of the measurements….. start target =0, measured =0, next target = .5" actual measured = .499" so, error at .5" target is -.001" CNC12 adds .001" to compensate. That .001” is a PITCH error. The .001” error compensation is then spread out over the distance between X0 to X.5" So for example, at the x.25" location cnc12 is adding turns to result in an additional .0005" to the commanded position of .25" so that the machine reaches that actual position of .25" etc... Test it yourself, program moves to any position after setting up the table and look what the DRO reads.. and I think you will be shocked at how well it works. Often one measurement pass is all it take, but feel free to make another run for super fine tweaking. Note: sometimes finer (smaller distance between points) measurements might actually produce worse screw pitch measurement results as it is a good idea to get somewhat of an average pitch error over a distance (1/2”). We have settled on 1/2” increments in our 20 year experience with both Laser and Manual table software screw compensation for a number of good reasons. Note: Centroid screw comp is taking care of two things, #1 is pitch errors of the screw in BOTH directions! as the pitch on each side of the screw can be and most likely is different (going one way the nut presses on one side of the threads and going the other way the nut press on the other side of the threads, which are different) and also backlash in the system on the turn-around. Moving .5" between measurement point gives a good reading of the pitch error. Also, backlash can come from several locations, most common are the screw nut to screw connection and more often and often more pronounced is the screw to machine tool connection made by the support bearings. For best screw comp results, its always a good idea to reduce backlash mechanically as much as possible from all sources starting with the screw support bearings (without adding drag!) and if your machine tool has an adjustable tension nut to minimize it there as well before running the screw comp measurements. Try it, it works!
Subbed! What model switch are you using? I'm running a Tormach and my home switch has 5 thou or more in variance in the home switch! It drives me nuts!
Hi Sir I just got set up Centroid Arcon. But I do not know how to run a program. can you help please. I have the pipe chuck in 4th Axis. And Need to drill hole on different angle and distance with X travel . Thanks you very much for your help. Tony
Tony, are you using a cam software for your g code or trying to use the internal software for your cutting? If you want to shoot me an email insted of going though RUclips, amhknives at gmail.com
Nice informative video Alan. I'm having trouble with the incremental G-code command. Acorn says the G91X-.5 line is a 502 Invalid G-Code and won't continue running it.
Hello Alan I need help. I dont have precision. I have cnc router with dmm servos. I followed your video of ballscrew mapping. I do the complete travel I return to make the home every time, I make a step as the software asks me to, but when I return to zero and verify again the travel loses the precision again. I don't know what to do. It seems I lost buying acorn cnc. Could you help me please. Hector.
Two quick checks, on the screen that has your motor steps and backlash make sure the ballscrew comp column has y in it turning on the comp for that axis, also after you do the comp save and restart the software, then verify that your comp table is there, you can do it in the software or in c: of your computer. If these dont work ill shoot you my email and we can try a few others.
@@AMHKnives I did the process again but the result is the same the precision is lost. be happy if lost by .0015 or.002 inc is nice. but it is lost by .005, .006 up to .008 inc that does not work. I have experience in diy cnc previously I used path piloth software it worked very well but when I made the change to servos the path piloth did not want to work with the dmm servos. I made the change with mach3 but almost always lost communication with the computer. I made the change to centroid looking for improvement and I find this problem very bad. Could you help me with a more detailed video of how you did the process
@@deborahcarattini765 a few questions, have you made sure your pulse per rev is dead on for each screw and that your backlash is completely 0 or compensated to 0. Also how are you measuring your movments? Have you checked to make sure all your compensation numbers are in the software and a at least .5" for each measurement. You have to be sure 100% that you have no mechanical issues before you even try to comp. Now when i did my final compensation on my machine u used a laser interferometer for the movement, i talked about it in another video. If you have more questions its easyer to message me at amhknives at gmail dot com
thanks Alan , i would like to know how to cofigure the wizard using DYN2 and DYN4 servo driver from DMM, I was trying but when selet that config. on the wizard, I can"t use MDI into the centroid software to enter gcode directly.
sorry gabriel i know nothing about the dyn2 and 4 servo drivers, I recommend you contact centroid directly or go on the centroid forum for help with this problem.
1 μm! Sweet! Would you recall the price and do they require any signal translation to connect to CNC 12 board, or it’s just straight NC/NO switch. May I ask which laser interferometer you incorporated in your design? Thank you very much! Best!
When i bought them they were $90 each and they connect directlyto the acorn board. I was running an older hp 5517d laser with all the mirrors and control board from sams laser faq but i have just upgraded to an optodyne laser.
@@arthursargsian thank you, im actually not running a metrol tool setter, i have a renishw ots tool setter and omp400 probe with a renishaw optical eye controling both.
@@crookedriver2079 sorry about that, you use the scales to measure the movement and the ball screw mapping software uses thes numbers to adjust movements of the machine to make it more accurate. Say you want it to move 1in and it moves 1.005" you can have the software fix the .005" deviation and then the mill moves 1.000 correcting the fault in the ballscrew movement.
Alan, can you provide the model number for the laser interferometer that you use? The Dro on my x-axis (lathe) is not accurate enough for me and would like to go the laser route if it's worth it.
I'd like to see this repeated going the other way. Something seems awry. Home, go to your extreme position, move back .5", then zero the DRO and move back.
![EST Gee - NEVER NEED (Feat. Selfpaid Savage) [Official Video]](http://i.ytimg.com/vi/f2cRtr52HXI/mqdefault.jpg)
Really starting to like this acorn board. Emailed Centroid and they responded same day and they have a solid post processor for Fusion.
I use fusion for all of my cad cam for the mill and the post processor has worked great for the centroid software.
Manny Calavera did centroid send you the post processor for fusion or is it in the autodesk website?
bostondan77 it's in fusion. M400 I believe from memory.
We have a Centroid M400 on out millport Runs 7 days a week for the last 10 years. Never missed a beat. Well it did go down a few years back, The CMOS battery dies. One call to Centroid and we were back up in 15 minutes.
Excellent! Thank you for taking the time to share this info!
Thanks, glad to help.
Very interesting vid! You've got me looking at the MCV-500 now. I didn't understand your comment yesterday about needing the laser, but I do now since this vid and the comments caused me to go do a bit of research on the laser. :) Now to find one.
I really need to try this. Thanks for the video. Good stuff
Love the Acorn info...def going that direction on my next cnc ;)
Hi Alan,
Thanks for sharing the video on your progress. Can you elaborate on Metrol tool seater and short range sensor that you used in your project, please? Thank you.
I use the metrol PTP5M1CB for homing switches. And in the video i used a linear scale from mitutoyo. Since the video i have upgraded to a laser interferometer for the mapping.
When taking additional passes and fine tuning do I need to go back into motor and turn the screw comp back off each time I am taking a new set of measurements? I have been turning it off when I take my .5" readings, enter new data, and turn back on. Am I just chasing my tail doing that way?
@mt6865 I have not done a comp in a long time but if I remember correctly, I would do the first comp off and tge fine tuning with the comp on and small adjusting the numbers plus or minus until it was dialed in, if you shut it off each time you can't see the difference your making.
@@AMHKnives Thanks, that is exactly what I ended up doing. Ran it once, turned on the comp and started adjusting the corrections little by little. Was able to get the first axis dialed into within .00015" across the 18" of travel. Should have done this 2 years ago.
Thanks for sharing Alan. Can you post the part number on the Metrol Switches you use for your Home switches?
Stephen Brown yes im using toolsensor.com/switch/ptp5m1cb/
Awesome thanks!
Bonjour s'agit t'il d'un réglage par zone ou des valeurs n'importe où sur le déplacement
Am new to cnc,still learning,your video was very helpful, how do you make the g code for the acorn,have never made a simple non part g code. To run on my acorn
You could use the intercom software in cnc 12 to try some simple moves or just some simple x+ or - moves to run the axis. To make all my parts i use fusion 360 for my cad cam and g code for thd acorn.
Do you have a video on how to input your ballscrews?
I'm sorry I do not, I believe Marty has a video or two on that subject, if not the centroid forum can help you.
Hello. , before buying acorn cnc I want to know if the interface has multiple languages .... French for me. thank you
I'm im not sure bud. You would have to ask centroid on there forum.
How do I get the Y/N selectable? I’m trying to turn on screwcomp but it’s not allowing me to turn it to yes??
I am trying this and I do have it turned on for the axis and I enter the new values for the first 1/2" and re run and there is no change. What's Up
Nice demo, thanks.
Did you calibrate your inches/rev and backlash first? Does this not also affect the screw compensation or do the screw compensation table override these values? Kind of seem redundant in spots... Great Videos. Thanks!
You do revs per inch first and backlash. Ballscrew comp is only for the error in the screw itself. And it uses the table you make to do small changes in the pulses to take out the error.
@@AMHKnives Thank you. Calibration is on my list of todo for the evening. I'll follow those steps in order. Trying to get my "circles" to be a bit more circular.
Alan, you choose .500 per step, curious if mapping half that would be more accurate if you wanted to get crazy- how many spots for the map are available ?
.5 is the standard setting but you can change it to small increments if you wanted to it would just take way longer to map and put in the information. I don't know how many spots are available, the paperwork does not say, you would have to email centroid cnc for an answer.
No need to go finer! Remember screw comp is compensating for screw pitch error! When trying to wrap your head around this...think in the terms of pitch not absolute positional error. Screw Pitch error is fairly consistent due to the manufacturing process equipment. The Centroid CNC12 software interpolates the measured error between the correction points. In other words the pitch error compensation is spread out over the .5" distance.
for example: from the beginning of the measurements….. start target =0, measured =0, next target = .5" actual measured = .499" so, error at .5" target is -.001" CNC12 adds .001" to compensate. That .001” is a PITCH error. The .001” error compensation is then spread out over the distance between X0 to X.5" So for example, at the x.25" location cnc12 is adding turns to result in an additional .0005" to the commanded position of .25" so that the machine reaches that actual position of .25" etc... Test it yourself, program moves to any position after setting up the table and look what the DRO reads.. and I think you will be shocked at how well it works. Often one measurement pass is all it take, but feel free to make another run for super fine tweaking. Note: sometimes finer (smaller distance between points) measurements might actually produce worse screw pitch measurement results as it is a good idea to get somewhat of an average pitch error over a distance (1/2”). We have settled on 1/2” increments in our 20 year experience with both Laser and Manual table software screw compensation for a number of good reasons. Note: Centroid screw comp is taking care of two things, #1 is pitch errors of the screw in BOTH directions! as the pitch on each side of the screw can be and most likely is different (going one way the nut presses on one side of the threads and going the other way the nut press on the other side of the threads, which are different) and also backlash in the system on the turn-around. Moving .5" between measurement point gives a good reading of the pitch error. Also, backlash can come from several locations, most common are the screw nut to screw connection and more often and often more pronounced is the screw to machine tool connection made by the support bearings. For best screw comp results, its always a good idea to reduce backlash mechanically as much as possible from all sources starting with the screw support bearings (without adding drag!) and if your machine tool has an adjustable tension nut to minimize it there as well before running the screw comp measurements. Try it, it works!
Subbed!
What model switch are you using? I'm running a Tormach and my home switch has 5 thou or more in variance in the home switch! It drives me nuts!
Im running metrol PTP5M1CB switches from japan, i believe you can get them through amazon now.
Awesome! Thanks for sharing, I'll have to add those soon!
Hi Sir
I just got set up Centroid Arcon. But I do not know how to run a program. can you help please.
I have the pipe chuck in 4th Axis. And Need to drill hole on different angle and distance with X travel .
Thanks you very much for your help.
Tony
Tony, are you using a cam software for your g code or trying to use the internal software for your cutting? If you want to shoot me an email insted of going though RUclips, amhknives at gmail.com
Nice informative video Alan. I'm having trouble with the incremental G-code command. Acorn says the G91X-.5 line is a 502 Invalid G-Code and won't continue running it.
Are you running that with a space between the G91 and x-.5?
@@AMHKnives I've tried both and to no avail. Same error code.
@@adrewfis925 sorry bud, I don't know the answer then, I would post on the centroid acorn forum, they should be able to help you there.
Hello Alan
I need help. I dont have precision. I have cnc router with dmm servos. I followed your video of ballscrew mapping. I do the complete travel I return to make the home every time, I make a step as the software asks me to, but when I return to zero and verify again the travel loses the precision again. I don't know what to do. It seems I lost buying acorn cnc. Could you help me please. Hector.
Two quick checks, on the screen that has your motor steps and backlash make sure the ballscrew comp column has y in it turning on the comp for that axis, also after you do the comp save and restart the software, then verify that your comp table is there, you can do it in the software or in c: of your computer. If these dont work ill shoot you my email and we can try a few others.
@@AMHKnives
I did the process again but the result is the same the precision is lost. be happy if lost by .0015 or.002 inc is nice. but it is lost by .005, .006 up to .008 inc that does not work. I have experience in diy cnc previously I used path piloth software it worked very well but when I made the change to servos the path piloth did not want to work with the dmm servos. I made the change with mach3 but almost always lost communication with the computer. I made the change to centroid looking for improvement and I find this problem very bad. Could you help me with a more detailed video of how you did the process
@@deborahcarattini765 a few questions, have you made sure your pulse per rev is dead on for each screw and that your backlash is completely 0 or compensated to 0. Also how are you measuring your movments? Have you checked to make sure all your compensation numbers are in the software and a at least .5" for each measurement. You have to be sure 100% that you have no mechanical issues before you even try to comp. Now when i did my final compensation on my machine u used a laser interferometer for the movement, i talked about it in another video. If you have more questions its easyer to message me at amhknives at gmail dot com
thanks Alan , i would like to know how to cofigure the wizard using DYN2 and DYN4 servo driver from DMM, I was trying but when selet that config. on the wizard, I can"t use MDI into the centroid software to enter gcode directly.
sorry gabriel i know nothing about the dyn2 and 4 servo drivers, I recommend you contact centroid directly or go on the centroid forum for help with this problem.
Do I turn off backlash comp before I do this ?
Yes, do all the calcs before turning on comp then test the movements afterwards with the comp on.
@@AMHKnives Thanks for the reply
1 μm! Sweet! Would you recall the price and do they require any signal translation to connect to CNC 12 board, or it’s just straight NC/NO switch. May I ask which laser interferometer you incorporated in your design? Thank you very much! Best!
When i bought them they were $90 each and they connect directlyto the acorn board. I was running an older hp 5517d laser with all the mirrors and control board from sams laser faq but i have just upgraded to an optodyne laser.
Thank you Alan! Much appreciated. Would love to see the knives you making :)
P.S. Which Metrol tool seater did you use?
Best of luck!
@@arthursargsian thank you, im actually not running a metrol tool setter, i have a renishw ots tool setter and omp400 probe with a renishaw optical eye controling both.
Alan,
is the metrol PTP5M1CB push button activated or optical or what? Thanks.
They are nc switches that repeat .0005mm
@@AMHKnives
Thanks Alan!
Alan,
what does the ka200 do in this technique of yours? Thanks.
I use the ka200 and the glass scales to meause the movement, now i use a hp laser interferometer for more accuracy.
@@AMHKnives
I guess what I'm not getting is why you need the Mitutoyo _and_ the glass scales/KA200 for this test? Thanks, Alan.
@@crookedriver2079 sorry about that, you use the scales to measure the movement and the ball screw mapping software uses thes numbers to adjust movements of the machine to make it more accurate. Say you want it to move 1in and it moves 1.005" you can have the software fix the .005" deviation and then the mill moves 1.000 correcting the fault in the ballscrew movement.
Alan, can you provide the model number for the laser interferometer that you use? The Dro on my x-axis (lathe) is not accurate enough for me and would like to go the laser route if it's worth it.
Very interesting XD
I'd like to see this repeated going the other way. Something seems awry. Home, go to your extreme position, move back .5", then zero the DRO and move back.