A couple of days ago another Maslow user from my town asked me, whether I could take a look at his machine. He guessed that someting went wrong with the calibration of his M2. Since I’m a happy user of the original Maslow (controlled by WebControl), I’m not so expierenced with the M2.
At the moment the guy controls the machine with a local app on his Mac. Since he have seen my Maslow with the WebControl on the Raspi, he wants to upgrade his machine.
#1 | If we get the M2 calibrated with the actual setup, I guess we can switch from the local app to the raspi without any hassle. Right?
#2 | Does the Makerverse run on a Pi 3?
We went through the calibration process twice and were not able to improve the result.
By now we did not touch the chains. So should we start from the very beginning and reset the chains and sprokets too?
@Orob What do you think? Should we do a fresh start and reset the chains?
if you have a center point mark on your board, drive to that spot. I may have mentioned this previously, but when my system was really far off, I adjusted the beam and very carefully measured the offset and motor distances. My Y offset was really far off to the point that the sled was driving off the work piece to punch the hole. Resetting those and getting an accurate measurement made a huge difference with the calibration results. It is like 3 mm now. I realize it isn’t 0.8 like some have been able to get, but it is good enough for me. Is there some error (or “uncertainty” if you are into being technically correct) in your measurement? Is your work space exactly 48 x 96? that you are using for measurement? Do you have an exact middle spot marked on it? I’ve been using webcontrol most recently, so my makerverse isn’t fresh. I was frustrated with the random offsets and trying to keep it calibrated. I have two controllers and I just swap the Due for the Mega and the raspberry pi runs whichever one I start. Here is the video I made on it that might be useful regarding calibration.
the classic to M2 calibration change makes the frame build exactness matter more. Generally speaking, a poor quality build is a bad idea every time, but I’m talking about “is the beam exactly level. Is it parallel to the floor, is it exactly parallel to the top of the workpiece.?” In my simple and limited opinion, I think the M2 frame build matters more because you are using the frame as reference for the calibration vs counting chain links and using a series of holes in space with some offset. The holey calibration took the distances from each other vs the distances from the edges. You better have really good edges or draw some if you want makerverse to give you good results. My small 4x4 system has a 5’ high work piece and I had to draw a 4’ line on it and use that to measure. If that line isn’t exactly straight, we are talking 2 mm offset from just that. the holey calibration on the older setup where you measure from hole to hole is less about your frame build quality and about getting the chains to a known position… just my meandering thoughts. In any case, the beam needs to be rigid and not deflect when the sled is at the top center. I was using a 2x4 mounted to the side of a 2x6 and the 2x4 was bending… not good. Make changes. The measurement chart here is useful. M2 to me means “Measure and Measure again.” Once you get it, you should be good to go. Generally speaking your measured offsets should be within a couple mm. if they are not, then check center location and your motor-motor spacing and y offset distance.
Hope I didn’t lose you in the ramble. Let me know how it works out.
Thank you for your support. We were finally able to calibrate the machine and the guy is very happy with the accuracy (3.4mm).
Now he will move from the local installation to the Raspi.
That is great news! I think the raspberry pi on this machine is one of the best uses of a single board computer.