Version 1.14, calibration with 1500x750, 7x7, concrete floor
everything looked good. Yesterday I was indignant, because I would have thought why, when he finds anchors, he still has problems with the axis scales and a person adjusts hundreds to thousandths of a percent in the axes, so that 1000mm cuts 1000mm.
since everything looks ok, I decided to cut today (fortunately only templates made of hardboard)
none of the attempts worked, when testing I made reference points on the spoil board about 1x2m apart, so when moving to the first one, the strips were so loose that it broke.
now it stopped in the middle of my work, I’m adding the serial file.
I’m tired of it. I don’t understand how I was able to cut anything at all in the beginning. sometime in the summer. now even the same settings in the previous version don’t work for me anymore. it just keeps cutting a few mm differently, or now it doesn’t cut at all.
where is the problem? in my HW? connection? or firmware?
The issues that you are seeing are absolutely caused by this error message.
[MSG:WARN: Position error on Bottom Right axis exceeded 15mm while running. Error is 15.020mm Counter: 1]
Unfortunately there are a couple of different things that can cause that message.
Basically what it’s telling us is that something is causing the machine to not be where it wants to be and it’s shutting down to protect itself.
With the 4.0 hardware this issue was often caused by a bad connection between the encoder boards and the controller board because the connectors could get dirty. The update to JST connectors in 4.1 largely fixed that, but if your machine has been sitting for a long time it might be worth checking those connections.
From looking at the log I’m also seeing that it took a couple tries to get the machine to be happy with the frame size:
[MSG:INFO: Center point deviation: TL: 0.000 TR: 0.000 BL: -25.334 BR: 5.258]
It looks like maybe there is something going on with the bottom left or top right arm.
One good test to see what is happening there is to do Retract All → Extend All → Retract All and post the log from that. What we’re looking for is to see that the offsets are small (close to zero) indicating that after extending and retracting the machine knows that each belt came back to the same spot where it started
thanks for the help. Unfortunately I’m still not sure what to do, but I’m making progress.
In the afternoon I had a new calibration done with a larger range and grid. I noticed that the anchors were found in a much different place than in the previous case.
can you grab a tape measure and manually measure the 6 distances (including diagonals)
If you put a hex bolt in the anchor holes, you can measure flat to flat instead of trying to eyeball the center of the hole (just use left edge to left edge for consistancy)
the belt extension/retraction looks good, but if you found that the connections were loose, that could easily have been the biggest part of your problem.
i won’t start a new topic, my problem is probably dragging on.
i just tried several futile attempts to cut something into hardboard.
it always stopped in the middle of cutting. i’m attaching the serial code and also a video of what was happening during cutting and when it stopped.
please try to advise me how to proceed, i’m starting to get desperate that i won’t be able to cut anything anymore. this was a part of approx. 1.7m x 0.8m, part of a template for other boards to fine-tune the dimensions.
i’m on version 1.15 and it cut test rectangles 900x100 and 100x900 with an accuracy of 0.5mm.
it just doesn’t give these larger parts…
I would like to ask about manual moving, I don’t like that behavior, as if it moves, loosens the motors a little and then moves another 1-2cm somewhere, I don’t know where exactly, maybe to the right place, but is it supposed to slide into position like this?
I had that problem, and it was because the spool was building up tension and closing so tightly that it wouldn’t turn. To keep up with the other parts, the motor was exceeding 2200 RPM, and the machine would stop, leaving a messy cut. My solution was to smooth and sand the spools, especially the one that was drawing too much power.
At first, I had no intention of disassembling the arms, so I would cut a maximum of two small pieces, extend all the belts until I saw the bare spool, and retract it. Then I would make two more small cuts and so on until I finished, because I thought the motor was faulty. But I discovered that the steel belt accumulates tension inside, and the motors require more power to move. That’s why I decided to sand the spools until I could work with a retraction force of 800, and I saw the light.
the value of 2200 was only shown after the machine stopped during the process, I took a picture of the motors and there were about 600 to 700, it alternated between different individual ones.
also notice how one belt is completely loose during cutting. that’s why I actually filmed it.
normally I have a retracting force of 1200 and it always retracts properly.
of course I’ll start disassembling the machine, I’ll delay it a bit.
in some places on the surface it feels like the machine is clicking somehow, I think it could be the belt protector, I think I’ll sand it down too.
yes the whole time the engine values were up to 700. , some were much less. obviously those in the opposite direction than the direction of the machine.
it stopped 2x in the same place and with the same purpose. Then in the second case I noticed that one engine was over 2200.
basically I can try it again before I start disassembling the machine and documenting it more. That is, if it really happens again.
If it always happens in the same place (especially if you can make it happen there by using just the jog buttons) that is a strong sign that it’s not a mechanical issue.
The issue happening in just one place is a good indicator that the issue is that the belts could be pulling too hard which would be an issue with where the machine thinks that the anchor points are
I found out that it is not a specific arm motor, but it is always the motor that is pointing towards the corner where I want to move. So all the motors do it, always at their corners of the 2.5 x 1.25m work surface.
I drove the machine manually along the outer lines and watched how the motor current changes from one extreme to the other.
e.g. here in the upper left
and lower left corner
So when cutting it is even more and it stopped again at the same place of the cut part again due to overcurrent
any ideas?
otherwise retracted and extended belts work without a problem. I have 1300 and 1200 retracting forces.
will grinding the coils help? or are these values in the corners normal? or is it Z-offsets? should I raise the anchors to the coil planes?
why are these forces not always the same on all motors anywhere on the surface? horizontal mode.
I think that is a good idea and certainly wouldn’t hurt
How close to the corner are we talking? It does get to be more and more of an issue as you get closer to the corner. Check out this simulator to get a sense of how close to the corners it is reasonable to go: Anchor Point Layout Simulator
That is an excellent thought! If you have the ability to test and see if that helps that would be really valuable information
hi, I’m all in the green field in the simulator. I have a 4400x3600 frame, I don’t know exactly now.
I tried to raise the anchors so that there are no offsets.
The motor currents may have decreased a little by about a few tens of mA. However, the extreme positions go above 2000mA. But I managed to cut off the test section where I stopped.
I also noticed some instability of the balance on the sled.
I caught it on video.
I’m going to disassemble the machine and grind the coils. Check the overall mechanical condition.
I’m a bit surprised by this play. The lower parts of the plastic rollers are broken. I hope the belt can’t jump.
also the belt sagged when moving to the other side
Any ideas about tilting?
Why is the motor current actually increasing? I can’t imagine it, I think that if I moved the motor with 700mA across the entire surface to the anchor, I would pull the machine there.
In the extreme corners of the area when there are three belts pulling all in one direction against the one belt pulling in the other direction it makes sense that the current would be higher.
Maybe we need to detect that situation and reduce the tension in the belts?
