Modifying Homing String

[Skullbot]

So after a while I think I’ve got pretty much everything done. The table is built, robot is set up and able to draw patterns. The only issue is that my centering is still not working. I have to manually move the arms into the correct positions whenever I want to draw a pattern which is obviously not that great.

When I try to ‘home all’ the lower arm is able to move into position properly. Due to the nature of the sensors I’m using, the upper arm sensor is slightly out of place, causing the upper arm to home an inch or so past where it should be. I thought it would be a simple fix to modify the homing string from
"FR10;A+10000N;B-10000;#;A+5000n;B-5000;#;B+10000N;#;B+5000n;#;A=h;B=h;$"
to
"FR10;A+10000N;B-10000;#;A+5000n;B-5000;#;B+10000N;#;B-250;#;A=h;B=h;$"
where B-250 moves the upper arm back enough to be centered after it first starts triggering the optical sensor.

The problem with this is that the amount that the upper arm moves past the center position varies depending on where I begin homing from. I’m not sure why this is since visually the lower arm appears to be homing to the same place each time. My only guess is that the lower arm is moving slightly more/less depending on where it starts to home from which makes very little sense to me, but I’m not sure what else could cause the upper arm to be more/less off center on different homes.

Does anyone have any advice on how to go about fixing this issue? I feel like this is the last problem I need to resolve before the table is fully functional.

[Mattyp]

Have you tried the following

“FR10;A+10000N;B-10000;#;A+5000n;B-5000;#;B+10000N;#;B+5000n;#;B+250;#;A=h;B=h;$”

Then play with the 250 untill you hit home.

[Skullbot]

I tried that in the past and for some reason it made it so the machine wouldn’t accept any commands at all. Tried it again now and it’s working, so that’s something.

Still having the same issue with the homing being inconsistent though which makes it impossible to find a number which works. What I mean is that if I set the arms at, say, 10:00 and press home, putting the last command at B-150 makes it home perfectly. However, if I home when the arms are at 3:00 then the lower arm goes slightly further and then moving the upper arm 150 steps back doesn’t bring it back far enough. If I home when the arms are at 8:00 then the lower arm goes slightly less far and then moving the arm 150 steps back brings it back too far. So on with homing in any other position. Also, I’ve noticed on occasion that the lower arm will be homing, hit the optical sensor and cause it to light up, move past it until it stops being lit up, and then ‘bounce’ back slightly causing the optical sensor to relight. This of course makes it so the upper arm doesn’t stop when it should since the lower optical sensor is still active.

Something I found was that the lower arm has some play in it even when it’s moving. When the machine is actively drawing something, the upper arm is unable to be moved by me which seems like what I should expect due to the holding torque of the stepper motor. The lower arm however has a slight amount of play; I can wiggle it back and forth about a quarter inch before encountering the resistance which stops me from moving it further.

I suspect this is the issue causing the inconsistencies in the lower arm’s homing which, consequently, is affecting the upper arm’s homing. Any insight into what could be allowing me to wiggle the lower arm while it’s being moved?

[Mattyp]

Surely this movement is in the arm build not the motor.

[Skullbot]

Yeah, the problem ended up being a combination of the slight amount of slack in the lower belt and the fact that the belt was wrapped around a screw unbeknownst to me. Seems like when the lower arm reached the optical sensor and stopped there would be a slight amount of bounce due to the slack. This bounce would occasionally push the arm far enough that the belt would be turned far enough to catch a tooth on the screw and prevent it from bouncing back into position. This then made it so the upper arm had to move further than normal to hit the upper optical sensor.

I ended up taking apart the whole thing and putting it back together. It seems like it’s homing properly in air now. On sand it’s less consistent and occasionally the lower arm won’t move far enough to stop triggering the lower sensor causing the upper arm to not be able to trigger the upper sensor. That seems like a fixable error though, and even if it isn’t, all I have to do is make it home again after the first attempt fails to get it to work.

[Author]

Posts