As I was redoing the calibration (vertical orientation) I noticed that when the cable was being slackened it very often did not completely go out through the encoders (instead it bulged out on the spools), so I gave it a quick tug in the split second before it would tighten up again. I’m not sure whether this has any impact on this first pass of calibration or not. But it could explain some other problems people have with cables being chewed up etc.
1 Like
You are correct. This is the primary reason that Bar changed the calibration routine to only do a small 3x3 first to get enough fitness to make the belts tighter on the next phase, and spiraling out, getting better along the way. Initially it was doing the whole calibration that way and a lot of people ended up with tangled belts. You do have to “manage” the belts during calibration in just the way you described. In vertical orientation, this is just the bottom two belts.
1 Like
I just finished building my arms and was testing retraction and release of each arm using 24V power supply and the same release behavior bulges on the spools unless you pull on the belt.
Though I swear I saw a video where Bar shows calibration and the belts were spooling out without having to managing them, so I was also going to ask this same question.
1 Like
Mine do that too. As long as the 3x3 grid is small enough I find that it won’t spool out so much that I need to stress about it, but it’s 100% there and that’s absolutely the reason we changed the system to work the way it does now. It was a big mess before 