Glowforge recently replaced my original unit, and for the most part the new one is working fine. (It definitely doesn’t have the same problem that my first one did.)
However, the new unit is exhibiting an odd behavior on startup. Almost every time, when first powered on and beginning calibration, it makes a loud rumbling sound as the head attempts to move, and the Y-axis belt slips. It repeats this behavior every few seconds until I power it off.
If I manually position the head under the camera before turning it on, then it calibrates normally. But the next time I power it on without manually centering it, it again fails to calibrate and makes the loud belt-slipping noise.
My first unit didn’t do this, but I know that others have encountered the issue:
How do I resolve this issue permanently, so that I don’t have to manually position it every time?
It’s probable that during the initial calibration routine the camera S/W can’t figure out where your laser head is positioned. It thinks the head is in a different position than it is and tries to move in the wrong direction. In your case this seems to be to the left against the stops?
Possible causes:
Shadows or reflections from overhead lights
Magnets on the bed or material with outlines that confuse the camera
Cut outs in material left on the bed that confuse the camera
A dirty camera lens
Image artifacts outside of your control
Some of the above you can control. The last you can not. There may be many other causes but those seem to be common. If you can’t control what the overhead camera sees then manually moving the head at startup to make the position obvious is one solution.
That probably happened to me 6 or 7 times in 7 months. It seems to be related to confusion of where the head is at power up.
It bother me enough that I am in the habit of hand centering the head before powering it on. No big deal for me.
Yes, the problem occurs when the head is in the default position at the rear left.
What’s weird is that the new unit is located in the same place as the old one was, so environmental lighting isn’t the issue. The bed is typically empty when this happens, so it’s not magnets or debris. I’ll double-check that the lens is clean, but I don’t think that’s the issue either.
You might fall into the category of the camera being a nano-dust particle in the wrong direction to reliably recognize the head. Some have no problems, some have many. It will get better eventually.
It seems that 95% of the folks with bumping issues see the unit travel to the left stops from the park position. Not sure why they don’t have the head move to the right and down just a couple inches closer to the field of view of the camera at the start of calibration. But certainly not going to be one of those folks that always think something is easier, or a higher priority than it may be. Anecdotally, Glowforge has a history of trying to fix a problem completely rather than provide patches. Frustrating to us but not going to pretend I make better decisions.
I have this issues each time the printer is turned on.
I have placed a large sheet of foamboard over the top while calibrating since it has it’s own lights so environmental lighting cant affect it and it still grinds. I can’t imagine a spec of dust on the lenses could do this.
I too manually home the head before starting, not a huge deal since it’s early, but hopefully this is addressed in a future software update? The machine should know where the laser parks itself once a cut is complete since it homes itself.
Yeah, this has happened every single time that I have turned on my machine the past couple weeks. I think I have turned it on and off about 5 times within that span. I have even left the gantry in place let it calibrate and wait till it says “ready”, then turn it off and on again without even opening the lid, to the exact same thing.
Like you, it seems like it would be easier to just leave the x axis alone when the Y axis moves down for the first time. I am not sure why they have it move the Y axis down while moving the x axis slightly to the left. This almost seems like a small error that they just haven’t bothered to fix for some unknown reason. I don’t think it will eliminate the issue with it losing it’s position, but it would at least stop the head from ramming into the x min while only the y axis is moving. That just causes unnecessary wear.
Where is the open source firmware? I only got a 6 month warranty, that will be up in a couple months. I want to know that I can fix this thing if it goes belly up after a couple months. Such a small warranty for something that I have put less than 30 hrs on…sorry. Rant over. I just cant believe they haven’t fixed this issue yet.