I’ve now attempted to cut my first design about 12 different times. 2 of them have finished successfully, but 10 of them failed in various ways. A few of those were my fault, like using tab settings that were too small, etc. But the vast majority were from various failures of the Maslow.
I know there’s an ongoing theory with ESD having a vacuum hose attached, but I’ve tried running it without and that doesn’t seem to make any difference. In fact I’ve tried a lot if different ways to attempt to alleviate any ESD potential. This most recent failure, I did have the vacuum hose attached and running, but I have a bare copper wire running inside the length of the hose, that is ultimately attached to ground on the vacuum’s power cord itself. The copper wire isn’t attached to anything on the maslow, though it could be.
Most of the failures, like the one I just experienced, was the result of an “Encoder read failure” ( see attached log ), but that’s not always the case. I’ve had the machine just shoot off in a random direction and error out as well.
what firmware version are you running? the ‘shoots off in a random direction’
was something we had early on due to a network related bug, but it should have
been fixed long ago at this point.
The versions I’ve upgraded to since I got the Maslow was v0.74, v0.77, v0.79, v0.80, and today v0.82. The shooting off in a random direction was one of the 0.7x versions for sure, but I’m not sure which exactly.
The “Encoder read failure” from today was after updating to v0.82.
We’re working on a more permanent fix for that issue, it seems like the cause is dust and vibrations messing with the wires connected to the encoders. Have you tried the hot glue trick? That fixed that issue for us in testing.
Thanks! This issue could be very costly for me! I use cast acrylic to make LED neon and a cut messing up midpoint would not be good! Currently doing lots of testing first so I’m glad I found this issue now!
I will test out the hot glue approach this evening.
For what it’s worth, I have an ethernet cable tester and all 4 cables passed about a dozen tests each while flexing both ends in different directions. Fairly confident I can rule out cable defect or quality at least.
I started getting some “Motor current on Top Left axis exceeded threshold of 4000” warnings toward the end of the last cut and it left me with a wobbly edge. I already determined in another thread that I need to take my arms apart and be a bit more generous with the lubrication
Using this thing just went from extremely frustrating to extremely satisfying, all thanks to some hot glue!
Given that I have to take this whole thing apart now for the aforementioned lubrication, I may peel off the hot glue from one side or the other and see if the issue is isolated to either the ethernet plugs on the encoder boards or the main board. For Science!
The difference is night and day! Two more successful cuts in a row!
great to hear
I started getting some “Motor current on Top Left axis exceeded threshold of
4000” warnings toward the end of the last cut and it left me with a wobbly
edge. I already determined in another thread that I need to take my arms apart
and be a bit more generous with the lubrication
the other thing you can do is slow down the feed rate. The ‘current hits 4000’
is saying that the machine is trying to move, but can’t move as fast as it
thinks it needs to.
that leads to the wobbly cuts as it’s trying to catch up, and overshoots…
Given that I have to take this whole thing apart now for the aforementioned
lubrication, I may peel off the hot glue from one side or the other and see if
the issue is isolated to either the ethernet plugs on the encoder boards or
the main board. For Science!
thanks, experimentation from users out in the world is how we find these things.
Bar did not see these in his pre-release testing (or he would have fixed them
then).
you can also see if you can shift the motors a hair out from the idler. There is
play in those screw holes and just a little shift can result in a significant
decrease in the power used by the gears meshing.
Mostly just a good small test for dialing in the Maslow, plus I’ll save a bunch of plastic in the process. I still printed the wall mounts, after widening them to fit the 3/4" plywood.
Those 2 ends ( and dividers ) were the first and only two successful cuts before today. Adding another row below this one with an additional wall support in the middle of both.
