Reviewing this thread it looks like soldering the boards in the arms provides some level of solution but also allowing for static build up to reduce also helps. That said there is no data on what is causing the behaviour.
I’ve Telnet’ed to the Maslow but the output doesn’t give any detail. Is there a log (or log increase setting) that shows how the belt sensors/motors, etc., are being controlled?
I’m still seeing the push to extend issue on my bottom left arm. I’m on 0.75.
Thanks,
frank.
Oh yes, the issue had happened off and on several times and this last time around I could never get the chips to switch back. Saturday I tried unplugging overnight without success.
That is excellent feedback! Thanks for confirming that!
I ordered some new boards that had the directional pins grounded and the Maslow is back in action and working much better.
@bar - These were the boards that I ordered last week that were pre-soldered. So far so good - will keep an eye on them and update if I see the behavior where belt extension is not functioning. Just an update/FYI.
Excellent, glad to hear it!
I have never had this issue before today and have gone through the setup dance easily 50-100 times. Updated to Firmware 0.78.1 and it started happening immediately.
I see people are leaning towards this being a physical grounding issue but based on my experience it feels more software related. It’s the only thing that has changed for me.
After a few power cycles it seems to have resolved itself for now. Tried it 4-5 times today
Don’t do what I did! Important note!
You dont actually have to extract the screws holding the motors on to do the soldering……
I basically wrecked all the holes in all the arms just trying to get the stripped Allen heads screws out, used plenty of thread locker of course, so after using, superglue, a Dremel with grinding stones and diamond bits, drill bits, extractors etc. etc I got one set out, then only to notice that you can just take a bearing out to get access to the board to make the solder bridge 
.
Now to see if it holds together after I reassemble everything, AND to see if my horrible soldered ground bridge works.
Thank you for this insight, I will definately keep the motors on when I try to solder the bridges!
threadlocker comes loose with heat, maybe put the soldering iron on the remaining pieces of the bolts to get them out?
Got them out with pliers by gripping and turning, good note about the heat though.
Woop woop, after pulling it apart, mashing the arms to get the screws out to solder to ground, putting it back together 3 times as had found 3 bearings left over 
and then remembered I forgot to tighten the grub screw on the gears, it’s all back together and running. Got calibration going and it’s alive! No more push to extend bug!!! Everything worked as intended…fitness is too low but will work on that tomorrow
Hi,
Yesterday I had also the same bug. I soldered the legs to the capacitors and printed overnight the belt guards (For soldering I took a very small piece of copper wire, actually only one strand of wire). Today morning I assembled everything, run calibration and now everything works as intended. Perfect!
I also soldered the leg without removing the motor. Only the bearing needs to be removed.
I think, that maybe humidity is a factor. During summer my workshop stays really cool. Because of that, the relative humidity can reach up to 80%. clintloggins had a storm before this bug happened. Could also be related to humidity rising. Maybe a temporary fix could be to warm the encoders a little bit with a hairdryer and thus reduce the humidity. Maybe someone can try this out for further investigation.
Bye Markus
Awww man, wish I had seen this before I attempted to remove the motors. 2 stripped bolts on 1 motor and I mangled the plastic pretty good. 2nd motor got lucky. Will do the grounding through the bearing hole for the last 2. Thanks for pointing out the better technique!
