changed to: the svn version from tonight, and using the same settings, except:
have_heated_bed=1, and bed temp set to 0.
Display is normal, and it worked.
Tried a second print now, bed temp set to 50. The printer rebooted as it had reached both temperatures and was moving Z down to start the print:
The printer is set to 1/32 microstepping. Z axis steps/mm is 8000.
With Z feedrate (and homing feedrate) set to 5 the display gets a bit sluggish - I guess the little processor has too much to do and I guess the watchdog kicked in and rebooted it. This didnt happen with the previous version.
Tried again with z feedrates set to 2, and this time it worked.
Babystepping is a very nice feature. I use it on every print now, instead of adjusting the z probe height to the correct value
(And yes I know 1/32 is a bit pointless, and 8000 steps/mm is stupid, but I built this printer just for fun and to test the "Diamond Hotend", so I try lots of strange things. Will go back to 1/16 step on the Z and extruder axes the next time I open the "electronics box")
I have repetier server on one of the older PI models (B+ I think its called). Was going to use it for my two printers, but moved them back to the PC to debug this problem. If you need testing, let me know.
Thinking about it, many of my problems may be due to a very busy processor, I guess dual X, Y Z steppers with 1/32 microstepping on all of them, and auto bed leveling that continously moves Z increases the amount of work it has to do in one "cycle". What happens when it runs out of time?
1. Your com error i sthe real com error as it looks. These chars make no sense in responses so i think you are realling having some bytes changed. The frequency should be no problem so nothing really to worry about.
2. Too high feedrates can trigger watchdog. You spend 99,9% in timer interrupt not letting enough room to reset watchdog in time. Fix max z speed and your problem is gone. New versions have it a bit more difficult since watchdog reset requires temperature manager run every now and then plus interrupt being called. So a blocke dinterrupt can not cause an error.
1. My Z probe connection was either loose/faulty or something strange was going on. After opening the electronics box the display kept flickering when auto leveling. I reversed the Z probe connection, and flickering stopped. (Swapped ground/signal the other way around). I know it makes no sense, and this shouldnt matter as a microswitch is either connected or disconnected. But it helped, so I guess it was a loose connection.)
2. A thin 12V wire going to one of the fans (the always on hotend cooling fan, I think) was trapped between VOUT and GND on one of the DRV8825 stepsticks (under the stepstick PCB). I removed the stepstick and made sure no wires were trapped between its pins.
3. I set the two Z steppers to use 1/16 microstepping.
The result is that there is absolutely no slowness in the display/control, the com errors happen every 4 minutes instead of every minute (or more), the servo is less "jittery", the auto leveling is more accurate (still not 100%, there seems to be 0.1 to 0.2 mm difference between left and right side of print bed, it used to be up to 0.5mm even with 5 probes per point), and the display is no longer flickering.The "extruder stops extruding issue" is most definitively solved for me, so I will stop posting in this thread now
Today I learned an easy way to "clean" the nozzle/hotend (nylon cold pull) and I am now trying to force the heatbed to "hold on to" nylon.
Comments