i have a k8200 with stand alone controller board, printing from SD card.
I have recently been experiencing a new phenomenon, that is, the print job starts and goes okay for some time (usually 2 hours +) and then the bed shifts - only on one axis but by varying amounts from as little as 0.5 mm up to 25mm, the printer continues on as if nothing happened.
I have spent some time testing all of the motor stepper drivers and all four x,y,z & e are set to 0.55 v (+/- .02v),
the rest of the machine is working as it should - the events are too random to diagnose, but too frequent and destructive to ignore.
small objects rarely shift, but that could be down to chance rather than definitive.
You didn’t say what axis they were shifting on.
If you are running V1 firmware and the shift is happening on the Y axis you should check the wires for the Y end stop.
If they are sticking out and rubbing on the back of the frame this may cause an open on the end stop and cause the shift.
If this is the case you need to do a few things.
Check the solder joints on the switch and repair.
Zip tie the wires so they don’t stick out.
Upgrade the firmware to V2.
V2 will ignore end stop hits after the printer has started printing.
I have updated to version V2 but unfortunately I still have this unwanted shift X direction. I tried to change the cables of the endstop switch. I double checked the voltages in each motor driver, but everything seems to be ok.
I apprecieate if you can help me with this issue because I don’t know what else to try.
Some linear ball bearings show serious stick slip properties. I have replaced some of them which seems to be effective. The resistance can be tested by disconnecting the belt from the pully and moving the flatbed very slowly by hand. In my case I could feel moments of resistance. I have measured a force of about 0.4 kgf to overcome the resistance. This is about 1/3 of the force which can be delivered by the stepper. But in combination with a fast acceleration of the flatbed I think this can result in loosing steps.
I have the same problem Sometimes Y axis Sometimes X and sometimes both axis at once. If you ever hold the belt during a print the amount of torque the steppers are putting out is very sizable. Unless you have an extremely sticking bearing I doubt if that’s the problem. I am wondering if its a burp in the G-code feed from the computer. I assume the controller board does not have a buffer.
I am running the stepper voltages at .55 and by feel when this happens the stepper drive chips are not hot. Very frustrating after a three hour or more print and has become an real problem.
[quote=“calumetflyer”]I have the same problem Sometimes Y axis Sometimes X and sometimes both axis at once. If you ever hold the belt during a print the amount of torque the steppers are putting out is very sizable. Unless you have an extremely sticking bearing I doubt if that’s the problem. I am wondering if its a burp in the G-code feed from the computer. I assume the controller board does not have a buffer.
I am running the stepper voltages at .55 and by feel when this happens the stepper drive chips are not hot. Very frustrating after a three hour or more print and has become an real problem.[/quote]
I am having exactly the same problem. The jump is in X and Y simultaneously. This happens three or four times in the process of a very simple rectangle print 20 x 30 mm x 2.5 thick which I am using as a test piece.
I printed the board cover successfully and one other file I created but since then I have had nothing but these random moves which ruin every print.
Check the wires for a bad contact (especially the plugs on the mother board - A bent contact is enough) and/or for a broken wire/solder in the flat cable.
The steppers do not have much reserve capacity. So be care that every runs as smooth as possible. Beside I have replaced some bearings I have also paid attention to the pulley, guiding the toothed belt. To minimize the friction I have made a design which allows free rotation of the guiding rings.
When I was replacing the bearings I also observed that the position of the clips keeping the bearings in place is important. When the opening is directed too much down wards it will touch the axis in x-direction.
This sounds like you skipped over the part of setting the reference voltage.
Out of the box they are set to 0.80
This will cause the drivers to over heat
Set them to 0.55
that the bed is free to move - It is
the reference voltage is set to 0.55 (was at 0.425)
all cables appear to be good - no bent pins
What puzzles me is that the bed shifts in X and Y simultaneously and at random times in the print.
I am wondering if the serial connection is too fast for my computer or that the buffer overflows or something else associated with teh communications with teh computer.
I’ll try resetting the jerk and acceleration. My voltages are fine. .55 belt tension free running bearings ect. Wonder if a brown out in power could cause this. Am going to provide a separate 24 volt power supply for the bed after I try the new jerk and acceleration settings. I only like to try one new thing at a time else I have no idea what may have solved the problem.
Some of the things I have done to try and rectify this issue, it hasn’t reared its ugly head in a while so I suspect one of these has fixed it:
[ul]Printed a new sturdier X Motor Mount that sits within the aluminium extrusion perfectly http://www.thingiverse.com/thing:253741[/ul]
[ul]Lubed up the linear bearings and rods with lithium grease (will be switching to inox soon)[/ul]
[ul]Put heat sinks on the stepper drivers[/ul]
[ul]Added a belt tensioner to the x axis http://www.thingiverse.com/thing:130235[/ul]
[ul]Decreased the acceleration to 1000 instead of the default 3000[/ul]
[ul]Decreased XY Jerk to 15 instead of the default 20[/ul]
[ul]Reduced max feedrate to 300[/ul]
Things that I will be doing in the next week or so
[ul]Replacing all the belts and pulleys with GT2 timing belts[/ul]
[ul]Replacing the stepper motors with newer lower current versions. The ones that came with mine are 2.5A motors, the max current of the drivers is only 2A. Although we don’t operate anywhere near that value[/ul]
[ul]Replacing the controller board with a smoothieboard. This board comes highly recommended and is 1/2 the price of the original board. It runs on ARM technology rather than the very outdated and limited AVR technology[/ul]
I will make a post once I have done all these things plus some other stuff, so people can learn from my mistakes and triumphs!
