Headed bed and noise issues
Posted by Ralf
|
Headed bed and noise issues February 13, 2013 04:45PM |
I have been having "fun" the last coupple of days trying to get some decent quality out of my new prusa.
Very long story short, I wanted to upgrade the printer, so I ordered a heated bed, and some alu pulleys from T3dP3d. For a start I did the first coupple of test prints with the HBP hooked up to my bench PSU, and the prints looked fine.
Then I took the printer apart, upgraded the belts, and hooked up the HBP to my sanguinololu, and then the fun started. All my prints had a "vinyl record" texture to them, I tried a lot of different things, like slowing the print speed down, adjusting motor current, and different combination of settings, but nothing worked. Though I did found out that I had a problem with filament supply, but that wasn't the main problem. Even reverting to the old T5 belts and printed pulleys didn't made any difference..
Then I thought about my first test prints, wich was done with the HBP hooked up to my bench PSU, so just to try something different I tried that again, with a dramatic increase in print quality as a result.
Then I set out to test things further, and hooked up my oscilloscope the supply for the HBP, and there it all was:
When the HBP FET turned off, the HBP behaved like an inductor, and send noise back to the sanguinololu.
This issue was solved with a capacitor and a diode in parallel with the HBP, and then it looked like this:
However print quality didn't improved, it was still crap, the right one is when the HBP is connected to the sanguinololu
The supply is a decent quality Chieftech hacked ATX supply capable of 12V 17A, I am not 100% sure, but I think that I used it for my first reprap project, also with a heated bed, but of my own design.
Have anyone expeirenced similar problems?
Very long story short, I wanted to upgrade the printer, so I ordered a heated bed, and some alu pulleys from T3dP3d. For a start I did the first coupple of test prints with the HBP hooked up to my bench PSU, and the prints looked fine.
Then I took the printer apart, upgraded the belts, and hooked up the HBP to my sanguinololu, and then the fun started. All my prints had a "vinyl record" texture to them, I tried a lot of different things, like slowing the print speed down, adjusting motor current, and different combination of settings, but nothing worked. Though I did found out that I had a problem with filament supply, but that wasn't the main problem. Even reverting to the old T5 belts and printed pulleys didn't made any difference..
Then I thought about my first test prints, wich was done with the HBP hooked up to my bench PSU, so just to try something different I tried that again, with a dramatic increase in print quality as a result.
Then I set out to test things further, and hooked up my oscilloscope the supply for the HBP, and there it all was:
When the HBP FET turned off, the HBP behaved like an inductor, and send noise back to the sanguinololu.
This issue was solved with a capacitor and a diode in parallel with the HBP, and then it looked like this:
However print quality didn't improved, it was still crap, the right one is when the HBP is connected to the sanguinololu
The supply is a decent quality Chieftech hacked ATX supply capable of 12V 17A, I am not 100% sure, but I think that I used it for my first reprap project, also with a heated bed, but of my own design.
Have anyone expeirenced similar problems?
|
Re: Headed bed and noise issues February 14, 2013 01:01PM |
Spot on.. 
I measured with a dial indicator, which showed approx 0.1mm movement up and down within the first second after the heater turned on or off.
Reason.. The glass plate I am printing on is missing a corner, so I rearranged the attachment to 3 point instead of 4, and away from the broken corner. Then I set the bed temp check interval to 0.5 sec, and now the movements is in the order 0.01 to 0.02 mm, so now I guess that I'll have some other tolerances taking over..
I measured with a dial indicator, which showed approx 0.1mm movement up and down within the first second after the heater turned on or off.
Reason.. The glass plate I am printing on is missing a corner, so I rearranged the attachment to 3 point instead of 4, and away from the broken corner. Then I set the bed temp check interval to 0.5 sec, and now the movements is in the order 0.01 to 0.02 mm, so now I guess that I'll have some other tolerances taking over..
Sorry, only registered users may post in this forum.