Inconsistent calibration results

[calabus]

OK...so I have gone with a 3rd party hotend (E3D) and started the calibration process. I plugged in my Z height numbers and started the calibration process. I did my measurement from hotend to bed. I run the G code and bring the head down to 10mm and jog it down until contact. Take the displayed Z height number and subtract like always in the printer settings and the firmware. I bring the hotend down again and it is still off by 1 to 2mm. WTF? I adjust nothing and bring the hotend down again. Still off by the same amount. OK... I subtract from the Z in firmware and in printer settings. Run it again and the hotend hits the bed way before Z 0. Rinse and repeat. I have checked the head and it is tight. The stops are tight. The belts are tight. No slack anywhere. Everything is how it should be. Before replacing the hotend and using the OEM hotend I ran this calibration without issue. It was dialed in perfectly. The only thing I have done software wise was update RepetierHost to the latest version. Any suggestions? Am I missing something?

[lordbinky]

If everything is TOO tight, you're stepper motors could be missing steps. You can either try to check the tightness again or turn up the stepper motor's power in the firmware temporarily to see if it alleviates your problem.

[calabus]

I don't think hardware is the issue. I say that only because the only changes I have made since I had it perfectly calibrated before was the hotend, which has no play in it, and RepetierHost. It seems like making the firmware and printer settings change it isn't getting properly applied between tests. If that makes sense.

[foshon]

Try this once, plug your changes into the firmware, upload. Open rep host and make the same changes to your eeprom. Disconnect from the printer and then reconnect. I have to do it every time I make changes to have them stick reliably.

[calabus]

foshon

I should have been a bit more clear in what part of the config I am adjusting. When I said firmware I meant the EEPROM config dropdown in Repetier and not the arduino firmware adjustment. WOuld the same process you described still hold true?

[calabus]

So just for the heck of it I tried foshon's suggestion. Don't know if that actually fixed the problem or it was just luck, but I can run through scripts 1, 2 and 3 and get consistent results that seem to stick. Running script 4 brings me down about .5 too close to the bed. I fired up Arduino and and made my #define printer_radius (I'm not looking at the code but I'm sure you know which setting I'm talking about). Doesn't seem to matter what amount I add o subtract in .5 increments the head seems to remain at the same height. I have tried foshon's suggestion for this issue as well, but no luck. Sooooooo close to getting this thing up and running.

[calabus]

After a LOT of numbers juggling I have it calibrated. This thread, gcode and spreadsheet help a GREAT deal. http://forum.seemecnc.com/viewtopic.php ... lper#p8409

I am now wrestling with hotend and bed temps. Since I can't print a fan yet for PLA I am starting with ABS. I see a lot of info on ABS temps. A friend at works says he prints at 105 bed temp and 190 hotend for ABS. The reading I have done says 220-230 hotend. I went to 200 and saw a lot of smoking and hissing. I went down to about 190 but the start of the print was a mess. Rolled up plastic and such.

[jmick]

I did't do mine that way. I measured the height and put it in to eprom and printer set. I then came down to within 5mm and jogged down one mm at a time until I was close and then 0.1mm until the paper rubbed. If I was off or too tight on any I turned the screw clockwise 2 turns like it said to do and started over. I kept this up and it only took me 4 or 5 turns on each position until I got within 0.2mm on all points. Some of the screws I had to back off of a few turns during this operation and I only had to change the overall dim 2 or 3 times. I ended up with a dim of 366.9mm in the eprom and printer shape.
I only have the spacers, the onyx and the glass for a platform.