Really irritating RepRap problem fixed

The Y carriage movement on my new printer has been very weak. Enough so that though I finished a few small prints, when I tried larger stuff (2 or 3 hour prints) inevitably it would eventually lose torque and destroy the print.

I completely tore the Y carriage down to try re-lubing the horrid linear bearings. No good. I went ahead and fixed the LED and resistor that caught fire under there a couple of weeks ago while I had it open. I broke out the other printer and spent a couple of hours adjusting settings and printing a set of replacement bushings (they print fast but they have to be very exact dimensions).

None of this helped. The motor still was very weak; just slight finger pressure could cause it to jump and miss steps. I replaced the motor, no good. I adjusted the drive current probably 50 times in the last few days. I built a new cable for that section. I replaced the motor driver. None of this helped.

Finally after talking it over on IRC I decided these motors just aren't up to this application, and went out to a RepRap supplier and looked at their motors. Their windings are 2.4 ohms per phase. Mine are 60. There's the problem.

I bought these motors off eBay. They were originally part of some vinyl cutting printers that were damaged (new) and parted out, and probably were intended to run on 24 or 36 volts.

However, I realized that these are center-tapped, so I tried re-wiring one using the center tap 30 ohm coil to make them draw twice the current. This helped immensely, enough that I can now print (I think - I have a 2.5 hour print going as I get ready for bed, we'll see in the morning). The Y motor is getting pretty hot though (67 degrees, too hot to touch comfortably). I will probably order a pair of the proper motors for this application from lulzbot.com because though I *CAN* print now, I have to go slow. I'm printing at about 60mm/second. On the other printer I've hit 150 Also I think it may be possible that by using only half the coils I'm getting less resolution.

I also ordered a couple of 5# spools of PLA filament so I can start working on a few prototypes for Duckon stuff.

RepRap problem HALF fixed

OK, the print I left running last night finished without Y axis problems. The X axis flaked out instead. Serves me right for jumping the gun and not fixing the X axis as well while I was at it. It was way past my bedtime, but I shouldn't have started the print.

Actually I think I will re-wire all 5 motors. The extruder hasn't skipped since I cleaned the hot-end but it has in the past.

Trying to decide now whether to order 2 new motors or all 5. It's possible that the re-wiring will make the whole thing fine, but I think I'd like to replace at least the X and Y motors.

EDIT: just ordered 5 new motors. I've spent way too much time screwing with these to justify continuing to mess with them. I've updated my price sheet that I've shared with a few people to say "just buy the right motors, this is not a place to save money."

More research on stepper motors

According to this page on stepper motors, the six-wire motors that I have can be legitimately used the way I'm wiring them (both the original way and the change I made last night).

I was concerned that perhaps there were four discrete poles, two adjacent ones for each winding (one on either side of the center tap) and that they were just connected for convenience. However, that reference indicates that there is only one pole per set of windings and the center tap is actually JUST a tap in the center of one big winding. So using only the half between the center tap and one end is totally legit.

It seems that the 6-wire steppers are used because it allows control with super-cheap electronics; just four 2n2222 transistors could easily drive a stepper. Otherwise you need two H-bridges, with 8 transistors. If you're doing it in custom silicon like a Pololu driver, then the electronics parts becomes basically free and you go full bipolar to make the motors a few cents cheaper, but if you are doing it with discrete electronics, it probably swings the other way and spending an extra few cents per motor reduces the parts count by 4 transistors, 4 diodes and maybe some other stuff.

It's not as strong as the right motors would be (30 ohm versus about 3 ohm coils, so 12/30=.4 amp max versus the amp they usually run at) but given that it should be at least twice as strong as what I have now and that's very nearly working, it should be totally sufficient for at least the extruder and the Z axis. I will still drop in new motors for X and Y since having them be as strong as possible will significantly affect the maximum printing speed.

I will keep the other 3 motors on reserve and replace the rest of them at the first sign of trouble from any of the other motors. Or maybe when I'm just bored some day.

For future reference, Pololu electronics sells 200 step NEMA 17 motors with 2.7 ohms per phase for $14.95, so that's probably the place to get them, since a builder would probably be buying driver modules from them anyway.

QUITE happy with motor changes

With the wiring change, the X and Y axes both went from being able to stall the motor with slight pressure from one finger to being able to move the entire printer around if I try to hold them steady.

The only odd thing is that the motors now emit this weird hissing noise when they're on certain phases of the PWM microstepping. Changing the drive current affects the noise. And they're running kind of hot (67 degrees) but that's pretty normal for these printers.

I may not even bother changing the motors when they come in, unless I actually have a problem.