Sudden overextrusion

[SyntaxTerror 344]

Yesterday my UM2+ finished a print with a sample of Colorfabb nGen. I was going to do another one straight away with PLA. Did a quick cold pull to get any leftovers out of the nozzle - it's something I've pretty much gotten used to when changing material types - and got the next job on the way.

Halfway through, though, it started underextruding pretty badly. I figured I might have a bit of higher temperature material still in there somewhere and tried a couple more cold pulls to see if I could clear something. Absolutely nothing came out. I restarted the print and got more underextrusion straight away. Since it's been awhile since I checked the coupler, I took the hot end apart and had a look... nothing to be seen. Unscrewed the nozzle and cleaned it. Disassembled the feeder to see if I might have some junk rattling around in there. There were the usual bits of filament ... but really nothing I'd worry much about if I'd seen it at another time. So I cleaned those out, put it all back together and restarted the print.. again.

And this is where the fun begins. The underextrusion was gone, now replaced by overextrusion. Completely baffled, I disassembled the entire thing again, checked everything again, put it back together and now it works perfectly once more.

Now, I've had underextrusion before.. but overextrusion out of the blue? I haven't changed my e-steps or done anything to make that happen.

I'm puzzled. Can anyone come up with an idea as to what happened to make it overextrude?

[gr5 2,005]

Very curious!

[SyntaxTerror 344]

I think I figured it out myself now.

The overextrusion is back - or so it would seem at first glance.

My best guess is the underextrusion was just your run-of-the-mill thing. I've just been unlucky enough to see it in combination with z-motor trouble. It looks like the step motor is intermittent and doesn't turn every time. It's also getting seriously hot.. so I'm going to have a look at it tomorrow, see if it can be salvaged. Could be a bad bearing or something has come loose and blocks the rotor sometimes.

In any case, it's not overextrusion, it's just the bed not moving and the nozzle pushing around what's already printed.

[neotko 1,417]

If is the z it can be the lack of grease on the z (or dust) , or the square flanged bearings or the z nut.

Can you download printrun/pronterface and do a home and move the z 0.02 looking at the bed to see if it 'jumps'?

Is a free software http://www.pronterface.com/

Connects by usb. Hit the connect button the z home then move z to 1mm and statt moving it down with the program. Also if possible record it with a camera.

The third possibility is stepper driver overheating and making the z skip movements. But since we are in winder it shouldn't be the cause unless your room temp is 28C+

[SyntaxTerror 344]

It's a bit hard to get consistent results as the problem comes and goes, but good idea.

As the Z-bearings are maybe 2 months old and I lubricate everything religiously, I'll assume for now they're okay.

What I did yesterday was unscrew the trapezoidal nut and rotate the z-motor manually. It definitely has uneven, position dependent friction.

[neotko 1,417]

I know sounds weird but with a bad nut or bad z everything can print and give perfect prints, unless the flanged squeak bearings ain't right. Please try the printrun and move the z slowly and record it.

[SyntaxTerror 344]

I hear ya

There's definitely a shot bearing in the motor though. I disassembled it after work and it runs about as smoothly as if it's lubricated with gravel. Probably a moisture problem during fabrication. Should be able to find a replacement when the nice people in the instrument workshop show up tomorrow.

[gr5 2,005]

There's a known overheating issue on the Z axis driver with newer UM2 and UM2+ PCBs. The servo itself is unlikely to overheat. The servo can get up to 80C no problems.

The official fix was to lower the current to the axes. Another fix is to remove the cover and add a fan. This problem happens even in coldish environments (not just when the air is at 28C).

Some versions of Marlin (tinker maybe? latest maybe?) let you change the z current here:

Z axis max current adjustment:

Maintenance -> Advanced -> Motion settings -> Current Z

I think it will lose this setting when you power cycle maybe. There might be a save command so you don't lose it?

You can also adjust with M907

M907 Z1000

M500

If you don't do the M500 the printer will go back to previous setting on power cycle.

That lowers the max Z current to 1000ma (I think the default is 1250).

[neotko 1,417]

 

@gr5 the Z motor amps used on the firmware was changed before the um2+ was released

[gr5 2,005]

@gr5 the Z motor amps used on the firmware was changed before the um2+ was released

 

Hmm. I'd check it anyway. I'd also remove the cover and add a fan just to see if that helps. I think UM lowered the current twice.

[SyntaxTerror 344]

Thanks guys

Of course I couldn't find a replacement bearing that fit.. but being kind of in a hurry I cannibalized the motor of my other printer that's waiting for new bearings. All is well in the world again as of a couple of hours ago.

[neotko 1,417]

Thanks guys

Of course I couldn't find a replacement bearing that fit.. but being kind of in a hurry I cannibalized the motor of my other printer that's waiting for new bearings. All is well in the world again as of a couple of hours ago.

 

Hey post a photo! I love cannibalized modtricks