Help - extruded filament has tiny ridges every mm

[rocketwiz]

Can anyone help? I was trying to diagnose why my prints have faint vertical and angled ridges across the the vertical faces. I checked the extruded filament and found little ridges space about 1mm apart.

As per my previous post using the original QU-BD extruder with Barts spring/bearing mod as well as a proper Mk8 drive gear , with ABS. Ramps 1,4 and drivers from reprapdiscount.

I have tried lowering the driver current but the ridges got spaced out more but were still there. The driver current setting was already spot on so the motor wouldn't move after less than a quarter turn of the driver pot. Increasing the current made no difference (although I only went a quarter turn clockwise). The drive gear is clean (no sign of shredded ABS), the filament exiting the gear (prior to hot end) looks good as well. The only thing is the extruder motor runs quite hot as it's right against the black chassis block (I measured the temp and it runs at over 100degC.

The filament is quite old although it's been stored in zip loc with dessicant. Could it be moisture? Although I read that moisture issues don't manifest themselves in such a fixed way.

Alternatively would switching to a 0.9 deg stepper improve things? If so can I use my current A4988 type drivers (if I don't go to 1/32 micro stepping)

[rsm5178]

Did you re calibrate the extruder since you changed the diameter of the extruding gear?

A picture would speak a thousand words.

[rocketwiz]

Yes - did the calibration when I installed the new drive gear.

I've attached a pic of the issue. You can just see the ridges on the free air extrusion which then manifest themselves clearly as bumps on the perimeter layer, and finally as banding on the single wall print. Interestingly the banding is angled on the usual thick wall test cube and on irregular edge prints I'll see bumps instead of bands.

I don't think the gear is slipping on the filament as it happens even straight from cold. I'm starting to think these are step pulses showing up on the 1.8 deg stepper. What do you guys think?

[cvoinescu]

That looks like an artefact of the drive wheel. Because of the way the teeth engage the filament, it does not push it uniformly. That, at least, would be my first guess.

Other non-uniformities can cause effects like that. You may see the cycling of the extruder temperature control (which can be fixed by increasing the temperature), or you may see motor steps, or non-linearities in the drive train (if your extruder is geared). The trick is to figure out what correlates with the frequency of the bulges.

[rocketwiz]

I'll take the extruder apart to have a look - the extrusion in the photo was done at 230C, and I'm really loathe to go higher.

I think issues with the gear should be easy to spot by observing the filament after it exits the gear. if it's a drive gear artefact I'm not sure what can be done other than to swap the gear. I did have a look at the filament when I was doing the extruder calibration and the recalled it looked ok (ie a uniform 'bite' pattern as expected) although I didn't examine it closely.

In any case I'll get a 0.9 deg motor before the weekend to see how much improvement I can get from it.

In the meantime I also found some y axis backlash due to the belt clamp not being tight any more after a couple of years of sitting there

[cvoinescu]

There's not much you can do about the gear, if it's indeed that. It's not that it's bad, it's just that it inherently pushes the filament a little faster when the tip of a tooth is in contact with it, and slower when two teeth contact at the same time (or vice-versa, I'm not sure). A larger wheel, and/or one with more (smaller) teeth would probably push more smoothly, but it has disadvantages.

[rocketwiz]

Thanks for the help. In that case the only solution is to switch to a geared extruder. What extruder do you use?

[cvoinescu]

Any extruder with this type of drive wheel (and pressure wheel) will probably have the same artefacts, no matter whether it's geared or not. And I would not make changes until I made sure it was this, and not something else. If the period of the ridges matches the period of the drive wheel teeth, then yes, you know it's that, and it pays to switch to a geared extruder with a larger, finer-toothed drive wheel. Extrude, say, one full turn of the extruder motor, and count the ridges. If you get as many as there are teeth on the drive wheel, it's that. If you get 50 or 100 or 200, you know it's the stepper motor (or possibly the driver).

I used to use a not-very-well-printed Greg's Wade Reloaded with a clumsily hobbed bolt. It was very reliable, but it had some artefacts that looked similar to yours. It did not matter for what I was using it for at the time, and I also had fairly large oscillation artefacts after sharp corners, so I did not bother to try to fix it. (I have not used the extruder for many months now.)

[rocketwiz]

Will do - thanks again.

[rocketwiz]

Ok this is interesting. A Mk7 gear has an effective diameter of about 10.6mm, which is a circumference of about 33mm (ignoring the radius of the filament itself for a sec).

I extruded 33mm via pronterface at the default speed of 50mm/s which resulted in an extruded string of about 250mm (free air). The number of ridges at the beginning of the extrusion (which pooled on my bed) was about 6/cm and at towards the end was about 10/cm. Assuming 10/cm then it looks like very much closer to the stepper resolution than the teeth on the gear (I couldn't find any information on this but estimate this to be no more that 40 teeth.

So it is starting to look like the stepper motor. I originally turned the driver pot down by about less then 1/8 turn (from where I originally set up the driver) but stalled the motor so I'm really at about the sweet spot for the driver already. So I'll definitely swap over to a 0.9 deg motor as a test. I do have a spare driver so I might use this instead as well.