Pressure advance causing underextrusion/motor madness



  • @CCS86 doesn't a jerk setting that low really slow down your print? I see lots of other people using 3000 jerk on the same extruder is the thing too.



  • @Dezdoghound there is a balance to be found between stepper jerk and real-world capability. IE stepper motors attached to extruders can't accurately "jerk" as fast as some people wish them to. So, if speed is your main concern, and slightly inaccurate movement is acceptable to you, go for higher jerk. The inaccuracy of the movement is quite low -- it's still gonna be pretty damn close to what you asked for, perhaps just a bit noisy and sloppy.

    If accuracy is your main concern, then slow the jerk down to the point that you are guaranteed accurate movement every time, with little noise.



  • @Dezdoghound said in Pressure advance causing underextrusion/motor madness:

    @CCS86 doesn't a jerk setting that low really slow down your print? I see lots of other people using 3000 jerk on the same extruder is the thing too.

    If you are specifying jerk with M566, it is in mm/min. So, 3000 mm/min = 50 mm/s; which is still very high. It is probably higher than the max speed for E, so that effectively disables any attempt for planned acceleration.

    I don't see how that is a benefit at any time. It just increases the chances of skipping E steps and chewing up filament.

    Once you turn on PA, the demands on E change dramatically. High jerk is even less appropriate.



  • @CCS86 exactly this. To make matters worse, with the standard jerk policy (and perhaps even jerk policy 1, I'm not sure) the E axis almost never uses the jerk value. Jerk normally ONLY comes into play with pressure advance. So, having an excessively high E jerk value can be hidden until PA is enabled.



  • I still struggle to find a use case other than pressure advance, when extruder jerk actually gets applied - unless it's set really low so that it is the limiting factor in any G1 XY E move. Extruders generally run a lot slower than axes during print moves, so it's likely that the lower of either X or Y jerk would determine the speed at direction change - unless extruder jerk is really low and the combined XY E move has to slow down to the extruder jerk speed, rather the X or Y jerk speed. That's mainly why I set extruder jerk "silly high" - so that it doesn't impact on carriage speed. One could say that extruder jerk gets applied to retraction moves. BUT, in RRF (at least until very recently) jerk is/was never applied to the start of a move from rest - only on direction change. I seem to recall having this discussion with DC and I think (but could be wrong) that jerk does not get applied to retraction moves. If all of that is true (and it may not be), then extruder jerk is never used unless it is so low that it affects XY E direction changes. BUT, something strange happens with the interaction between extruder jerk and PA which cannot be explained by the "normal rules" for instantaneous speed threshold (jerk), because even a moderately low value will slow down the overall print speed more than can be explained by the application of extruder jerk alone.



  • @bot Yes, I think we are in agreement on this.



  • I think so. Except, the slowdown of E jerk in combination with PA is exactly calculable and accurately applied. It is a pretty sharp slowdown that happens, once it begins happening. It's at that point that the user must balance the noise of the extruder vs the speed needed and if that noisiness of the extruder affects the print quality in any way. Usually it doesn't seem to affect anything, but at extremely "precise" extrusion rates (IE tiny amounts for tiny traces) the inaccuracy of the system can begin introducing noticeable artifacts.

    Also, of course, if the E stepper is stalling out, the Jerk is impossibly too high. But I figured that goes without saying...



  • @bot TBH, my printer has been pretty much out of action since about last August when I started to convert to Duet 3, and more recently since I started trying to develop a new hot end. So any findings I have are a few firmware versions old (to say nothing of the hardware - much of which has also changed). Way back then, it appeared that setting extruder jerk "silly high" had no adverse affect on PA (or anything else), but setting it low (or even what would be considered "normal") had detrimental affects - especially carriage randomly "stuttering" during arcs or curves when using PA. This was all with multi-input, multi extruder, mixing hot ends, which are strange and fickle beasts at the best of times). But a lot has changed since I last attempted any sort of in depth testing or analysis.



  • @deckingman I think you'll find not much has changed in that regard (I don't think). The behaviour of PA has remained unchanged since early 2.0x versions, if not earlier.

    With some stepper/extruder combos, the achievable jerk speed without stalling the motor may indeed be incredibly high. For setups like mine, with high gear ratios and relatively small stepper motors, the actual achievable jerk speed is quite low compared to values typically found in users' config.g files. Thankfully, it's not quite low enough to be ridiculous, but I'm certainly balancing the thin line.



  • @bot said in Pressure advance causing underextrusion/motor madness:

    @deckingman .................The behaviour of PA has remained unchanged since early 2.0x versions, if not earlier.

    Hah - my cynical side won't let me believe that I'm afraid. 🙂 (Unintended consequences and all that stuff.....)



  • @deckingman True, something may indirectly affect the behaviour. The added jerk policy 1 may affect it.


  • Moderator

    @bot said in Pressure advance causing underextrusion/motor madness:

    The added jerk policy 1 may affect it.

    Yes I've found that I must lower my extruder jerk a bit more now with P1, but I haven't done any deep testing.



  • So after some experimentation, I've found that I don't get issues when extruder jerk is at 120mm/min and accel is at 3500 mm/min. This however slows down a print a fair bit (calibration cube took 19min20 instead of 15:42).

    I've tried lowering microsteps to x8 and x4 but it just makes such a lot of noise that I don't think I want to have it like that even if it works!


Log in to reply