Pressure Advance. Wow!!
-
Thanks for the feedback!
-
Why pressure advance fixed the jams? I'm not sure to understand…
-
Read the desertion in the online manual. It buffers the variances in filament feed. Especially prevalent in bowden setups.
Just finished another 32 hour print in the previously troublesome PLA with no problems at all. -
From the online manual:
What is pressure advance?
Pressure advance aims to compensate for the elasticity of the filament and the extruder system. There are at least three sources of elasticity:- The filament in the Bowden tube behaves as if it is compressible, because its diameter is typically 0.25mm smaller than the inside diameter of the tube. When it is under tension, it will take the most direct path that the tube allows. When it is under compression, it will snake from side to side, so the length of filament in the tube will be greater.
- The filament itself is slightly compressible.
- In order to produce torque, the angle of the rotor of a stepper motor must lag the angle commanded by the current in the coils. The more the filament resists being fed, the greater the lag angle.
These factors cause under-extrusion whenever the extrusion rate is increasing, for example at the start of a straight line when the nozzle has to accelerate from zero or near-zero speed. This is because some of the filament fed at the start of the move is used to counter the elasticity and build up the pressure. Similarly, you get over-extrusion when the rate of extrusion decreases, for example when the nozzle slows down at the end of a straight line. This is because the pressure in the Bowden tube continues to push filament through the nozzle even after the extruder drive slows down or stops.
Pressure advance compensates for elasticity by feeding additional filament through the extruder drive when the extrusion speed is increasing, and feeding less filament through the extruder when the extrusion speed is decreasing, This may result in filament actually being retracted during the last part of the deceleration phase of a move.
Mathematically, it works like this:
actual_extrusion_speed = requested_extrusion_speed + (K * current_extruder_acceleration)
The constant K is the amount of pressure advance you configure.In my case, the problem was twofold. I had what seemed like over-extrusion after a couple layers that would cause the "jamming", yet under-extrusion when starting new rows in infill causing gaps in the beginning of the infill layer.
-
It works like a charm!
Even in direct drive systems. I have a Bondtech extruder with a E3D hotend. And also on the short path the filament has from the hobbed gears to the nozzle ( aprox. 7cm ) it compresses a bit, but enough to cause blobbing issues.
It requires very small values like 0.05, but it's a clear improvment for precise extrusion on start / end of lines, corners, infill, etc.
If you want to get rid of the blobbs by under extruding, then you will get gaps at the start of the line. Esspecially on short lines.
So the pressure advance is a must if you want flawless surfaces and dimensional accuracy.But for it to work properly you MUST have the extruder calibrated spot on. A lot of over/under extrusion problems are originating in poorly calibrated extruders. And a lot of jammings from over retractions (retracting too much filament, too slow).
Thank you for the great work, David.
This is the best controller I ever worked with. High quality electronics and great software and support (and I'm an electronics designer myself by trade, so I've seen alot). -
I just enabled it last night with my Titan Aero and the corners are crisper–-noticeable improvement. I set it at .05 since the Aero has such a straight path.
-
Can anyone post pictures? Before/after pressure advance enabled?
-
Can anyone post pictures? Before/after pressure advance enabled?
Here is a video that I did a while back when I was exploring high speed printing. Pressure advance certainly made a huge difference but on my printer, with my hot end, at those speeds, I needed a lot of it. https://www.youtube.com/watch?v=lnYYNfVoxmQ&t=341s
Edit. I did a write up on my blog at that time too https://somei3deas.wordpress.com/2017/06/25/duet-pressure-advance-experiments/
-
deckingman, thank you. I'll try pressure advance again.
-
Very impressive! Thanks for sharing.
