Pressure advance quick question
-
I had, I think, the same issue with a Hemera Direct Drive extruder using Prusa Slicer.
Check the gcode to see if there are retractions occuring in the area first. If there are none then this suggestion will not help you.
I increased extruder max acceleration (from 3000 to 5000) which helped a bit. I then added 0.15mm of 'Extra length on restart' (after retraction). Those two things combined seemed to solve the problem.
You will need to make valid changes for your machine of course but hopefully focusing on those two items will fix your problem too.
I should also note that two days of testing/setting/testing and some luck were required for me to resolve this problem.
-
@tas , thanks for your input. I did not see any retracts and extruder acceleration is already set at 5000. After very careful examination, I found other spots that have the extrusion issue - hard to see when it only happens on a single layer.
I am doing another test print with temperature increased to 210 with a number of slicer settings changed and different filament loaded. Of course if this print happens to come out good than I have to remember what all I changed which will be problematic
If I still have the same issue I will try a different extruder altogether just in case there is something blocking things. I guess I should mention that the extruder is a Dragon HF and an earlier print done at a volumetric maximum flow of 16 mm3/sec produced all kinds of skipping which really surprised me. I was under the impression that I would be good to 20 mm3/sec with that hot end and PLA.
The print I was asking about was done with a maximum flow of 10 mm3/sec which should not cause a problem but maybe something in the filament path is not quite kosher. I will see how this next print comes out. -
@jens55 said in Pressure advance quick question:
The calibration was done with PETG and I am currently running PLA - would pressure advance change drastically with a change in filament?
Pressure builds up when the filament exits the extruder faster than the rate that it can be melted and flow freely out of the nozzle. Of course, we need some pressure and it only becomes problematic if the print head starts to move before that pressure builds up, or if the pressure decays too slowly as the print head decelerates and/or starts a non-print travel move. Molten PET-G is much more viscous than molten PLA so it will need more pressure to commence flowing out of the nozzle, and that pressure will decay more slowly at the end of any non-print move. So yes, PET-G will need more compensation (pressure advance) than PLA. Ideally, one should tune PA for every filament type.
On my ToDo list is an investigation in the relationship between molten filament viscosity and pressure. If we know what that relationship is, we could probably tune PA for one filament, and then simply apply a scaling factor for others.
-
@deckingman , Hmmmmm .... while I knew that there was a small difference between PETG and PLA, I was under the impression that it was a SMALL difference (because of viscosity changes like you said) and that the majority of PA is needed for Bowden setups.
So what you are saying is that I need to put custom g codes into every filament in order to set pressure advance (or change config.g every time filament is changed) wow ..... -
@jens55 said in Pressure advance quick question:
@deckingman , Hmmmmm .... while I knew that there was a small difference between PETG and PLA, I was under the impression that it was a SMALL difference (because of viscosity changes like you said) and that the majority of PA is needed for Bowden setups.
So what you are saying is that I need to put custom g codes into every filament in order to set pressure advance (or change config.g every time filament is changed) wow .....For sure, you need a lot more PA with Bowden setups, but the same principle applies. That is to say, a higher viscous filament will be harder to force through a nozzle thus causing higher back pressure in the system - from the nozzle back to the extruder that is doing the pushing. I would have thought that the proportional increase would be similar. So if you had a long Bowden set up which needed say 0.5 PA for PLA and 0.7 for PET-G, then that's about a 40% increase. Then if you had a direct drive which only needed 0.1 PA for PLA, then it could be that 0.14 would be needed for PET-G. This is all just conjecture on my part which would require a more detailed investigation to prove or disprove the hypothesis, but it seems plausible to me.
EDIT. I'm eager to do some testing with my new Bondech LGX Ace/Slice Engineering Mosquito combination. One of the reasons for choosing it was that it has just about the shortest filament path between the extruder drive and the hot end input known to mankind. That makes it ideally suited to printing flexible filaments but should also mean that very little PA is required (if any at all).
-
@deckingman said in Pressure advance quick question:
One of the reasons for choosing it was that it has just about the shortest filament path between the extruder drive and the hot end input known to mankind. That makes it ideally suited to printing flexible filaments but should also mean that very little PA is required (if any at all).
The "elasticity" of the stepper motor will need some pressure advance even if the filament path is zero. So you might like to experiment with how the extruder motor current affects the required PA.
Duet WiFi hardware designer and firmware engineer
Please do not ask me for Duet support via PM or email, use the forum
http://www.escher3d.com, https://miscsolutions.wordpress.com -
I have started a print in PLA with a pressure advance of 0.04 compared to the 0.13 as used for PETG. The new PA was based on what I was using some time ago and was not derived from recalibration. While it is early in the new print, I think that a lot of my issues went away.
As I said before, I was not aware that a change of filament would be a completely different PA. I was expecting maybe a 10% or so difference that, while not optimized, would not completely mess up my prints. I wonder if anybody has published figures about change in PA vs nozzle size and change in PA vs print temperature.I wish I could find a PA calibration routine that can directly work with a multi tool printer (Jubilee) without having to do all kinds of code re-work.
-
Just to close this topic out - it was pressure advance! Totally different figure between PLA and PETG. 0.13 vs 0.04. Now in all fairness, I also upped the temperature to 215 from 190 .... but I believe PA was the cause of all my grief.
Thanks everybody! -
@dc42 said in Pressure advance quick question:
The "elasticity" of the stepper motor will need some pressure advance even if the filament path is zero. So you might like to experiment with how the extruder motor current affects the required PA.
That's an interesting concept and one that I hadn't considered. Can you elaborate?
-
@jens55 I'm not surprised at the difference you found between PLA and PET-G. Other people on this forum have complained that they are unable to print at high volumetric flow rates when using PET-G but can do using PLA which indicates to me that PET-G must be highly viscous. Add to that it's propensity for "stringing" and sticking to the outside of nozzles, the evidence starts to stack up. Having said that, a temperature of 190 degC is pretty low for PET-G so increasing it will have significantly reduced the viscosity. I would not be surprised if your 0.13 PA for 215 Deg C needed to be around 0.2 for 190 Deg C.
-
@deckingman, the temperatures of 190 and 215 were for PLA ... sorry if I wasn't clear.
