Sanity check: Mosquito & BMG-M pressure advance 0.2?
I'm trying to dial in my settings after performing Duet Wifi brain surgery on my Flash Forge Creator Pro. I'll have a more significant post on the tuning process as I get further into it, but first, just a Sanity check:
I'm driving a Mosquito (non-magnum) directly with a BMG-M, and doing a pressure advance sweep with PETG (50mm/s, 245C, 0.4mm nozzle, 0.2mm layers). The optimum value seems to be coming out as 0.2, which is significantly higher than I had been led to expect for a direct drive system. The attached image is a 2 wall tower sweeping from 0.3 to 0.025.
Does this result seem reasonable, or does it suggest some kind of mechanical/flow problem?
JamesM last edited by JamesM
What do you have for your extruder acceleration and instantaneous speed change setting? For a combo like that I would expect them to be fairly high. For my bondtech with a v6 hotend clamped in it(direct drive) my instantaneous speed change is at 1200 and acceleration is 2000 for my extruder with an E3D 0.9degree pancake stepper. With my setup my pressure advance is 0.06 for pla and around 0.08 for petg.
I've been trying to tune things in the get DRR working well, (worthy of a separate thread), this is what I am working with at the moment:
M203 X7200.00 Y7200.00 Z800.00 E1200.00 ; set maximum speeds (mm/min) M566 X300.00 Y300.00 Z20.00 E900.00 ; set maximum instantaneous speed changes (mm/min) M201 X1500.00 Y1500.00 Z50.00 E1500.00 ; set accelerations (mm/s^2) M593 F31 ; DRR ringing reduction at 31Hz
I'd probably run that cube again but with finer resolution around the 0.15 and 0.25 range. And then maybe even once more to get closer in still.
0.2 seems high to me for a direct drive as well, but the starting values are pretty conservative. The proof of the pudding is in the eating, so I'd take empirical results over theory any way.
I'd probably bump up your extruder speed, accel, and jerk values by double or triple what they are.
Also do some other testing with more varied models to verify that the values isn't have unexpected consequences.
I would also look into increasing your Z axis jerk and acceleration in order to limit the time spent on layer changes and z hopping. Low values can also cause stuttering with mesh compensation active.
Thanks for the input. I'll certainly try upping the extruder and z axis accel/jerk values. (The reprap defaults seem to be very low.)
In other news, I ran the same test on some PLA, and the best result was somewhere between 0.05 and 0.1, closer to 0.05. Perhaps I'm running this PETG too cold at 245C, and it's excessively viscous? I haven't calibrated my temperature sensor so the true temperature could be somewhat below that.
Perhaps I'm running this PETG too cold at 245C, and it's excessively viscous?
Well it does behave a bit differently than PLA, so don't expect the results to be exactly the same, however, in my own experience, I haven't found the values to be THAT far off, so much so that I basically leave it at 0.04 and forget it. Though occasionally a filament will be better at 0.03 or 0.05, but that's still pretty close in range.
But the results don't lie. Go with whatever gives you the best print.
zapta last edited by
@Luke_D3D, you can try printing a heat tower with PETG, pick the temperature you like and optimize PA for that temperature.
I calibrated PA just for PLA and am using the same value (0.08, using direct BMG and ESD hot end) also for PETG but reading here I may try calibrating PA also for PETG.
Yep, temperature tower is certainly next on the list.
I find it such a pain dialing in PETG to give good overhang performance at the same time as good layer adhesion. If anyone has any magic bullet, I'd love to hear it.
good overhang performance at the same time as good layer adhesion. If anyone has any magic bullet, I'd love to hear it.
I use a custom version of Cura that allows you to set an overhang angle and then set a custom print speed and fan speed for overhangs. It's been a huge game changer in how well I can print overhangs.
Now that sounds cool, definitely a reason to dip back into Cura. Is it your own customization or someone else's?
It's by @burtoogle