1.2.8 experimental

Anyone tried this yet? This new feature sounds interesting:

•New experimental autospeed feature that chooses speeds automatically in order to keep volumetric speed and nozzle pressure constant; just set all speeds to 0 to enable it

Sounds great...but how does it detect the nozzle pressure?

I doubt it detects it, but rather calculates it.

FYI- it does not appear that the Linux tarball version was built with the GUI included. I had to build from source on Ubuntu, which went fine (I am far from a Linux expert).

I guess I am thinking that it would need some kind of feedback in order to get the fastest speeds with the best print quality.

Nozzle pressure will be related to extruder speed and nozzle size, therefore it can be calculated. But the exact pressure is not necessary to know, just the limiting extruder speed will do. If the limiting factor for print speed is the extruder capability (i.e. the extruder skips if the speed is too high), just a set limit (which can be found by experimentation) on the extruder speed. The software then calculates the fastest move speeds that will result in the limiting extruder speed being reached based on the move length and amount of extrusion required. That means that thin extrusion widths and small layer heights will print faster than wider extrusion widths and coarser layer heights.

I use only 2 different widths (perimeter & infill), and have found by experimentation what maximum speed I can have for the perimeter and infill for the two layer heights I use (0.2mm and 0,25mm), and thus just have two profiles that will have the same effect as that automatic method.

Dave

Ok, that makes a lot more sense now. Thank you for taking the time to explain it to me.

Is that including the functionality of the pressure advance setting? From how I understand both features they have similar functionality.

No, pressure advance is a seperate feature.
This is just for automatically determining the same speeds in mm³/s.

Edited 1 time(s). Last edit at 06/14/2015 06:36PM by Mikk36.

Aha, do think the autospeed considers the slow down from the pressure advance?
If the autospeed has the strict directive to keep constant speed/volume the slowdown from the pressure advance could conflict with that.

Just started playing with 1.2.8 autospeed. Have a 0.4 E3D v6 direct drive extruder on a Frankenstein Solidoodle. Layer height set at .1552 with infill every 2 layers. Set all speeds to 0 to test autospeed. Print time estimated at 15+hours.. Print time per my original settings was about 6.5 hours.

I suspect computed pressure is less than half of what I'm normally running my extruder at. Either I'm a speed demon or the pressure calculation is too conservative.

Roger

Update: On my small .75 X .75 X .188 (inch) test block Autospeed was 8.41 versus 8.13 for normal. Very strange. Late here going to bed will think about it in morning.

Edited 1 time(s). Last edit at 06/14/2015 11:24PM by Miamicraft.

1.2.8 tested and found a problem. Setting the pressure advance to a non zero value printing will randomly paused, but pastic keep on extruded and destroy the object.

@miamicraft: Interesting. I printed a model which with old setting showed a time of 52 minutes and with autospeed 50, so that was ok. I can imagine that infill every two layers can mess with the auto speed. have you tested what happens to normal infill?

@Samuel: I had something similar happen but in the other direction. When pressure advance at 0.2 the retract was ok, be the push pack did not happen so the filament pulled out of the nozzle rather quickly. here also I can imagine an interference with the autospeed function.