microstepping with interpolation

spllg

M350 X16 Y16 Z16 E128 I1
M92 X200 Y200 Z400 E3282.72

results in slight over extrusion which in turn can result in model knock off

while
M350 X64 Y64 Z32 E128 I1
M92 X800 Y800 Z800 E3282.72

results in reasonable extrusion but i feel sometimes (not reproducible) might result in 'wild moving' in x- or y-direction.

what did i miss?

my config:
FIRMWARE_NAME: RepRapFirmware for Duet 3 MB6HC v0.6 or 1.0 FIRMWARE_VERSION: 3.01-RC2 ELECTRONICS: Duet 3 MB6HC FIRMWARE_DATE: 2020-02-18b1

ii duetcontrolserver 1.2.4.0 armhf Control server application for Duet 3 series
ii duetruntime 1.2.4.0 armhf .NET Core runtime libraries for the Duet software framework
ii duetsd 1.0.5 all Virtual SD card directory for the Duet software framework
ii duetsoftwareframework 1.2.4.0 armhf Meta package for the full Duet software framework
ii duettools 1.2.4.0 armhf Optional tools (code examples)
ii duetwebcontrol 2.0.7-1 all Official web interface for Duet electronics
ii duetwebserver 1.2.3.1 armhf Optional web server for Duet 3 series

Phaedrux

@spllg said in microstepping with interpolation:

3.01-RC2

I suggest you update to RC5 first.

spllg

@Phaedrux haven't go the time to upgrade and check everything again - waiting for next release.

thanks.

Danal

Odd... you did not change E micro or s/mm... yet extrusion looks better?

That implies the calculation of XY and E into a 'coordinated move' either works better (unlikely, it is just math, and should produce the same end result), or that the pulse generators or driver chips work differently at 16 v 64 micros.

Of course, all this is pure "thinking out loud" speculation. No hard facts.

And, have you thought about sticking with 16x and adjusting the extrusion factor in the slicer a bit? IMHO, adjusting that factor is a pretty normal thing, for different filaments, etc.

deckingman

@spllg You haven't given us much information to go on like what speeds you are running but it's highly likely that you might be running into the step pule frequency limit which I believe is currently lower in RRF 3 than it was RRF2. I'd imagine that for normal print moves the axes and extruders are OK but extruder only retraction and un-retraction moves, those high micro-stepping values will almost certainly cause hiccups which might explain your problems.

For info, you can do this:

M350 X16 Y16 Z16 E16 I1
M92 X200 Y200 Z400 E410

Then change M350 to whatever values you want to experiment with and the firmware will automatically re-calculate the steps per mm for you - no need for you to enter any M92 commands again. Note that I have calculated the steps per mm for the extruder by dividing what you had by 128 and multiplying the result by 16.

But ideally and as @Phaedrux has pointed out, you should still be running the latest RC otherwise who knows if you have found a new problem or one which has already been fixed?

spllg

@spllg said in microstepping with interpolation:

'wild moving' in x- or y-direction.

reduced travel speed to 120 mm/sec which seems to have eliminated 'wild movings'.