M592 - Non-Linear Extrusion Comp - Cant get it to work...

timschneider

@FurbyDealer27 said in M592 - Non-Linear Extrusion Comp - Cant get it to work...:

; Miscellaneous
M501 ; Load saved parameters from non-volatile memory

I guess you have an M592 in your config-override.g which is called via M501 at the end of your config.

; Miscellaneous
M501                                               			; Load saved parameters from non-volatile memory

deckingman

@FurbyDealer27 How are you doing the under extrusion testing?

droftarts

@FurbyDealer27 Can you check if M592 is configured after a reset? Send M592 D0 in the console. I just tried this, with your M592 command right at the beginning of the config.g, and it responded with the correct extrusion compensation, so position in the config.g file doesn't matter. I didn't actually run any extrusion test to see if it is compensating, though. I'm also using RRF 3.5.0-rc1, rather than 3.4.6.

If it is configured after reset, but not once you start extruding, ie send M592 D0 while extruding to check if it is still active, please check your gcode to see if it is being reset. Check your config.g is not reporting errors by sending M98 P'config.g' too.

@timschneider As far as I'm aware, M592 is not saved in config-override.g with M500.

Ian

deckingman

The reason I asked about the test methodology was that from the documentation we have .............."Nonlinear extrusion is not applied to extruder-only movements such as retractions and filament loading".

So unless the extrusion move also corresponds with an X and/or Y carriage move, extrusion compensation isn't going to do anything.

FurbyDealer27

@deckingman

Fantastic question. I am using CNCKitchen's Blob Test.

I had to modify an excel sheet i found online to make it work for a delta printer, but i am moving up 10mm while extruding 200 mm of filament.

Example: G1 Z10.5 E200 F12.473

This makes me wonder if a Z axis movement is not enough to activate the compensation....

deckingman

@FurbyDealer27 said in M592 - Non-Linear Extrusion Comp - Cant get it to work...:

............................This makes me wonder if a Z axis movement is not enough to activate the compensation....

Highly likely I'd have thought, but only a moderator or someone who can dive into the code could confirm.

FurbyDealer27

@deckingman

I checked my Config-override.g and i do not call M592 (only in my config.g)

When i request "M592 D0" in the console, i get a return of "Drive 0 nonlinear extrusion coefficients: A=0.044, B=0.0000, limit=1.00".

Thanks for the asking me to check that.

droftarts

@FurbyDealer27 I've asked @dc42 to have a look at this.

Ian

timschneider

@droftarts
yep, you are right about the M500 - i thought that he maybe added it manually.

@FurbyDealer27
it looks like the non-linear-compensation is only active for printing moves, which are defined as XY and E+ moves.

if (Tool::GetXAxes(nextMove.tool).IsBitSet(drive) || Tool::GetYAxes(nextMove.tool).IsBitSet(drive))
{
    flags.xyMoving = true;				// this move has XY movement in user space, before axis were mapped
}

flags.isPrintingMove = flags.xyMoving && forwardExtruding;

// Add the nonlinear extrusion correction to totalExtrusion.
// If we are given a stupidly short move to execute then clocksNeeded can be zero, which leads to NaNs in this code; so we need to guard against that.
if (flags.isPrintingMove && clocksNeeded != 0)
{
...

deckingman

@FurbyDealer27 said in M592 - Non-Linear Extrusion Comp - Cant get it to work...:

@deckingman

I checked my Config-override.g and i do not call M592 (only in my config.g)

When i request "M592 D0" in the console, i get a return of "Drive 0 nonlinear extrusion coefficients: A=0.044, B=0.0000, limit=1.00".

Thanks for the asking me to check that.

It was the other Ian (@droftarts ) who asked you to do those checks - confusing I know.

FurbyDealer27

@timschneider

Thank you everyone!!!

I changed the Blob test to extrude 200mm of filament while moving 30mm on the X axis. This now makes amazing looking weld beads.

I can confirm that this fixed my issue. The M592 compensation does require an XY movement.

Looks like i can push a crappy hotend and a Titan Extruder to around 10 mm3/s.
Thanks everyone for your help!, I really appreciate the immediate responses and feedback.

Kiolia

@FurbyDealer27 said in M592 - Non-Linear Extrusion Comp - Cant get it to work...:

@deckingman

Fantastic question. I am using CNCKitchen's Blob Test.

I had to modify an excel sheet i found online to make it work for a delta printer

Would you be willing to share the modified sheet, or just explain what needed to be changed? I run a delta and this is testing I need to do (EDIT: actually, scratch that, I figured out quickly what needs modding and I'm already done. It didn't have a round bed mode, mainly.)