<E> variable placeholder broken ?

[Davide Ardizzoia]

Hello All,
As I am trying to add adjunctive retraction when swapping extruders, I am fiddling with the "deselect extruder" gcode section.

So far, the best working solution *should* be

G1 E<E-5> F1000         ;suck 5mm of feed stock
G92 E0                  ;zero the extruded length again
M104 S<TEMP>

..unfortunately, KS seems to keep swapping extruder positions, no matter if i use or not the T<EXT+0>  instruction.

For example, here are some snippets from a test gcode where I changed the "deselect extruder" as follow to higlight the <E> value

G1 E<E> F1000    
G92 E0                  ;zero the extruded length again
M104 S<TEMP>


(SNIPPED GCODE)
.
.
.
.
G1 X119.18 Y88.83 Z1.29 E17.7964 F360
;
; *** Cooling Extruder 2 to 260 C ***
G1 E0 F1000         ;suck 5mm of feed stock
G92 E0                  ;zero the extruded length again
M104 S260
.
.
.
G1 X131.22 Y85.35 Z1.29 E9.1277 F360
;
; *** Cooling Extruder 1 to 250 C ***
G1 E17.7964 F1000         ;suck 5mm of feed stock
G92 E0                  ;zero the extruded length again
M104 S250
.
.
.
G1 X124.38 Y81.33 Z1.69 E5.5899 F360
;
; *** Cooling Extruder 2 to 260 C ***
G1 E9.1277 F1000         ;suck 5mm of feed stock
G92 E0                  ;zero the extruded length again
M104 S260



...am I missing something or I just found a bug ?

Best Regards
Davide Ardizzoia

[frozen...@yahoo.com]

remove the <> from the deselect code, should be

G1 E-5 F1000
G92 E0                
M104 S<TEMP>

[Davide Ardizzoia]

Frozen,
your suggestion would rewind extruder at -5 (absolute coordinates) !!!
Let's say that current extruder position is 90, with your approach you would retract 95mm !!!

Best Regards
Davide Ardizzoia

[Franz Scherz]

Did you try:

G91 ; Relative positioning 

G1 E-5 F1000

G90 ; Absolute positioning

[Davide Ardizzoia]

Tried, but after 2-3 layers the retraction is no longer performed.
Tried also slowing the feed rate (F500), but seems not to change anything-.
After a few layers, extrusion simply works WITHOUT the respecting the "deselect extruder" relative+retraction+absolute.
The GCODE is correct, but Marlin is not honoring the negative extrusion requests.

What am I doing wrong?!?

Best Regards
Davide Ardizzoia

[frozen...@yahoo.com]

right right, try this

G91
G92 E0

G1 E-5 F1000
G92 E0                
M104 S<TEMP>

This is what would work with my Mach3 setup, don't know much about firmware based controllers.

[Davide Ardizzoia]

Seems that I've bumped into a Marlin bug....

:-(

Thank you,
Davide Ardizzoia

Il giorno sabato 26 luglio 2014 17:09:27 UTC+2, frozen...@yahoo.com ha scritto:

right right, try this

G91
G92 E0

G1 E-5 F1000
G92 E0                
M104 S<TEMP>

[Davide Ardizzoia]

...But still, <E> behaviour seems broken to me.

Best Regards
Davide Ardizzoia