Finally completed long print - had to edit the pings in the g-code

I am using Canvas/KISSlicer/In Place Side Transition. I am using the 22.02.09.0 firmware - things are much worse with 22.03.29.0. My historical average is 100.4%.

I have been trying to get through a ~13 hour print, but every attempt fails due to missing pings. Sometimes it is right away, sometimes it is after several hours, but as soon as I miss a couple pings it quickly snowballs until transitions are happening inside the object and not the purge. The transition length is around 100mm, so there is not a lot of room for error.

I have never seen any sign that my extruder is slipping or missing, so the random appearance of missing pings always seemed odd. I know that there might be something physical causing it, but what I have found is that there is also a possible software reason.


One thing that I have learned is that as long as I do not miss any pings (or maybe 1?), the firmware does a great job of keeping the transitions showing up where they should, even if a ping is a little higher or lower than usual. If you look at the failures stacked on the left you will see that the firmware was even able to get things back on track, but the damage was already done. The 2 on the right are the good ones.

Block Image

After multiple failures I now have 2 successes in a row and would like to share what I had to do (#3 is the the most important!)

1) In Place Side Transition is missing the g-code to move my nozzle to the transition location (x=-9, y=0). I had to add the g-code G1 X-9 Y0 F7200 to the BEFORE SIDE TRANSITIONS sequence.

2) The Ping extrusion is using an xy travel speed instead of an extrude only feed rate. The ping extrusion is:

G1 E20 F4200

which is 70mm/s. No hot end is going to extrude 20mm of filament in 2/7th of a second! I had to manually update the g-code and replace the ping extrusion with one at 3 mm/s:
G1 E20 F180

3) A ping is a 13s pause, 20mm extrusion, 7s pause. When I look in the g-code I see a lot of pings that look like this:

; move to side

G1 X-9 Y0 F7200

G1 E0.7 F2400 ; destring prime

G1 F2100

; ping 1 pause 1

; totalE = 385.688416

G4 P4000

G1

G4 P4000

G1

G4 P3724

G1 E20 F7200

; ping 1 pause 2

; totalE = 405.688416

G4 P4000

G1

G4 P1588

G1 E-0.7 F2400 ; destring suck

; leave side

I am sure there is a reason that the G4 pauses do not add up to 13s and 7s, maybe it is trying to account for what happens before and after the ping? Either way if I have a ping that fails it almost always looks like this.

I manually updated the g-code file, changing every ping so that it starts with a 1mm extrusion, does 13s pause, 20mm extrusion, 7s pause, and then another 1mm extrusion:


; ping 3 pause 1

; totalE = 1126.489746

G1 E1 F180 ;start with 1mm extrusion

G4 P4000

G1

G4 P4000

G1

G4 P4000

G1

G4 P1000

G1 E20 F180 ;the ping extrusion

; ping 77 pause 2

; totalE = 27580.345703

G4 P4000

G1

G4 P3000

G1 E1 F180 ;end with 1mm extrusion

G1 E-0.7 F2400 ; destring suck

; leave side

I know this adds 2mm to every ping (so my printer is pulling an extra 2mm every 350mm), but this is easily accounted for by the firmware. I figure if the firmware misses pings that looks like this then I might be doomed to nothing but failure.

After 5 failures before my g-code surgery I have now made it through the print twice. A total of 26 hours, over 700 slices and 300 pings, with no missing pings. I have to admit that maybe I just got lucky twice, but it seems unlikely - I guess time will tell…

Now on to the next problem - the KISslicer print quality. I don’t know if there is anything I can do to improve it, and I am pretty sure that I can do better with my superslicer profiles and P2PP. In the current version of P2PP, Side Transitions with Accessory Mode are not generating pings, but Tom knows this is missing and once he get’s the work done I will do the print again and compare print quality. I have a strong suspicion that superslicer/P2PP is going to be my workflow in the end, but at least I will know that I gave canvas/KISslicer a fair shot.



Answer this question I have this problem too

Is this a good question?

Score 1
Add a comment