• Welcome to the new Distech Automation Forum!
  • Feel free to post questions, comments, and feedback.
  • Ask us anything!
Hello There, Guest! Login Register


Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
KISSlicer + Post-Processor
#1
Hi Guys

If anyone is interested in testing KISS with the Prometheus System, there is a post-processor available, originally designed for the Prusa Mk2/S MMU but adapted to work with a range of SNMM systems.

Details available on the KISSlicerTalk forum here: http://www.kisslicertalk.com/viewtopic.php?f=27&t=1988

I currently own 2 x Prusa + MMU printers so I have not had the opportunity to test with other systems so any feedback/fault reports will be gratefully received!

Thanks in advance

Peter
 
Reply
#2
Nice to see another tool to enhance multi-color printing. I finally got time to test this out and seems to do it's thing, but there are few issues. I use the custom unload/load strategy as the Prometheus, especially 4x version, needs different unload/load lengths.

Extruder steps seem to be set wrong; in the program I set the steps to 500, but in the result gcode:
;MMPP Set Extruder steps
M92 E344.82759

In unload, i've set Pause After to 2000 ms. But in result gcode:
M104 S205
G1 X66.75 Y95.75 E-20 F1500
G1 X56.75 Y95.75 E-15 F1500
G4 P0
G1 X66.75 Y95.

I'll fix the gcode by hand for now and do a test print
 
Reply
#3
(12-06-2017, 01:20 PM)spegelius Wrote: Nice to see another tool to enhance multi-color printing. I finally got time to test this out and seems to do it's thing, but there are few issues. I use the custom unload/load strategy as the Prometheus, especially 4x version, needs different unload/load lengths.

Extruder steps seem to be set wrong; in the program I set the steps to 500, but in the result gcode:
;MMPP Set Extruder steps
M92 E344.82759

In unload, i've set Pause After to 2000 ms. But in result gcode:
M104 S205
G1 X66.75 Y95.75 E-20 F1500
G1 X56.75 Y95.75 E-15 F1500
G4 P0
G1 X66.75 Y95.

I'll fix the gcode by hand for now and do a test print

Thanks for the update; I will check out these faults and get them fixed.

Peter
 
Reply
#4
Ok the print. I did aim quite high as I tried printing 3DBenchy with profile that is allowed to use layers from 0.1mm to 0.3mm. With Benchy, most of the layers are below 0.2mm due to the steep angles. Sadly I ran out of blue filament during the print due to massive purge tower (as expected with low layer heights). But the print was going on strong so the post processor is doing it's job nicely Smile

[Image: 2017-12-09%2020.46.06.jpg?dl=1]
[Image: 2017-12-09%2020.46.22.jpg?dl=1]
Purge tower, which suffered some bulging at two corners. Some material buildup was happening...
[Image: 2017-12-09%2020.46.39.jpg?dl=1]
[Image: 2017-12-09%2020.47.27.jpg?dl=1]
 
Reply
#5
(12-11-2017, 08:38 PM)spegelius Wrote: Ok the print. I did aim quite high as I tried printing 3DBenchy with profile that is allowed to use layers from 0.1mm to 0.3mm. With Benchy, most of the layers are below 0.2mm due to the steep angles. Sadly I ran out of blue filament during the print due to massive purge tower (as expected with low layer heights). But the print was going on strong so the post processor is doing it's job nicely Smile

Many thanks for the update.  Been very busy (wifey just had an operation on her foot) so haven't had a chance to look at program.  Planning on some time tomorrow/Wednesday.

With those colours, I think you could reduce the red purging by 50% and the blue by 20% (Red filament "PURGE=25" and Blue "PURGE=40" in filament names)

On load, you may want to reduce the final load distance by 2mm, so you unload 2mm more than you load.  There's a lot of filament left behind in the nozzle and equal load/unloads tend to build up the excess.

What material, KISS PreloadVE and retraction settings are you using?  You should be able to eliminate those wispy strings.

Peter
 
Reply
#6
Yeah there's tuning needed. I recently bough some Capricorn bowden tubes and my retraction settings are too low due to testing how well the tubes are working. I'm not sure about the preload, I'll need to check that. Material is PLA, two different brands.
About the load distance: currently load is 5mm less than the unload. I did a test print that had something like 0.2mm minimum layers and it did not show any signs of material bulging.
I'll post a pic of my settings later.
 
Reply
#7
Basically at the start and end of the purge tower, the previous/next PreloadVE values are taken into account and if that is set too high, there may be some over-extrusion at the start.

However, I just didn't worry about it; it's waste so it really didn't matter to me.  If it is a concern or causes bad prints, I could reduce the extrusion at the start.

Peter

P.S. I have a few metres of Capricorn for testing as well; so far seems to need lower PreloadVE values than regular tubing.
 
Reply
#8
Ok, I haven't really tested the preload that much, just ran the wizard some time ago and left it. I'll need to adjust the settings after changing the tubes, maybe run the wizard again.

Here are the PrusaPP settings I've tested.
Oh and the file preview doesn't seem to work, I only have grey screen.

[Image: Screenshot%202017-12-12%2020.58.55.png?dl=1]
 
Reply
#9
(12-12-2017, 07:07 PM)spegelius Wrote: Ok, I haven't really tested the preload that much, just ran the wizard some time ago and left it. I'll need to adjust the settings after changing the tubes, maybe run the wizard again.

Here are the PrusaPP settings I've tested.
Oh and the file preview doesn't seem to work, I only have grey screen.

Looks like you have some settings incorrect.

The extruder calibration (145) can't be right - unless your firmware is very wrong.

This is designed to correct small differences in the extruder drive pulley of the extruder motors, and based on the length of filament extruded when a "G1 E100 F150" command is issued (via Pronterface etc).

It requires the firmware E-Steps in order to set the correct E-Steps for each extruder - Marlin (certainly PR Marlin) only stores a single E-Steps value which is used for all extruders.

The post-processor calculates the revised E-Steps based on 100/Length_Extruded * E-Steps (in your case, 100/145 * 500 = 344 steps)  If both your extruders are around the 100mm for the "G1..." command, leave them set to 100.

The Original PR Mk2 MMU used single Brass drive pulleys with between 155 and 180 E-Steps/mm.  The later use of Bontech dual drive geared pulleys is pretty much exactly 140 steps/mm for all 4 extruders.

Any chance that you could upload your Benchy G-code (before post-processing) so I can have a look at a few things?

Thanks for your help with this!

Peter

V1.5 RC2 is now available on my FTP.

This fixes the G4 P0 problem.

I hope I managed to explain your E-steps issue (subject to you agreement) so it should now be good to go.

I would still like to see your G-code...

Many thanks again for your assistance with this.

Peter

Just one more thought

The model/tower preview generally works OK for me; on occasion (maybe 1 in 20 times) the preview isn't displayed. Don't know why this is, but on my old PC it may have something to do with the graphics drivers.

Peter
 
Reply
#10
Ah okay, that makes sense. I was wondering what the value is, thought it was some kind of feed length. Yes that explains the E-steps being wrong.
I haven't measured the actual feed vs. expected with ProStruder's but I'd imagine it should be quite close, the teeth of the pulleys bite quite well, especially with PLA. But I'll check that too.

G-code for the Benchy: https://www.dropbox.com/s/emf5qequs6ajx1...gcode?dl=1
 
Reply
  


Forum Jump:


Browsing: 1 Guest(s)