Page 1 of 2

Filament Calibration - Hatchbox PLA

Posted: 17 Jun 2019, 01:40
by jenovaEU
Hello Everyone,


Recently I decided to give KISSlicer a try. I have heard of KISSlicer in the past from Michael Hackney but never gave it a real try. After going through several of Michael Hackney's videos and trying to understand KISSlicer, I switched to KISSlicer for my 3D printer. [Ultibots D300VS]. The preconfigured profile prints amazingly well. Better than other slicers I have used in the past.

In an effort to better understand how KISSlicer works, I decided to follow the filament calibration steps [tuning wizard] to tune my Hatchbox Wood PLA. After a day of running test prints I got the stringing down to as little as possible. However I am still trying to tune it out completely.

The following are the material settings I settled on [for now]. Any feedback or tips are appreciated.
Hatchbox Wood PLA Material Settings
Hatchbox Wood PLA Material Settings
Capture.PNG (42.11 KiB) Viewed 1117 times
Destring: Currently my Destring is set to 4mm. I am aware that for my hotend [E3D V6 DirectDrive] this is more than recommended. According to what I have read on other sites it is not advised to go beyond 3mm. However after several Destring Length tuning prints I concluded that this was the minimum length for the least amount of stringing. The following is a picture of my last Destring Length test. This test starts at 2mm and goes to 5mm at the top. I also did several of these prints with wipe lengths ranging from 1mm to 3mm to see if this would improve and reduce the destring length required but wiping did not seem to make any difference.
Destring Length Test 2mm to 5mm
Destring Length Test 2mm to 5mm
Destring Speed [velDstr]: I do not fully understand how to tune Destring speed. I did some experimentation with speeds from 25mm/s to 60mm/s but saw no noticeable change in the amount of stringing. If someone has an explanation for how this should be tuned, I'd love to hear it.

Wipe: Another tuning point where I could not find a good explanation for how this should be tuned. Therefore I experimented with 1mm to 3mm wiping. I did notice that when the wipe length is set to 3mm,a second wisps of filament would appear but would also disappear once the Destring Length was at least 4mm. Does someone have an explanation for how to tune wipe length?

Z-Lift: This value was easier to figure out. The higher the value the more stringing it created. The default value used for PLA in the preconfigured profile was 4mm. I reduced this to 1mm and noticed an improvement. I have thought of trying less than 1mm but figured that might be a bad idea.

Here's a test print I did after a few hours of testing, calibrating, and verification. This print looks good. I see some small defects. However being a perfectionist, I want to tune out the wisps and get the print to be as clean as regular PLA.
Aztec Chief 1
Aztec Chief 1
Aztec Chief 2
Aztec Chief 2
Aztec Chief 3
Aztec Chief 3
Aztec Chief 4
Aztec Chief 4
Any feedback or tips are greatly appreciated.

P.S. If this is in the wrong category on the forum, Let me know and i'll move it to the right place.


Sincerely,
Jenova

Re: Filament Calibration - Hatchbox PLA

Posted: 17 Jun 2019, 02:38
by pjr
You raise a few points, but here are my thoughts:

I have never been able to stop *fill (wood, copper, bronze etc) stringing. There's always some stringing, just the nature of that type of filament.

Destring speed is not particularly important with PLA; you have to retract fast enough to empty the nozzle threat, but not too fast so that the melted filament separates, leaving some behind. I tend to use 25 to 35, but for *fill I think slower may be better; you could try 20 or even 15.

The V6 will cope with retractions up to about 6mm, but may start to clog if you have too many retractions in a short time frame.

Other things you can try:

Preload will help and you can reduce retraction distance significantly.
Turn off wipe. With some filaments, this can encourage stringing
Ensure your travel speed is as fast as possible
If your model will allow, turn off Z-lift

The idea here is to finish a path and then get to the start of the next path as quickly as possible. If you delay anywhere (wipe, Z-lift, retraction), you are more likely to get stringing.

Hope this helps.

Peter

Re: Filament Calibration - Hatchbox PLA

Posted: 17 Jun 2019, 06:51
by jenovaEU
pjr wrote:
Other things you can try:

Preload will help and you can reduce retraction distance significantly.
Turn off wipe. With some filaments, this can encourage stringing
Ensure your travel speed is as fast as possible
If your model will allow, turn off Z-lift

Peter

Hello Peter,


During my calibration of the filament I did the preload. I repeated the preload with Wipe turned off and Z-Lift turned off. I do like how the layers turned out but as you can see in the images below, At no point from 0x to 2x does the stringing become less. That is why in my previous post I had the preload set to zero.

The following are the preload test results. Bottom is 0x and top is 2x.
20190617_134340_2.jpg
20190617_134332_2.jpg
20190617_134318_2.jpg
I will repeat the Destring Test now with 0 wipe and 0 Z-Lift. I'll post those test results shortly.

Re: Filament Calibration - Hatchbox PLA

Posted: 17 Jun 2019, 08:11
by pjr
It would be interesting if you could zip up and post some G-code.

There are still some faults in the alpha versions which may affect/contribute to stringing.

Thanks

Peter

Re: Filament Calibration - Hatchbox PLA

Posted: 17 Jun 2019, 08:56
by jenovaEU
pjr wrote:It would be interesting if you could zip up and post some G-code.

There are still some faults in the alpha versions which may affect/contribute to stringing.
Peter
Hello Peter,

I'm running the stable release. Version 1.6.3
Here's the G-Code for the PreloadVE I posted previously.
PreloadVE Wizard_Wood PLA_0-2.zip
(70.13 KiB) Downloaded 51 times
I am doing a few Destring tests now at diffent retraction speeds to see if the results change. The stringing is much cleaner now with wipe and z-lift set to zero. I will post my testing results soon.


Sincerely,
Jenova

Re: Filament Calibration - Hatchbox PLA

Posted: 17 Jun 2019, 16:15
by pjr
FWIW, in that model there is no destring operating, only preload.

Not sure whether that's a fault in 1.63 of just the way the model is sliced. In any event, without the destring, there will be some stringing.

Peter

Re: Filament Calibration - Hatchbox PLA

Posted: 17 Jun 2019, 21:34
by jenovaEU
pjr wrote:FWIW, in that model there is no destring operating, only preload.

Not sure whether that's a fault in 1.63 of just the way the model is sliced. In any event, without the destring, there will be some stringing.

Peter
Hello Peter,


Maybe a little miscommunication. The G-Code I posted was for the Tuning Wizard - PreloadVE. So yes its supposed to have only preload and no destring.


Sincerely,
Jenova

Re: Filament Calibration - Hatchbox PLA

Posted: 17 Jun 2019, 21:46
by jenovaEU
Hello All,


Based on the tips from Peter I reran some test. The following Tuning Wizard Destring tests were done with 0mm Z-Lift and 0mm Wipe.

The following test was done at 60mm/s retraction speed. After 35mm [or when the retraction is 4mm] the stringing stops. I also checked the third/last pillar in the row for defects to confirm this.
60mms Retraction Speed
60mms Retraction Speed
60mms Retraction Speed - Last Pillar
60mms Retraction Speed - Last Pillar
Becuase Peter mentioned that slower retraction might be best for filaments like wood PLA. I repeated the test with retraction speed set to 20mm/s. After 35mm [or when the retraction is 4mm] the stringing stops just like the previous test. I also checked the third/last pillar in the row for defects to confirm this.
20mms Retraction Speed
20mms Retraction Speed
20mms Retraction Speed - Last Pillar
20mms Retraction Speed - Last Pillar
Based on these test results. I concluded that the retraction length should be 4mm and the speed doesn't seem to matter. I repeated the test but set the retraction to start at 4mm and stop at 5mm. The test results were clean. No stringing at all.


Sincerely,
Jenova

Re: Filament Calibration - Hatchbox PLA

Posted: 17 Jun 2019, 22:21
by jenovaEU
Hello All,


Now with the settings for Hatchbox Wood PLA dialed in as tight as I can get. I did a few test prints. The first print was done with Z-lift set to 0mms. This was very interesting to watch but also nerve wracking. I was constantly worried the hot end might bump into the print and it would fail. I sat next to the printer for the whole print. The Benchy turned out amazing. Ignoring the very fine wisps the quality is great. The Surface finish is amazing!


Print 1 - 0.2mm Benchy
0.2mm Benchy 1
0.2mm Benchy 1
0.2mm Benchy 2
0.2mm Benchy 2
Benchy 0.2mm Scar
Benchy 0.2mm Scar
Circled in blue, You can see the seam scar.


In KISSlicer I did set the seam to be at the front of the model and jitter set to 0. However the seam was across the side like a scar. This might have been a bug in the slicer as the 0.1mm Benchy doesn't have this issue.

With the successful print of the 0.2mm Layer Height Benchy, I wanted to push the quality further and decided to reprint at 0.1mm Layer Height. For this print I set the Z-Lift 0.2mm. I did this becuase I feared the hotend crashing into the print.

Print 2 - 0.1mm Benchy
0.1mm Benchy 1
0.1mm Benchy 1
0.1mm Benchy 2
0.1mm Benchy 2
0.1mm Benchy - Seam
0.1mm Benchy - Seam
I am very satisfied with these settings and I do not thing I can completely remove the remaining thin wisps of filament but I can live with that. :lol:
The following are the settings I used for these prints.
Hatchbox Wood PLA [1.75mm] Settings
Hatchbox Wood PLA [1.75mm] Settings
Material Settings.PNG (27.54 KiB) Viewed 1092 times
From this whole filament calibration process I learned a few new things.
- Keeping the Z-Lift as low as possible helps with the print quality and reduces stringing.
- As peter mentioned and my testing confirmed. When using filaments with a fill like wood, Wiping can introduce stringing.

I have included the G-Code for these prints here. In case someone wants to have a look.
Benchy - Gcode.zip
(2.76 MiB) Downloaded 40 times

Sincerely,
Jenova

Re: Filament Calibration - Hatchbox PLA

Posted: 18 Jun 2019, 02:34
by pjr
A couple of other things.

Retraction speed. This may be limited by your firmware and when you have set it to 65 mm/s it may only be happening at the firmware limit (which was 50 mm/s in early "original" Prusa printers)

Parts warping with Z-lift = 0; generally, if you have sufficient part cooling and don't print too fast, you will not have a problem.

Peter