KISSlicer 1.6 Beta  

Re: KISSlicer 1.6 Beta

See attached for destring tuning...

Peter

Thanks, that was easy and fast 🙂

Embarassing I didn't think of creating a new stl file with the models placed as required 😳

Thanks a lot Peter!

Re: KISSlicer 1.6 Beta

Thanks a lot Peter!

Pleasure 😀 😀 😀

Peter

Re: KISSlicer 1.6 Beta

Sorry for bothering you again Peter, but I have a couple of questions to your settings.

I don't have the MM upgrade so I adjusted the settings and GCodes according to your recommendation:

For the single filament printer, you will need to take one of the MM versions and copy that to a new printer, reduce the number of extruders and change the purge line in the start GCode section.

Also in the start GCode, there are three printer specific lines.

M301 - extruder PID (remove or change this line)
M304 - bed PID (remove or change this line)
G80 - bed level correction ((remove the additional parameters)

The purge lines of:

G1 Y-3 Z0.250 F7200.000
G1 X50.0 E75.0 F1000.0
G1 X160.0 E10.0 F1000.0
G1 X220.0 E6 F1000.0
G1 X240.0 E2 F1000.0
G1 Y-1.
G1 X40 E15 F1000
G1 X0 E2 F1000.0
G1 E-4 F4800
G92 E0.0

need to be changed to this:

G1 Y-3 Z0.250 F7200.000
G1 X160.0 E5.0 F300.0
G1 E-4 F4800
G92 E0.0

Or something similar.

Peter

So after the 120s PINDA calibration the purge line starts. But this line does not stick to the head bed and is not a straight fine line. Instead it is a line consisting of dots connected by small strings of filament (hard to describe, I can post a picture this evening...) I can hear the extruder motor making sounds representing the dots.
I have limited knowledge of GCode, but
1. The problem the line does not stick is because of the Z0.250 (maybe not the reason, but adhesion is decreased, right?).
2. Extruding 5mm on a 160mm path doesn't sound much (does it?),so that may be the reason the line is not a real line, right?
3. After the purge line the model is printed. I have the problem that most of the time the perimeter of the first line is not printed properly. I think it's because of the G1 E-4 command, correct? So 4mm are missing for the model... Is this the problem or is using a skirt always a good idea?

Could you elaborate a bit about these commands? Are there maybe better values? Is there a better way of finding the right values except trial and error?

Sorry for asking so many questions 😳 The learing curve for KS is steep and I hoped I benefit from your knowledge

Regards

Manu

Re: KISSlicer 1.6 Beta

Manu, not a problem.

Ensure there is a G92 E0 before the prime line is printed - I think I missed that out, sorry.

5mm of filament - at 2.4mm^3/mm that's 12 mm^3. Extruding a line 0.25mm high over 160mm = 0.3mm linewidth, which is sufficient to prime the nozzle. You could use maybe 6mm, but you would then have greater pressure in the nozzle leading to possible ooze when moving to the part.

The RC2 of KISS (just released) includes a <DESTRING> token. Retract at the end of the prime should now be:

G1 E-<DESTRING> F4800 ; NOTE the minus sign after E...

KISS G-Code should include a retract prime before starting the skirt/model.

Seems that this is a constantly-moving situation, and I am having difficulties keeping up. Sorry...

Peter

P.S. Also check the Settings/Printer/Extruder Hardware/DrvLen setting. Also ensure that NozLen is set to 0.6 (I received updated info from E3D last week).

Re: KISSlicer 1.6 Beta

Ensure there is a G92 E0 before the prime line is printed - I think I missed that out, sorry.

So G92 before and after the prime line?

So a good point to start from would be

G92 E0.0
G1 Y-3 Z0.250 F7200.000
G1 X160.0 E5.0 F300.0
G1 E-<DESTRING> F4800
G92 E0.0

Is lifting the nozze by 250 microns is necessary? Or whats the intention?

The RC2 of KISS (just released) includes a <DESTRING> token. Retract at the end of the prime should now be:

Will definitely try this one!

KISS G-Code should include a retract prime before starting the skirt/model.

Seems that this is a constantly-moving situation, and I am having difficulties keeping up. Sorry...

not a problem at all. I am very thankful for your help.

Best Regards

Manu

Re: KISSlicer 1.6 Beta

Is lifting the nozze by 250 microns is necessary? Or whats the intention?

No that's lifting the nozzle by 100 microns to a height of 250 microns (sorry; I am being pedantic today...)

I find that the priming line at the start needs to be able to flow nicely without building up too much pressure in the nozzle. As mentioned, you could increase the flow a little to feed 6mm of filament and increase the width of the line a little.

When a print finishes, it should do a retract suck at the end, usually 1 mm or less. So at the start of the next print, you will need to replace that plus any ooze that has occurred during cool-down of the last print. 5mm should be enough, but may not be.

Peter

Re: KISSlicer 1.6 Beta

No that's lifting the nozzle by 100 microns to a height of 250 microns (sorry; I am being pedantic today...)

Oh yes, absolut values...

I find that the priming line at the start needs to be able to flow nicely without building up too much pressure in the nozzle. As mentioned, you could increase the flow a little to feed 6mm of filament and increase the width of the line a little.

Seems legit. That's a good point to start. I will adjust the values that my purge line sticks to the bed...

When a print finishes, it should do a retract suck at the end, usually 1 mm or less. So at the start of the next print, you will need to replace that plus any ooze that has occurred during cool-down of the last print. 5mm should be enough, but may not be.

Got it, thanks! I will have a look into this in detail and try this by myself!

Thanks again

Manu

Re: KISSlicer 1.6 Beta

Alright, did some testing with the latest RC. Works quite good.
I don't move the nozzle during prime to .250, otherwise the purge line doesn't stick. But it isn't squished very hard, so its ok for me 🙂

I'm having some trouble with the seam on .3 layer heights, but that's something I'm willing to find out on my own...

One minor thing bothers me: My STL files are located on my computer and I store my GCodes on my Raspberry Pi running OctoPrint. Everytime I save a sliced gcode file to my pi and want to open another STL file the default is the last location I stored the gcode file to. So input and output folder are treated as the same location. Is there somewhere a checkbox I can (un)tick to say KS that these should be different folders (Slic3r handles this that way). Or am I just using it wrong? I'm pretty sure I'm not the only one working with this setup.

Cheers

Manu

Re: KISSlicer 1.6 Beta

You may find that you need different preload values when printing thicker layers. That's my only problem when printing variable layer heights...

I have never used File/Open STL Model. Always drag/drop files onto KISS (or the shortcut). And I always save to my desktop and drag the GCode files to Octo. Just my way of doing things...

Peter

Re: KISSlicer 1.6 Beta

I've started using KISS last night by doing the PreloadVE tuning wizard. I'm using black filament so it is not easy to see were is the best Z. Moreover, the printing quality is already pretty good...
But I found a value which is not so far from the value I've computed through the oozing experiment.
Peter, which value should I trust more? The value through the wizard or the value through the oozing experiment?

Next step tonight with the destring wizard, but so far so good, I think that I'm already in love with this slicer although the first contact is a bit rough (the UI is one of the main reasons).

Re: KISSlicer 1.6 Beta

Your eyes are the best judge...

But in all honesty, I really don't know. The ooze test should be fairly accurate, but it's not a "real world" test. The nozzle keeps moving over the part; the ooze keeps going a lot longer than in the test, so the actual value may be significantly more than the ooze test value.

Peter

Re: KISSlicer 1.6 Beta

Mmmhh I now have a doubt... The first wizard I made was between 0 and 0.5 since the experimental value was around 0.389. Maybe should I have tested it on the full 0-1 range?

Re: KISSlicer 1.6 Beta

Yeah, I think I would have run it with 0.2 to 0.8 and then again maybe +- 0.15 from the best visible.

Peter

Re: KISSlicer 1.6 Beta

Ok, thanks for the advice. I guess that I will start with this tonight before the destring wizard.

By the way, those wizards are something I really like about KISS.

Re: KISSlicer 1.6 Beta

Reprinted the PreloadVE tuning part with the 0-0.8 range... If I've had read a second time the associated help, I would have done that the first time: "PreloadVE Coefficient [µs] (0 - 2x Experimental Wizard value)".
But I've made another mistake last night: I though that the material G-Code was automaticaly added to the G-Code, but you have to add the <MATL> token to the pre fix G-Code, so my extruder wasn't properly calibrated.

But my issue is now how to choose the best Z... The print is pretty good all the way up and the seam seems to be about the same. What are your criteria to choose the proper Z Peter?

edit: I may have mispoken about the seams, I can see a tiny little difference. A part of it the less wide than the rest. Is it the sweet spot?

Re: KISSlicer 1.6 Beta

The destring wizard after the preload one is quite impressive! After less than 7mm (on 50mm as advised by the help), there is no more strings at all knowing that I've choosen a 0-1mm range for this test. So this means that 0.14mm are now enough to be compared to the default 0.8mm given by Prusa in Slic3r!

Re: KISSlicer 1.6 Beta

A part of it the less wide than the rest. Is it the sweet spot?

Yup.

Peter

Re: KISSlicer 1.6 Beta

The destring wizard after the preload one is quite impressive! After less than 7mm (on 50mm as advised by the help), there is no more strings at all knowing that I've choosen a 0-1mm range for this test. So this means that 0.14mm are now enough to be compared to the default 0.8mm given by Prusa in Slic3r!

Not really a surprise; what that actually means is that you can increase the Preload a little and set Destring to zero. That's the ideal but you really are better off with a small destring and slightly lower preload values.

In theory, Preload should do the destring prime and suck by itself, but especially when printing variable layers it can't quite get the calculation 100%.

Peter

Re: KISSlicer 1.6 Beta

Ok, perfect, thanks for the tips again 🙂
I've already tried to slice some models with the variable height just to take a look, and now I can't wait to finish the tuning so I can print one!

Re: KISSlicer 1.6 Beta

I think I am still half drunk but I cant figure out how to make the changes to add the Single extruder version of the Printer

Also I see that the Release Candidate is out now so I downloaded that