UPDATED! Slic3r PE over extrusion and cooling solved  

Re: UPDATED! Slic3r PE over extrusion and cooling solved

Since nozzle size is originally 0.4mm i really don't see any usefull need for 0.35, since all "decent" manufacturers and experts strongly advice that layer height shouldn't be bigger than 50% of dozzle diameter. At that height layers just won't stick together that good as they do at lower heights.
And, as leonardo says: you can always add 0.35 profile yourself, it's not a big deal...

Re: UPDATED! Slic3r PE over extrusion and cooling solved

You can just replicate the profile yourself if you really need it 🙂

Technically true but if 0.35 is dropped, all official support for 0.35 will go away for the MK3, making this printer a major downgrade for some users.
It's not just like a small adjustment everybody can live with and I really think dropping 0.35 could become a serious problem.

Re: UPDATED! Slic3r PE over extrusion and cooling solved

Since nozzle size is originally 0.4mm i really don't see any usefull need for 0.35, since all "decent" manufacturers and experts strongly advice that layer height shouldn't be bigger than 50% of dozzle diameter. At that height layers just won't stick together that good as they do at lower heights.
And, as leonardo says: you can always add 0.35 profile yourself, it's not a big deal...

I have printed a lot of items in PETG 0.35 and have had no issues with layer adhesion at all.

0.35 on the MK2 is fast, strong and looks surprisingly nice.

Re: UPDATED! Slic3r PE over extrusion and cooling solved

...

As for 0.35 profile, doesn’t it depend on perimeter to infill ratio of model? E3d won’t keep up on infill but I think it can on perimeter. I need to calculate. If it can’t handle perimeter speed at 0.35 I agree to drop it.

I agree and even if the model is only printed at the same speed layer for layer, it still prints nearly at double speed in z-direction.
Models printed in 0.35mm "vase" mode, should print faster and for larger vases print times could be more important than layer height.

I did some quick math and 0.35 is still a win in most cases. So I would like to keep that profile.

Re: UPDATED! Slic3r PE over extrusion and cooling solved

You can just replicate the profile yourself if you really need it 🙂

Technically true but if 0.35 is dropped, all official support for 0.35 will go away for the MK3, making this printer a major downgrade for some users.
It's not just like a small adjustment everybody can live with and I really think dropping 0.35 could become a serious problem.

Ok.. it's your opinion and everyone is entitled to one so i respect it, but honestly i think you're making a big deal out of it... A major downgrade??

Re: UPDATED! Slic3r PE over extrusion and cooling solved

Sure, we can hack together any kind of profile we want on our own. But I put alot of faith in the Prusa provided profiles, as they all have produced fantastic results. I am no wiz when it comes to calculating retractions, and accelerations, so I am very grateful for Prusa providing those. If the .30 & .35 profiles are dropped, they wouldn't be updated with any new improvements that other profiles would receive.

Nothing says you have to use the profile, but it sure is useful for those of us that have been using it.

Re: UPDATED! Slic3r PE over extrusion and cooling solved

I didn't get an answer for my questions so i'll try again.

How does M221 S95 work in relation with the tweaks in slic3r flow settings?

Is there any way to see the current flow percentage when a print is running?

/Patrik

Re: UPDATED! Slic3r PE over extrusion and cooling solved

I did some quick math and 0.35 is still a win in most cases. So I would like to keep that profile.

As I understand the question was whether to keep the 0.35 mm FAST profile. This profile violates the volumetric extrusion limit of the hotend quite a bit, asking for hotend jams (mine cost me a few days to fix). So my suggestion would be to dump this profile. In addition, I'd delete all nonworking historic settings or at least mark them as "experimental". I don't see the point of having outdated stuff around. It's frustrating for new users and not necessary for pros, as they anyway can create their own.

As for a working 0.35 mm profile - some users think it makes sense so why not. However, only if it works - which means reducing the speed to stay within the extrusion limit of the standard hotend.

- Martin

Re: UPDATED! Slic3r PE over extrusion and cooling solved

I did some quick math and 0.35 is still a win in most cases. So I would like to keep that profile.

As I understand the question was whether to keep the 0.35 mm FAST profile. This profile violates the volumetric extrusion limit of the hotend quite a bit, asking for hotend jams (mine cost me a few days to fix). So my suggestion would be to dump this profile. In addition, I'd delete all nonworking historic settings or at least mark them as "experimental". I don't see the point of having outdated stuff around. It's frustrating for new users and not necessary for pros, as they anyway can create their own.

As for a working 0.35 mm profile - some users think it makes sense so why not. However, only if it works - which means reducing the speed to stay within the extrusion limit of the standard hotend.

- Martin

Slic3r automatically throttles speed to not exceed the volumetric setting. So the FAST profile still worked fine. It was just not as fast as advertised. But for many prints it is still faster.

Re: UPDATED! Slic3r PE over extrusion and cooling solved

Sure, we can hack together any kind of profile we want on our own. But I put alot of faith in the Prusa provided profiles, as they all have produced fantastic results. I am no wiz when it comes to calculating retractions, and accelerations, so I am very grateful for Prusa providing those. If the .30 & .35 profiles are dropped, they wouldn't be updated with any new improvements that other profiles would receive.

Nothing says you have to use the profile, but it sure is useful for those of us that have been using it.

+1!
Very good summary!
Keep the profile.

Re: Slic3r PE over extrusion and cooling solved

Thanks Joe. Any chance the extrusion issue may also apply to the MMU since it uses bondtech as well?

You can try it out easily yourself by adding M221 S95 to the start gcode. I am last at the office and too exhausted to test it today. We will run test tomorrow 🙂

I like the idea of using the M221 so you can adjust the gcode on the fly for different materials but I noticed an issue which I wanted to ask. First I printed a benchy with the M221 in the start code and it worked fine, I then went to print another file that I forgot had the extrusion multiplier modified from the filament setting to 0.95 . I noticed that it was under extruding is this because the the S95 was saved in the memory from the last print so was doing 95% flow on the new print which was already set to 0.95?

If that is the case would it be best to add M221 S100 in the end code to reset the flow back to 100% incase someone tries to print an old gcode file?

Re: UPDATED! Slic3r PE over extrusion and cooling solved

You can just replicate the profile yourself if you really need it 🙂

Technically true but if 0.35 is dropped, all official support for 0.35 will go away for the MK3, making this printer a major downgrade for some users.
It's not just like a small adjustment everybody can live with and I really think dropping 0.35 could become a serious problem.

The 0.35 FAST is limited to 6 mm3/s for perimeters per example.
If you choose 0.20mm 100mms Linear Advance, you go up to 10 mm3/s and reach the maximum for PETG on a E3Dv6. Then this profile is potentially faster than "0.35 FAST".
Personally I tend to use 0.25 layer height to be closer to the maximum for ABS (13mm3/s).
I would never go back to "0.35 FAST" because:

Re: UPDATED! Slic3r PE over extrusion and cooling solved

Folks,
I'm not sure that the overextrusion has been resolved with Slic3r 1.38.7. (ETA: I used the wizard to update profiles, but it didn't insert the M221 S95 command into the code. Should it have?)

I believe that Slic3r is still overextruding, which is particularly evident on top surfaces with numerous location changes and small fills. Simplify3D prints a very reasonable top surface.

In the image below, the left fish head is printed at 200c, middle at 190c. No material difference. For the right head, I dropped my extrusion multiplier by 0.08mm; there is no significant improvement.

Whenever I see this type of surface I immediately think over extrusion. However, the extrusion steps on my Prusa MK2 are perfectly calibrated to extrude 100mm, and the extrusion multiplier calibrated to produce single perimeter vertical walls that measure exactly 0.45mm. I've adjusted my Z steps/mm so my calibration tests measure exactly the expected height.

I've done upwards of 50 test prints, with every different setting imaginable, and every single top surface is extremely rough.

Here's what I've tried; there is no material difference with the top surface at all.
-Tightened all bolts/nuts/screws/belts on extruder body/carriage
- 5 different colors of Hatchbox filament, including a brand new roll
- Extruder Temps @ 210, 200, 195, 190, 185
- Extrusion multiplier of 1.04, 1.0, 0.96
- Extrusion width of solid infill and top infill values of 0.4, 0.35, 0.3
- Filament size of 1.74 (average for Hatchbox), 1.72
- Linear advance K30, K45, 0 (commented out M900)
- Top surface speed of 70mm/s, 40, 30, 20mm/s
- Decreasing height to 5 layers (1 bottom, 1 infill, 3 top) to avoid micro stepping to odd layer height
- Printing 2-up to give layers a chance to cool
I've recently started running a MicroSwiss 0.4mm nozzle, but this problem pre-dates the nozzle, if I remember correctly.

Here is a link to the STL file: https://www.dropbox.com/s/objodpnbsfewxk4/Top%20Surface-fish%20head.stl?dl=1

And to the Slic3r .ini file: https://www.dropbox.com/s/rwnxgduwyq4h1am/Top%20Surface-fish%20head.ini?dl=1

Re: Slic3r PE over extrusion and cooling solved

If that is the case would it be best to add M221 S100 in the end code to reset the flow back to 100% incase someone tries to print an old gcode file?

Jason, did you ever get an answer to this? If anyone uses S3D or Cura as well and the M221 is persistent, it seems like a good idea.

Re: Slic3r PE over extrusion and cooling solved

If that is the case would it be best to add M221 S100 in the end code to reset the flow back to 100% incase someone tries to print an old gcode file?

Jason, did you ever get an answer to this? If anyone uses S3D or Cura as well and the M221 is persistent, it seems like a good idea.

I didnt but looking at the menu on the printer after it starts it just tweaks the flow setting and it does seem persistent so in my end code I just added m221 s100 to set it back to 100 after every print

Re: UPDATED! Slic3r PE over extrusion and cooling solved

https://github.com/prusa3d/Prusa-Firmware/pull/482 should end up in a new test firmware scheduled for next week.