Does a Mk3s, in November 2020, need or benefit from an e-step calibration process?  

RE: Does a Mk3s, in November 2020, need or benefit from an e-step calibration process?

Boingomw, 

My Printers are still standard... 

Some people who start changing things like E setps, end up putting them back to standard, 
I have One Printer that is not standard, it has a 3:1 gear extruder, so Esteps are 3x normal... 
so effectively, still standard... ! Lol

Joan

RE: Does a Mk3s, in November 2020, need or benefit from an e-step calibration process?

I suspect if you slice with an older version of slicer and firmware  that your tolerance parts will fit again. Try it as an experiment.
I think the new K factor tinkering has some issues and that it will need to be tweaked over time to get back to the reliability it had. Which is why I’m running older firmware as there’s nothing in the new versions I had to have. 

RE: Does a Mk3s, in November 2020, need or benefit from an e-step calibration process?

If you have a moment, can you tell which firmware and slicer version you are running?

RE: Does a Mk3s, in November 2020, need or benefit from an e-step calibration process?

@boingomw

I have firmware 3.9.1 on my Mk3S and
Prusa slicer 2.2.0 is my gotta work... slicer.
I also have slicer 2.3.0Alpha4 which is my 'Play with the new features ' slicer...   
I also have Kissslicer and Simplify3D loaded but haven't used those for ages.

I have had a few issues with Alpha4 crapping out at slice time, or refusing to respond to mouse clicks any time soon....   so I am not happy with that version...
Alpha 1, 2 & 3 seemed better to me... 

Hope that helps, regards Joan

RE: Does a Mk3s, in November 2020, need or benefit from an e-step calibration process?

There is a high degree of consistency between Prusa printers of the same type, so you generally won't see much variation unless something has been changed or is wrong. When I tried the procedure, I came within 1-2% of the default setting.  On other printers, there is a lot more variability.

RE: Does a Mk3s, in November 2020, need or benefit from an e-step calibration process?

Mostly Conclusion:

The Prusa Mk3S most likely does NOT need an extruder calibration done, and it is likely NOT the cause of whatever problems you are having.  The testing process is easy to do, but you will likely find a very small variance.  Feel free to adjust, but you will probably end up adjusting back later.

What is VERY likely to affect things is the slicer version, firmware (unverified personally, but anecdotal evidence suggests this), and any G-code changes you have done to your start code.  Specifically, if you use the older slicer (2.1.1) and have the M221 S95 g-code "fix" in place, things will change drastically.  Test parts varied by almost 10%, depending on slicer version and g-code.  Even when the filament, printer, and print settings were visually identical.  For the record, I am on firmware 3.9.1.

What *IS* most likely to help is an Extrusion Multiplier calibration, which is a per filament type modification.  Also, it is probably a bad assumption (which started this whole quest for me) to think that just because you're using Prusament, you don't need to mess with the Extrusion Multiplier.  This SHOULD be the case, but apparently this IS NOT the case!

-----------------------------

A bit of a departure from the original question, but:  The original problem I was trying to fix was part tolerance (pieces fusing, rather than being able to print in place and move afterwards).  When I first got my printer, I ran this, https://www.thingiverse.com/thing:4322510 , and could get all but the .1 piece to move.  When I ran it again recently, everything below .22 was fused together.  Both were using Prusament; not the same exact spool (obviously) but "standard material".  I was confused.  Why had the gods forsaken me and broke my printer?  Had I not wasted days of time, and pounds of material to appease them?  Jerks!

So, I pined, and researched, and pined some more, and eventually asked questions, and got bad answers, and then got reasonable answers. And tested.  Here are my findings.

Extrusion Multiplier FTW.

I ran this test, https://www.thingiverse.com/thing:1622868, and plugged in the numbers for test parts run with different settings. The variance was kinda shocking, given my incorrect assumption that Prusament = standard.

The results were as follows:

"Muppet code" is the code from bobstro ( http://projects.ttlexceeded.com/3dprinting_prusaslicer_gcode.html#start-g-code).   I originally replaced all but the last two really odd lines at the bottom of the custom start g-code section that began with the frightening comment, "; Don't change E values below. Excessive value can damage the printer."

Removing those two lines ended up with a .04 smaller change in EM, so I'm leaving them out from now on (unless bobstro tells me that's dumb)

At the conclusion of this test, and once I modified the EM to .92, I can now once again print that original tolerance test part with the .13 and above tabs able to rotate.

Now back to printing random things for random projects!

Thanks,

mw

RE: Does a Mk3s, in November 2020, need or benefit from an e-step calibration process?

Just a thought, 
Are you using the same nozzle now, as you were when you ran the initial tests? 

Is it possible that the nozzle diameter has changed with use? 

regards Joan

RE: Does a Mk3s, in November 2020, need or benefit from an e-step calibration process?

Posted by: @boingomw

[...] "Muppet code" is the code from bobstro ( http://projects.ttlexceeded.com/3dprinting_prusaslicer_gcode.html#start-g-code).   I originally replaced all but the last two really odd lines at the bottom of the custom start g-code section that began with the frightening comment, "; Don't change E values below. Excessive value can damage the printer."

Just for clarification: Those scary lines were added by Prusa and are not in my startup gcode. I don't use them at all! I've never had the need to adjust extruder motor current, and prefer to adjust retraction and other settings to avoid the problem with extruder motor heat.

Agreed 100% that it's always worth doing an extrusion multiplier calibration with any brand. There can be minor hardware deviations and you always want your slicer settings to match the real-world values of your filament and printer hardware.

Removing those two lines ended up with a .04 smaller change in EM, so I'm leaving them out from now on (unless bobstro tells me that's dumb)

What gives the right results is the right answer! I flip-flop back and forth on having gcode extrusion rate modifiers. I find thicker layers tend to over-extrude, so do reduce the flow for layers over 0.32mm ... sometimes.

RE: Does a Mk3s, in November 2020, need or benefit from an e-step calibration process?

@joantabb

I AM, and that was actually on the list of things to change, but I don't have a spare handy.  I ordered one, and will swap it once it gets here.

RE: Does a Mk3s, in November 2020, need or benefit from an e-step calibration process?

Best wishes. 

regards Joan

RE: Does a Mk3s, in November 2020, need or benefit from an e-step calibration process?

Quick question - you altered your EM but have you checked the filament diameter at all ?  As in the 1.75 value, did you actually measure with calipers at several points along the filament to check its actual real world diameter ?  and if it differs plugged that number in ?

Personally for me I find that making it match means I end up not changing the EM at all, its just sat at 1.  YMMV

RE: Does a Mk3s, in November 2020, need or benefit from an e-step calibration process?

@neophyl

I choose to believe that Prusa isn't lying about the material diameter, so yes, I checked it.  You can too!!

https://prusament.com/spool/2258/28b79dfe

So, its a bit low, which if anything would indicate a likelihood of under-extrusion 39,661mm of material, so I should be right around 1.747mm diameter.

Hey, since you're here... what slicer version and firmware are you running?

RE: Does a Mk3s, in November 2020, need or benefit from an e-step calibration process?

Posted by: @neophyl

Quick question - you altered your EM but have you checked the filament diameter at all ?  As in the 1.75 value, did you actually measure with calipers at several points along the filament to check its actual real world diameter ?  and if it differs plugged that number in ?

My process with any new filament is:

1. Measure the filament diameter at several locations, average, and enter this in the slicer filament settings.
2. Print the extrusion multiplier test print with a 1.0 extrusion multiplier.
3. Average and calculate the multiplier adjustment and enter this into the slicer.
4. Print the test print again to verify results are within expectations.

This usually gives very good results and I don't have stringing or other common issues. I'm not after super-precise accuracy, so I leave it at that.

Personally for me I find that making it match means I end up not changing the EM at all, its just sat at 1.  YMMV

That's typically the case. I've rarely had to make more than a 5% adjustment with most quality brands.

RE: Does a Mk3s, in November 2020, need or benefit from an e-step calibration process?

Final update.  Nozzle matters.

Before and after getting a new nozzle (p3-d, Apollo, https://www.p3-d.com/products/e3d-style-v6?variant=17614735236) the extrusion multiplier did end up changing slightly... .92 before, .95 after, so a measurable difference.  That matches the non-scientific thoughts I had that the nozzle is letting out more filament than it used to, so the EM had to be reduced to account for that.  Doesn't quite explain why I need to adjust at all on a Prusament filament, but life is strange.  Everything from Prusa is perfect in every way, right??? 😉

Adapt to reality, or just sit there an whine, ya know?  I guess I'm done whining!

I did verify the size of the filament, and it matched what Prusa said it would be.

Thank you for all your support in these trying times!

boingomw

RE: Does a Mk3s, in November 2020, need or benefit from an e-step calibration process?

@boingomw

Maybe the nozzle has worn larger!

regards Joan