100mm coming out as 99.45mm  

  RSS
stigern
(@stigern)
New Member

Is this within expected ranges? Or should it come out closer to 100mm?

Printed a 100mm flat object, measured with my digital caliper.

Is there anything I can do to increase precision?

Posted : 16/06/2020 8:00 am
Kiwo
 kiwo
(@kiwo)
Eminent Member

I think, it is called PID calibration. How much the extruder must rotate to extrude extract length of filament.

Posted : 16/06/2020 8:40 am
crawlerin
(@crawlerin)
Honorable Member

@kiwo

PID tuning is essentially how much current needs to go to heater element in order  to reach and maintain desired temperature. What you probably mean is e-steps calibration, but I don't think it's needed on MINI with standard motor. Adjust extrusion multiplier in slicer instead.

https://help.prusa3d.com/en/article/extrusion-multiplier-calibration_2257  

Posted : 16/06/2020 11:00 am
Oxygen
(@oxygen)
Estimable Member

Have you tried several objects? I have made a square 30x30x30 that is almost perfect 30.02 x 30.01 x 29.99 and a tube of 30mm that is 30.03mm

IMHO it also depends on the filament (and hotend temperature), the idler screw, heat-break (did you re-align it) etc (the usual suspects in the MINI forum)

Mini with FW:4.1.0 + Bondtech Heatbreak...
Posted : 16/06/2020 12:34 pm
wh49ersfan
(@wh49ersfan)
Active Member

@crawlerin Thanks for posting this. I am looking to print a bunch of board game organizers, a lot of which need to be square or close to square. I will have to do this whenever my Mini actually gets here. sigh

Posted : 16/06/2020 6:45 pm
Karl Herbert
(@karl-herbert)
Famed Member
Posted by: @stigern

Is this within expected ranges? Or should it come out closer to 100mm?

Printed a 100mm flat object, measured with my digital caliper.

Is there anything I can do to increase precision?

...  or try this: https://mattshub.com/blog/2017/04/19/extruder-calibration

Statt zu klagen, dass wir nicht alles haben, was wir wollen, sollten wir lieber dankbar sein, dass wir nicht alles bekommen, was wir verdienen....
Posted : 16/06/2020 7:15 pm
oliver
(@oliver)
Eminent Member

Hi, there is a long thread on this topic here: Prusa Mini : difference between CAD sizes and Printed sizes...

Posted : 17/06/2020 2:17 pm
jsw
 jsw
(@jsw)
Reputable Member

This thread got me thinking, and I realized that I had never actually checked the dimensional accuracy of my unit, other than eyeball sanity checks.

Now that I have some slack time on the printer, I just printed a calibration cube, ABS, because that's what was loaded for my last print.  I sliced with Prusaslicer, close to defaults. .1mm detail.

The first one had a slight but noticeable 'elephants foot', and then I remembered that I forgot to switch the profile back to smooth1 from textur1, so I reprinted it, and the elephants foot disappeared.

Using the digital caliper, everything is very consistently between 19.97 to 20.02, or a fractional percentage point in error.

I am currently printing another cube in PLA for comparison, and I plan to do one in PETg as well, to get a good handle on what I can expect for accuracy.

99.45/100 is still well within 1%.  Before dorking around with any settings, I would suggest printing a calibration cube, paying attention to detail and technique, as this is a known model which should be very close to the specified dimensions.

This post was modified 2 months ago 2 times by jsw
Posted : 18/06/2020 12:02 pm
chocki
(@chocki)
Prominent Member

You will struggle to get exactly 100mm as different filaments shrink more or less, I tried to adjust my Mk3 at firmware level, but the limitations of an 8 bit micro controller stopped me being able to achieve the exact dimensions.

Maybe when the platform moves to a 32 bit board, this may be possible.

Normal people believe that if it ain’t broke, don’t fix it. Engineers believe that if it ain’t broke, it doesn’t have enough features yet....
Posted : 18/06/2020 5:05 pm
chocki
(@chocki)
Prominent Member

Sorry, Rambo mini thread!, don't know if the steppers resolution in firmware allows it, but maybe on the mini you can increase the steps per revolution and still be able to print at speed, this would then allow a finer degree of calibration.

Normal people believe that if it ain’t broke, don’t fix it. Engineers believe that if it ain’t broke, it doesn’t have enough features yet....
Posted : 18/06/2020 5:08 pm
nikolai.r
(@nikolai-r)
Noble Member
Posted by: @stigern

Is this within expected ranges? Or should it come out closer to 100mm?

Printed a 100mm flat object, measured with my digital caliper.

Is there anything I can do to increase precision?

Digital caliper does not mean better accuracy. +/- 0.5mm for 100mm is what you can expect from the most inexpensive ones. Beside of that you have to add also measurement errors.

I wouldn't recommend to change anything with your results. Any adjustments for the specific print can be done in the slicer.

Often linked posts:
Going small with MMU2
Real Multi Material
My prints on Instagram...
Posted : 18/06/2020 6:12 pm
orjan.w
(@orjan-w)
New Member

I have a similar problem with my Mini and haven't got a clue how to solve it!

Tried different filaments and always end up with a result that is smaller than expected. This is in scale though. A 20mm calibration cube ends up in 19.8 on all axis concekvently! I have a different printer at home that prints the same files correctly in scale. Not as good as the mini as it's 8bit but still correctly. How to calibrate the axis to do it right?

Posted : 11/07/2020 10:39 pm
premyumga
(@premyumga)
Active Member

Try this here its called calibrating the Esteps

 

https://all3dp.com/2/extruder-calibration-6-easy-steps-2/

Posted : 20/07/2020 2:32 pm
Oxygen
(@oxygen)
Estimable Member

Did someone already did the above calibration? and how much is the value from PR off?

Mini with FW:4.1.0 + Bondtech Heatbreak...
Posted : 20/07/2020 2:49 pm
charles.h13
(@charles-h13)
Famed Member

This value might change filament to filament.  Some shrink more than others.  

Posted : 20/07/2020 3:07 pm
Oxygen
(@oxygen)
Estimable Member

that is why I asked the question -- I have seen big differences and should determine the E-Steps per brand -- check

Mini with FW:4.1.0 + Bondtech Heatbreak...
Posted : 20/07/2020 3:11 pm
Karl Herbert
(@karl-herbert)
Famed Member

@oxygen

Basically it's not bad if you calibrate the e-steps at least once. I did the same when I converted to Bondtech gears. The deviation was 5mm to 300mm. I have now corrected this to +- 0.5mm and printed a test cube. The values were 19.92 in X/Y and 20.05mm in Z. I corrected this in the slicer with the multiplication factor and extrusion width. Of course every filament behaves differently. For this reason I tested the settings for each filament and saved them into a profile.
If you get a bad filament, the values can differ, because the filament flow is not even and therefore the print quality varies.
To cut a long story short: With a result of 99,45 you don't have to calibrate anything.
By the way a question: How could you determine the length 99.45 so exactly? Do you have your own measuring instrument for this?

Statt zu klagen, dass wir nicht alles haben, was wir wollen, sollten wir lieber dankbar sein, dass wir nicht alles bekommen, was wir verdienen....
Posted : 20/07/2020 4:51 pm
Oxygen
(@oxygen)
Estimable Member

I will do the calibration (one brand of filament gives me troubles) I borrowed a precision caliper

Mini with FW:4.1.0 + Bondtech Heatbreak...
Posted : 20/07/2020 5:15 pm
premyumga
(@premyumga)
Active Member

Esteps are not depending on the Filament it is depending on the Extruder Steps/mm and the Tension of the Gears in the Extruder(if it compresses the filament while pushing it)
You dont actually measure the Extruded (Molten) Plastic lenght
what you  are doing is measure that if you Push to extrude 10 CM of Plastic (from the Extruder to the hotend) it will actually move the Filament you marked by 10 CM
So this should not be depending on the filament maybe if the filament is super soft and you dont grip hard enough or vice versa like gripping filament too softly so you cannot move it.
Otherwise this should be the same for every filament.
I think you are referring to the Extrusion Width (This actually changes from Filament to Filament and should be calibrated aswell for every new roll)

Posted : 20/07/2020 9:20 pm
crawlerin
(@crawlerin)
Honorable Member

@premyumga

I'd look at extrusion multiplier in Slicer rather than width, extrusion width just says how wide extrusion filament trace will be. Sure it may influence some dimensions, and that's something that slicer needs to calculate when it interprets geometry into gcode e.g. on number of perimeters (plus how much filament extruder needs to push, too ... which is again influenced by extrusion multiplier).

Posted : 24/07/2020 3:59 pm
Share:

Please Login or Register