default x,y scale factor?  

RE: default x,y scale factor?

I don't see a way to do that in PrusaSlicer but in any case, wouldn't it only mask the underlying issue? If the model doesn't come out right, it seems there's something not correct in your printer settings. If you zip and upload one of your .3mf project files, we can look to see if there's anything unusual there. Specifically, I would check the extrusion multiplier for the filament you're using. Check out @bobstro's excellent notes on this topic: https://projects.ttlexceeded.com/3dprinting_mk3_dialing_in.html.

RE: default x,y scale factor?

I can say those scaling factors have been needed since I first got the printer and did a test print of a large cross of certain dimensions. In order to make it come out to the correct actual size, the x,y had to be scaled.  As far as I know I haven't changed anything regarding extrusion multipliers or any of the other advanced settings. Also, the scaling is quite small so I also supposed that was the normal variation of tolerances one would expect.

I will look over the notes you referenced.

Please find attached a 3mf file.

Top.emf (renamed to .zip)

RE: default x,y scale factor?

Is your printer a Prusa? This sounds like your e-step firmware settings are off.

RE: default x,y scale factor?

Posted by: @bobstro

Is your printer a Prusa? This sounds like your e-step firmware settings are off.

It is a Prusa i3 MK3S.

Is that a configuration setting or do you have to modify the source code?

RE: default x,y scale factor?

Posted by: @hdt551

It is a Prusa i3 MK3S.

Is that a configuration setting or do you have to modify the source code?

I didn't understand.  I guess the e-step is the extrusion multiplier. 

Shouldn't the x,y scaling be off by the same amount if that was the problem?

RE: default x,y scale factor?

Posted by: @hdt551

I didn't understand.  I guess the e-step is the extrusion multiplier. 

Sorry, I wasn't clear. If you previously were making adjustments to your printer settings using old documentation, it might have directed you to adjust the steps-per-mm. E-steps is the name for the extruder setting, but you can do the same for the XYZ axis. If you use the M92 gcode command, you can adjust the number of motor steps required to move a given distance. This lets you do corrections for printers that are exhibiting dimensional issues. This was more of a problem with home-built printers, and it usually NOT a problem with Prusa. However, if you used M92 gcode commands and saved the settings with M500, it's possible to change the Prusa settings. I have seen people get into problems doing this when following directions for other (older) printers and am wondering if you might have done so when trying to adjust your large cross print.

If you print one of the Prusa sample gcode files, the logo perhaps, are the dimensions still wrong? Have you tried another slicer? If every print is off by a consistent amount in all axis, it really sounds like this might be your issue. A factory reset might be all that you need.

RE: default x,y scale factor?

Posted by: @bobstro

...However, if you used M92 gcode commands and saved the settings with M500, it's possible to change the Prusa settings. I have seen people get into problems doing this when following directions for other (older) printers and am wondering if you might have done so when trying to adjust your large cross print.

If you print one of the Prusa sample gcode files, the logo perhaps, are the dimensions still wrong? Have you tried another slicer? If every print is off by a consistent amount in all axis, it really sounds like this might be your issue. A factory reset might be all that you need.

 

I have never done anything like modifying the gcode (M92 or M500).

I haven't tried a different slicer.

The prints will always be off if I don't change the x,y scaling so that is always done.

It's too bad there isn't an option to just set it as default, but I think I will just continue to modify the settings since that has worked for almost 2 years.

Thanks for all the suggestions.

RE: default x,y scale factor?

@hdt551

You should read up on setting the steps as once you have determined what they should be you write the new values to the printer and then save them to the eeprom.  That way they are there on everything.  About the only time you may need to re-save the values is after a firmware update as sometimes an update can overwrite the eeprom.

Normally with a prusa printer you dont have to do this process as the defaults are set up for them but if you have consistently had undersized prints from day one then its worth doing it to find out what the values for your setup need to be.  

Purely as an example if a particular stepper has say 492 steps to move a mm but its only moving 0.995 then you would find out what value you need to change the 492 to be to make it move that 1mm.  Afterall the slicer tells the printer to move 1mm.  The printer firmware /eeprom values knows that means pulse the stepper 492 times.  Once you 'accurize' the values stored in printers eeprom then the printer should actually move the full amount the slicer tells it to.

I have run this process on my extruder stepper to make sure it was extruding the correct amount (the extruder is the most common one the procedure is carried out on) but the same process can be used on the other steppers too for x/y/z.

RE: default x,y scale factor?

If the reset that @bobstro mentions does not help you, then an M92 entry in your startup gcode may be the solution you need.  In theory, most Prusa printers don't need this, but it solved my small, but consistent size errors in my prints.

My startup gcode contains this setting: M92 X100.26 Y100.44 Z399.2.  This puts me withing +/- 0.01 mm on most prints (assuming I've properly accounted for filament brand and spool-to-spool variations).

There are various sources for how to calculate M92 settings.  I used https://weitling.info/d3p/index.php, be sure to select the "XYZ Calibration" button at the top of the page.

There isn't much to loose, if you try this and find it does not help, it is easy to remove the gcode command from your startup. 

RE: default x,y scale factor?

I don't think I will do a factory reset since I have never made any changes to any eeprom values.

I didn't know about using custom g-code commands to modify the steps so that seems very promising since as you say, it will be easy to undo it it doesn't work.

Thanks.

RE:

I realize this is an old thread. However, it seems many people struggle with the calibration of the final size of their prints. While the printer may be 100% accurately producing the desired size while printing, we normally measure the prints when they are done, and cooled down. Thermal expansion (or contraction) must be considered before blaming the printer. Plastics have widely varying coefficients of thermal expansion. Check this site I found:

https://passive-components.eu/coefficient-of-linear-thermal-expansion-on-polymers-explained/#:~:text=The%20nature%20of%20most%20plastics,plastics%20are%20most%20commonly%20used.

I use Prusament PETG for nearly everything I print. According to the table in the site above, PETG has a thermal expansion coefficient of 8x10-5 /°C. When we print with PETG, the bet temperature is usually 90°C. If we measure the prints at room temperature (25°C), the difference is 65°C. 65x.00008 is 0.0052 mm/mm.

This number is very consistent with the results of our measurements. It would be very beneficial if PrusaSlicer had an advanced option on the Filament Settings window to account for this.

RE: x,y scale factor? Consider the Casting use-case.

Its not just the scaling of a model to compensate for printing factors thats at issue.  When Casting type PLA or PolyCast is used to make models for investment casting, it may be important to make allowance for expansion/contraction of the molten metal used to make the cast.
If the molten metal has expanded by (say) 5%, it will fill the mold, and when it cools will shrink by 5%, so it will be smaller than the mold. One might assume that this is uniform.  It depends on the model, but if dimensional accuracy is important then it must be allowed for.  At this moment, my thinking is to up-scale the model by the expansion factor of the metal at the temp being used for the casting.