What to expect from printing resolution ?
Hi all,
I've been using my MK3S for a while ( 1 year ) and enjoyed it a lot.
For a print of a 5x5x5mm box, I wonder what is the resolution expectation we can get from the Printer?
I have tried with different step travel, and the best box I could print was 5.01x5.1x4.95mm.
Is this the best the printer can produce?
Can you try and see if you're getting better results?
Thank you,
Roye
RE: What to expect from printing resolution ?
Can you Post a zipped copy of the Project file (.3MF)
regards Joan
I try to make safe suggestions,You should understand the context and ensure you are happy that they are safe before attempting to apply my suggestions, what you do, is YOUR responsibility. Location Halifax UK
RE: What to expect from printing resolution ?
I print the 20mm cube every so often as a sanity check and I'm always well within 1%. I'm happy at that point.
The only issues I've had are with closely mating parts and with those I'll do test prints and adjust as needed for best fit. There are lots of variables involved.
RE: What to expect from printing resolution ?
@jsw
Which step size ( layer height ) do you use for the 20mm cube test?
RE: What to expect from printing resolution ?
[...]
For a print of a 5x5x5mm box, I wonder what is the resolution expectation we can get from the Printer?
I have tried with different step travel, and the best box I could print was 5.01x5.1x4.95mm.
Is this the best the printer can produce?
The printer itself has a good degree of accuracy, although I've never seen an official specification from Prusa. I am able to regularly print snap fit parts with a 0.2mm tolerance, and threaded bolts designed in Fusion 360 print and fit together cleanly. Since this is a physical process, giving the slicer the most accurate information to work with will help ensure your prints meet your expectations. One of the keys to success with accuracy is matching up your slicer settings with your physical printer & filament characteristics. Only then can the slicer produce accurate output.
If found that the biggest bang for the buck for both accuracy as well as reducing myriad other problems like stringing is to calibrate your extrusion multiplier. Average your filament diameter (once per manufacturer and type) and enter that in your filament profile. Print test cubes to adjust the extrusion multiplier. Once these steps are done, the slicer matches the real world.
Some users take this a step further and tweak the XY & E firmware extrusion rates. This isn't really required on the Prusa since they have a high degree of consistency between printers, but you may find it interesting. There are also XY compensation settings in most slicers.
This page has a nice summary for a variety of 3D printing technologies. For FFF:
- FDM/FFF: ± 0,5 % (with a lower limit of ± 0.5 mm)
FDM printers usually print from ABS and PLA plastics. During cooling, the ABS shrinks to about 8%, while the PLA shrinks only about 2%. These are only guide values and may vary depending on the PLA or ABS used.
A distinction must be made between systems. While desktop or semi-professional systems work with tolerances in the above-mentioned range, high-end FDM printers (e.g. Stratasys systems) deliver lower tolerances of up to 0.1 mm. However, the printing costs of these systems are much higher.
and miscellaneous other tech projects
He is intelligent, but not experienced. His pattern indicates two dimensional thinking. -- Spock in Star Trek: The Wrath of Khan
RE: What to expect from printing resolution ?
@jsw
Which step size ( layer height ) do you use for the 20mm cube test?
Actually, I don't use the same each time. I'll usually just reslice for whatever settings and filament I'm currently doing a sanity check on.
RE: What to expect from printing resolution ?
One key factor many seem to miss is that if your belt tensions change after calibration, the X and Y scale factors also change. It is imperative you maintain a constant belt tension after a calibration. And yes, Gates belts do stretch.
ps: the last cube I printed was better than 0.1 on all dimensions.
RE: What to expect from printing resolution ?
@3Dprint-2,
Hi, I used your .3mf file with Carmine red (glitter) Prusament, and got 5.00mm x 5.00mm x 5.01mm I had to bring the model upstairs to the vernier calipers... so I don't know whether it was X or Y that was 5.01, but Z was 5.00
regards Joan
I try to make safe suggestions,You should understand the context and ensure you are happy that they are safe before attempting to apply my suggestions, what you do, is YOUR responsibility. Location Halifax UK
RE: What to expect from printing resolution ?
I'm seeing .197x.197x.201 on my better SAE calipers. With the 0.201 being the top layer goobers that usually need to be trimmed on high accuracy parts. If I press hard on the X I can get down to 0.196 ... The fact an $800 printer can print within 0.1 is pretty amazing.
RE: What to expect from printing resolution ?
@joantabb
Amazing results.
Now the question, if we both using the same printer, what might be the cause that mine is off?
Any recommendation?
RE: What to expect from printing resolution ?
@tim-m30
How do you maintain a constant belt tension?
RE: What to expect from printing resolution ?
@3dprint-2
I set it up during assembly but I used twin bearing toothed idlers which are non standard.
the Prusa belt normally runs on the teeth with a smooth idler... it is possible that this is causing the belt to go a little slack. to be honest I don't know...
I have had two single bearing idlers collapse on me, so maybe I overtighten the belts... again, I don't know.
So far, I have never had a twin bearing idler collapse, I currently have ten of them on five printers.
regards Joan
I try to make safe suggestions,You should understand the context and ensure you are happy that they are safe before attempting to apply my suggestions, what you do, is YOUR responsibility. Location Halifax UK
RE: What to expect from printing resolution ?
@tim-m30
How do you maintain a constant belt tension?
I measure my belts using the deflection method. Put a weight on the belt and measure how far it deflects. 200g should deflect X 15/64ths, and Y 12/64ths. This yields about 7 pounds-force belt tension : Gates specs the belts between 2 and 25 lbf - but the steppers want less than 8 lbf radial load... so 6 to 7 lbf is a good compromise.
ps: you want to measure deflection on the longest length of free belt, so use the lower section on our printers.
RE: What to expect from printing resolution ?
I measure my belts using the deflection method.
How much deviation are you finding over time? I'm using the "strum" method of belt calibration (as long as it doesn't slip, it's good) and haven't noticed any issues. Then again, I'm happy with 0.1mm resolution.
and miscellaneous other tech projects
He is intelligent, but not experienced. His pattern indicates two dimensional thinking. -- Spock in Star Trek: The Wrath of Khan
RE: What to expect from printing resolution ?
I measure my belts using the deflection method.
How much deviation are you finding over time? I'm using the "strum" method of belt calibration (as long as it doesn't slip, it's good) and haven't noticed any issues. Then again, I'm happy with 0.1mm resolution.
My last check about a month ago showed a subtle but measurable drift towards looser on X and Y. I don't recall the exact values, X about 10% and I made no adjustments, Y was looser, and I trimmed the tension screw a bit (Mk3). The prior adjustment was at least six months back, perhaps longer, maybe even a year.