how accurate is good enough
So I finished assembling the MK2S kit a few days ago and have been getting good results with it.
The selftests declared the geometry as en enough and stated that the skew could be corrected for.
In the support menu on the LCD, I can find the actual measurements:
Y distance from min:
Left: 1.67mm
Center: 1.18mm
Right: 0.70
Measured skew: 0.32
Slight skew: 0.12
Severe skew; 0.25
Now, 0.3 degrees does not seem like that much to me, but if I were to want to dial this thing right into zero, I dont think I know what any of these values actually mean.
The first one, for instance, 1.67mm of..... what?
So, do you think I should even bother to try to dial it in, and where should I start?
Thanks.
Re: how accurate is good enough
The printer's software will compensate for this, so I would not worry too much about it.
Any advices given are offered in good faith. It is your responsibility to ensure that by following my advice you do not suffer or cause injury, damage or loss. If you solve your problem, please post the solution…
Re: how accurate is good enough
Hi Mat,
the Y axis frame is intended to be located 100 mm back from the main aluminium frame.
I believe these measurements are relating to that.
it's a positive number, so I assume that the value one side is 101.67 mm and the other is 100.70 mm
the M10 thread pitch is 1.5mm,
so can I suggest that you adjust the main frame so that it is closer to the rear of the printer
in the picture above the top rod is the left hand side of the printer Y frame, looking from the front
the bottom rod is the right hand side of the printer
If you were to slacken off the double nuts on the left hand side of the printer so that they were half a turn closer to the back, then lock them in position, you would be able to tighten the main frame single nut, half a turn to tighten it up again (making sure that you keep the 10 mm screwed threaded bar in the bottom of the slot in the main frame( I usually put something like a coaster under the main frame to hold it up whilst adjusting the assembly))
this should move the main frame backwards approximately 0.75mm on the left. if you then re calibrate the printer you should that the two sides are within 0.17mm of each other... which is probably good enough for anyone...
If you want to sort both measurements out, you can adjust both sides, adjust the left side by marginally more than one complete turn, and adjust the left side, by marginally less than half a turn... (remember to make sure that the threaded rods are in the bottoms of the slots in the main aluminium frame
good luck
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: how accurate is good enough
You have to be careful with these numbers, joan. I do not know where they have been measured on X. The pinda probe points have the following distances: left to middle: 102mm, middle to right: 101mm. The rods have axial distance of 170mm.
When I construct a triangle with ɑ=0.32º, β = 90º and a= 1.67-0.7=0.97 then c would be 173.676...
Saying that I do not know what these values exactly mean.
But otherwise I think you way to get this perpendicular is sound, however I would move it a little bit less.
Re: how accurate is good enough
So I finished assembling the MK2S kit a few days ago and have been getting good results with it.
The selftests declared the geometry as en enough and stated that the skew could be corrected for.
In the support menu on the LCD, I can find the actual measurements:
Y distance from min:
Left: 1.67mm
Center: 1.18mm
Right: 0.70
Measured skew: 0.32
Slight skew: 0.12
Severe skew; 0.25
Now, 0.3 degrees does not seem like that much to me, but if I were to want to dial this thing right into zero, I dont think I know what any of these values actually mean.
The first one, for instance, 1.67mm of..... what?
So, do you think I should even bother to try to dial it in, and where should I start?
Thanks.
Where do you find these measurements? I'm running 3.0.11 and in the Support menu I don't see this data. Facing the same issue as you where the axes are slightly skewed. All of my measurements with calipers look good, but would be nice to see what the firmware thinks it sees.
Re: how accurate is good enough
Nevermind I found it, it's a new addition to 3.0.12 RC firmware.
Re: how accurate is good enough
@joan.t ah! That's super helpful, and makes a lot of sense. The 100mm gap at the back, is 100mm - at least to my eyes and callipers, but the machine knows better in this case! Now that I understand what the firmware is trying to tell me, it is reasonably trivial to make this adjustment.
Of course, it currently prints fine, so I'll not do it right away, but next time I'm doing something like servicing it or checking all bolts are still tight I'll address it then. Changing that would adjust the skew values obviously and would bring everything closer to zero. Yay!
@john.r10 I didn't realise this info was only available in the current RC. I installed 3.0.12RC right after printing the first test print.
Re: how accurate is good enough
... so I'll not do it right away,
who am I kidding, it's 4am, perfect time to play with the printer! Just tried and retested to:
Left 1.31mm
Center 0.82mm
Right 0.32mm
Skew: 0.30 degree
Slight 0.12 degree
Severe 0.25 degree
So I didn't move it enough, but I'm ok with that, was more worried about moving it too far! But it proves the test case, and so I'm now more confident to adjust it further. 100% accuracy will be achieved!
after some sleep.
Re: how accurate is good enough
Guys, I'm trying to print at the very top edge of the printer and it looks like the printer is cutting off the top edge of the printable area.
One of the things I noticed is the "Y distance from min" from the XYZ cal details screen:
Left: 4.37mm
Center: 4.41mm
Right: 4.44mm
Less than 0.1mm difference between them but the values are significantly higher than what (little) examples I've been able to find. Most people seem to be <2mm Y distance from min, but I can't find very many samples at all. More the point, when I run the Y-axis I can go to -2, and if I look at how much of the top edge of the print is cut off, it seems to be about 2-3mm, although it's really hard to say because the lines get printed on top of each other and all smooshed up.
This is not a kit printer, it was an assembled printer from Prusa. Anyone know if those numbers are within the normal spec? It'd be nice to get the entire 210mm on the Y as some of the calibration routines utilize that. I'm going to ask Prusa support as well, just curious what other people have.
Thanks.
Re: how accurate is good enough
Guys, I'm trying to print at the very top edge of the printer and it looks like the printer is cutting off the top edge of the printable area.
One of the things I noticed is the "Y distance from min" from the XYZ cal details screen:
Left: 4.37mm
Center: 4.41mm
Right: 4.44mm
Thank you! 😀
You made me realize the root cause of my problem!
I could not understand why I was having cropped top edge prints. I compiled the FW with a larger Y_MAX value (215) instead of the correct 210, just to be able to print full scale.
I had Y distance values similar to yours. So, I moved the Z frame backwards 4 millimeters: 4mm / 1.5mm (screw pitch) = 2.67 screw rotations. Now, i'm at 0.7mm and I can print to the very end.