RE: Failed Y-calibration of a new CoreOne
Turns out I made a rough assembly mistake and reversed the X motor pulley, which caused the trapeze like form. Now I am at that parallelogram form.
RE: Failed Y-calibration of a new CoreOne
Turns out I made a rough assembly mistake and reversed the X motor pulley, which caused the trapeze like form. Now I am at that parallelogram form.
Thanks for clarifying that. I was like "That should be impossible, if the printer is assembled correctly".
RE: Failed Y-calibration of a new CoreOne
After mounting and having failing Y axis I tried the arachnoid.com/3D_Printing_Prusa_Core_One/index.html#Core_XY_Axis_Alignment to get everything square, and I thought I had it square but it still failed the test. Someone said "filament sensor cable" and turns out it somehow was in the way of a smooth return so the check failed on that. Straigtened the cable and tightened the clip made it work.
So thanks everyone for your great input!
RE: Failed Y-calibration of a new CoreOne
This worked for me, but I had to put in a LOT of force in order to correct the problem. I tried with light force dozens of time with literally no improvement. Once the left and right side had NO play, both in the back and front, at ~85Hz on both belts, I tried calibrating again and it passed first try.
Extremely, extremely annoying problem, and the belt calibration is also a tedious process. I wish Prusa had put more effort into this, I honestly expected more from them.
RE: Failed Y-calibration of a new CoreOne
For what its worth - I had the same problem. I assembled my Core One but after about a week, I was getting a fail on the Y axis calibration. Part of this is that my belts stretched a bit after a one-week break-in, the other is that my gantry was not 90 degrees (Right side was about 3mm away from the front, while left was touching.
Prusa's knowledge base on Y axis fail ( https://help.prusa3d.com/article/selftest-failed-core-one_837189) it says to check this alignment when assembling and if it is off, you need to forceably bend the gantry by placing an object like a pliers or wench opposite the side that is not touching and bending it (refer to the Core One Assembly instructions) Make sure you completely loosen your belts (I broke my Idler Pulley Tensioner part and had to print another one because I did not loosen my belts all the way and when forcebably skewing the gantry, it broke). Once I was able to get both sides touching at the same time, I then tightened my belts to 85 Hz and again checked that both sides were touching when pulled fully to the front.
RE: Failed Y-calibration of a new CoreOne
I'm sorry not to have responded before. Yes, that does indicate that the linear bearing bar is not square with the Y-axis rods. Looking at your image it shows that the righthand side is too far back. Use the method above to correct it.
RE: Failed Y-calibration of a new CoreOne
This worked for me, but I had to put in a LOT of force in order to correct the problem.
Did you make the belts (and especially the one getting really tight, when you pull on the side that needs correction) completely SLACK?
I had my belts almost flopping from the pulleys/idlers and when correcting the gantry, one belt still got really tight and needed even more loosening of that tensioner.
If you didn't do that, you probably almost snapped one belt. Bending the gantry takes way less force, than the resistance of one belt having the slightest amount of tension.
RE: Failed Y-calibration of a new CoreOne
I tried with both slack belts and tensioned belts, same thing. I don't know if I got somehow thicker metal brackets in mine (it's one of the later kits), or they are more elastic, but they are easy to deform elastically and difficult to deform plastically.
RE: Failed Y-calibration of a new CoreOne
I tried with both slack belts and tensioned belts, same thing. I don't know if I got somehow thicker metal brackets in mine (it's one of the later kits), or they are more elastic, but they are easy to deform elastically and difficult to deform plastically.
Just to be clear: Did you try it with both belts slack and then also checking the tension of the one belt, that becomes tensioned, when trying to bend the brackets?
My belts were both flopping around, but right before the brackets would start to deform plastically, one belt got extremely tight!
Plucking on it gave me a similar tone, as it gives at 85 Hz!
Overall I agree though, you have to bend it A LOT to actually bend anything plastically. I didn't feel fully comfortable doing it, but it worked.
RE:
I am new here, but i may have found an solution to the problems.
I have build my own core one and have Zero issues.
Heres what i did.
The only good option:
1. Loosen both belts tensioners so they are wobbly.
2. Make sure both belts are at both ends 5 teeth inserted(engaged into each other) in the holder behind the nextruder.
(I checked and measured my belts and both belts are equal length exact the same tooths.)
3. Check for gantry touching front and rear at the same time. Otherwise bent it a little until striatght. WITHh belt tension this step is useless because it is the uneven tension that is pulling it out of squaryness.
4. Move print head to center and back (or front i can't remember.
5. Start to tension belts. a little. This is very important: ONE or HALF turn at the time on each end. DO NOT tighten one side more then the other because again you will pull gantry out of whack. Look for a side that is not touching the front or the back (i can't remember) Stop turning on the side where it does not hit the front anymore.
5. Start measuring with the app. When one belt is lower DO NOT tighten it a lot more to compensate for the other, becuase again, you will pull gantry out of square. I measured on Galaxy S23 2-3cm away from belt, but stright in front of the belt. Not at an angle. There is a little delay because it needs to sense a look for the good sound.
Till this day it is still 85hz with print head middle back. When belt are loosened before i start at 90-92hz and after a few prints check.
You can fiddle around with tensioning and the gantry hitting the front and the bakc equally, but keep one tihng in mind. TENSIONING ONE BELT AFFECTS THE OTHERS TENSION. Find A Middle way betwee 82-88 hz where it is hitting the front equal. That is more important then tension. I have printed on 70hz, 80 and 90hz belts and the difference is not that big.
But the gantry was always square.
After re-run x and y calibration.
Auto home takes 3 to 6 hits in the corner. But mostly 3 times and it is going to print.
another thing you may want to look at is the bed level.
1. First do an Z cabration.
2. Hover the printhead 1mm above the bed. In each corner and the middle drop the head to 0 and look if you see it not hitting the sheet. (DO not go 0mm over the bed)
For the corner that is not hitting the bed, unscrew the t10 screw in the corner pillar outside the housing. (The screws are at the exact height of the cory xy plate.)
Press the core Xy frame up if it was not hitting the nozzle. Mostly front left corner.
Of the rear is not hitting you can do the same thing, but the rear leadscrew holder has to be loosened behind the rear cable cover. (It even has an slotted hole for adjusting)
My printer prints liek a dream this way. It is really reliable. I can press print and walk away.
Happy Printing.
RE: Failed Y-calibration of a new CoreOne
I had a lot of typo's but i can't edit the post.
I am new here, but i may have found an solution to the problems.
I have build my own core one and have Zero issues.Heres what i did.The only good option:
1. Loosen both belts tensioners so the belts are wobbly/loose.
2. Make sure both belts are at both ends 5 teeth inserted(engaged into each other) in the holder behind the nextruder.(I checked and measured my belts and both belts are equal length exact the same tooths.) Carefully screw back the nextruder, the 3x screws behind fan are threaded in plastic nuts so don't overtighten!
3. With loosened belts check for gantry touching front and rear at the same time. Otherwise bent it a little until they hit both front and rear equal. WITH belt tensioned this step is useless because it is the uneven tension that is pulling it out of squaryness.
4. Move print head to front and center.
5. Start to tension belts. a little. This is very important: ONE or HALF turn at the time on each end. DO NOT tighten one side more then the other because again you will pull gantry out of whack. Look for a side that is not touching the front. Stop turning on the side where it does not hit the front anymore.
5. Start measuring with the app. When one belt is lower DO NOT tighten it a lot more to compensate for the other, because again, you will pull gantry out of square. I measured on Galaxy S23 2-3cm away from belt, in the middle, but straight in front of the belt. Not at an angle. There is a little delay because it needs to sense and look for the good sound.
Till this day it is still 85hz with print head middle back. When belt are loosened before i start at 90-92hz and after a few prints check.
You can fiddle around with tensioning and the gantry hitting the front and the back equally, but keep one thing in mind. TENSIONING ONE BELT AFFECTS THE OTHERS TENSION. Find A Middle way between 82-88 hz where it is hitting the front equal. That is more important than tension. I have printed on 70hz, 80 and 90hz belts and the difference is not that big.But the gantry was always square.
After that re-run x and y calibration.
Auto home takes 3 to 6 hits in the corner. But mostly 3 times and it is going to print.
Another thing you may want to look at is the bed level.
1. First do an Z calibration.
2. Hover the printhead 1mm above the bed. In each corner and the middle drop the bed to 0(all the way up) and look if you see the nozzle not hitting the sheet. (DO not go flying 0mm over the bed, because you will damage the sheet)For the corner that is not hitting the bed, unscrew the 2 t10 screw in the corner pillars outside the housing. One front one side. (The screws are at the exact height of the cory xy plate.)Press the core XY frame UP (With nozzle 5mm above bed) if it was not hitting the nozzle. Mostly front left corner. There is some play in the screws.Or lower the right side. (press down hard and tighten screws again) That's the same as lifting the left side.If the rear is not hitting you can do the same thing, but the rear leadscrew holder has to be loosened behind the rear cable cover. (It even has an slotted hole for adjusting/compensating)
Another option is to bend the front left or right POM leadscrew holder flaps a little. It is a straight steel plate, but when it is not level a straight plate is worthless anyhow.
Probing is way way faster. It is like the printer knows were Z=0 should be and it probes much faster.
My printer prints like a dream this way. It is really reliable. I can press print and walk away.
Happy Printing.
Check belts for both 5 teeth engagement.
See if your tensioner are almost the same. (Mine are not really the same place, but tension is ok and i checked for equal belt length
RE: Failed Y-calibration of a new CoreOne
Three and a half months later, it's finally fixed (mostly).
Quick update for what was actually changed:
Prusa customer support sent me new brackets, changed and the problem not only didn't go away it got worse. Now, instead of ignoring calibration and printing, I couldn't even unload filament.
Customer service sent me a pair of new x/y motors. Changed and fault cleared. I'm passing every calibration for the first time since I first powered up late April of this year. Get everything back together and I get a chamber thermistor error...
I guess I must have pinched something putting everything back together. I measured the ohms of the thermistor and they came out to be over 1m at the connector so it's bad.
Note to customer service if you're reading this: If your customer mentions they took the liberty of taking measurements and the measurements seem off (like I did 02May, page 12 of this thread, and you got the exact same information in your email as well), take them a bit more seriously please. This problem could have been resolved two full months ago.
RE:
I also noticed that if Y calibration is failing, the recalibration is failing also every time until I press reset. Then it passes first try (with correct aligned Y-axis). Seems a bit odd to me. Maybe an issue with the firmware?
And be aware that the belt tension frequency changes when the printer gets warm / hot due to thermal expansion. So let the chamber cool down to ambient temperature. This point has a higher impact on the Core One because it has a much longer belt comparing to the MK4S. However, this should not have any impact on XY-precision because the belts have the same length.
And third: The tensioning pulleys (the screw) can't be "locked" / fixed (like für MK4S / MK3S). So it's possible that the screw will losen itself due to vibrations when printing. If this is an issue a nut for locking could be used.
Greetings M
GreetingsM
RE: Failed Y-calibration of a new CoreOne
Prusa CoreOne Y-Axis Calibration Fix
The Problem: Y-axis calibration fails on new CoreOne printers.
The Solution (Simple but Critical):
Initial Belt Tension - Start LOOSE: First, loosen the belt tensions enough so the Y-axis can move freely and hit the frame stops without resistance There should be NO play or slack in the system (as shown in the referenced images) The axis must be able to reach its full travel range unimpeded Run Calibration: Only after achieving this free movement should you run the Y-axis calibration The calibration will complete successfully with proper loose belt tension Final Belt Tensioning: ONLY after successful calibration should you tension the belts properly Critical technique: Alternate between left and right sides when tightening Important: Each screw tensions BOTH belts simultaneously - just stronger on one side than the other Never focus on tensioning just one belt per side
Key Insight: The calibration routine needs to detect the mechanical limits first with minimal belt resistance, then you can apply proper working tension afterward.
LEFT SIDE:
RIGHT SIDE:
