Ever since I got my MINI I've had this issue: first layer calibration shows that the left side of the bed is about 0.10 mm "higher" than the left. This means that, to get a good first layer on the left part of the bed, I get poor adhesion on the right side (which isn't great, because printing things closer to the vertical arm - i.e., on the right side of the MINI's bed - actually reduces vibration and ringing). But, to get good adhesion on the right, the nozzle scrapes along the bed on the left (which is worse).
Why the sensor reads slightly different heights, I have no idea (there's some vertical deflection to the arm depending on the X position of the head, but that deflection is there both while printing and while doing the auto levelling, so that shouldn't interfere with the mesh). I'm guessing that either the sensor's reading is slightly affected by the proximity to other printer parts, or the issue is that the sensor is reading the distance to the bed (ignoring the sheet?) and the sheet itself doesn't have uniform thickness (i.e., it's slightly thinner on the right).
Regardless of the reason for the error, this would be trivial to fix if there was an option in the printer's menu to manually adjust (i.e., add a correction factor to) the auto-level mesh. As there is in other Prusa printers (and as I'm pretty sure we were told there would also be for the MINI). So, why hasn't this been implemented yet? Is there some reason why it can't? Is it a "crippleware" conspiracy to make other models more desirable? Assuming it can and will be added to the MINI, is there an ETA?
If I move one side of the bed up by 0.1 mm (or if I tilt the arm to make it 0.1 mm lower on that side), the sensor will read it as 0.1 mm higher, and will still apply the same (incorrect) correction to it.
My problem is that, after the automatic bed levelling (which completes just fine), the first layer is always slightly "thicker" (i.e., the print head is higher) on the right side, leading to poorer adhesion. If I use Z-adjust to make it stick properly on the right side (which is the only control available for first layer calibration on the MINI), then the nozzle touches the plate on the left side.
Other Prusa printers (ex., MK3) have a mechanism to deal with that, so it's clearly a known problem, that was "solved" a long time ago by giving users the "bed level correction" option, as you can see here:
The same reasons that led to that being implemented in the MK3 can also affect the MINI. But the MINI (using the official firmware) doesn't let users correct it; it only supports a single Z-adjust value, that affects all points equally.
I don't have any insights into Prusa's plans whatsoever, but my gut feeling is that this is close to the bottom of their priority list, if it's on there at all. With the much smaller size of the print bed of the Mini compared to the Mk3S, MBL should be able to easily compensate for unevenness, if the Mini is properly adjusted in the XZ plane. Why that doesn't seem to be the case in your situation, no idea. Perhaps a broken PINDA sensor? Or a warped sheet (although I've never seen one). One cause of first layer issues I've experienced myself was caused by a tiny bit of debris that managed to slip under the sheet and got baked onto the heated bed. Beyond that, I'm running out of ideas.
Again, people like @diem are just trying to help but none of us know anything about Prusa's plans. If you feel this would be a valuable feature to add, I'd head over to the Prusa github site and create a ticket for it. More likely to be seen by the folks that write that stuff than here.
I already have (multiple times). Everything is as straight as I can measure it, considering the printer itself isn't completely rigid. I even measured the deflection of the printer's horizontal arm when the print head moves across it (it bends down very slightly as the head moves left) - which shouldn't make any difference, anyway, because the sensor moves down by the same amount as the nozzle.
Because some other user may have read something about it that I didn't. I don't expect to know everything every other user knows. Hence, I asked. If that bothers you, you don't need to reply. Especially if the reply is to a different question.
Again... the problem isn't that auto bed levelling isn't "compensating for the unevenness", it's that it's introducing an unevenness. That's what the "bed level correction" option is for, as described on that official Prusa page.
What fixed it for me was tightening the fudge out of the belts, and adjusting the X axis several times until it just worked. I may have even gone a micron or two overcompensating for the harsh squish on the far left back corner. I've noticed when fixing my MMU that something that shouldn't effect something else, still does. Obviously this is due to my lack of knowledge. I can't explain why the mesh bed leveling doesn't fix this but the squaring did, and I was thinking all the same thought you have posted above.
I do own a MINI but I have not experienced that problem but I can tell you every time I contacted support with a problem they helped me solve it quite quickly, this is what they do 24/7, and they are quite experienced at doing it right.
In my experience, it is the unexpected source of a problem what catches us. Sometimes when we do not see the problem right in front of us, some other person with a fresh view can tell us what we are failing to see.
One problem I have experienced with both MK3 and the MINI is that sometimes I have placed the steel sheet over one of the guiding screws in the back of the bed so it was tilted. The printed realized there was a problem and refused to print, leaving me puzzled until I figured it out.
Did you solve this problem? I have something similar. I used nylock mod for perfect bed leveling. Currently, Octoprint shows deviation 0.05-0.1mm, but even that my prints in one edge (and specially in one corner) don't look so good. Even, if I try to "unbalance" this edge by adding tape between table and plate it still print first layer incorrectly. For me, it looks like software issue - in that specific corner, printer adds too much correction.
Before trying them though I tried to simply level my bed by sticking some aluminum foils underneath the sheet where I have the issues showing up. But as I was raising the center of the bed (where I have the biggest issue) and the right side of the bed (like in your case) I realized that the firmware was automatically over compensating and was the root of the issue.
I had some intermittent bed leveling issues originally, and what fixed it was a factory reset followed by re-flashing the firmware. This was suggested by Prusa support. I highly recommend contacting them if you are having trouble figuring something out.
5) Company takes no responsibility for any loss including but not limited to loss of data, financial loss, or personal loss) that might arise from the use of, or the inability to use, the manual library or the specified software.
But maybe I could be cured. Maybe, on a closed course with a proper instructor (sorry Brock), I could figure out how to massage the clutch, how to engage the throttle just enough, and how to turn those steps into something fluid that resembles real driving.
So in addition to offering more manuals, Mini is working to convince people that driving these cars; like, really making them hum and pur, is a fun thing to do. Not so in my memory, but I could be convinced otherwise, and I am willing to try to learn.
BMW, which owns Mini, is a member of the club. Its property there comprises the BMW Group Performance Center West, where you can whip around BMWs and Minis on a couple different chunks of track. There is a Mini stunt driving school ($750) and a Mini manual driving school ($499) that just opened this month.
Best case scenario: I am heel-toeing in Belgian loafers a la Ayrton Senna. Worst case scenario: I fail again but in private, away from the anxiety that trying to learn to drive a manual in a public place, with a friend as a teacher, induces.
When we actually had to make the cars move, starting them up and then easing off the clutch to give them enough oomph to roll along, I was surprised at first how smooth it felt. Shifting between gears felt like playing a mini-game in a video game, where I had to find the right window of time to finesse a couple of quick motions.
"It is necessary for every urban guerrilla to keep in mind always that he can only maintain his existence if he is disposed to kill the police and those dedicated to repression, and if he is determined to expropriate the wealth of the big capitalists, the latifundists and the imperialists."
During the 1980s, the United States Central Intelligence Agency (CIA) sought to better understand and combat terrorist movements and anti-government armed militancy more effectively. In order to achieve this, the CIA produced English and Spanish translations that were distributed among intelligence services worldwide. These translations also served as teaching materials at the CIA-run School of the Americas in Panama.
Carlos Marighella wrote the Minimanual of the Urban Guerilla with the intention of popularizing guerrilla tactics. His goal was "not only for people to read this manual here and now, but to spready its content far and wide." According to Marighella, an urban guerrilla is "a person who fights against a military dictatorship with weapons, using unconventional methods." There was an explicit desire to instigate an armed revolutionary struggle against Brazil's military dictatorship, characterizing the book as a practical action manual. For Marighella, the fundamental characteristic of an urban guerrilla was participation in the armed struggle and the expropriation of lands belonging to major capitalists and landowners.
For Marighella, the key characteristics of effective guerrilla tactics were an aggressive nature, hit-and-run attack actions, and the ultimate objective of developing the guerrilla force itself. The goal was to demoralize and wear down the police and military forces embodied by Brazil's Department of Political and Social Order (DOPS) under the military dictatorship. Marighella argued that a defensive posture would lead to the guerrillas' destruction, as they were "inferior to the enemy in firepower," making "defensive action" tantamount to death.