Inconsistent First Layer Calibration  

Re: Inconsistent First Layer Calibration

same here. I'm in contact with Support, but no resolution yet.

here is a copy of my post from https://shop.prusa3d.com/forum/assembly-and-first-prints-troubleshooting-f62/print-bed-low-in-center-t23716-s10.html

my bed if flat, but my spring steel plate is bent, the center is higher than the edges. if I lay it on the desk and turn it with the center "hump" down, I can easily rotate the spring steel sheet (it keeps rotating). Ruler measurement also shows the spring steel is not flat.

My problem is, that when I put the spring steel on the magnetic heatbed, it is attracted to the magnets, thus the Bed Mesh Leveling points are in the same height. That "hump", however, must go somewhere - so when the steel sheet is on the heated bed, the "hump" is between the magnets = not reachable by the PINDA probe since it does only 9 points and the hump is between them.

it's only a minor issue because the overall effect is that my first layer is not perfectly squished between left side and center of the steel sheet. Is this to be considered a lower quality product like I'm considering it? I already contacted support with this, just wanting to hear your opinions

if the powdered steel sheet is bent - concave/convex (depends from which side you look at it) - it will become flat on the magnetic heatbed. but the magnets are not everywhere. There is a tension in the steel sheet when you are trying to make it flat, which results into a "hump" somewhere between the magnets. My heated bed is flat, printer says I have 100% perpendicular axes, belt tension is correct and I still have same problem like you have.

take your steel sheet off the bed, put it onto a table and try to place a ruler diagonally. I bet you will find it not flat, just like me. Can you please?

when you try to set the nozzle distance from magnetic bed (without sheet on, by rotating Z screw rods), do you experience differences in nozzle-heatbed distance when moving the nozzle across X axis?

Re: Inconsistent First Layer Calibration

Well here's some news:

Support told me to use mesh bed correction and lower the back of the bed by -20 microns.

However, I have expected the printing surface will be flat in third version of i3. Previously I was printing on glass and it was flat. Now I paid 770eur for a 3d printer while getting a non-flat printing surface. The printbed is flat, it's the steel sheet or hardened rods which are bent, or have some humps. Am I the only one thinking the resolution is to use mesh bed correction? I need a completely flat bottom of the print, I cannot achieve this with this printer.

I cannot see many people complaining about this problem, but for me it is a problem.
Is anybody from Prusa Research watching these forums? Is this a batch problem with powdered steel sheets? How can I achieve completely flat surface of my prints? This is not acceptable solution, what Support told me to do

Re: Inconsistent First Layer Calibration

more pictures

Re: Inconsistent First Layer Calibration

and more pictures

Re: Inconsistent First Layer Calibration

sorry something turned my photos upside down, but it's clear that I've got different measurements across the test pattern. To clarify, I haven't touched the dial when doing the calibration - it was left at -0.814mm, which works me at center.

is this really an acceptable thing when buying this printer? I haven't expected that, to be honest...

Re: Inconsistent First Layer Calibration

Hi,
Welcome to the club. If my heatbed was so flat like yours ....
You can use 3mm shim washers 50um or 100um to level the bed
You can check the bed with a dial gauge, attached with adapter to the print head
Other option is Pronterface with G80 and G81

Re: Inconsistent First Layer Calibration

well, how can this pass a quality check?? is it really me who has to investigate? I was thinking about buying parts from Aliexpress (which are completely availble for the half of the price) and build the printer myself. Instead of this I paid 777eur for something I thought be better to get directly from Prusa, because I expected 100% quality of the parts. I still have had to build it myself.

I understand there is lot of things to screw up during Kit building, but printer's self checks shows zero problems. This has to be sorted by the Support in the first place. One of the reasons I bought MK3 was because my glass (with a kapton tape, which i used to print on) got bent when heated and I could not get a 100% flat bottom surface on large prints. Now I got a machine for 2x more money and I get a similar problem.

Support just seems to ignore my emails and they only reply when I ping them again, but with some dummy questions just to keep me busy, or not even questions. That conversation is going nowhere. That's my feeling of the Support when it comes to my problem.

I would expect a clear statement from Support, they should ask me to perform measurements, take pictures, videos and then provide an analysis and a solution which will be satisfactory for both sides. I don't want neccessarily to get new parts (seems to be a ridiculous to ask for it), I just want to get what I paid for - the best 3D Printer which is leading most of the comparison tables. Instead of this I got a printer that has issues which I could not affect during building. The only thing I could do better was not buying the MK3.

thanks a lot ... I could save the money if I knew, or spend a bit more and invest into some commercial printer with better quality check when selling.

Re: Inconsistent First Layer Calibration

I had to make shims and place them under the bushings on my heatbed. Even then, I had to min/max my mesh correction values. It's still not quite perfect, but it's acceptable.

I also noticed when doing the initial Z height calibration my smooth steel rods were not parallel (once the z motors hit the max limit to calibrate it cocked the X axis rods noticeably even after adjusting them to be parallel with the frame).

Re: Inconsistent First Layer Calibration

Support staff has changed upon Twitter contact, they now respond, let's see how we resolve it together. It happens in every company. Thanks!

Re: Inconsistent First Layer Calibration

So i have the same problem... so i sent them of my bed... it is a bit bent in the middle so i have been getting unsquished filament in the middle but squished at the left and right... they are sending me a new bed... lets see...i was able to get it to work with the correction... but it was a hassle..and iguess i need to redo the correction for each recalibration...

Re: Inconsistent First Layer Calibration

Yes same here. Let's see how the new parts will work!

Re: Inconsistent First Layer Calibration

https://imgur.com/a/8ngiBvn here was my problem...

Re: Inconsistent First Layer Calibration

Well the gap is visible, mine was also slightly bent at the end, just like yours. But the gap is so small that one could think it's not important.

I'll update this thread when I install my new parts. Now I say Support is really helping, although it required an escalation. I was dealing with them for weeks now

Re: Inconsistent First Layer Calibration

I don't know what should I think. I got new parts: new bed and new powdered sheet.

- New powdered sheet is concave again, but not that much, so it's better.

- The bed is clearly bent, more than my old one. But surprisingly I get better results with the new one than with the old one. The only thing I tried to do different is that I put the bed on the Y-carriage with loose screws in. Then I heated it and fastened screws when it was hot. This seems to work better.
I still have slightly inconsistent first layer thickness over the sheet surface, but I can live with it. One of idea how to fix it is to put some kapton tape layers under places where it's too low.

But that's for long winter evenings, as I said, I can live with what I have now. I'll send the old parts back to Prusa next week, they sent a pre-paid UPS label