PLEASE HELP!! - Still not able to print on the full build surface after 3 years!

[DdeWaard]

Hello,

About 3 years ago I bought the Rostock MAX V1 and sadly i've never been able to truely use the full size of the bed.
All i've been able to print really has been in a circle of say 15 cm MAX. I've always just accepted this small build area, because i got tired of trying to calibrate it.
Since this tiny build area is not what i bought this large and expensive printer for, i just started to recalibrate the machine about a week ago after having adopted the new ball cup arms, because the old style arms started to wear out and get sloppy.
I've spent 4 evenings calibrating it on sight, by printing a single layer high 120mm circle and adjusting the endstop screws if i see a part of the outlines is too thick or too thin. This all after doing the paper under the nozzle trick i've done so many times, which i dont find accurate enough.
I still haven't managed to get it to print an outline with a consistent thickness.

The way i think this calibrating of endstops in relation to the distance of the nozzle to the platform both near the tower and opposite of the tower works is as follows: (and please oh please correct me if i'm wrong)
I see it in such a way, that in the software, the build plate can be seen as a 360deg. seesaw pivoting on the center of the build plate. so if i turn the screw of tower X in such a way that it moves closer to the bed, it moves the opposite of tower X(in between tower Y and Z) away from the bed.
Therefore I need to use the screw to find the balance between the point in front of the tower and the one opposite of the tower in order to calibrate this "axis"
//This all after i've already calibrated the printer_radius in the eeprom//
This same applies to the other towers.
I've been fooling around with this by printing circles of 120mm diameter on blue tape (more on that later) and seeing how the outline's thickness look both directly in front of- and in opposite of each tower.

Could you please tell me in as much detail as possible if i've got the right idea about this, or how i'm wrong and how i can accurately level the bed.

I do not want to use a dial indicator, as i will have to go through a lot of trouble detaching the hotend and swapping it out for a mount with dial indicator (i'm using a E3D Chimera with a homemade, supersturdy and dimensionally accurate milled aluminium mount, plus the wires are soldered, rather than using a bunched up collections of connectors dangling around the hotend. This makes removing it from the effector quite a pain)

I do have to say as last, for some magical reason the boro plass plate, which i've taken very good care of the past 3 years, has suddenly cracked while heated up from 20 to 80 degrees celcius, so now i'm using a waterjetcut, sanded and brushed 2mm aluminium plate with the previously mentioned tape.
This could be the big problem, I know, as the plate might not be perfectly flat.
If you think the leveling technique is correct, and would work with a glass plate, then this alu plate could be the big problem here, and i'll make sure to get it off the printer ASAP. In that case, I'll go back to using a glass plate, which i'll probably have to have ship from Seemecnc to the Netherlands.

I sincerely hope some of you Rostock forum wizards could help me with these problems, please take some time to understand what i'm trying to explain and answer my questions, as I badly need them after 3 years of not having unlocked the full potential of this machine.

Thanks in advance,

Demy de Waard

[Xenocrates]

So, inconsistent thickness in the outlines. Where on the printer? Across from the towers, next to them, as you move farther out from the center, just randomly?

If it's in relation to the tower, it is likely an endstop offset issue. I would use something like the attached spreadsheet instead of the screws, and enter the offsets into the EEPROM. That will likely be more precise. You can also screw the endstops all the way down to gain some more Z height.

If this doesn't get you a flatter layer, and it isn't a radius issue (moving out from the center), there are 2 possibilities I can see left. The first is that you have a tilted effector or other mechanical issue such as backlash (make sure the belts are tightened and all that jazz). This could be checked somewhat by, since you have an off-center nozzle setup with the Chimera, printing two disks without adjustment, one for each nozzle and comparing, swapping to the Cyclops (if you have the legends pack), and comparing that to the previous results. The second is the possibility that the AL plate is warped. Since you've had it waterjet cut and say that you have a milled mount that you home-made, I would use either a Z probe on the mill (This is presuming it's a CNC of some sort, rather than a bridgeport, although one with glass scales could work), or re-face it flat with a large face mill. Alternatively, apply layout dye to the surface, take a flat reference like a surface plate, and place a very mild abrasive sheet between (I would use 2K personally, since I have it). Any dips would remain more strongly colored, while high spots would lose the dye almost immediately and be slightly reformed.

Finally, there are options to ensure a good calibration and a better surface finish. The duet family of controllers (More specifically for you, the .8.5 or Wifi, as they support dual extrusion), have excellent delta calibration baked into the firmware, and support almost any sensor (I suggest FSR's for your purposes). This is expensive and rather involved, which is why I mention it last, and only to head off those who would push you to it slightly more aggressively.

I wish you the best of luck with calibrating, and pictures of the issue would be helpful.