Rostock Max V3 size issues

[Cusackmusic]

OK, I searched, and can't find the info I'm looking for, so I'll put my question here. I just bought a used Rostock Max V3 that I spent some time fixing (It was put together wrong). My prints all come out about 5% too big. I adjusted the Diagonal Rod Length to 305.34, and now parts are still big by almost 5%. That seems like a pretty big change (from 290.8) to have almost no affect on size. Is it possible something is put together wrong? Is there any good places for help here? I did calibrate the extruder. I lowered it from almost 92 Steps/mm down to 90 based on my measurements. That seems to feed correctly now. I did that before changing the rod length.

I have a Rostock Max V2 that I've only ever calibrated the table on once (when I upgraded to the HE280, and all the other V3 parts), and that one is great. I didn't even do the multiple point cal on that one, since it prints fine.

Also a curiosity question. The V3 sounds louder AND different than my V2. Does having the motors up top make it louder?

Thanks,

Jon Cusack

[geneb]

The arm length should be 290.8 - that's why your scaling is off.

I'd recommend that you clear the EEPROM then re-run G29 to do the basic calibration. Once that is done, print a test cube and check the dimensions.

Keep in mind that if you run the web-based calibration wizard, it's going to screw with the arm length and likely introduce a scaling error - make sure you do a test print before and after to see what the effect is. If you use OctoPrint, you can run my calibration plugin - it only does 6 factor and won't fiddle with the arm length.

g.

[Cusackmusic]

I changed the arm length back to 290.8 and did a G29 PRIOR to doing the web based calibration wizard. In BOTH bases, my prints are 5% big. I went back to 290.8 and did a G29 again, and the prints are still 5% too big. I'm either missing something obvious, or there is a goofy issue here.

The calibration wizard shrunk my arm length. I manually increased it using the method I found in a thread by measuring the part error. In ALL cases, it doesn't appear changing the arm length has much affect on the part size, which doesn't make sense.

Jon

[Cusackmusic]

I re-aligned the z tower, it was very slightly not perpendicular to the table. Doubt that was my issue, but fixed anyway. Now I notice that the cheapskate for the Z tower has a lot more "play" in it. If I grab it and try to "twist" it, it moves much easier than the X and Y. Nothing is broken, but it is not as tight as the other axis. Could the spring (built into the plastic pieces) be "bad"? I mean, I can barely get the other two to move, and this one wiggles all around. Any known issues with this?

[geneb]

Did you clear your EEPROM first? Note that the arm length shouldn't ever be changed - because it's an injection molded part, it's virtually guaranteed to be the specified length.

If you can, use the OctoPrint plugin - it won't fiddle with the arm length.

g.

[Cusackmusic]

OK, finally got around to just about completely rebuilding this machine, and guess what? I have the EXACT same problem. I re-installed the firmware three times. I did G29 20 times. I manually changed the Horizontal Arm Length back to the default, but everytime I run G29, it changes it back to somewhere around 150. I'm assuming that's whats causing my oversize prints, but I'm not sure. I've tried the delta calibration tool. That doesn't work.
Is there a good MANUAL way to calibrate the table after I manually change the Horizontal Arm Length, or am I looking at the wrong area?

I did notice that the nozzle is right on the edge of the bed at the "Y" pillar when doing a G29, but at least 10mm away at the other two. Is there another build issue that would cause this?

[geneb]

Um, there's no "Horizontal Arm Length" parameter. A v3 has arms that are 290.8mm in length. 150-ish sounds right for the Horizontal Radius.
After you've done G29, finish the calibration with this: https://www.seemecnc.com/pages/delta-calibration-wizard.

If you've got OctoPrint, you can use my calibration plugin that does the same thing, but it's automatic.

g.

[Cusackmusic]

Sorry, I meant Horizontal Radius. I have tried the Delta Calibration Wizard three times. Always the same result. A cube that is 21.5x21.1x10.5. No matter what I do, I get the same results.

I JUST installed the latest firmware (92.2) and now I can't get through a G29, because it says "OFFSETS OFF BY TOO MUCH".

I also noticed that when I open the EEPROM in Matter Control it doesn't show all the EEPROM settings, some descriptions are blank, and every time I load it changes the order of the EEPROM items it displays. Something is REALLY wrong with this stupid thing, and I'm about to throw it out the window. Now I understand why the person who built it sold it.

[pouncingiguana]

The Offset error you are getting means your endstops are requiring too much offset. Make sure the screws coming out the top of your carriages are approximately the same length and try running G29 again

[Cusackmusic]

Yeah, I adjusted the Endstops to make them more even, and it calibrated fine. I re-installed firmware again and now suddenly all is fine. I must have missed a step in the process before, but I'll probably never know which one. My 20x20x10 with a 2mm wall is now about 19.9x19.9x10 with a 2mm wall, so I'm happy with that. Now onto the Mendel rebuild...