Fighting nozzle gap at bed edges

[Matt Gunn]

Maybe that's not the most clear title? Allow me to explain.

Newb to printing. I built the Rostock Max V2 as my 1st printer and Ive had great luck with it on small parts no wider than about 3" total. Now as I stretch her legs and try to print a quadcopter frame (or anything that fills the outer edges of the bed) I get excessive gap between the nozzle and the bed in certain quadrants of the bed. Here's what Ive done multiple times:

- Readjusted the towers so they are at right angles to the bed.
- Done the tower cal macro so many times my head hurts. Getting great results with X,Y,Z nozzle/bed gaps being uniform.
- Adjusted Z height so many times as well as required after tower cal.
- Adjusted Horizontal Radius value a little and settled on 139.8 as it showed identical heights between center and xyz during tower cal macro.
- Removed bed and tightened screws.
- Turned bed leveling on (and setup) and off with no change in results.


Attached is a picture of what's happening, you can clearly see either the nozzle is lifting or the bed is dropping in the quadrant between X and Z, and squishing close to X. However, the center is always good and I can print well around the center out to about 3" from center. Im open to any ideas to get this right. Am I doing something wrong with this delicate dance between end stops, horizontal radius, and z height?

[Matt Gunn]

Is anybody able to assist?

[BenTheRighteous]

Try following some of the advice in the OpenDACT thread. There are more variables in play than just horizontal radius.

My personal experience was in order to get my printer perfectly flat across the whole bed, I had to not only modify horizontal radius, but also delta arm length and the tower rotations, as well as physically shim the bed.

I think OpenDACT helps you determine those values somehow and maybe more. I don't know, I never used it; I figured mine out through trial-and-error.

[Jimustanguitar]

It's actually just the government trying to keep you from printing dangerous drones

Look for posts that include the word Perimeter, Z0, edges, etc. There're several.

[nebbian]

Guaranteed your steps per mm is slightly off.
I had the same problem.

Get your steps per mm exactly right by using a digital caliper, and measuring how far your carriages move with a vertical movement of 50mm or so.
Then measure your diagonal arms as accurately as you can (between centres) and plug that number into your software.
Don't touch those two numbers again.

Now do your normal calibration (paper test) again, you need to do this for four points: centre, and near the three towers. Adjusting the delta radius (which is calculated from the smooth rod offset and other variables) will make it print flat.

Once you've done this, verify the height of three points at the perimeter in between your three towers.