Odd Thicknesses on Brim on Large Part

[lbarger]

I seem to be having some difficulties getting my recently built Rostock V2 properly calibrated with regards to setting z max travel, the ends stops for each tower and printer radius. I thought I had it dialed in pretty close, but I just failed in an attempt to print the first layer of a large part. The part failed shortly after printing the 5mm brim. I have measured the thickness of the printed at several points and am confused by the results. I am therefore wondering if anyone can help me understand what is going on.

( x y Thickness)
(-115mm 70mm 0.51mm) (0mm 70mm 0.25mm) (115mm 0mm 0.59mm)
(-115mm 0mm 0.43mm) (Open center, no measure) (115mm 0mm 0.45mm)
(-115mm -70mm 0.34mm) (0mm -70mm 0.36mm) (115mm -70mm 0.34mm)

(Yes, that is 134.6mm on the diagonals.) Across the front of the machine (between x and y towers), the first layer thickness variation is just 0.02mm, however, across the back of the part, there is wild variation. The 0.59mm at the back right corner may be even greater as the material does not seem to have been flattened at all.

Looking at just the four corners, I would think my z tower stop was to low causing the platform to be too high. However, the thinnest part of the brim is mid way between these thick corners. If I lowered the z skate, I would risk crashing into the bed at the mid point.

Having typed this up, and looked at the print location again, I see the two thick corners are both outside the triangle formed by the towers which is getting into known problem areas. I've tried the calibration method in the manual along with the 'quick calibration' someone posted using points at 120mm radius. I've checked that my towers are square to the bed and my belts are tensioned. Any suggestions as to what I could check (or change) would be appreciated. (My nozzle is stock - 0.5mm)

[lbarger]

I designed a 'quick' calibration triangle that has 7 1.95mm tall, 5mm diameter, posts in the following locations.
1) Centered (0,0) {1.88mm}
2) Base of X Tower (-112.6, -65.0) {1.96mm}
3) Base of Y Tower (112.6, -65.0) {2.01mm}
4) Base of Z Tower (0,130.0) {1.97mm}
5) Midway between X & Y Tower (0,-65) {1.98mm}
6) Midway between Y & Z Tower (56.3, 32.5) {1.89mm}
7) Midway between X & Z Tower (-56.3, 32.5) {1.90mm}

In brackets after each location is the height I measured on the printed part. It appears that I might be a little high on the Y tower and my radius may be a bit large. This could explain why the head is lifting up quickly on the corners of my part (mentioned at start of thread) that are outside of the triangle formed by the towers. Guess I was not as close on calibration as I thought I was, but it was all within .1 mm which is about the thickness of the paper I was using the calibrate.

With the Delta configuration, it seems the height at any location is driven by the three skate positions, length of arms and the so called delta radius. The arm length is pretty much fixed with the injection molded parts of the V2 Rostock. This leaves four unknowns. With four heights at different locations on the bed, it seems like one should be able to derive equations to prescribe adjustments to the three end stops and delta radius. Has anyone seen such equations?

I also left a 2mm wide, 0.75mm tall wall between each of these posts. I noticed the first layer was considerably thinner when the print head was being pushed toward the post in the corners than when it was being pulled back away from it. I have a light that shines into the edge of the glass plate, thin layers are bright, and its really dark if it did not stick to the glass. The difference was measured at 0.2mm when the print was done. Is there any way to adjust the stock arms? Are the carbon fiber Trick Laser arms better in this respect or should I be looking at the magnetic ball joints. (For that matter, can would use magnetic ball joints with the Trick Laser arms?

[lbarger]

Replaced stock arms with carbon fiber version to eliminate play in end effector. Major improvement. Currently have full bed variation in height of just 0.16mm and about to reduce that with minor (~0.02mm) adjustments to end stops and 0.1mm change to horizontal radius. Problem was play in the end effector created too much 'noise' in my nozzle location to accurately measure/calibrate. Wanted to update post so anyone following it with similar issue might find a solution as well.

[Eaglezsoar]

Replaced stock arms with carbon fiber version to eliminate play in end effector. Major improvement. Currently have full bed variation in height of just 0.16mm and about to reduce that with minor (~0.02mm) adjustments to end stops and 0.1mm change to horizontal radius. Problem was play in the end effector created too much 'noise' in my nozzle location to accurately measure/calibrate. Wanted to update post so anyone following it with similar issue might find a solution as well.

I am glad you found the fix!

[bot]

Just FYI the carbon fiber arms themselves dont offer much, if any, advantage. It is simply the manner in which try are fastened to the carriages and effector that improves performance. The same can be achieved by adding tension to the arms to prevent lateral play.

[lbarger]

Just FYI the carbon fiber arms themselves dont offer much, if any, advantage. It is simply the manner in which try are fastened to the carriages and effector that improves performance. The same can be achieved by adding tension to the arms to prevent lateral play.

I am agree with you on that. I got the arms more for the joints and included tension bands than the material. Probably should have commented on that.

My issue was play in the system. I tried adding some rubber bands to the original arms. It seemed to help but did not eliminate my issues. I probably did not use enough tension as I did not know how the plastic arms would react to tension over time (aka slowly deforming?).

[geneb]

The arms are glass-filled nylon. I doubt they'll deform over time, just don't apply so much pressure that you can see the arms deform.

g.