Poor 1st Layer Adhesion Between X & Z

[Christian79]

I recently completed my v1 to v3 conversion and am amazed at how much better the entire build process and final product is in just 3 short years.

For the first time since I bought my v1 in 2013 have I been able to print the 265mm circular test print with decent results. However I am still getting issues between the X and Z tower. For example if I print the test pattern the nominal thickness of the print is around 0.30mm but between X and Z (from 7 to 10, if Z tower is 12) the thickness increases to 0.42mm to 0.46mm but this effect is not mirrored to the same degree 180degrees apart. The opposit side of the bed (from 1 to 4) is only reduced from 0.3mm to around 0.28mm.

I have verified my bed is flat at 20C and 80C, my endstop offsets are (1,2,1), my Horizontal radius is 143.62, and my diagonal rod length is still stock. All other settings are still stock. I have ran G29 about 10'times to get my offsets dialed in but didn't use the online calibration as it previously changed my arm lengths and required a change of over 1.0 to my horizontal radius. And even after that I still had the same issue. I have also measured my tower lean with X and Y being less than 0.2mm off at the top and Z being about 0.65mm tilted in at the top.

Any advise would be greatly appreciated!

[mhackney]

This sounds like the "bed tilt" issue that many of us experienced earlier this year/late last year. Using the web based calibration based on David Crocker's implementation either manually (on the SeeMeCNC site) or via the Octoprint plug in significantly improved things. Which version of calibration are you using?

[Christian79]

I have used the online option. I have held off on Octopie thinking I may go duet, but I wanted to keep the accelerator probe.

Bed tilt does seem an good way to describe it aside from the error is not equal across the centerline of the bed.

[Christian79]

I had some time this weekend to work on my printer before I head off to Germany on a BT next week. It seems my layer adhesion trouble has more to do with thermal variation through out my bed than nozzle spacing. Using an old non contact temp gun I was able to find out two things, number one 80C on the LCD is more like 65C on the surface of the PEI coated glass. Number two, depending on the area of the bed I am testing, the temp towards the edge can be 25+ degrees lower than the middle. I have been able to successfully bond ABS in the middle with the printer set to 100C (92C wit my gun) but still no luck at the edge were temps can drop as low as 67C which even PEI wont help when printing with ABS.

So I now have two questions:

1) How do I even out the bed temps? I already let the printer warm up for 15 minutes...
2) How do I keep the calibration constant(ish)? I calibrated to 0.03mm per the online tool but two days later I was seeing 0.15mm to 0.20mm of "proud" work surface causing the extruder to grind in the filament until the next layer.

As a followup question to number two, does everyone recalibrate every time? I know mhackney does but I don't know if that is just a G29 calibration or some Duet or Octoprint calibration. I really love the MAX but I don't want to have to spend an hour each time I want to use it to tweak the values.

As always any advice is welcome and appreciated!

FYI, my printing environment is a finished basement at a steady 65 degrees and a regulated relative humidity of 40% to 45%. However the printer does sit below and to the left of a supply vent and also in the path of exhaust for the dehumidifier. (150pt/day). Maybe I will get some cardboard and make an simple box for 3/4 sides of the printer.....

[Xenocrates]

First thing: Aluminum/copper foil or sheet. Two layers or so of heavy duty foil starts making an impact on temperature evenness. I personally use an AL plate, either 1/8th with PEI on top, or 1/16th under glass. Copper is better if you can find it, because of a higher thermal conductivity, but it's harder to find and work with to a degree.

Second: Enclose the printer, and ensure every fastener is tight and threadlocked. How Mike recalibrates is that he has built the gcode to call the script into his start gcode, and that is a bit of a Duet thing, although the V3 should be able to do it too.