Z-Axis Circles out of round at one corner of bed

I thought I had things well calibrated, but I just discovered that vertical circles are printing as ovals at one corner of the bed. That is, if I print a tube from the lower left corner to the upper right corner, the upper
right end has a good circle, but the lower left is squashed down into an oval. Given the variance between the corners, I'm not sure where to look for a calibration or other problem. Any ideas appreciated. Thanks very much!

More detail on your printer is in order, as well as pictures of the problem. Could be that one of your z motors is skipping steps and not rising far enough, causing a squished circle. Only a guess though.

I'll have to get a couple of photos. For a cylinder test print, you can see it start to taper down past the center of the bed, until when you get to the corner the circular cross-section has been squished down to a noticeable oval. I assume that binding could cause this too, perhaps if the x-axis has drifted out of a level alignment between the two z-axis lead screws? Thanks.

move the X gantry up a way and measure the distance from the Y smooth rods to the X smooth rod, make sure they are equal
I allso found it usefull to draw a vertical line on the 2 Z motor couplings (when they are alligned) helps to see if they have missed any steps and got out of sync.

Here are links to four associated photos (this forum is fighting my attempts to upload them).

You can see how the rod rather abruptly starts to narrow down towards one end -- that's the end with the oval, and that end was positioned near 0,0 on the bed. You can see the shape of the oval in another photo, with a bit of the support structure still underneath. You can also see that the other end (positioned near x max, y max on the bed) has a fairly decent circular shape. Thanks.

[drive.google.com]

[drive.google.com]

[drive.google.com]

[drive.google.com]

When it's that bad I'd say it's fairly certain one side of your z axis isn't moving up enough. Several possible causes: binding, too little (or too much) power to that motor. Did you try making sure the motor on the skewed circle side has approximately the needed amount of current?

I'm looking into alignment issues that might cause binding now. I'm pretty sure both Z motors are being driven from the same driver in parallel, so I probably can't increase the current on one without increasing it for the other. There aren't obvious adjustment pots on the board, so the only way to change that might be in the firmware itself -- and since I don't have the original Configuration.h for the build this is not a trivial exercise, given that I'd have to figure out new temperature tables for unknown thermistors as well.

For the record, I just did the following:

1) Matched height of both sides of the x-axis gantry on the z-axis motors the best I could with my un-great calipers.
2) Put a sliver of tape on the front of both z-axis motor couplings
3) Ran the gantry up almost to the top, then back down to the bottom

To the eye at least, the motors stayed in sync visually as far as the tape markers were concerned.

Back at the bottom, remeasuring the gantry height on both sides, showed them at around 0.2mm difference, which is within the margin of error for these calipers.

Can you check the X Y and Z are 90° to one another.

If the Y is skewed relative to the X it would oval the print at one corner.

Ed