Hey Asmbly! In the summer of 2020 I acquired a Taz Workhorse with an intense number of modifications. The person I purchased it from had replaced the control board with a Duet 2 Wifi and added a PanelDue to it. I have since replaced the 1.75mm hotend with a Matrix (Chinese Hemera knockoff) although I’m waiting for my Hemera Revo to come in to put on the printer and added an EZABL.
In the year and a half I’ve had this printer, I’ve never managed to get a consistent first layer. I’ve rebuilt the gantry twice, replaced the z belts, swapped the bed platform a number of times until finally installing the lulzbot flex plate.
I’m quite confident the Duet configs are correct and that this is in fact a mechanical issue. At this point I’ve been staring at this printer in a vacuum by myself for so long - I’m not sure it’s possible for me to have an objective eye any longer.
If anyone is interested in getting into the guts of this printer with me, I’m happy to compensate with money, food, free filament, printing, cad help, or really anything you want, I just need some help!
I’m about 70% certain its not an ABL issue because the issues are present whether or not the bed mesh is active during the print.
I don’t have any current pictures, but if the coordinates of the bed are normalized, at 0,0 and 1,1 (front left and rear right corners), the nozzle is the correct height from the bed. However, at 1,0 (rear left) the nozzle is so far away that a standard .2 mm first layer is being extruded into thin air, and at 0,1 (front right), the nozzle is so close that it is preventing material from being extruded.
Hmm have you checked the squareness of all right angles to see if something is skewed in the frame? are the acme screws all straight? I’m just tossing out those easy but forgettable things to check for, apologies if you have gone through all of that.
Squareness is my thought as well. I’m not familiar with this particular printer, but wonder how the person who assembled it went about it.
It’s generally not good to insert one fastener at a time and tighten completely before moving to the next - instead, insert as many as possible and go a little less than finger tight. Then finger tight, intermediate, and finally fully tight. Assemblies will be much less likely to be warped that way than if each fastener is tightened by itself.
If that really is what’s wrong, simply loosening the fasteners on the offending assembly, wiggling the pieces a bit, and re-tightening will make a big difference.
Other things to check for would be washers or other spacers in the wrong place, or pieces with borderline tolerances that are stacking up and becoming even worse.
HA! Thank you @mark999 and @Devmani ! I’ve discovered an issue with the squareness that I wouldn’t have thought to check until some folks really pushed me to verify. The printer is out of square in 2 dimensions, z and y. What I didn’t realize is that when I square the printer in the z axis, it pulls the y out of square ever so subtly, and when I square the y axis, it pulls the z axis out of square. For whatever reason, I’ve verified squareness in a single axis numerous times, but I didn’t think about how that would impact the squareness in the other axes.
Now, to figure out how to go about getting the frame fully squared is going to be a slightly different task.
You’ll need to make small adjustments to both the z and then the y. So like a quarter turn on one then a quarter turn on the other. It’s tedious for sure or you can start on one axis and almost get it tight then start tightening the other axis and then bounce between the two until both are tight and squared.