I’m putting the LR4 beta system back together after getting a couple systems together for my friends. They got the skrs I had. One runs klipper. The other marlin. I kept the oddball parts from the old 3d printer the motors came off which leads to the issue:
The old trigorilla board has 5 of the 2209 removable drivers that are hand wired to the servo pins for rx communication. It is operated by a mega 2560. X moves. Y moves and only one z channel works… The one labelled E1 (X, Y, Z, E0=Y1, E1=Z1)
I think the micro pin for Z is bad because
Same motor and cable work on the x driver while X motor and coil do not work on this driver
Swapped the 2209 for z and x. Problem stays with the z position.
Z motor and z cable and z driver all work on x channel. I’m running klipper. There are no errors, just nothing. The Z1 channel moved independently and the Z channel does nothing.
I gave both skrs I had with the new builds I just completed. With everyone going to jackpots, anybody got a spare skr 1.2 or 1.4 turbo they will part with? I have had great luck with those and if they are on the excess pile, let’s talk. This build is mostly all from extra parts so I’m not buying new for it.
Trying to add an ender control board (mini E3V3) with 4 channels and move an axis or two to that board to split the motion load between two microcontrollers and have 8 working channels total in klipper… this could be cool.
UPdate: the falling z axis on the left may have cooked the z channel. LR4 brakes anyone? I’m going to set them up now. anyone have an skr with one cooked board channel - not a showstopper.
fried Z channel from the Z on X min side spinning up like a jet engine when the power cuts off.
Move Z, Z1 and X over to a miniE3V3 along with 2 relays for Z axis brakes (1 coil mode only on each z motor) This doesn’t entirely stall it either.
used multi_pin in klipper to tie X axis, both z axis and the coil shorting relay pin to all enable at the same time.
One note of importance I gleaned from the multi MCU option in klipper: make sure you have all the pieces of one axis on the same MCU. both motors and their endstops if Y or Z. X and its endstop if X is moving. for the second controller, every pin is called out from the [mcu extra_mcu] label where the communication is declared for the second mcu. main mcu is this:
step_pin: PA4
second mcu is this
step_pin: extra_mcu:PA4
The Y1 axis failed as well. not sure what the story is on the trigorilla board. Moved the Y1 from E0 to X channel and it works now.
system will home all axes on demand (assuming the left side doesn’t bury the bit first while the power is off).
not shown: the rat’s nest that is the board stack on the back to run this beast.
I really like the idea that anyone’s old control boards with a working 2209 channel could be added as spare channels or agglomerated to make a functional system. “one man’s trash…”
Random cuts last night for a project: got the thing leveled and homing. had the brakes wired in wrong, so now they hold position. Made a “go home” macro so after homing, it goes to a reasonable location. After a while the X axis was skipping terribly. I swapped the x motor for a Ryan’s full size motor and that didn’t fix it. Had to reverse the direction config so +X moved the right direction. Then tried dropping all the accelerations… what are typical XYZ accelerations you all use for CNC? cut the speed, changed the router rpm… no fix. Before I threw anything explosive at it, just stopped for a minute. then clink. the bit fell out. Guess that will do it. Swapped the bit and snugged it down - back in business.
the tmc driver survived. I still wonder if something wasn’t configured correctly. Moving the tmc driver, cable, and motor to another location worked, so it points to the pin or the config. I copy/pasted the config and it worked on the last printer this board ran 6 months ago, so I’m pretty sure it is the pin.
(photo before final routing. Lower nubs on the slots were manually cut off because they chipped. ACX ply didn’t like the tight corners)
there was a several mm difference in gantry height from left to right. noticeable on the upper far left with the layer color change. I measured with calipers and it was 0.2 mm difference, but that was before the workpiece mounted and it wasn’t level. Whether it was late and I was tired or I fat fingered the endstop when I did the original measurement I don’t know exactly. All are possibilities. The result ended up useable, but it could have been much cleaner. Had to get it out in time for christmas and allow time for final color application. I believe milk paint is on the menu. Will be fun to see how it turns out.
second carve of a 2x6 - as cut with the vacuum shoe attached. This was how much debris remained on it. Kudos on the TPU bristles for a setup that just works. The janky taped adapter insert shown in the first picture is all on me:
pretty impressive. used a 60 deg v bit at 3 mm step depth. dropped initial start position to -0.5 mm and rezeroed and ran again so the small letters showed up better.
Klipper z probe does not work yet for me. Homing +Z with endstops works fine, but how to get a third channel with the plate to work as the z probe is not yet understood. Any insight or macro code for this would be appreciated.