v0.2 Smol Printer Build: V0.2-S1r1

[claudermilk]

Yes, that one. Original Github link here.

[claudermilk]

A little progress after a vacation hiatus. I had hoped to make a bunch of progress over vacation, but wifey kept us all busy and away from the printer.

Prepping the bed.

Wiring up the printer side and loading the cable chain. Aha, now we see why they didn't insert the pins into the connector shells. We also had an amusing minor roadblock: one of the chain links had some molding flash that "blocked" the path. I pulled it out and replaced with a normal spare link, then found that the flash was very thin.

 

Mounting the bed

 

She was intimidated by tackling the wiring, but with assistance got it done.

Next up is the tool head, so we saved that for a dedicated build session.

[claudermilk]

Some more progress. It's slow but steady, contingent upon the time available and desire of my co-builder to tackle the next step. This time it's the tool head.

First up, the extruder (still Pocketwatch?)

The fans were a bit of a challenge, so she punted over to me.

 

I also handled mounting the tension knob and making the initial setting. With commentary on what it's doing, why, and how the parts interact. Then comes the motor; she got it mounted and I finalized gear lash--again with instruction commentary and hands on feeling what proper lash is.

Then comes the hotend. I'm totally tearing these apart later and stealing the blue heat sink for my blue-framed Trident and giving this guy the red Voron edition part. 

Lastly, tool head mounted! It's really looking like a printer now. Next up is electronics.

 

[claudermilk]

Beginning electrical parts now! It took a while to get a time window we could work on it (and she was motivated to do at least a little); I suspect she's a little intimidated by the idea of working on PC boards and wires.

Toolhead board.

  

I'm an idiot. Looking at @YaaJ's post just now, I see where the motor cable is supposed to go. I'm pulling the toolhead board off later and re-routing the cable correctly.

Picobilical frame PCB

  

The peels!

 

Finally, power supply and inlet switch.

 

I managed to forget to print the controller/Pi mount parts, so we couldn't get those done--and she was about done at this point anyway. Next up: mounting the controller and Pi and starting to attach wires.

[claudermilk]

As promised, I tweaked toolhead wiring routing. It's much cleaner now. I think this will pass Sanity's serial checks.

Also, thanks @YaaJ for making me check those Picobilical Frame PCB jumpers. They were indeed set to 5V. Fixed now.

[Sojin]

3 hours ago, claudermilk said:

As promised, I tweaked toolhead wiring routing. It's much cleaner now. I think this will pass Sanity's serial checks.

Also, thanks @YaaJ for making me check those Picobilical Frame PCB jumpers. They were indeed set to 5V. Fixed now.

That looks much cleaner than mine. Might have to revisit my routing. I probably cut the left side PCF too short to route through the middle, but the others might still work. Glad the Picobilical Frame PCB jumper PSA helped 

Edited September 21, 2023 by Sojin
spelling

[claudermilk]

Big progress! Wiring is done! Next is making it move--that is tomorrow.

Mounting the electronics cooling fan.

 

Prepping the controllers, and mounted.

 

Connecting the power supply, and the nice safety cover.

  

First exposure to these tiny darn connectors. Yes, I'm totally cheating using her small hands for this stuff.

 

Wiring montage!

     

A full cable channel, and covered!

 \

Connecting the umbilical. This took some care--enough pressure is needed that I was worried about breaking things.

How it sits now. Next up is inserting the flashed SD and starting software configs and printer checks. Getting close!

 

[Perkeo]

Wow, that is some serious clean wiring in the back of that Fusion Geomagnetic Mauve machine!  To get the A+ for looks in the electronics bay, a Fusion Geomagnetic Mauve or Galaxy Teal coloured cable tray would finish the job.  

[claudermilk]

 

I just ran into a brick wall. I am getting no 5V power out of the Picobilical Fram PCB. Has anyone else run into this?

The SKR Pico is getting 24V. The outputs for the wire to the picobilical is getting 24V and I see 24V at the input end on the picobilical board. I have nothing coming out of the 5V outputs for the Pi. Am I missing something, or is this possibly a bad board? There have been no sparks or magic smoke released.

It's wired up exactly like the LDO docs show. The only difference is I'm using a Pi 3B, but the GPIO blocks are the same across all Pis, so that won't be an issue unless I'm missing something.

 

[Penatr8tor]

@claudermilk Dude, I feel your pain. I struggled all weekend getting the VzBoT configured to home/move in the right direction. Got it homing tho. I'll post up the details in my build diary.

Sorry that I can't add anything to help you Chris, but I have to say your build looks awesome so far and what's equally as awesome is that you're doing the build together with your daughter. You're a good Dad my man. 

[Sojin]

Sorry to hear about the brick wall.

Do you have a multimeter to check if you're getting 5v out of the Picobilical output? I'm assuming you've checked the Pi3B standalone and all is well there? 

[Perkeo]

The LDO Picobilical is very different from the much simpler Formbot Umbilical, which is just a feed through of all connections into the 14 pin connector.

If I read through the LDO documentation, you actually need to flash the firmware onto the Picobilical board.

If the board receives the 24V power, but shows no output, my assumption is that somehow the flashing of the firmware did not work correctly and therefore it does not process the information from the RPI correctly.

Other option is to try to change to a different USB cable. I had this problem when I could not see that the board was in DFU mode. Changing my USB cable to another one fixed this problem. 

Hope you find the solution and get things working! 

 

 

[claudermilk]

I did probe out the power. I'm getting what I expect everywhere except the Pi 5V. I did go through the LDO docs site a couple of times and it sounds like the board should be flashed already. It certainly leaves me the impression it should just power things up--there's no instruction for powering the Pi separately first. I also took another look at the main V0.2 manual and I'm thinking I could build a jumper to power the Pi the normal way shown from the Pico controller.

[Perkeo]

Why not simply try re-flashing the board? I read or saw YouTube about a CAN installation with pre-flashed boards that also did not work,  but after re-flashing the board it suddenly worked?!

[claudermilk]

I have to get the Pi up and running to be able to do that. It was a chicken-and-egg situation.

I went ahead and built a quick jumper for the 5V from the SKR Pico to the LDO power hat. It worked! KIAUH is installed, and Klipper is installing as I type this. From here I'll be able to get that Picobilical PCB fixed.

[claudermilk]

OK, made some progress. Pi is set up, Klipper, Moonraker, and Fluidd are installed. It appears the SKR Pico is flashed ok--I have a serial ID. The  Picobilical frame PCB remains a PITA. I think I got it flashed. I did get a serial ID once. Now I cannot get that to show up any more. I went through the steps here: https://docs.ldomotors.com/voron/voron01/Picobilical#compiling-the-firmware. I tried the UF2 mode, which didn't seem to do anything--I didn't find any kind of drive to mount. So tried the Klipper Make which seemed to work. But no serial ID after the first time I saw it. What the heck am I supposed to do with this thing? I'm sure i'm missing something stupid, by my brain is tired now.

[claudermilk]

Got it solved! It was stupid. I had a think and tried the PC method, connecting the board via USB to a laptop; that still didn't work. So in pulling the USB cable I noticed the 24V power leads moved. Huh? It turns out that they were loose for some reason, so the board was only getting power while in DFU mode.  Both 24V and 5V leads carefully checked again that they are good and secure. I now have happy LEDs on all three boards. Pi is powered up off the 5V buck from the Picobilical board. ls /dev/serial/by-id is giving me both serial IDs. I can get back to moving forward again tomorrow.

Obviously, my daughter didn't stick around for all the gory details of troubleshooting this, but I did make sure to walk her through the broad strokes of what I was doing, and what the intended flashing steps are.

[claudermilk]

First hit on the Fluidd page after getting the Pi set up.

And as of last night.

Time to reprise this gif!

While we melted plastic, no print just yet. It was getting late and I somehow mis-set the z-offset; I'll get that redone and we will print a cube later today.

[claudermilk]

I printed a thing!

The finish line is in sight now. Just panels, skirts, and tophat left to build. Then all the software/macros tweaking and tuning.

While I kept my daughter's attention long enough to walk her through the broad strokes of installing Klipper and setting up the config files, it's not holding her attention. So I'll tackle the tedium of tuning--she's ready to just start slicing and printing stuff.

The interesting bit with that cube is I used the default V0 config for SuperSlicer (though I turned on firmware retract) and my generated filament tuning from the Trident. PA & EM testes pretty much tracked exactly with this printer, so I don't see a need to redo all that for now.

Edit: Oh, and bed leveling and getting z-offset right is killing me. I now really appreciate the z_tilt_adjust and Tap on the Trident.

[claudermilk]

Getting really close to the finish line. Panels went on this time. Tophat is next, and that will be it.

The peel!

Bottom panel going on. This will make Steve happy.

Back panel with all the pixies safely contained.

And where we sit now.

 

[claudermilk]

Fin.

Now for tuning, software tweaks , and mods! Oh, and recording a serial video.

She got distracted (and had some more homework to do), so I mostly handled sorting what subtly different bracket went where on the tophat for the panels. She did make sure to record herself doing the peels--she's a pro now.

[Sojin]

Are those tophat handles there on the front? For lifting open the tophat?

[claudermilk]

FYI, trying to get input shaper set up was giving me fits. Unfortunately the instructions are scattered in several places and the current last instructions for installing numpy are incorrect. It throws an error that no BLAS libraries can be found. After much searching, I found the solution on the Klipper discourse. Using ~/klippy-env/bin/pip install -v numpy==1.25.2 allowed numpy to install and now input shaper is off and running.

[claudermilk]

3 hours ago, Sojin said:

Are those tophat handles there on the front? For lifting open the tophat?

Nope, those are little nameplate parts from LDO's Github. They give you a kit serial plate which will go on the left one. The right one is just a Voron and LDO logo. Not necessary at all, but why not?

[Perkeo]

Congrats on a nice build and getting your daughter interested in the 3D printing hobby. 

Serial number and input shaping is on my ToDo list. Then I urgently need to start printing the first parts for my hopefully shortly arriving Formbot 2.4 Kit