Project Logs

Collapse

• Firmware Quirks

  CriptasticHacker • 09/23/2024 at 00:40 • 0 comments

  When I first powered this up, the screen came on for a couple of seconds, and then shut off.

  Only after translating the manual and picking my brain did I discover that this is by design!

  The unit doesn't give any indication of this, but it needs to be calibrated before first use.  So the person has to hold down the Button 1 for a few seconds to go into the calibration mode.

  Then the probes need to be attached to for resistor values (ideally at 0.1%) - 0 ohm, 1 ohm, 2 ohm, and 8.2 ohm.   You'll need to assemble a test rig for this.

  That's as far as I've gotten so far.  The unit is working now, so I passed it onto my friend. He said he would let me know later of the updates on using the firmware and squashing any software bugs!  Personally, I would like to see some text added to the screen on boot that says "need calibration" or something like that.  I thought I had broken the power circuit!
  

  I also tried to download most of the documentation that is stored in the cloud on EasyEDA - which is not in the original Git repository from the author.  Gerbers, board layout PDFs, etc.  I'll try to upload those later.
  

• Battery terminals and final parts

  CriptasticHacker • 09/20/2024 at 04:45 • 0 comments

  The battery terminals slide in nicely (as do the batteries - it's a cool design) but no accomodation was made to keep the terminals at a safe distance from shorting.  What's more, they have to have their solder tabs cut.  Personally, I would have opted for a 2mm thicker pen:

  
  Again, I ended up using insulating polyimide over the contacts and under the main probe. Sadly, the positive terminal can short to the motherboard if you're not careful - it was not designed with any unpopulated courtyard on the PCB to avoid this, the way consumer products usually are.

  
  After this, you'll need to connect 2 wires (ideally insulated together - like from an old audio cable) to PB1A and PB2A.  these are the through hole solder joints under the display.  This cable, or rather, those 2 conductors, travel through the 3D printed back nut and then get soldered to another externnal pinpoint probe.  This is what completes the circuit - 2 wires (to their own test probe) and the built-in test probe  So, a 3-wire Kelvin setup!
  

  
  At the end of the day, it does come together and it does work.  and it's free and open source!
  

• not all OLEDs are created equal!

  CriptasticHacker • 09/19/2024 at 21:16 • 0 comments

  So...it turns out these cute lil' 0.96inch SSD1306 displays have many manufacturers, each with slight variations.  Make sure to get the color of the pixels you want as well! (white is best in pretty much all cases for the visibility on black background).

  
  
  There is a frame the designer included that they recommend all sorts of glueing and stuff to get aligned.  The case only fits a particular make of the display though, and if printed in PLA it is the same proble of tolerances being too tight - the buttons wouldnt work.

  Printing in TPU, or making a 1mm acrylic lasercut cover would be nice.  I decided I didn't care enough for this project.

  (I tried the glue thing...it was wack)
  

  I simply insulated PB1A and PB1B pads with some polyimide and used 32 AWG inductor solid wire to connect the OLED data and power...and this holds it in place just fine.

• 3D print updates

  CriptasticHacker • 09/19/2024 at 20:56 • 0 comments

  So a friend asked me to print a case for this cool project.  Easy peasy, yeah?
  Sometimes open source stuff can be an undocumented headache though, and this one had some road bumps for sure.

  
  For starters, the entire manual was in Czach.  There was an English "README" so that threw me off for a while.  The Youtube video from the maker is also in English...and doesn't mention the manual.  See how I got lost here? :P

  Anyway, the manual is now translated and uploaded in a .zip here.

  After that was done...I see the info on 3D printing is scant, to say the least:

  • there is no info on print settings at all
  • there are only 2 .scad models, neither are the primary case
  • other than that, there is only a Cura .3mf file

  Fortunately (and this is really cool), Cura .3mf files can export individual part STLs - pretty neat!  I stupidly did a 6 hour print of everything first though, and they mostly failed.

  There are fillets where the case meets the bed, and this is never ideal for good print adhesion.  I opted to update the models for easier printing:

  Imported everything into TinkerCAD as you see above.  The tolerances were also unecessarily tight, and even with my highly calibrated dimensional accuracy settings, the case would not snap shut.

  This is a bug that will happen when tolerances are too tight and walls are too thick, so there is no "give" in the plastic.

  I reduced the highest Z edges in the Main Case B model (my naming scheme) to be 0.5mm away from the edge.  Now everything fits nicely, if a little loose.  And that is ideal!  Because there are 2x bulky nuts to keep this thing tightly closed anyway.

  I "squared" the base of each model, removing the fillets.  This makes for a slightly less ergonomic design at the gain of much easier printability.
  

  After that, I needed to fix the buttons.  The original prints didn't fit at all.  I changed tolerances, increased Z-height (so your finger can press them more easily), and combined them with a connecting arm - always a good idea with multiple buttons.  I updated Main Case A to fit this new arm as well.
  

  print supports are needed for the threaded ends of Main Case A and Main Case B
  

  Fortunately, the screw on tip and back nut pieces work fine as-is, just make sure you print them one at a time.  The tip might fill in the top hole a bit too easily, so poke it through if needbe.  I may upload a model with a wider opening later (again, tolerances too tight here).
  

  ---

  Print Instructions:

  Make sure to print Main Case A and B with supports!  Use my provided STLs here.

  (no other parts need supports)
  

  Print the main case with your favorite standard PLA settings (200c nozzle / 60c  bed roughly and 50mm/s speed / 30mm/s first layer).

  
  Buttons should be printed very slow, same temp settings.  I went with 10mm/s.  Still only a 9 minute print on a regular old bed slinger.
  

  The back nut and screw on tip can be printed with regular settings, but I reduced speed down to 30mm/s just to be safe. This gives extra cooling time for those threads to come out great, since they have slight overhangs.

  No need to print the OLED bracket or ground wire holder pieces - they aren't useful imho (more on that later)

  (there is also the nut and the other half of the case to print - these are just some reference pictures)
  

View all 4 project logs