Ahhh man maybe can I touch it just a little bit tho?
November 26, 2025 at 12:44 AM
Ahhh man maybe can I touch it just a little bit tho?
Still waiting on some chemicals, so kicking off the first part of a via direct metalization process with graphite powder. Just graphite in the vias, no copper plating yet.
It's not pretty, but it seems to be maybe functional?!!
About 300ohm resistance between sides.
It's not pretty, but it seems to be maybe functional?!!
About 300ohm resistance between sides.
November 21, 2025 at 7:02 PM
Still waiting on some chemicals, so kicking off the first part of a via direct metalization process with graphite powder. Just graphite in the vias, no copper plating yet.
It's not pretty, but it seems to be maybe functional?!!
About 300ohm resistance between sides.
It's not pretty, but it seems to be maybe functional?!!
About 300ohm resistance between sides.
these pretty traces!
November 15, 2025 at 3:40 PM
these pretty traces!
just living the dream around here
October 24, 2025 at 4:18 PM
just living the dream around here
I set off the fire alarm so many times today.
October 17, 2025 at 11:26 PM
I set off the fire alarm so many times today.
it worked in real life!
- Frontier scan to detected a feed tape
- YOLO bounding boxes for all parts (model tuned on LumenPnP cams)
- Ran a capture at each detected part
- Re-ran YOLO to get the nearest bounding box
- Calculated the offset, reran capture
cc @thea.codes
- Frontier scan to detected a feed tape
- YOLO bounding boxes for all parts (model tuned on LumenPnP cams)
- Ran a capture at each detected part
- Re-ran YOLO to get the nearest bounding box
- Calculated the offset, reran capture
cc @thea.codes
October 12, 2025 at 4:03 AM
it worked in real life!
- Frontier scan to detected a feed tape
- YOLO bounding boxes for all parts (model tuned on LumenPnP cams)
- Ran a capture at each detected part
- Re-ran YOLO to get the nearest bounding box
- Calculated the offset, reran capture
cc @thea.codes
- Frontier scan to detected a feed tape
- YOLO bounding boxes for all parts (model tuned on LumenPnP cams)
- Ran a capture at each detected part
- Re-ran YOLO to get the nearest bounding box
- Calculated the offset, reran capture
cc @thea.codes
deduped, with highest confidence score. I'm going to guess that there's an offset, but probably a constant one
October 11, 2025 at 3:19 PM
deduped, with highest confidence score. I'm going to guess that there's an offset, but probably a constant one
A map of all the components on the board, by mm. The overlap is a little disappointing, guessing an affine transform thing.
Yellower color is a confidence score weighted by distance to the center of the image.
Yellower color is a confidence score weighted by distance to the center of the image.
October 11, 2025 at 3:15 PM
A map of all the components on the board, by mm. The overlap is a little disappointing, guessing an affine transform thing.
Yellower color is a confidence score weighted by distance to the center of the image.
Yellower color is a confidence score weighted by distance to the center of the image.
Direct YOLO -> Pixel Calibration
scale_x = 0.01219138 mm/px
scale_y = 0.01153833 mm/px
idk, maybe it's right?... ish?
@thea.codes @opulo.io
scale_x = 0.01219138 mm/px
scale_y = 0.01153833 mm/px
idk, maybe it's right?... ish?
@thea.codes @opulo.io
October 11, 2025 at 3:13 PM
Direct YOLO -> Pixel Calibration
scale_x = 0.01219138 mm/px
scale_y = 0.01153833 mm/px
idk, maybe it's right?... ish?
@thea.codes @opulo.io
scale_x = 0.01219138 mm/px
scale_y = 0.01153833 mm/px
idk, maybe it's right?... ish?
@thea.codes @opulo.io
And combined with the image splits, did a frontier scan to figure out regions of the build plate that look promising.
It's only step 1, but it significantly cuts down the search space for automatically configuring feeders.
It's only step 1, but it significantly cuts down the search space for automatically configuring feeders.
September 29, 2025 at 4:54 PM
And combined with the image splits, did a frontier scan to figure out regions of the build plate that look promising.
It's only step 1, but it significantly cuts down the search space for automatically configuring feeders.
It's only step 1, but it significantly cuts down the search space for automatically configuring feeders.
Did a simulated grid scan over the entire build plate to see regions that might be worth exploring more.
September 29, 2025 at 4:53 PM
Did a simulated grid scan over the entire build plate to see regions that might be worth exploring more.
From here, built a simulator in a colab to stitch together images from any position. Messy, weird projection, stuff, but okay for this purpose. I split images into 9x9 grids and ran embeddings on these splits. This tells us "looks like there might be more parts top-left from here"
September 29, 2025 at 4:52 PM
From here, built a simulator in a colab to stitch together images from any position. Messy, weird projection, stuff, but okay for this purpose. I split images into 9x9 grids and ran embeddings on these splits. This tells us "looks like there might be more parts top-left from here"
Been having a blast with the @opulo.io LumenPnp. But my overall goal is to throw a bunch of junk a build plate and assemble a circuit.
First step, I wrote a plugin to do a 1cm x 1cm scan of the entire build plate. Collected images, got embeddings, manually classified them.
First step, I wrote a plugin to do a 1cm x 1cm scan of the entire build plate. Collected images, got embeddings, manually classified them.
September 29, 2025 at 4:50 PM
Been having a blast with the @opulo.io LumenPnp. But my overall goal is to throw a bunch of junk a build plate and assemble a circuit.
First step, I wrote a plugin to do a 1cm x 1cm scan of the entire build plate. Collected images, got embeddings, manually classified them.
First step, I wrote a plugin to do a 1cm x 1cm scan of the entire build plate. Collected images, got embeddings, manually classified them.
cooking up some raspberry pi
September 23, 2025 at 6:00 PM
cooking up some raspberry pi
'void in negative space', 1 of 3
September 13, 2025 at 8:04 AM
'void in negative space', 1 of 3
Same!
Re-discovered at a surplus store for $5 the other day. Art is supposed to be controversial.
Re-discovered at a surplus store for $5 the other day. Art is supposed to be controversial.
August 11, 2025 at 11:15 PM
Same!
Re-discovered at a surplus store for $5 the other day. Art is supposed to be controversial.
Re-discovered at a surplus store for $5 the other day. Art is supposed to be controversial.
New dual purpose drum machine / cat scratchers are in.
August 11, 2025 at 1:43 PM
New dual purpose drum machine / cat scratchers are in.
feeling shiny
July 24, 2025 at 3:31 AM
feeling shiny
$5 case at the university surplus store!
Art is supposed to be controversial.
Art is supposed to be controversial.
July 21, 2025 at 1:23 AM
$5 case at the university surplus store!
Art is supposed to be controversial.
Art is supposed to be controversial.
Made it through two weeks of travel and salt. Lost a diode so F4 stopped working, and finally lost the USB-C at the end. But managed to get through a re-write in Rust, and now have an working FM synth.
Next, figuring out flash. ESP32 Rust FAT support seems iffy.
Next, figuring out flash. ESP32 Rust FAT support seems iffy.
June 14, 2025 at 2:20 PM
Made it through two weeks of travel and salt. Lost a diode so F4 stopped working, and finally lost the USB-C at the end. But managed to get through a re-write in Rust, and now have an working FM synth.
Next, figuring out flash. ESP32 Rust FAT support seems iffy.
Next, figuring out flash. ESP32 Rust FAT support seems iffy.