btz
added to the journal ago
More soldering
Yesterday started exciting and ended depressing.
Here was my todo list:
reflow all joints
Wire a new mt3608 bodge with the new module. You'll need to bodge across, swapping + and - on both vin and vout.
Cut the corner pins off of the board. Melt the plastic and cut the metal?
the EN ESP PIN should lead nowhere.
The gp23 ESP PIN should lead to the pad under the VIN pin
The VIN ESP PIN should lead to the pad under the 3v3 pin
The 3v3 ESP PIN should lead to the pad under the en pin.
In kicad-speak:
Traces that currently go to vcc, go to en.
Traces that currently go to gp 23, go to vin.
Traces that currently go to vin, go to 3v3
Change the pins in the header file to match the scam mirrored pinout.
solder in the lcsc parts
cross fingers!
I did all this (except for the header file and lcsc parts).
Also, even with the flipped polarity, my second mt released mystical magical smoke. So I'll put lipo battery charging on hold (and, as it turns out, 3
xAA straight to VIN (through a switch) is gonna work just fine. )
This is how it's looking!

update:
(below not included in time tracking btw)
Through various mishaps, I am no longer able to flash my ESP32 (and even if I did once, it'd probably be unreliable, though I'm open to solutions)
The reason why I started messing with the esp pins and adding bodge wires in the first place was because, despite self-review and blueprint review, the pinout on my devboard footprint was really wrong (not just mirrored, like really really wrong; it's a long story).
I'll probably need a new esp32 and pcb for my project to work cleanly.
And I guess I'll scrap my lipo charging logic because I fried both mt3608s, even when there was no battery. A battery pack will suffice, I guess.
shipping's ofc going to take forever now, and I'll have to wait again.

I've now refootprinted the ESP and removed the charging logic, and rerouted everything.
Kids, let this be a reminder to never trust devboard footprints off of the internet without physically verifying it!
btz
added to the journal ago
soldered most of the pcb
everything came except for the lcsc parts.
Here's the before photo:

and here's after:


a bunch of issues came up. I realized my caps are polar, so I had to figure out how to orient them right. The quadruple resistor didn't quite fit right, so I had to mess with that. Also, the mt3608 from amazon (more on that later) is... different from the footprint. It's smaller, so I had to invent some funny things out of bent headers to enlarge it.
There's some sus wiring on the pn532 too, since the headers are too tall coming out of the esp.
I had to make some... modifications to the switch legs as well because my homemade footprint was too small.
So I got excited. I turned on battery mode and... magic smoke came out of the mt3608. that's one of the few parts I have spares of, at least. And it looks like these freaking amazon modules have a different pinout. So polarity of vin, vout is spatially flipped. At least I know what to do for the other one...
And I guess I'll have to pray it didn't fry anything else. So fun to trust the end user (me) with power management logic!!
man, charging sucks. This is otherwise an easy module build.
Anyways, tried flashing the firmware and the screen didn't turn on. The backlight's not on either, so it's just not getting power. A classic symptom of my sucky home soldering iron. I honestly should have just been patient and done ts at school tmr. Oh well. c'est la vie.
because guess who (lucky someone) doesn't have a multimeter at home!!
the pn532's power light is on, so at least the gnd+3v3 on there is good. That's not saying much though.
I predict many, many hours of integration hell in my future...
btz
added to the journal ago
Ordered all the parts

aliexpress didn't like hcb cards so I had to retry a few times. For the mt3608, in the end, I used amazon
I also ordered pcbs in green instead of black accidentally, but it's ok.
alexren ⚡🚀
approved Flopper one ago
Tier approved: 4
Grant approved: $43.00
post a video in #blueprint once you build this thing! seems pretty awesome
btz
added to the journal ago
misc pcb improvements

did small things while waiting for review. Added rounded corners (above)
also optimized the routing a little bit, and ended up using vias instead of a really gnarly route. idk
btz
added to the journal ago
Automated blender renders
learned blender scripting for fun. Works pretty well, now I don't have to click ANY guis to update my docs, which is pretty sweet.
stumbled upon the wonderful bpy.ops.pcb2blender.import_pcb3d( filepath=os.path.join(os.path.dirname(bpy.data.filepath), "flopper.pcb3d") )
also edited the materials so the pcb is black and the esp is too.

btz
added to the journal ago
dummy journal entry to change banner
See the title text.
Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat. Duis aute irure dolor in reprehenderit in voluptate velit esse cillum dolore eu fugiat nulla pariatur. Excepteur sint occaecat cupidatat non proident, sunt in culpa qui officia deserunt mollit anim id est laborum.

btz
added to the journal ago
fixed some problems
turns out the display is spi, not i2c, so added some traces. Edited the firmware to be spi. otherwise pretty minor. Also didn't realize that I was allowed to journal while under review, I've made quite a few changes back when it was in prelim review. Like made the battery traces thicker and stuff
Also added a hole in the case for IR.


schematic looks SO COOL now
also added a blender render

🙏 praying for review acceptance
btz
submitted Flopper one for ship review ago
alexren ⚡🚀
requested changes for Flopper one ago
hi there - im returning because its missing 3D model files, you can get them for almost any part from https://grabcad.com/!
btz
submitted Flopper one for ship review ago
btz
added to the journal ago
lowk wrote the firmware, bom, and readme, and redid power
firmware first, mostly a test script getting set up with platformio. It compiles on the esp and throws I2C errors ofc (because no i2c plugged in)
Then I checked slack again and realized that my power supply is wrong, but also discovered this crazy hack where vin is also 5v out. So I redid that and made a switch footprint.

Then I uh made the BOM and wrote the readme and stuff.


btz
added to the journal ago
muck around with the pcb some more
cleaned lots of things up, cut a hole out of the solder mask. I added 3d models for nearly everything (not the screen bc I couldn't find one)
In adding a model to the MT3608, I realized that I had it mirrored, so I had to fix up the footprint.

and uh yeah
hopefully case + firmware + ship soon
btz
added to the journal ago
fixed the battery stuff x2
Thanks to some feedback from some wonderful folks, I realized that I was doing battery charging completely wrong
so I uh fixed it. Basically you need a boost converter, so I got a premade MT3608 module. And it turns out that's pretty big, so suddenly all my free space went away and my footprint doesn't seem so generous anymore.

I also did some more aliexpress research and saved $15 by looking for parts with free shipping
btz
added to the journal ago
mucked around with the pcb

added the hack club logo, thought about routing, asked for feedback in the blueprint slack.
Reduced the number of vias needed in the battery charging rats nest.
PCB is honestly almost ready
btz
added to the journal ago
made battery stuff


researched, added to preliminary bom, and asked on Slack. Looks pretty good
the pcb is getting pretty crowded atp so I guess I'm proud that it's as compact as it is
btz
added to the journal ago
routed the pcb fr

ZERO VIAS USED LETS GO
also it got a lot easier bc I used labels

I also researched parts. The cumulative cost of parts I own is like $8 but shipping from china is $20. US vendors from aliexpress was more expensive. This means that if tier 1 I will have to supplement with my own money, so I hope tier 2.
time to add battery charging I guess
btz
added to the journal ago
learned to use labels

yay!
also I most likely should do the battery stuff before doing any more routing. So now I'll have to research that... But it'll be fun. Unfortunately, unlike the xiao, the esp does not come with a battery charging circuit.
btz
added to the journal ago
routed the switches

I love procrastinating on schoolwork with blueprint
anyways discovered the insane technique of rotating switches for easier routing
the last few traces are a bit circuitous, but making everything on red will make my life a lot easier down the line.
btz
added to the journal ago
started planning placement for routing

uh yeah. Current design is ~80x60 mm which is kind of big but not too bad. Mainly the breakout modules are REALLY big, and there are only two sides (front and back), so I'm constrained in that way. minimum that has to be on the front are the screen and buttons. I couldn't find a smaller footprint with the esp on the back, so the esp's on the front now
btz
added to the journal ago
Researched stuff and routed the pcb


I got this idea a couple of weeks ago, but now I actually got the motivation to do stuff
I already own an esp32. This adds: IR tx/rx, nfc/rfid, screen, buttons. ESP has bluetooth built in
This as the potential to be a really cool project tbh
btz
started Flopper one ago
2/5/2026 8 PM - Researched stuff and routed the pcb


I got this idea a couple of weeks ago, but now I actually got the motivation to do stuff
I already own an esp32. This adds: IR tx/rx, nfc/rfid, screen, buttons. ESP has bluetooth built in
This as the potential to be a really cool project tbh
2/5/2026 9 PM - started planning placement for routing

uh yeah. Current design is ~80x60 mm which is kind of big but not too bad. Mainly the breakout modules are REALLY big, and there are only two sides (front and back), so I'm constrained in that way. minimum that has to be on the front are the screen and buttons. I couldn't find a smaller footprint with the esp on the back, so the esp's on the front now
2/6/2026 - routed the switches

I love procrastinating on schoolwork with blueprint
anyways discovered the insane technique of rotating switches for easier routing
the last few traces are a bit circuitous, but making everything on red will make my life a lot easier down the line.
2/7/2026 8 AM - learned to use labels

yay!
also I most likely should do the battery stuff before doing any more routing. So now I'll have to research that... But it'll be fun. Unfortunately, unlike the xiao, the esp does not come with a battery charging circuit.
2/7/2026 10 AM - routed the pcb fr

ZERO VIAS USED LETS GO
also it got a lot easier bc I used labels

I also researched parts. The cumulative cost of parts I own is like $8 but shipping from china is $20. US vendors from aliexpress was more expensive. This means that if tier 1 I will have to supplement with my own money, so I hope tier 2.
time to add battery charging I guess
2/7/2026 12 PM - made battery stuff


researched, added to preliminary bom, and asked on Slack. Looks pretty good
the pcb is getting pretty crowded atp so I guess I'm proud that it's as compact as it is
2/7/2026 3 PM - mucked around with the pcb

added the hack club logo, thought about routing, asked for feedback in the blueprint slack.
Reduced the number of vias needed in the battery charging rats nest.
PCB is honestly almost ready
2/7/2026 6 PM - fixed the battery stuff x2
Thanks to some feedback from some wonderful folks, I realized that I was doing battery charging completely wrong
so I uh fixed it. Basically you need a boost converter, so I got a premade MT3608 module. And it turns out that's pretty big, so suddenly all my free space went away and my footprint doesn't seem so generous anymore.

I also did some more aliexpress research and saved $15 by looking for parts with free shipping
2/7/2026 7 PM - muck around with the pcb some more
cleaned lots of things up, cut a hole out of the solder mask. I added 3d models for nearly everything (not the screen bc I couldn't find one)
In adding a model to the MT3608, I realized that I had it mirrored, so I had to fix up the footprint.

and uh yeah
hopefully case + firmware + ship soon
2/8/2026 - lowk wrote the firmware, bom, and readme, and redid power
firmware first, mostly a test script getting set up with platformio. It compiles on the esp and throws I2C errors ofc (because no i2c plugged in)
Then I checked slack again and realized that my power supply is wrong, but also discovered this crazy hack where vin is also 5v out. So I redid that and made a switch footprint.

Then I uh made the BOM and wrote the readme and stuff.


2/17/2026 7 AM - fixed some problems
turns out the display is spi, not i2c, so added some traces. Edited the firmware to be spi. otherwise pretty minor. Also didn't realize that I was allowed to journal while under review, I've made quite a few changes back when it was in prelim review. Like made the battery traces thicker and stuff
Also added a hole in the case for IR.


schematic looks SO COOL now
also added a blender render

🙏 praying for review acceptance
2/17/2026 8 AM - dummy journal entry to change banner
See the title text.
Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat. Duis aute irure dolor in reprehenderit in voluptate velit esse cillum dolore eu fugiat nulla pariatur. Excepteur sint occaecat cupidatat non proident, sunt in culpa qui officia deserunt mollit anim id est laborum.

2/20/2026 - Automated blender renders
learned blender scripting for fun. Works pretty well, now I don't have to click ANY guis to update my docs, which is pretty sweet.
stumbled upon the wonderful bpy.ops.pcb2blender.import_pcb3d( filepath=os.path.join(os.path.dirname(bpy.data.filepath), "flopper.pcb3d") )
also edited the materials so the pcb is black and the esp is too.

2/22/2026 - misc pcb improvements

did small things while waiting for review. Added rounded corners (above)
also optimized the routing a little bit, and ended up using vias instead of a really gnarly route. idk
3/8/2026 - Ordered all the parts

aliexpress didn't like hcb cards so I had to retry a few times. For the mt3608, in the end, I used amazon
I also ordered pcbs in green instead of black accidentally, but it's ok.
3/15/2026 - soldered most of the pcb
everything came except for the lcsc parts.
Here's the before photo:

and here's after:


a bunch of issues came up. I realized my caps are polar, so I had to figure out how to orient them right. The quadruple resistor didn't quite fit right, so I had to mess with that. Also, the mt3608 from amazon (more on that later) is... different from the footprint. It's smaller, so I had to invent some funny things out of bent headers to enlarge it.
There's some sus wiring on the pn532 too, since the headers are too tall coming out of the esp.
I had to make some... modifications to the switch legs as well because my homemade footprint was too small.
So I got excited. I turned on battery mode and... magic smoke came out of the mt3608. that's one of the few parts I have spares of, at least. And it looks like these freaking amazon modules have a different pinout. So polarity of vin, vout is spatially flipped. At least I know what to do for the other one...
And I guess I'll have to pray it didn't fry anything else. So fun to trust the end user (me) with power management logic!!
man, charging sucks. This is otherwise an easy module build.
Anyways, tried flashing the firmware and the screen didn't turn on. The backlight's not on either, so it's just not getting power. A classic symptom of my sucky home soldering iron. I honestly should have just been patient and done ts at school tmr. Oh well. c'est la vie.
because guess who (lucky someone) doesn't have a multimeter at home!!
the pn532's power light is on, so at least the gnd+3v3 on there is good. That's not saying much though.
I predict many, many hours of integration hell in my future...
3/18/2026 - More soldering
Yesterday started exciting and ended depressing.
Here was my todo list:
reflow all joints
Wire a new mt3608 bodge with the new module. You'll need to bodge across, swapping + and - on both vin and vout.
Cut the corner pins off of the board. Melt the plastic and cut the metal?
the EN ESP PIN should lead nowhere.
The gp23 ESP PIN should lead to the pad under the VIN pin
The VIN ESP PIN should lead to the pad under the 3v3 pin
The 3v3 ESP PIN should lead to the pad under the en pin.
In kicad-speak:
Traces that currently go to vcc, go to en.
Traces that currently go to gp 23, go to vin.
Traces that currently go to vin, go to 3v3
Change the pins in the header file to match the scam mirrored pinout.
solder in the lcsc parts
cross fingers!
I did all this (except for the header file and lcsc parts).
Also, even with the flipped polarity, my second mt released mystical magical smoke. So I'll put lipo battery charging on hold (and, as it turns out, 3
xAA straight to VIN (through a switch) is gonna work just fine. )
This is how it's looking!

update:
(below not included in time tracking btw)
Through various mishaps, I am no longer able to flash my ESP32 (and even if I did once, it'd probably be unreliable, though I'm open to solutions)
The reason why I started messing with the esp pins and adding bodge wires in the first place was because, despite self-review and blueprint review, the pinout on my devboard footprint was really wrong (not just mirrored, like really really wrong; it's a long story).
I'll probably need a new esp32 and pcb for my project to work cleanly.
And I guess I'll scrap my lipo charging logic because I fried both mt3608s, even when there was no battery. A battery pack will suffice, I guess.
shipping's ofc going to take forever now, and I'll have to wait again.

I've now refootprinted the ESP and removed the charging logic, and rerouted everything.
Kids, let this be a reminder to never trust devboard footprints off of the internet without physically verifying it!