Yes! I messed that one up π«£
05.02.2026 08:09 β π 1 π 0 π¬ 0 π 0@gregdavill.bsky.social
Electronics engineer. RISC-V ASICs at IQonIC Works π§ββοΈ (He/Him) I also take macro photos of electronics
Yes! I messed that one up π«£
05.02.2026 08:09 β π 1 π 0 π¬ 0 π 0QFN silicon chip sitting upside down on a ruler, showing that it is 2mm across.
QFN silicon chip held in tweezers.
Small USB PHY from WCH. CH123C.
Who needs symmetry in footprints π€ͺ
Circuit board with QC sticker pointing to an oscillator which has been installed incorrectly.
Oscillators don't like begin installed upside-down π
23.01.2026 08:50 β π 16 π 0 π¬ 5 π 0Green light emitting diodes on a circuit board
LEDs βοΈ
These are about all on at about the perfect brightness for visual indication.
But my standard macro photography process (off camera flash) just floods so much extra light that these appear really dim.
Saleae cables connected to a PMOD
Debug time
17.01.2026 03:03 β π 15 π 0 π¬ 1 π 0Circuit board. 0402 capacitors. A bodge wire connected neatly between two vias.
Another day. Another bodge β¨
14.01.2026 07:42 β π 28 π 2 π¬ 3 π 0Not if I want 400kHz.
The internal pull-ups are equivalent to 50kohm in this case. π
Resistors soldered to the edge of a TSSOP package, with a bodge wire connected to them.
It has been 0 days since I forgot i2c pullups. π₯²
12.01.2026 00:56 β π 56 π 2 π¬ 5 π 2solderpaste applied to various footprints on a green circuit board.
Some BGA footprints are distinct enough to act like a fingerprint for the part. If you've used this part on a board design before you'll likely recognise it. π
06.01.2026 21:46 β π 13 π 0 π¬ 2 π 0GC-10. Sure, Iβm not planning on testing alternatives given how well it works for me.
06.01.2026 06:24 β π 1 π 0 π¬ 0 π 0solder paste applied to a green circuit board
solder paste applied to a green circuit board
Solderpaste π―
06.01.2026 04:02 β π 53 π 3 π¬ 3 π 0I wondered if anyone would notice. Yes. The stock fans donβt actually push that much air, and theyβre very noisy when this gear is right on the desktop.
I need to whip up a small inline 24V to 12V adapter to make the conversion a bit neater.
Benchtop test gear with GPIB adapters installed
Benchtop test equipment
Now all my benchtop gear that had GPIB can now connect via USB π
"Weekend" project complete!
3d printed enclosures around a circuit board and connector
Printed off some enclosures. Made a small modification around the M3 nut tolerances, which was easy as the f3d file was provided in the repo π
26.12.2025 23:50 β π 11 π 0 π¬ 2 π 0AVR programmer attached to a circuit board
AVR programmer attached to a circuit board
It's not stupid if it works π
Avoiding soldering down a connector for the one time operation of burning fuses, and loading bootloader/firmware for some UsbGpib adapters.
github.com/xyphro/UsbGpib
Are they still doing this? I had bought the same part maybe 5 years ago and had the same slightly bent pin issue π₯²π
22.12.2025 07:06 β π 1 π 0 π¬ 1 π 0A linux client spite coding session in the near future? π π
18.12.2025 06:13 β π 1 π 0 π¬ 1 π 0Circuit board macro photo showing capacitors, resistors, and other components ontop of solder paste ready for reflow
Surface mounted components atop solder paste
16.11.2025 01:44 β π 29 π 1 π¬ 0 π 0Board manufacturer date of wk10 2024. Wild that this issue is still making it into designs.
28.09.2025 23:08 β π 2 π 0 π¬ 0 π 0Circuit board with bodge wire neatly snaking between passives. Connecting two vias.
Terminal output from nextpnr showing an error that T16 is not available on this package.
nextpnr output showing device usage around 15% of the logic for a HS USB core and QSPI logic to implement a boot loader.
I'd forgotten that the ECP5-25F doesn't have all the I/O, and 4 or so pins that are on the 45/85F are N/C on the 25F. I just happened to use one of these for the ULPI interface. π₯²
Bodge wire to the rescue! Now we're enumerating again!
USB boot loader in gateware, takes up ~15% of the device
If by fab house, you mean my a toaster oven in my fabulous houses garage? Then yes. π
01.09.2025 22:41 β π 6 π 0 π¬ 0 π 0Assembled circuit board with soldering defect. A tombstoned part. In which a 2 terminal passive is only soldered to one side so it raises up like a tombstone. A small yellow sticker with an arrow points at the defect on the board.
Tombstone time
01.09.2025 22:34 β π 12 π 0 π¬ 2 π 0Small update to PCB and I forgot to buy a new stencil π
01.09.2025 22:09 β π 2 π 0 π¬ 0 π 0Circuit board with a large BGA footprint. Some pads on the BGA package are larger than others.
Circuit board with a large BGA footprint. All the pads on the BGA package are the same size.
These two boards have the same BGA geometry defined in their gerbers. For some reason JLCs DFM adjustments have expanded all the "via in pad" connections.
Makes for an interesting visual effect.
Circuit board with solder paste applied. The photo is close enough so you can just start to make out the tiny spheres that make up solder paste. There are 2 components which don't have solder paste applied.
Circuit board with solder paste applied. The photo is close enough so you can just start to make out the tiny spheres that make up solder paste. There are 2 components which don't have solder paste applied.
How's my solder paste application?
I think I missed some spots. π«£
When I started using brand name AMTECH flux. I realised it had a βsweetβ smell, then immediately bought a fume extractor, because I realised if I was smelling that I was 100% inhaling the rosin and additive vapours π¨π΅βπ«
01.09.2025 01:46 β π 2 π 0 π¬ 0 π 0ECPBreaker FPGA development board featuring a Lattice ECP5 FGPA. Being programmed via JTAG from an ORBTrace, that is powered by a Lattice ECP5.
FPGAs programming FPGAs π’
23.08.2025 02:53 β π 20 π 1 π¬ 0 π 0Saleae trace with QSPI signals shown
The bodge works!
A DFF in front of the ECP5's FPGA SPI configuration port before a mix leading to FLASH and PSRAM. Enabling an initial jump through FLASH where the clock frequency is bumped and QSPI entered.
This now fixes an issue violating the PSRAM's tCE<8us requirement.