Finally got my wrist computer wearable working after months of tinkering

Oct 11, 2026 Last reply: 17 hours ago 4 Replies
Jump to replies (4)

Spent the better part of six months on this thing and it finally works well enugh to actually wear around the house. It's a forearm-mounted unit with a flip-up screen, a full keyboard underneath, and a little round dial on the side that shows the time. Printed the housing myself over about four failed attempts before the hinges stopped snapping off.



The keyboard was the worst part. I wired each row by hand and my first revision had two columns swapped, so typing anything felt like a bad dream. Rewired it last weekend and now I can actually type notes on it without wanting to throw it across the room. Right now it pulls the moon phase and some satellite pass times, plus a clock face on the dial. Battery life is the next problem to solve since it dies in about three hours. Not exactly practical yet but it turns heads at the coffee shop, which honestly was half the point.


Six months rings true, mine took closer to eight before I stopped being embarrassed by it. Mine was a belt-clip unit with a little OLED and a chunky keypad, and I also fought the printed hinges for ages before switching to printing them in PETG with more perimeters. The row-column wiring thing got me too, except mine was a shifted row so everything came out one key to the left. Took me a week to spot it because I kept blaming my code instead of my soldering. On the battery, that three hour figure sounds familiar. My first setup ran the screen at full brightness all the time and I was swapping packs constantly. What actually helped was letting the screen sleep after ten seconds and only waking it on a button press, plus dropping the radio to check data every half hour instead of every minute. Those two changes togehter roughly tripled my runtime without me touching the hardware at all.

I also moved to a single 18650 with a small boost board instead of the flat pack I started with, mostly because spare cells were sitting in a drawer alredy. If your dial display is always on, that might be your biggest drain, so try giving it a sleep schedule too. Curious what screen and micro you ended up with, mine was an ESP32 on a homemade carrier board and the idle draw still bugs me.

I skipped the hand wiring on mine and went with a membrane keypad pulled from an old cash register keypad board I had in a drawer. Soldered it straight to the matrix pins and it typed right the first time. Feels less satisfying under the fingers than what you built, but after a previous project where I spent three evenings tracing a shorted row, I had no pride left to protect. On the battery side, I gave up on a bare cell and just strapped a small power bank to the strap with velcro. Ugly, adds real weight, but mine runs a full day now instead of dying by lunch. Yours pulling only three hours suggests something is staying awake when it should sleep, mine idled terribly until I got the display to power down between inputs.

For the hinge problem, I never did solve it in plastic. I scavenged the tiny hinge off a dead laptop screen and recessed it into the printed housing. It has survived about a year of daily flipping, which is longer than any printed version lasted me. The coffee shop turning heads part is real. Mine gets comments at the hardware store, which is its own kind of audience.

For the battery hunt, the thing that actually got me somewhere was a cheap inline USB power meter I bought for checking phone chargers. Sat it between the pack and the board, watched the draw while toggling stuff off one at a time, and found my always-on indicator LED was quietly eating half my budget. Ten minutes of measuring beat a week of guessing. Yours being forearm mounted, weight matters as much as runtime, so it might be worth seeing which parts earn thier milliamps before adding a bigger cell.

That dial is the first thing I'd look at. On my wrist unit I got sick of watching the main pack drain and swapped the clock face for an actual quartz watch movement scavenged out of a dead wristwatch, mounted right in the printed recess. Runs on a coin cell for close to a year and it never touches your main battery. Old habit from building prop gauges back in the day, mechanical solves what silicon won't. Also, if you wired the keyboard by hand you probly have more copper in there than you need. I shaved a full layer off my harness by rerouting the rows along the forearm instead of over the top, cut the bulk and made it sit flatter too.

Join the Discussion

Have something to add? Share your thoughts — no account required.

Didn't find your answer?

Ask the community — no account required