Conversation

Notices

1. Embed this notice
   Ryan Castellucci (they/them) :nonbinary_flag: (ryanc@infosec.exchange)'s status on Thursday, 23-Jul-2026 05:05:08 JST Ryan Castellucci (they/them) :nonbinary_flag:

   I got one of my fan controllers (I need two) put together.

   Its purpose is to connect to a motherboard PWM fan socket and then control up to three separate PWM fans, with I2C for configuration, control, and telemetry, along with serial for debugging.

   Have not yet powered it on, anyone want to tell me what I messed up before I let the magic smoke out?

   I also still need to write the code for it, lol.

   The unpopulated resistor footprints at the top are I2C pull-ups, which I shouldn't need, but can add if I do.

   • Embed this notice
     Ryan Castellucci (they/them) :nonbinary_flag: (ryanc@infosec.exchange)'s status on Thursday, 23-Jul-2026 05:11:18 JST Ryan Castellucci (they/them) :nonbinary_flag:

     in reply to

     • Ryan

     @fullywoolly It's probably just flux, but I haven't check with a multimeter.

   • Embed this notice
     Ryan (fullywoolly@mastodon.social)'s status on Thursday, 23-Jul-2026 05:11:19 JST Ryan

     in reply to

     @ryanc the secondary side is a little blurry looks like you might have some solder briding between 3v3 and RX?

   • Embed this notice
     Simon Zerafa (simonzerafa@infosec.exchange)'s status on Thursday, 23-Jul-2026 05:12:33 JST Simon Zerafa

     in reply to

     @ryanc

     You've reminded me about this project I need to work on!

     I have one in the pile of projects to attend too, somewhere 😉🤷♂️

     https://github.com/SasaKaranovic/OpenFanController

   • Embed this notice
     Fritz Adalis (fritzadalis@infosec.exchange)'s status on Thursday, 23-Jul-2026 06:11:52 JST Fritz Adalis

     in reply to

     @ryanc
     This is so cool. (Fans. Ha!) So are you able to do i2c over the fan connector to the mb?

   • Embed this notice
     Ryan Castellucci (they/them) :nonbinary_flag: (ryanc@infosec.exchange)'s status on Thursday, 23-Jul-2026 06:59:06 JST Ryan Castellucci (they/them) :nonbinary_flag:

     in reply to

     • Fritz Adalis

     @FritzAdalis The motherboard interface is ground, 12V (or 5V sometimes), PWM speed control, and tachometer for monitoring RPM. My board connects to that, and then has a separate connector for I2C. On the Radxa Rock 5 ITX boards I built it for, I can hook into the I2C bus via the touch panel connector. Alternatively, I have serial broken out.

   • Embed this notice
     Fritz Adalis (fritzadalis@infosec.exchange)'s status on Thursday, 23-Jul-2026 07:08:33 JST Fritz Adalis

     in reply to

     @ryanc
     Oh, I was thinking it was just for a PC, I need to broaden my horizons. Still neat though.

   • Embed this notice
     Ryan Castellucci (they/them) :nonbinary_flag: (ryanc@infosec.exchange)'s status on Thursday, 23-Jul-2026 07:15:55 JST Ryan Castellucci (they/them) :nonbinary_flag:

     in reply to

     • Fritz Adalis

     @FritzAdalis Nearly all PCs have at least one I2C bus. :-)

   • Embed this notice
     Ryan Castellucci (they/them) :nonbinary_flag: (ryanc@infosec.exchange)'s status on Thursday, 23-Jul-2026 07:16:07 JST Ryan Castellucci (they/them) :nonbinary_flag:

     in reply to

     • Fritz Adalis

     @FritzAdalis Nearly all PCs have at least one I2C bus. :-)

     The trick is finding a way to access it.

   • Embed this notice
     muddle 🥣 (muddle@infosec.exchange)'s status on Thursday, 23-Jul-2026 12:16:51 JST muddle 🥣

     in reply to

     @ryanc is it something more subtle like current flowing back from the fans due to inertia after main power is removed?

     Beyond seeing an microcontroller and reading the silkscreen, I'm really at a loss as to understand all the other components/tracings.

   • Embed this notice
     Ryan Castellucci (they/them) :nonbinary_flag: (ryanc@infosec.exchange)'s status on Thursday, 23-Jul-2026 14:50:14 JST Ryan Castellucci (they/them) :nonbinary_flag:

     in reply to

     • muddle 🥣

     @muddle Each of the fans has a big MOSFET to allow it to be powered on/off, a small MOSFET to drive the PWM, a diode to protect against voltage on the tachometer, and current limiting resistors to prevent the microcontroller from trying to source too much current.

     Similar circuitry is in place (except for the big MOSFET) but reversed to connect the microcontroller to the motherboard.

     To the right of the microcontroller is a 3.3V linear voltage regulator (normally, a linear voltage regulator to go from 12V to 3.3V would be a bad idea, but the circuit itself only needs to draw maybe a dozen milliamps, so :thisisfine: ).

     The component with a V on it is a self-resetting fuse.

   • Embed this notice
     Ryan Castellucci (they/them) :nonbinary_flag: (ryanc@infosec.exchange)'s status on Thursday, 23-Jul-2026 15:30:34 JST Ryan Castellucci (they/them) :nonbinary_flag:

     in reply to

     • muddle 🥣

     @muddle also, this particular design uses low side switching for the fans - to turn them off, the ground is cut.

   • Embed this notice
     Ryan Castellucci (they/them) :nonbinary_flag: (ryanc@infosec.exchange)'s status on Friday, 24-Jul-2026 02:07:27 JST Ryan Castellucci (they/them) :nonbinary_flag:

     in reply to

     • Simon Zerafa

     @simonzerafa I've seen those sorts of things, and they're great, but the difference here is I actually want to pretend to be a PWM fan with a tach to the motherboard. I also plan to make a single port version for e.g. control of fans in network gear where one might not be able to add software for USB control.