Reach Out And Touch Your Computer

cpu

[Carter Yagemann] found himself in a bit of a pickle. He uses his computer mostly for gaming, but would like to access it remotely from time to time to do a littler server work on the side. The problem is gaming computers eat up a lot of electrons and he didn’t want to waste them by leaving it on all the time. The obvious solution was to use the Wake on Lan function. Unfortunately, his motherboard did not support this technology.

Like any good hacker would do, [Carter] used an IoT board to connect the power button of his PC to the internet. He achieved this goal with a Particle Core board. His motherboard was an ATX variety, so wiring up two of the IoT board’s I/O pins to the power on pins on the motherboard was a simple task accomplished with the help of an inline resistor.

This hack is so easy that it’s a great alternative to the blinky LED first program we all know so well. Want to get started in the hacking community? This is a great way to get going.

22 thoughts on “Reach Out And Touch Your Computer

  1. I’ve always thought it would be nifty to make a quick firmware for the ESP8266 that does this on the cheap. You could even sell them for a buck over what you paid and make a tidy sum from people that couldn’t be bothered to flash them (or lack the kit).

      1. I agree with @gudenau. If you use this to access your computer remotely, then Remote Desktop and shut it down. An accidental back back back in your browser can send the command again which will trigger a shutdown while you are still busy.

        1. If you can switch the computer back on via the remote, then what’s the actual problem? Ok, you loose a few minutes waiting for the computer to finish booting. Well, that’s you punishment for accidentally sending a back back back. :)

  2. Did this a while ago with a digilent PIC32 Arduino-pinout board to power on a motherboard with similarly sketchy WOL, except it was over wired Ethernet for a computer that was usually going to be on the other side of the country and inaccessible 99% of the time. I recall having some issues with the power button on this particular computer, and ended up having an input that would reflect the state of the LED as well (occasionally, even when you did it by hand, the computer would turn on and immediately go into standby or something silly and flash the power LED instead of booting).

    Also probably a good idea to use an optocoupler rather than directly sending a logic signal back. I don’t think there’s any spec on which pin is supposed to be positive, and it just gives a bit more robustness against any possible ground loops, etc.

    I agree – deliberately coded against any hard reset capabilities because I didn’t totally trust the Microchip TCP/IP SSH stack, and didn’t want someone trashing my sessions on the outside world in case they ever manged to find a hole in the stack firmware. Worst case, someone could waste power and turn on my server when I wanted to leave it off, and monitor when it turned on and off.

  3. Seems a little silly doesn’t it? Can’t WOL be added with the addition of a sub $40 NIC?

    Are there really MBs that don’t support this out of the box these days? Perhaps lower end MBs? I can’t recall a single computer that I’ve built in the last 15 plus years that didn’t have that feature.

  4. A project like this without a doubt raises questions regarding usability verses security. I sacrificed security in exchange for usability because I’m not too concern about a remote attacker. If I were, for example, running a website on my server for mass consumption, that’d be another story.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.