I have two hard drives that draw approximately 1.1A / 1.5A on their 12V and 5V power pins. I have several 2SA1020 PNP transistors which are rated 50V 2A and was hoping to use them to switch power between the two drives - the idea is that when the computer is off, opening or closing a switch would allow the 12V and 5V power sources to flow to just one drive. I considered using a MOSFET but the drives draw only a little power so it seems like these transistors would be ok, right? If not then please let me know what the main pitfalls are.

enter image description here

In my previous days of analog electronics tinkering this would have been a much simpler concept - the first thing that came to my mind was an automotive relay switch because it has a N.O. and N.C. leg and can direct power between the two of them. But I didn't find anything when I was searching for a '2-way mosfet' with NO and NC pins. I'd also just like to see if I can build this using the transistors I already have.

The idea I came up with for this was to have a switch which can ground the bases of the first two transistors used for drive 1 (D1), normally a pullup to 5V circuit, and use an NPN transistor as a NOT logic gate to invert the instruction to the second set of PNP transistors for the second drive (D2).

Here's what I've come up with:



  • Basically, what would be the best / easiest way to accomplish this?
  • The circuit I drew is the best that I can come up with. Will it work OK?
  • If not, what are the caveats I'm missing?
  • And, is there some other way that is much easier / simpler to achieve this?

ps: I'm mostly curious about the 'switching' part so I intentionally omitted additional components that I believe would be good to add in, such as those which would prevent switching if the computer was powered on, etc.

  • 2
    \$\begingroup\$ why would you want this? \$\endgroup\$
    – kolosy
    Commented Jan 16, 2015 at 6:17
  • 1
    \$\begingroup\$ SATA supports hotswap but don't recommended to disconnect the power of the HDD contains your OS, when it is running. \$\endgroup\$
    – GR Tech
    Commented Jan 16, 2015 at 6:23
  • 1
    \$\begingroup\$ The end objective in this case is an inexpensive (using existing resources), headless, dual-boot system, having physically different drives, that allows only one of the drives to be turned on at a time, electronically, on a system that isn't spec'd with enough power for both drives to run at once. Regardless of how ridiculous you may think the actual application, can anyone provide guidance on the actual electrical engineering in question here? I have invested time in doing my 'homework' researching, writing a clear question, explaining what i've tried and creating a diagram to illustrate it. \$\endgroup\$
    – cwd
    Commented Jan 16, 2015 at 7:42
  • 1
    \$\begingroup\$ If you only need to do this drive switching while the system is off, why not just use a DPDT switch on the 5V and 12V rails? \$\endgroup\$ Commented Jan 16, 2015 at 8:34
  • 1
    \$\begingroup\$ I can see a use for this by hiding the switch and use the two HDDs for deception. One with the "clean" system and another one with stuff to hide. The user who knows about the switch can select the hidden drive, while turning it back to the clean when leaving the PC would not allow any other users to find the hidden stuff because the hidden drive is not physically connected anymore. Still, the DPDT switch would be the simplest way. \$\endgroup\$
    – dscharge
    Commented Jan 16, 2015 at 9:28

1 Answer 1


Your circuit concept is starting in the right direction but it requires some important changes for it to work.

You will require series base resistors for all of the PNP transistors. As drawn you will cook them with no base current limitation.

Secondly the NPN transistor is drawn backwards. Its emitter needs to connect to the GND.

Third you need to do some serious work reorganizing the way the bases of the various transistor are biased. You need to separate the 12V and 5V PNP bases so that the 12V one can shut off when the base drive goes away to the 5V one. After this fix it is necessary to provide also a separate pullup of the NPN transistor.

Don't cringe. Resistors are incredibly cheap.

Lastly this schematic, to be bluntly honest with you, is horribly drawn being that it does a bad job of following conventional standards for how a good schematic is drawn.

enter image description here

  • \$\begingroup\$ Thank you for reviewing this. I would love to get better at schematics. Would you have any additional details or links regarding 'conventional standards for how a good schematic is drawn' ? Is there a 'go to' industry reference or some personal favorites? \$\endgroup\$
    – cwd
    Commented Jan 16, 2015 at 21:15
  • 1
    \$\begingroup\$ @cwd higher voltages up, lower voltages down, signal flows from right to left. \$\endgroup\$ Commented Jan 17, 2015 at 22:56
  • 1
    \$\begingroup\$ @cwd here, I'd call this decent at least: imgur.com/F9TiPMz \$\endgroup\$ Commented Jan 17, 2015 at 23:26
  • \$\begingroup\$ and Michael, please see my schematic too, I believe you can omit R2 and R3 from yours. \$\endgroup\$ Commented Jan 17, 2015 at 23:27
  • \$\begingroup\$ @VladimirCravero - thank you so much for the info, example, and the note about R2/R3 \$\endgroup\$
    – cwd
    Commented Jan 18, 2015 at 6:12

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.