• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

What is -5VDC and -12VDC used for

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

WhitehawkEQ

Premium Member
Joined
Dec 6, 2010
I want to know what -5VDC and -12VDC is used for on todays PC MB's, I know what they were used for 30-40 years ago but technology has changed and inquiring minds want to know :)




MB=Motherboard
 
-5V isn't part of the ATX spec anymore, it was for ISA slots.
-12V is used for serial ports, not sure if anything else.
 
That's.... bizarre. If you look at the PSU plug you'll find an empty hole where -5V used to be.
 
But it's still listed on the MB's. Look in the manual for your MB and see what is listed :)

It may be listed, but I will bet my last dime it isn't used ;)

-5V hasn't been part of the spec for years and years now. -12V should have been dropped with it, if you ask me.
 
Some old chips used -5V or -12V to isolate their individual transistors from one another, by applying it to the overall chunk of silicon, I think, while newer chips do this by generating one of those voltages internally.

While RS-232 serial ports use about +12V and -12V, few motherboards made since the late 1990s have depended on the power supply for those voltages but have instead generated them from the +3.3V or +5V. I had an old motherboard with only ISA and PCI slots that used an Analog Devices charge pump chip to generate +10V and -10V from +5V, and when somebody plugged in a parallel printer while the power was on, the tiny ceramic capacitors in the charge pumps would explode.
 
-12V was also once used for audio opamps, but they're pretty much all single rail nowadays.

Interesting trivia: In the early days, audio stuff pretty much always used +-12V, then to +-5V, then to single rail 5V, and then to single rail 3.3V. PCs haven't used any lower voltages for analog audio (since there's no real point shaving off a few mW at most), but various mobile devices have went to 2.5V, 1.8V, and even lower rails. In order to get rid of the (relatively) big and expensive coupling caps, they often use charge pumps inside the amplifier chips to generate a negative rail. Headphone impedance has also been going down in order to accommodate lower voltages. (In the past, 32 ohms was the norm, now 16 ohms is the norm.)
 
Some old chips used -5V or -12V to isolate their individual transistors from one another, by applying it to the overall chunk of silicon, I think, while newer chips do this by generating one of those voltages internally.

Do you have a link to anything with more info on that? It's a fascinating concept and I want to read more about it now. Never heard of that before.
 
Some old chips used -5V or -12V to isolate their individual transistors from one another, by applying it to the overall chunk of silicon, I think, while newer chips do this by generating one of those voltages internally.
Do you have a link to anything with more info on that? It's a fascinating concept and I want to read more about it now. Never heard of that before.

A Wiki article about reverse biasing the substrate:

http://en.wikipedia.org/wiki/P–n_junction_isolation

The alternative is to put each transistor inside an insulator layer of silicon oxide:

http://en.wikipedia.org/wiki/Silicon_on_insulator

I think Intel still uses the reverse bias method for its CPUs, while AMD switched to using Silicon On Insulator (SOI) a couple of chip generations ago, apparently because it couldn't build chips that were as dense or fast as Intel's otherwise. I read somewhere that SOI adds $1,000 to the cost of each wafer.
 
Back