clueless1 wrote:
Can someone explain this concept in layman's terms? Why does the soundcard need a -5v rail? What is the difference between positive and negative in this instance?
Concept: 1.5V batteries and speaker cone movement.
Basically, while you can use just a single 1.5V battery and decide if you want to use it to control the speaker cone in or out from the neutral position, then you have to decide which way to connect the battery, but with computer power supplies you can't do this swapping of speaker connections so you have to have both +1.5V and -1.5V "batteries" (voltages) available simultaneously to drive cone both ways away from the center position.
So, take two 1.5V batteries, and wire them in series, so that effectively you have one 3V battery if you look at the whole thing. But if you artificially decide that your reference 0V or ground voltage is the point between the two batteries, you get +1.5V from the other battery, and because what you set as the 0V point is now positive side of the other battery, the negative terminal of that battery is 1.5V below 0V, or -1.5V. Same thing if you have a 3 feet stick, use a pen to draw a line at the center to divide it into two equally length halves, and shove it halfway to ground, it's still a 3 feet stick, but one end is 1.5 feet above ground (positive from the reference level), and the other end 1.5 feet below ground (negative from the reference level).
If you would demonstrate this with a loudspeaker and connect speaker black(gnd) wire to the center point of the two batteries, this enables you to move speaker cone out from the neural position by connecting the speaker red(positive) wire to +1.5V, and in addition, you can also move speaker cone in from the neutral position by connecting speaker red wire it to -1.5V.
So why (older) sound card needs -5V from the ISA bus is to power DAC chips and amplifiers as said by elianda. This basically enables sound chips to drive say speaker cones or whatever the equivalent is in your headphones both inwards and outwards from the center position with positive and negative voltages. But that's actually not the whole truth. Later sound cards did not require external -5V from the ISA bus because they added voltage regulators to make local -5V from -12V that comes from ISA bus, and that was a good thing because the locally generated -5V then had less audible noise from other devices and these sound cards can work with ATX supplies that do not provide -5V. And to be honest, on most sound cards, the speaker amplifier just used +12V supply voltage so their output "reference" neutral voltage was 6V, so the output could still go up by +6V to 12V or down by -6V to 0V, and the 6V DC bias voltage is removed by DC blocking capacitors on amplifier output, so at the connector the voltage can go between -6V and +6V and sits at 0V when there is no audio. But if the main speaker amplifier chip is not taken in to account, and the fact that basically all audio inputs and outputs have DC blocking capacitors, it's definitely better for the normal audio buffering chips to use positive and negative supplies so 0V ground potential can be used as the signal reference voltage. +5V and -5V are enough for the amplifier chips to work with audio signals that are approximately within +2 and -2 volts compared to ground.
So as a joke, the difference beween +5 volts and -5 volts is exactly 10 volts 😀
clueless1 wrote:
I've done a bit of reading on stackexchange on this topic, but it's just not sinking in. 😕
I found this comment quite hilarious at first, because I thought this was an intended pun - for example integrated circuits such as logic ICs or amplifier chips being able to sink current in... Sorry I just had to say that 😀