[RockysDad]

What if I got a +24VDC wall wart going to a two resistor voltage divider (ie. CMoy) with virtual ground then taking each +/- 12VDC voltage to a positive and negative 9VDC regulator going to my Jung multiloop amp. Would this work? Has it been done before? Just wondering....

 

[PeterR]

Like this?

 

[RockysDad]

Yep!

I'm not sure if the virtual ground buffer is required but "by George, I think you've got it".

So what do you think?

 

[PeterR]

I think it would work just fine.
The question is whether to use that wallwart at all? Building a proper dual PS from scratch isn't that much effort either...

 

[jeffreyj]

Quote:

Originally posted by RockysDad Yep! I'm not sure if the virtual ground buffer is required but "by George, I think you've got it". So what do you think?

If you ditch the buffer then you'll have to make the voltage divider "stiff" enough to accomodate the difference in current drawn from each rail at any given point in time. Generally, "stiff enough" = one-tenth the equivalent load impedance. If your load draws 50mA from a 12v rail (assuming the drain from the opposite rail is zero, which would be worst case), the equivalent impedance would be 240 ohms. Your voltage divider resistors would have to be 48 ohms max (the center-tap impedance is the two divider resistances in parallel). The divider will draw 250mA from a 24V input at all times?!

The whole point of using a buffer is to eliminate this staggering quiescent current draw, while still delivering an even lower equivalent impedance to the virtual ground rail (usually under 2 ohms!). Your divider can be composed of 10k resistors, for example, reducing quiescent current drain to 1.2mA. This is, perhaps, not so important if you are using a wall wart power supply, but the buffer still delivers a much lower impedance. To compete with the buffer's output impedance would require divider resistors of 4 ohms. The quiescent current draw is now 3 amps, but the output impedance is now 2 ohms. Obviously, the power supply capable of delivering this current would cost far more than adding a buffer to the ground rail, so there you go.

edited a stupid math mistake...

 

[Budgie]

jeffreyj- wow, man- excellent explanation! clear and so easy to understand, I even got it. I hope you post alot more around here!
Thanks!

 

[jeffreyj]

Thanks, Budgie. I pick up lots of interesting tidbits from forums like this so I try to return the favor when I can.