Single supply as dual with virtual ground

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

flintan

Well-known member
Joined
Feb 22, 2007
Messages
297
Location
Sweden
I,m putting a circuit that wants +-12V in a box that already has 24V available. Would this circuit do ok with a simple resistor voltage divider for a virtual ground? The 24V supply would be floating and this ground connected to the main ground.

Like this:
rdiv.png


And the circuit:
fig4.gif
 
In practice, you can't depend on a passive v/2 to be very solid. If you circuitry draws more current from the +12V as the -12V v/2 will drift around .

I count 7 opamps in your design, so if you have an unused section, that could be repurposed as a v/2 buffer, to make v/2 stable.

JR
 
A precision driver for a Magic Eye?

Why?

How are you going to put 10%, 20%, ...100% marks on the phosphor screen inside the bottle?

This tube: IC2A must pull to the negative rail (LM324) or R116 must be biased a few volts above IC2A's negative rail so that your cathode-drive circuit will force amp-triode to maximum current.

> Would this circuit do ok

Depends entirely on the parts values not shown on your plan. If impedances are in the GigaOhm zone, then sure, 4K7||4K7 is pretty solid. If C106 is taking 35mA peak pulses, then 220||220uFd will damp that, but not to-zero. If C106 jolts kick the V/2 point around and something amplifies that kick (the path can be subtle; I posted an analysis of the Cmoy headphone amp which shows heavy offset in some cases), results may be unexpected.

Hint: your to-"ground" caps C104 C106 may and probably should return to hard-ground, not V/2. With FET input on IC1A, R101 can be a Meg. Then all other V/2 stuff _appears_ to be opamp inputs or the zero pot. Small or constant loading on the V/2 supply. I do agree that, unless you make a million of these, you may just as well buffer the V/2... another <19 cent amp is cheaper (in coffee and brain-pain) than computing circuit sensitivity and leakage.

I assume the Magic Eye is a short cut-off type. Some made for US AM radios have extended cut-off and won't reach maximum deflection until 30V or so on the first grid. 
 
Because it feels and look good? The optical reading won't be as precise so maybe it's overkill..

Here is the parts list.
And the projects page.

Ok so i'd better buffer the v/2. What if i connect this buffered-middle-of-24V to hard-ground? The 24V supply is floating. Would one TL071 survive providing a stable ground to this circuit? I forgot to say it's in stereo so there is actually 14 OPA's, three quad and one dual.

I was planning to use a russian 6E1P (cyrillic letters looks like 6E1n) wich is equivalent to EM80
 
> maybe it's overkill..

I agree.

It's only 15mm of glow.

But don't let that stop you.

This is less linear (in a good way) and much simpler. Probably not cheaper. Unless your time is worth something.
 

Attachments

  • Eye-overkill.gif
    Eye-overkill.gif
    8.1 KB
Oh thanks! I'd like to try your idea but the thing is i had already perf-boarded this other thing before i started the thread. Mybe i'll just build a separate psu for this circuit.

I'm doing this to (hopefully) learn something and because it's a great hobby so time isn't really an issue. But of course time is worth something in the aspect that i want the projects to be finished someday before 2015 so i can use them for recording..

It's 15mm of precious glow :) I used to play with old radios when i was a kid and was specially fascinated by these glowing eyes. So the reason for me to use them now is more aesthetic/emotional but i'd like it to have a decent functionality of course.
 
Back
Top