Using comp as PSU
Aug 10, 2004 at 8:16 PM Post #16 of 20
A recharging circuit is definitely possible, but you'd need several poles on your power switch. In normal operation, ground is halfway between the battery voltage rails, so V+ is 6V above computer ground, and V- is 6V below. To recharge, you have to disconnect both rails from the amp, then reconnect them to your charging circuit, since V- is computer ground, and V+ is at most 12V above ground.
 
Aug 11, 2004 at 12:50 AM Post #17 of 20
Quote:

There's also people that claim you can hear the difference between different chemistries of batteries.


Why doubt it? There's solid science behind this one, unlike, say, wire differences. The impedance of alkalines is much higher than it is with rechargeables. High supply impedance equal worse sound. End of story, no argument possible.

Well, you can argue about the extent to which this is relevant, but anyway...

Quote:

but you have to isolate the virtual ground from any metal part of the case


The others are right, guzzler. Virtual ground only works when the power supply is isolated, so that the virtual ground can "float" to be equal to the source's ground. In a "CMoy-in-a-PC", audio ground from the sound card will be equal to PSU ground, but vground in the amp is trying to be +6V above PSU ground and the PSU isn't playing along.

Quote:

PRR had an AC coupled CMoy over at Headwize which might be of interest


Now that will indeed fix the problem, because the circuit PRR used doesn't have a virtual ground; it simply biases the input signal up to halfway between the power rails, while the op-amp runs single-ended. Calling it a PRR amp is inaccurate, but at least it doesn't tie the discussion to CMoy's design, which is quite different.

Quote:

A recharging circuit is definitely possible, but you'd need several poles on your power switch.


A DPDT is sufficient, if you can live with turning the amp off to charge the batteries. The advantage of this method is that the dirty wall power never enters the amp except when charging the battery, and the amp is turned off during this time. One pole of the switch runs the charger, the other the amp, and the poles are configured so that only one is on at a time.

I did this in my measurement preamp. See the second page of the schematic. Steal my circuit!
smily_headphones1.gif
 
Aug 11, 2004 at 9:30 AM Post #18 of 20
i agree with tangent about keeping the batteries as separate from other power sources as possible.

The PSU isn't the only problem.

Computer GND is shared with digital devices,
Fans generate a lot of electrical noise on their power rails,
RFI is huge inside a computer (but manageable if wiring is done properly and some basic design considerations are adhered to).
CDrom drives and HDD and anythng else adds noise to the power rails.

THe battery idea is awesom but if that doesn't work at the very worse whack a dc jack on the amp and use a wallwart connector.
If you REALLY want to use the PSU in the computer there's a -12V rail going into the motherboard which can be tapped, and in theory then the GNDs will all be at the same potential albeit a bit noisy. If you use this do as much filtering as possible.
 
Aug 11, 2004 at 1:26 PM Post #19 of 20
Quote:

Originally Posted by stereth
Worst case, toss an 8xAA holder in the bay. It'll last for about a week and a half of constant use.


For that matter, eight C cells would prolly fit. It would run for a very long time then.
The noise thing is pretty real. Even with batteries, I'd build it in a decent metal box for shielding, and have the controls come through a faceplate (versus just floating a PCB behind a faceplate). The batteries could be outside the little metal box still.

Maybe build the whole thing in an aluminum drive carrier. Slide it out to swap batteries.
 
Aug 11, 2004 at 7:22 PM Post #20 of 20
I was facing exactly this problem when making my DAC and CMOY in a 5.25" drive bay.

If you want to use the computer PSU, there is not a virtuall ground issue, you can use the +12V and -12V lines, or even the +5V and -5V, so your virtual ground is now real ground. (to do this you need to draw power from the ATX motherboard power connector, but if you can make a CMOY you can do that!)

However after much deliberation I chose not to do that, since I decided the possibility of noise on the power supply was quite high. Instead I used a high quality regulated wall wart. I made a screened cable that goes from my DAC/AMP to a socket on the back. I put the socket in a card slot cover so I can move it from one computer to another if I like.

With my set up only the ground of the amplifier is connected to the ground of the computer (and the rest of my house) so there is the theretical possibility of a noisey earth. I hear no power supply noise, even at very high gain. The DAC I used only has 96dB SNR, so I can hear the DAC noise when it is turned up very high, it is absolutly unchanging white noise. Hard drive activity, fan speed changes, cd drives produce no audible noise on the output what so ever.

So you may consider this solution tried and tested.

Luke
 

Users who are viewing this thread

Back
Top