I have a simple audio circuit which when I power from a power pack (5V DC) works fine. If I connect to a USB power supply (literally only supply + & - connected going to the same supply inputs) all hell breaks loose audio wise.
The circuit is isolated from the soundcard bias in both cases with a 470nF electrolytic cap.
Is there a simple way to clean this supply up or is it simpler to use a battery supply and keep it "clean" from the USB supply? To my simple mind, a USB DC supply is just +-5V DC but I guwss I'm mistaken.

Sorry for being a tad simple in advance - I thought I had a nice clean audio until now.

Here's the sound... (no pic cos I couldn't be bothered to plug in a camera).
Sound Example

 

It could be that the power from the USB cannot supply enough current.
What current are you trying to draw?

 

It could be that the power from the USB cannot supply enough current.
What current are you trying to draw?

it seems to be drawing about 70mA (but my multimeter isn't that trust worthy)

 

Even USB 2.0 covers that IF it's working properly...

 

Are you running from a laptop USB?
If so, try it without the mains power supply connected.

 

It's a mains powered PC (I want to use it for long periods so even when it goes into a laptop (there's a whole raft of other issues relating to USB hubs & USB sound cards) the laptop will be plugged in.

I might try and run my circuit with a battery charged via the USB - failing that I can add a colar panel to keep the battery charged (& keep it totally isolated from USB power)

 

I had trouble hearing the audio on the youtube clip, but it sounded like it was dropping out completely. Is this accurate?

Where is the 470nF electrolytic cap you mentioned? Is it in series with the audio line, to provide DC isolation? Or is it across the power supply rails?

You said your meter indicated the amp was drawing 70 mA, but that it may not be an accurate reading. Accurate or not, your meter is probably measuring the average current. If the amp or the USB power source is shutting down, then coming back up in a cyclical fashion, your meter would not indicate that.

Finally, I believe the USB power delivery system is not as simple as just providing a 0.5A, 2A, 2.5A, etc DC power source. In order to make use of the higher power delivery modes, I think there is some sort of negotiation that goes on between the source and the load before the higher current is delivered.

 

I had trouble hearing the audio on the youtube clip, but it sounded like it was dropping out completely. Is this accurate?

Where is the 470nF electrolytic cap you mentioned? Is it in series with the audio line, to provide DC isolation? Or is it across the power supply rails?

Could you hear the 1st part? (ie the seperate power supply). The audio ALMOST drops out completely but behind the noise you can just about hear me talking (just not what I was saying) The 470nF is on the audio output (in series) to knock out DC bias from the soundcard. I also have a smaller cap (330uF) from mid point of volt divider (going to the opamp) and to ground to get rid of a bit of noise). I'll stick up a schematic later today - for now I'm just running it on a 9V battery (the circuit is happy between 5 and 9V supply). It should last ages since I remeasured the current draw on the supply - I was a factor of 10 out - it's actually only drawing 8mA.

 

I did have to call for warranty repair on my fairly new tower when the USB failed to provide power. Serial communications were fine but would not power an Arduino on one of the ports. Also, there are many consumer products on the market now that provide a USB connector solely for power connection for battery charging, etc.

 

Also, there are many consumer products on the market now that provide a USB connector solely for power connection for battery charging, etc.

I might try one. I had hoped to power it from a USB hub within the project (since that's part of it's purpose). Part of me is now thinking that the battery option for this part of the circuit might be the better option anyway since it keeps preamp stage isolated from the other part of the project (the USB side) with the exception of the line out (which I have isolated by means of the 470nF cap). One bonus - using a 9v battery, I've got rid of a lot of clipping I get with the 5v supply to the preamp. It also makes it a bit more versatile since the preamp stage takes it to line level so I also have a little portable preamp compatible with a PA system if I take a switchable audio out at that point on the circuit.
I have to say, until I started playing around & chased down some of the issues, I was totally oblivious to the about of noise on a USB cable (not even connected to the PC) . Apart from the 50Hz (being in the UK) there's a lot of background high frequency within that wave - that'll be the downside to all the gadgets we all make
Here's the real irony - until I sort out a rechargable battery/solar panel it's first proper test tonight with a disposable battery is at a climate change meeting