Products > Other Equipment & Products

Is there a DC-blocking cable I can buy for a soundcard microphone port?

(1/2) > >>

I am looking at getting a cheap USB soundcard dongle, and it has only 2 ports, microphone-in, and audio-out. It has no line-in. I know that microphone ports have a DC voltage (from what I remember when I had another soundcard in another computer, this was about 2V, this is NOT 40V phantom power) coupled onto the signal line with a pull-up resistor on the sourd-board itself, designed to power an electret microphone. When I buy this USB soundcard dongle, I really don't want to actually cut a tracing on the board to disable the DC offset on the microphone signal line, as I will want to be able to use it for both a mic and an un-powered audio source (depending on what I need at any given moment). Therefore, I much rather have a DC-blocking audio cable. I know I can make one with either an RC highpass filter, or using a 1-to-1 isolation transformer. I actually made one once, a long time ago for a different soundcard I had in a different computer that I'd had at the time, but I don't remember where I put it, so I think it's lost for now. Also it wasn't exactly professional-manufactured quality. The soldered connections on the audio isolation transformer were covered over with electrical tape, clearly something that somebody just hacked together.

I want to buy (if one exists) a professionally made DC-blocking audio cable, designed with the intent of basically turning a mic-in port into a line-in port, so that non-powered audio sources can get their audio signal into a soundcard via its mic-in port. Ideally the cable will contain the isolation transformer or the RC lowpass filter directly in part of the cable itself (such as an end connector, or a tiny box in the middle of the cable that's professionally molded into the cable itself so it's not an obvious added-on part).

And one more question, not specifically related to the main topic of this thread. The 2 possible low-price USB soundcard dongles I'm considering are these:
This one from Sabrent
This one from Plugable
I don't know which is better though. I've heard of Sabrent before, but I don't remember if it was in the context of being a good quality manufacturer. or in the context of it being a maker of cheap Chinese junk. I've never heard of Plugable at all though, and it sounds so generic of a name that it makes me think they might be some Chinese company that just makes junk that would break if you so much as sneezed. So does anybody here know which is actually the better brand?

In general, a microphone input is very sensitive, perhaps as little as 1 mV.  A line input is more like a few hundred millivolts.  So all you need is an attenuator, which will drop the signal and the dc to miniscule values.

Yes, there are pullups often on computer mic inputs and if you design your attenuator with too much impedance, they may shift the DC level of the signal above the range of the ADC. If you drive the input with low impedance it may not be a big deal.

And there is a bigger problem. Those ultra cheap USB dongles generate a virtual ground at ~2V and reference all their audio signals to that. If you short the virtual ground to some grounded equipment you will damage the USB dongle. I'm not 100% sure if there isn't built-in AC coupling for the input signal, though, including the input port's ground being connected to USB ground.

I'm not aware of such cables existing, I think you would need to experiment with a DIY build tailored to a particular dongle at hand.

The ground difference wouldn't matter if an isolation transformer was used to couple the signal source to the soundcard.

By the way an attenuator between the signal source and the soundcard would only reduce the signal strength from the source reaching the soundcard, not the DC voltage level form the mic-port of the soundcard going back to the equipment and possibly damaging it. For that you need either an isolation transformer between the dongle and the signal source or else an RC lowpass filter configured to block DC from the dongle from reaching the signal source.

Also there's no technical reason for the audio ground floating on some DC level. I just used MS-paint to diagram a USB soundcard circuit that has USB ground being the same as audio ground. I've attached the circuit diagram to this post.

The components are the following:
R1 and R2 form a voltage divider to set up the virtual ground for the op-amp. These should be selected to be low enough to keep the virtual ground's voltage fairly constant but not so low as to draw excessive current from the USB port.
R3 and R4 form a voltage divider to put the voltage onto the mic-in port's signal line to power the pre-amp of an electret mic.
R5 is needed to isolate the hard virtual ground at the opamp's + pin from the floating signal line that goes into the amplifier feedback loop's first resistor.
R6 and R7 are used to set the gain of the opamp.
C1 is the AC coupling capacitor.
ADC is a 16bit analog to digital converter with 16 digital output pins (the 16wire data bus), one pin for each bit (parallel output).
USB Interface IC is a USB audio class interface IC with 16 digital input pins to get the digital audio signal from the ADC, its output serializes the data and formats it for the USB specification so it can interface with a PC.

Note that other components are NOT included in this simplified diagram, such as the 48kHz clock oscillator which would generate the sampling clock signal that would also be fed to the USB Interface IC for sampling the digital data it's getting from the ADC.

I think you vastly overestimate complexity of those cheap dongles.

A single series capacitor should solve your problem (getting rid of mic bias DC), you don't even need a resistor to ground afterwards but you need to anticipate a few kΩ input resistance in the soundcard and size the capacitor for that.

Virtual ground only becomes a problem if you try to bypass the built-in AC coupling to get DC operation.


[0] Message Index

[#] Next page

There was an error while thanking
Go to full version