开云体育

Hey all -- I have a project I'm working on that involves a raspberry pi, a , and a small electret capsule () powered from that interface via PiP.

Everything works, but there is 60Hz mains hum.

I believe the the audio interface is providing a stereo 1/8" jack, but it mixes to mono internally - the computer is only provided a mono capture by the audio interface. (If I use a different set of stereo PiP mics that I made which are wired to a stereo plug, there is no hum, and both capsules hear audio in the result, but the overall signal is weak.)

The mic capsule is wired to a stereo plug (using shielded lav cable), but the signal only goes to one channel of that plug, so I presume the disconnected channel is just feeding in hum, is mixed with the signal in the interface, and that explains the hum in the result.

Does my diagnosis sound correct? If so, what would you recommend? Should I just wire the signal from the mic to both the tip and sleeve of the plug, or will that mess up the interaction with the capsule somehow? Or would it be smarter to get a mono plug?

Thanks!

-Casey

开云体育

What about grounding the unused channel?

开云体育

Thanks Jerry and Scott! -- when I tie the unused wire from the plug to ground as suggested, the mic goes silent - meaning totally dead with no signal at all being recorded.

So, currently the live channel is going to tip, and ring is disconnected, resulting in hum, but joining ring to ground it goes dead, so I assume a mono plug won't help.

I have crudely tried joining signal to ring (in addition to tip) but it didn't seem to make a difference. I also crudely tried a 10k resistor between ring and ground (with no signal connection to ring) and it also made little difference (maybe like 2dB less hum, but didn't really help.) But I did both of these things in a hack way; just holding the exposed wires (using a plastic-handled tool so my hands weren't pinching the joint directly). I can try a more "real" version if it would seem useful.

Any ideas on a next move?

Some other notes in case they reveal that I am doing something dumb:

- mic cable is two conductors inside a braided copper shield; I use both of them joined to carry the signal from the mic (since I don't need both)

- mic is not shielded besides its factory metal enclosure, which AFAIK is adequate for PiP mics

- joint between mic cable and the pigtail leading to the plug (maybe 10mm) is not totally shielded: maybe this is my issue? Similarly, where the cable joins the capsule is not fully shielded (maybe 6mm). I had assumed these spots (which are small) might be irrelevant, but maybe I was naive.

- I mention again in case it's a clue: a different stereo PiP mic (one mic wired to tip, one to ring) works without hum, though the signal level seemed poor

- no idea what voltage the PiP supply is but I could measure if useful

Thanks for any further thoughts!

-c

Could be that the mic power gets shorted out.? You could try a big electrolytic cap instead, shorting the hum to ground but not the DC.

Or just connect the mic to both inputs.

I have tried to get rid of this hum many times, both with ultrasonic mic or normal audio. It is the raspberry Pi and the cheap sound card that will make it impossible. There are two solutions.

Or just connect the mic to both inputs.

Get a mono plug. It'safer.

How is the capsule shielded? Cover the capsule with an empty can. If the hum is gone - you have improperly or insufficiently shielded capsule.

powered from that interface via PiP.

?

When I connect ring to ground (which I think is what a mono plug will do, right?) the mic goes totally dead.

开云体育

Thanks Chris and Jerry:

Jerry wrote:

It seems to indicate that the capsule body is not properly grounded.
In 3-wire connection, there is no need to shield teh capsule, as long as the body is connected to teh shield. Check your connections.

Chris wrote:

It is not stereo, it is standard PC microphone wiring of a TRS jack, sleeve is shield + signal and power return, tip is signal, and ring provided power for a capsule with built-in FET.

If I use my separate stereo PiP mic, which is wired one capsule to tip and one to ring, both mics work in this interface (meaning, their input shows up in the mono output)... does that contradict anything that you said above? E.g. it seems that ring isn't just for power but is actually receiving a separate signal (which is mixed to mono before going to the computer).

Also, currently the mic is just wired signal-to-tip, shield-to-sleeve, so ring isn't connected to anything, and it works (with hum), implying that the bias isn't only on ring?

However, most of the information I could find about PC standard microphone connections indicated that both the resistor and the DC-blocking capacitor are typically in the audio interface, in which case I don't see why connecting tip and ring together at any point would not work.

I may have been unclear: connecting tip and ring does work, it just didn't eliminate any hum so it didn't solve the issue.

I am not familiar with"PiP," does it match that connection described above?

I admit being mystified at the myriad ways there seem to be for 1/8" jacks to be wired, so I don't know if it matches or not, but PiP is just "plug in power" referring to lower voltage supplies on (usually 1/8") mic inputs.

I see that there is another response which came in while I was typing which indicates that you did add better shielding using conductive tape.

Yeah that seems to solve it completely.

-c

开云体育

Thanks Chris and Jerry:

Jerry wrote:

It seems to indicate that the capsule body is not properly grounded.
In 3-wire connection, there is no need to shield teh capsule, as long as the body is connected to teh shield. Check your connections.

It's a two-wire capsule () -- AFAIK the ground terminal is connected to the capsule body in that capsule. I have continuity between capsule body and plug sleeve. It just seems like the connection between the cable and the capsule (maybe 5 or 6mm) where the wires are exposed is enough to admit hum?

Chris wrote:

It is not stereo, it is standard PC microphone wiring of a TRS jack, sleeve is shield + signal and power return, tip is signal, and ring provided power for a capsule with built-in FET.

If I use my separate stereo PiP mic, which is wired one capsule to tip and one to ring, both mics work in this interface (meaning, their input shows up in the mono output)... does that contradict anything that you said above? E.g. it seems that ring isn't just for power but is actually receiving a separate signal (which is mixed to mono before going to the computer).

Also, currently the mic is just wired signal-to-tip, shield-to-sleeve, so ring isn't connected to anything, and it works (with hum), implying that the bias isn't only on ring?

However, most of the information I could find about PC standard microphone connections indicated that both the resistor and the DC-blocking capacitor are typically in the audio interface, in which case I don't see why connecting tip and ring together at any point would not work.

I may have been unclear: connecting tip and ring does work, it just didn't eliminate any hum so it didn't solve the issue.

I am not familiar with"PiP," does it match that connection described above?

I admit being mystified at the myriad ways there seem to be for 1/8" jacks to be wired, so I don't know if it matches or not, but PiP is just "plug in power" referring to lower voltage supplies on (usually 1/8") mic inputs.

I see that there is another response which came in while I was typing which indicates that you did add better shielding using conductive tape.

Yeah that seems to solve it completely.

-c

开云体育

It's a two-wire capsule () -- AFAIK the ground terminal is connected to the capsule body in that capsule. I have continuity between capsule body and plug sleeve. It just seems like the connection between the cable and the capsule (maybe 5 or 6mm) where the wires are exposed is enough to admit hum?

Not in normal circumstances. Are you sure that the body in which you put the capsule is properly grounded?

No: there is no body -- well, no metal body; it's inside a bakelite antique wall-mounted telephone mouthpiece. I was under the impression that a 2-wire PiP capsule acts as its own shield so I didn't need shielding, but apparently I'm wrong and the join from cable to capsule does need to be shielded?

-c

开云体育

It's a two-wire capsule () -- AFAIK the ground terminal is connected to the capsule body in that capsule. I have continuity between capsule body and plug sleeve. It just seems like the connection between the cable and the capsule (maybe 5 or 6mm) where the wires are exposed is enough to admit hum?

Not in normal circumstances. Are you sure that the body in which you put the capsule is properly grounded?

No: there is no body -- well, no metal body; it's inside a bakelite antique wall-mounted telephone mouthpiece. I was under the impression that a 2-wire PiP capsule acts as its own shield so I didn't need shielding, but apparently I'm wrong and the join from cable to capsule does need to be shielded?

-c

开云体育

It's a two-wire capsule () -- AFAIK the ground terminal is connected to the capsule body in that capsule. I have continuity between capsule body and plug sleeve. It just seems like the connection between the cable and the capsule (maybe 5 or 6mm) where the wires are exposed is enough to admit hum?

Not in normal circumstances. Are you sure that the body in which you put the capsule is properly grounded?

No: there is no body -- well, no metal body; it's inside a bakelite antique wall-mounted telephone mouthpiece. I was under the impression that a 2-wire PiP capsule acts as its own shield so I didn't need shielding, but apparently I'm wrong and the join from cable to capsule does need to be shielded?

Yes indeed.

Got it, thanks for the education.

There is a splice near the capsule, now shielded, and a splice near the plug which is not shielded. The second splice does not cause hum -- is that because it is further from the capsule?

-c

开云体育

It's a two-wire capsule () -- AFAIK the ground terminal is connected to the capsule body in that capsule. I have continuity between capsule body and plug sleeve. It just seems like the connection between the cable and the capsule (maybe 5 or 6mm) where the wires are exposed is enough to admit hum?

Not in normal circumstances. Are you sure that the body in which you put the capsule is properly grounded?

No: there is no body -- well, no metal body; it's inside a bakelite antique wall-mounted telephone mouthpiece. I was under the impression that a 2-wire PiP capsule acts as its own shield so I didn't need shielding, but apparently I'm wrong and the join from cable to capsule does need to be shielded?

Yes indeed.

Got it, thanks for the education.

There is a splice near the capsule, now shielded, and a splice near the plug which is not shielded. The second splice does not cause hum -- is that because it is further from the capsule?

-c