Recording Problems using Griffin iMic and Audacity

Just bought a Griffin iMic and I’m having serious problems with the volume, I think its an input problem. I use a Windows 7 and a Shure SM58.

I’ve boosted and heightened all microphone and output volumes to no avail.

After looking through topics on the forum here I tried the amplify effect and lowered the gain, but although the volume is increased the audio quality is still very very low.

On previous occasions I have used the SM58 through an Olympus WS 550m dictaphone, then imported the audio file into audacity for editing and had absolutely no problem.

Is there any way of replicating the audio quality that I got with the Olympus through Audacity so I can just use Audacity to edit or is the iMic just not of as good quality as I had previously thought?

Any help would be greatly appreciated.


Giffin iMic, at least the one I had, was nothing to write home about. How are you connecting the SM-58 to the iMic? The two connectors are very different. If you get the adapter wrong, the volume of the microphone will be very low or zero.

Like this…

or this…

If your 1/8" connector has an extra ring on it, this one as an example (ignore the RCAs)

That’s not going to work.


One more. My iMic had a switch where I could configure it for Line or Microphone. Line will give you a very quiet performance.


Thanks for the help man.

The SM-58 is connected by a female XLR to 1/4in jack cable, which in turn is adapted into the iMic by a standard 1/4 in to 1/8 in adapter you see on headphones.

Like this:

But this 1/4in to 1/8in adapter was never a problem when recording into the dictaphone?

The dictaphone will have a mono input, whereas the the iMic has a stereo input. These two devices have the input socket wired up differently.
The SM-58 is designed to be plugged into a 3 pin XLR socket.
It may work if you use a mono 1/4" to 3.5mm adapter (but still not ideal even if it does work).

You see on the left side of your adapter, the plug has a metal tip, a ring, and then the rest of it is sleeve. The adapters you have put half the show on the tip and half on the ring. These adapters…

Are wired to put the whole show on the tip. They have to be specially internally wired to do that. You can build your own like I did, or you can buy one pre-wired. That’s this one…

One of these days I’m going to order one of these and see if it’s wired the same way mine is. I don’t know.


I may be off base here, but isn’t this likely to be a simple impedance matching problem? The Shure SM58 is a dynamic mic - so low impedance (150 ohms IIRC). Looking at the Griffin iMic I’d bet at least 5p that the microphone input is designed to work with an electret microphone. Could you plug in a cheapo desktop microphone (such as the good old logitech wand mic) and see if you get better quality just by using a high impedance mic?

I used an olympus WS-210S and a variety of electret mics for a couple of years to record audio books. The Olympus voice recorders are very, very good, but the recordings are a little noisy. So I don’t know if the iMic will improve things - depends on the digital noise produced by the iMic itself.


Absolutely fantastic help there guys, thanks so much.

… isn’t this likely to be a simple impedance matching problem?

Possibly, but I don’t think so. The SM58 will produce killer voltages at its output when you sing loudly. There are provisions for sticking fixed, resistive attenuators in the mic cable to knock down the signal and avoid overload of the electronics during rock bands. This is not a quiet microphone.

No, it’s more magic than that. The SM58 does not generate it’s signal hot pin to ground like an RCA connector. It generates its signal between two hot pins – XLR 2 and 3. The mixer or electronics is required to “know” what to do with both pins for music to come out.

If you’re missing either pin or manage them badly, the music will fall apart. The adapters in the post obediently deliver both pin 2 and 3 to the iMic. The iMic either sends pin 3 to the computer microphone battery system (in microphone mode) or it tries to share the show between left and right of a stereo show.

In the first case, the signal on pin 3 becomes damaged or missing and because they’re tightly linked, damages the other pin as well. In stereo line mode, it’s seriously magic. The microphone signal becomes as much as a thousand times too quiet and even if you do manage to get it to work by boosting (unlikely), it produces an out-of-phase show. It vanishes on a mono speaker or music player.

Outside of that, it’s fine.


I think I understand what you are saying. I understood how balanced line cabling worked, but some of the subtleties of amplifying signals from balanced line cables had eluded me, so bear with me here.

If I understand this correctly the pins 2 and 3 float relative to ground and act like a twisted pair in telephone lines. Pin 1 is connected to ground and a shielding mesh or sleeve around the outside of the signal pair. The actual signal is taken from the difference between the floating cables, so that the EMI can be filtered out - it will be picked up equally by the two floating lines and so is ignored by the differential amplifier.


So amplifying the signal between pin 1 and either pin 2 or pin 3 will effectively be amplifying the EMI (sort of, plus DC offset). Amplifying the signal between pin 1 and the sum of pins 2 and 3 will amplify the EMI and perhaps any phase difference between the signal pins (not really I know, but some weird undefined variation relative to ground). Grounding either pin 2 or pin3 so you can amplify the signal between them will destroy the balanced line operation and may introduce scads of noise.

So is the only real (high quality) answer a mixer or an XLR to USB adapter like the Blue Icicle or similar?


Quite correct - “common mode rejection”.

To connect a 3-pin output (Shure SM58) to a MONO input, pins 1 needs to be linked to pin 3. This looses the benefit of common mode rejection, but it means that pin 3 is tied to ground and not floating, so you will get the proper signal voltage across 2 and 3.