Defeating audio AGC?

Oct 21, 2007 8 Replies

This isn't really a design question as such, but I hope someone will have an idea about it.



I have an Canon Elura 100 miniDV camcorder, which has automatic gain control on audio. I'm sure Canon thought they were doing a good thing with this, but it causes motor noise to be picked up quite noticeably during times when the audio level is low and the gain is ramped up to the maximum.



This model also has an external mic input, but the mic needs to be a "self-powered" electret. So, I built a little amp for that, powered by a 9V battery. Unfortunately, it appears the AGC is still there. And while I don't get motor noise, it appears the amplifiers in the camera aren't exactly low-noise, so when they crank up looking for something to hear, Or it may be noise from my amp that gets amplified, but I don't hear that when I use this setup on the computer mic input.



Anyway, since I've already built this amplifier, I just wondered if there's anything I could add to its output that wouldn't be audible to humans, but would fake out the camera AGC. Or maybe center the output level a bit above ground - well, ok, that's not likely to work assuming the camera inputs go through capacitors.



Well, I don't have my hopes up on this, but just thought I would ask in case there was a solution.



Of course I could record the audio separately on another device, but that's a lot of trouble since the streams would have to be synced at some point.



Thanks for any ideas.



On a sunny day (Sun, 21 Oct 2007 11:20:15 -0500) it happened George wrote in :

Sure, couple of periods of 30kHz once a second MAY perhaps help... It all depends on the circuit.

If you can easily record the audio on another device then you could record them on both and then just replace the audio on the video. It should be very easy to sink up because its just a matter of matching up which would be extremely easy to do with the right software.

Generally, you will be out of luck, the camera audio was probably designed by the RIAA marketing team, even the professional sound guys complain about the really crap audio from cameras in the 20k$ ++ market, and the headphone returns can be somewhat mediocre as well.

Maybe carry something like a H4 (samson?)in your pocket, just for when you need a some better sound, maybe with a cheap lav.

A handclap is a good sub for a clapper board ( in vision of course) The advantage of a person recorder, is that you can do longshots.... and still hear a person talking

Martin

At least 20 years ago an employee of mine at OmniComp/GenRad had a friend who was a sound man for "60 Minutes"

We made a wireless synchronizer to put a sync tone on the audio when the video started.

I'll see if I still have a copy of the schematic.

IIRC we called it "bleep-bloop" ;-)

...Jim Thompson

| James E.Thompson, P.E. | mens | | Analog Innovations, Inc. | et | | Analog/Mixed-Signal ASIC\'s and Discrete Systems | manus | | Phoenix, Arizona Voice:(480)460-2350 | | | E-mail Address at Website Fax:(480)460-2142 | Brass Rat | | http://www.analog-innovations.com | 1962 | America: Land of the Free, Because of the Brave

There used to be a scheme with a tungsten bulb in the gate of a 16mm chemical cam that did the same, fortunately I have forgotten all about it, and don't want to remember.

Loads of people use 433MHz and ancient smpte linear timecode these days

Martin

My expeience trying to do that in the past was not good. Two devices, even if both are operating perfectly, will be operating off time bases which are inevitably just a bit different - because crystals aren't perfect.

So what I found was that, for a long clip, if the A/V was in perfect sync at the beginning, it no longer was at the end.

So then you are looking at, say, stretching the entire audio stream so that it's in sync the whole time. I know there is software to do that, but I think that whole process would get old pretty fast if I had to do it on everything.

Thanks, Martin, but an H4 is like $500, which is about $480 over my budget for this project.

It would have to be something like using my laptop to record the audio, but that wouldn't be very convenient. Plus, my laptop has lousy audio to begin with.

Another option, for motor noise, is to export the audio to an audio editor, say cooledit, do an FFT analysis, and notch out the offending frequencies. My Canon XL1 has a very nasty, but sharp 150Hz buzz when used with the internal mic, and a 30dB notch improves it no end. Tedious but it works

Martin

Join the Discussion

Have something to add? Share your thoughts — no account required.

Didn't find your answer?

Ask the community — no account required