Electromagnetic Interference or otherwise

[Tim Sills]

Hi All,

Have recently returned from a long holiday with my HV20 + Azden SMX-10 external mic.

Well, the video is stunning - can't dispute it.

The audio is fantastic, at least up until something went wrong.....

In the last 4 of my 23 tapes the audio system has picked up something like high frequency EMI, even at locations as far from a mobile tower as you could possibly imagine on remote islands in Scotland.

The interesting part? Well, this interference (or whatever it is) is recorded, at its own leisure, even when the external microphone is switched OFF. Hmmm...

All I can say is that when it occurs, it occurs for the whole length of the event that is being recorded, and may not be there for the next. And, it never comes into force during a recording - it's there for the whole event of not at all.

I have called Canon Australia and asking some queries but nobody seems to have a clue.

Anybody else heard/know about this?

If necessary I can post in an ac3 or wmv file (etc) with the sound.

Regards,

Tim

[Eugenia Loli-Queru]

Hmm, it could be interference, but might also be your camera just going bananas... Not sure how you could distinguish that, other than doing long recordings away from that place.

[Tim Sills]

Good suggestion - may just be true !

This problem occured in places as remote as the small islands of Scotland, through to places like the middle of the Singapore Zoo. It did not seem to occur with any degree of correlation with such things as temperature, humidity, time of day, or location near EM environments such as motors and mobile towers etc.

It's like the camera just got grumpy with me and something shorted out. It's a hit and miss affair - when I press record, it may be there, it may not. It's also not dependent on how long the camcorder has been used beforehand or what type of tape I am using.

I suspect something is awry within the beast - perhaps it's as simple as a loose screw?

Thanks,

Tim

[Eugenia Loli-Queru]

I had a problem with my G4 powerbook with the hdd doing weird noises every time I would visit the mountains. It just didn't like the altitude I guess, while it was working fine on sea level. Might be a similar case with your camera, not liking a particular place for some weather condition or another...

[Tim Sills]

I must admit, altitude is one of those things I can scrub since all of the problems occured pretty much at sea level, but I hear where you are coming from.

This camera even managed to pick up the paging of my mobile phone when it was searching for new towers - you know that dit-dada-dit-dada-dit sound like you hear from car speakers when driving around with a mobile phone? I thought that was most peculiar !

[Don Palomaki]

Post a short .WAV file (avoids compression artifacts that the ac3 format might add).

Any correlation with using an A/C power supply rather than battery only? Any other accessories attached?

[Tim Sills]

Hi Don,

I use Vegas Movie Studio Platinum and traditionally make ac3 files then convert to wav for file editing. This I have already done for example clips with the interference and the noise is identical to my ear whether in ac3 of wav format, so I might start by posting in ac3 format, but I hear what you are saying about formats.

This only happened with battery use and the only accessory (apart from a lens filter) is the Azden microphone.

Tim

[Tim Sills]

Ok - here we are. An attachment with the noise in wav form (which sounds to my ears absolutely identical to the orginally rendered ac3 form).

The amplitude seems to have come down somewhat with conversion from ac3 to wav so you may need to turn it up. You know it's at the correct volume however when you can imagine it almost dominates normal speech, etc.

Any thoughts?

Tim

[Chris Soucy]

Reckon there's a couple of things to try to see what, if anything, eliminates this, or at least fingers the culprit.

1.Try the mic on another camera. Plays up - it's the mic, doesn't play up, looks like camera.

2.Try the camera with another mic. Plays up - it's the camera, doesn't play up, looks like the mic.

3. Simply disconnect mic and use the internals. Plays up - it's the camera. Doesn't play up - it's the mic (probably).


My best guess?

Some exotic connection fault involving the mic lead shield, allowing the cable/ mic O/P circuits to act as an aerial/ rectifier/ coil/ capacitor all in one, thus feeding nicely recieved RF straight into the mic pre - amps. Depending on the circuitry in the mic circuits, it needn't actually be switched on to act in this fashion under fault conditions.

Of course, the fault could lie with the camera internals, which can only be proved by having another (external) mic create similar results.


Good luck, think you're going to need it.


CS

[Tim Sills]

I must admit, I didn't quite think of things the same way. But, as you say Chris, if there is a microphone fault, even if it's off, it still feeds into the camera and hey presto. (Just confirm that I understand you correctly though on that matter).

Perhaps I will check the warranty on the Azden mic before sending the HV20 into Canon since the turnaround time, especially at this time of year, could be rather large...

Many thanks.

[Chris Soucy]

Yep, I think you got me right.

If the mic is connected (whether on or off), if it is the sort of esoteric event I can envisage, it can introduce all sorts of horrors.

The bit BUT here is, just where is the fault?

If you unplug the external and go internal, the fault could still be there in the camera but not visible because of no external aerial feed to give it a signal.

If it is, indeed, the mic, well, it MIGHT show up on something else, but cannot be guaranteed, as other equipment may well have better shieding to prevent such occurances.

In short, hmm, you really need access to more gear to completely eliminate culprets (and, of course, everyone has access to three mics, two cameras etc etc etc.....).

Shall give this some more thought and get back to you - there must be a way to nail this SOB!.


CS

[Don Palomaki]

The noise in the .WAV file is about 55 dB below max record level, and appears to be all harmonics of 1100 Hz, up to ~23100 Hz (the limit of 48 kHz audio sampling) with 6600 Hz and 8800 Hz being about the strongest. What is interesting is that the noise "spikes" continue above 16 kHz, while the noise floor goes silent.

It is very regular, and thus can be greatly reduced by using noise filtering in an audio editing application such as Audition. This is good because it will help recover the effected footage.

Did you say that the .WAV file was created from an ac3 file? Could you share the full work flow you used to prepare the audio file? including the steps used starting with the camcorder record modes, how the audio was captures, etc.

Does the Azden mic need to be connected to get the noise?

Can you hear it in head phones when recording? or only when playing the tape in the camcorder? Or is it only in the capture video in your NLE.

[Tim Sills]

Hi All,

I'll first talk in response to the post from Chris:

I have recorded about one and a half hours of footage (in 1hour per tape mode) using the internal mic only and have heard nothing untoward. I have used it indoors and outdoors, under incoming flight paths, loud environments, quiet environments, etc. The odds are that if something were to go wrong by now, it would have since this one and a half hours I speak is pretty much straight after my holidays were the last four tapes were "corrupted".

Now, I decided to antagonise the system. Inside my home, without the external microphone, I placed my mobile phone next to the HV20 and rang from the (wireless) home phone. Not an audio problem at all (but the screen did shown signs of interference), even when I moved the mobile all around the HV20 up close when it was ringing. Not a problem when I moved the wireless phone all around the HV20 when it was calling out to the mobile.

When the external microphone was connected, and get this, with or WITHOUT a battery, switched on or off, under the same experimental circumstances, the HV20 always recorded the mobile paging sound before the mobile actually started ringing. And, on the audio level meter, it was close to maxing out even thought I have the little green line set under the -12db mark.

The microphone frequency response is up to 20000Hz, but mobiles work in the MHz (I think). Questions all round now. Has to be something simple Chris - keep the ideas coming !!


In response to the post from Don:

Thanks for chiming in on this one Don. As mentioned however, on the original tape, it sounds alot stronger than the wav files gives it credit for. The file that was posted was from a section where the external microphone was OFF (so as to avoid anything else in the track), but if you would like one with the microphone on to hear the ratio of interference to desired audio that's easily done as well.

I must be honest, I have learned to use WavePad to filter this out (and motor noise) using spectral subtraction, but the sacrifice is reduced vocal fidelity and so forth. Still, far better than the original, however my ears are open to suggestions that may be better.

The creation process is this: capture as m2t file using Vegas Movie Studio Platinum. Render out for DVD Architect which makes an mpg for video and ac3 for audio. Using Ciler's AC3 tool to convert to wav without any of the compression or volume changes, etc (or at least as best I know how based on the GUI), save and post to the forum.

The camcorder record mode was 16-bit audio, HDV, P mode and a few image changes like reduced sharpness, etc.

Yes, the external microphone must be connected for the problem to occur.

I must admit, I have not tried using headphones while recording, ever, so I can't answer that specific question. Upon playback, the noise does however come from both the HV20 speaker and is captured by Movie Studio Platinum so yes also out of the PC speakers.

Thanks for all the help - I really appreciate it !!!

Tim

[Don Palomaki]

I agree with Chris. If it only happens with the Azden is connected, that is the likely source. A cold solder joint somewhere is probably acting as a rectifier/demodulator and allowing an ambient RF signal to be picked-up as audio. The fact that it matters not if the mic is turned on indicates that it may be in the final output stage of the mic, or possibly the input stage of the HV20 connected to the mic jack. (The built-in mic may use a different preamp stage.)

If you can, try a different external mic and see whether or not you still have the problem. You could probably "fake" a different mic for test purposes by connecting a set of headphones to the mic input jack. Typically not very good as a mic, but plenty of "antenna" to pick up stray signals around you.

[Tim Sills]

Thanks Don,

I'll have a go at this if I can source some good headphones.

The mic. has been returned to the supplier under warranty and these good people will be "testing" the item. The sound clip has also been sent to the manufacturer in Japan for analysis.

Will keep you posted.

Tim