Tried everything to stop videopro mic from hissing

Started by 60dmicrophone, April 15, 2016, 01:18:41 AM

Previous topic - Next topic

0 Members and 1 Guest are viewing this topic.

60dmicrophone

Hello,

I have just recently installed ML on my canon 60d in an attempt to turn AGC off so i wouldn't have to deal with the annoying hissing sound that comes with it. This was done after I visited any forum I could find that dealt with this hissing issue. Before installing ML, I tried setting the audio to manual, lowering the gain all the way and then clicking it up 2 notches, and then turning on the 20db boost on the rode mic. This lowered the hiss but it was still very audible. This led me to installing ML and turning the AGC off and turning on the 20db boost on the rode mic, which still produced the hissing sound. I am all out of solutions and am looking for any advice at all. I can provide test footage if you'd like so you can hear the hissing yourself. Thanks.

simonm

First off, this probably has nothing to do with Magic Lantern. What follows is a long explanation because you're not asking a question with a simple answer!

I've tried to skip as much technicality as I can, but you can't escape it really. To get the best from audio, just like cameras and lenses, you do need to know stuff, and the knowledge needs to be correct, not folklore or marketing puff from a manufacturer or reseller.

. . .

The specifications for the Rode mic are a bit arcane and not very encouraging. Depending on which way it's calculated (power or voltage, and TBH Rode don't make this clear), it could be outputting as little as 0.5 mV/Pa.

Contrast this with a Sennheiser 416, which apparently gives 25mV/Pa. Both numbers come from manufacturers' web sites, although I roughly converted the Rode number to be in the same units. Mic manufacturers never make comparisons easy!

I'd be surprised if the Rode was anything other than noisy in comparison. There's a reason one costs around 5x the other. The basic approach is rather different: the Sennheiser uses RF resonance, and has a much lower noise system (right at the front end of the circuitry) than conventional designs, electret or otherwise. So, when you amplify it to the levels needed for a good recording, there's less noise to start with that you're making louder. It's not clear to me how the Rode one works, but it's probably not the same way, otherwise they'd be shouting about it.*

The noise you get on the recording comes mainly from two places: the analogue preamps in the camera, and the analogue preamps in the microphone system. The Sennheiser mic's preamp is probably MUCH quieter than the Rode, AND it has a much higher output (if my sums are right!) - it's a win-win, which overall means the recording will have a lot less noise. The less you have to wind-up the camera's gain, the less noise you get from the camera's preamps on the recording, generally speaking. Less gain needed for the Sennheiser (probably). It's in part what you pay the big money for!

. . .

That said, if you want to find out exactly what is going on with your own kit, you need to do some careful and systematic tests:

First off, you need to be certain your camera's audio circuitry isn't faulty (very occasionally it happens - it's analogue and analogue things do go faulty/noisy). Here's how to check it, and for both tests you need to use MANUAL level control, not automatic.

Testing the camera's audio inputs

Do this first, to eliminate the camera from your enquiries, otherwise you may be going round in circles.

You need access to a proper audio mixer, something like a decent CD player and some classical music (chamber music is good), and an output lead from the mixer to the camera with something like a 60dB pad in it (attenuating the audio from high professional line level down to the sort of signal you'd get from a microphone in practical use for, say drama dialogue).

The 'pad' can be made from three cheap resistors (per stereo side) costing literally a few cents each. You can go directly from CD RCA "jacks" to your camera. If you omit the audio mixer, the pad needs to be about 50dB instead (CD to camera), and you have no actual check on the signal level you're sending. Ideally, get an audio pro to help you.

The idea is to have known-good, clean audio, at mic level, available for the camera. You might get the CD to work into the camera without a pad at all (by turning the camera's gain right down), but this will not help you find mic-level problems. You need a resistive pad (a.k.a attenuator), to drop the CD/mixer level down to mic level for the camera. I've seen some cable suppliers charging really silly money for cables with pads in (mentioning no names). Honestly - find a keen teenager with a soldering iron and get the plugs/jacks and resistors from eBay or Radio Shack! It doesn't need to look pretty; it does need to work properly and for only as long as you need it to.**

Set your camera levels manually, so it's peaking properly on the meters. Record a bit. Play it back in proper listening conditions - HiFi headphones or speakers and a quiet room with few electrical fans or aircons to confuse matters. If it sounds unduly noisy compared to the original CD, there may be a problem. If not, you know your camera is OK.

Roughly comparing one mic's noise to a known reference

You'll need something ancient and simple like a Shure SM58 (with a suitable lead), and your Rode (set to no attenuation nor gain). Set up both mics across a tabletop in a quiet location, importantly at as close to the same distance away as you can manage. Record one at a time. Talk at them, normal voice, not shouting nor whispering, and in each case adjust the camera gain so they peak properly. Read a piece of text (poem or something from a newspaper perhaps), and leave a few pauses too. Listen back to the two recordings.

I would expect the Rode to have much less noise than the SM58, because the latter is quite insensitive and will make the camera's mic inputs (the mic preamps) work harder. I would hope the Rode's internal preamps would be quiet, and thus you need a lot less gain at the camera (overall less noisy).

The SM58 recording should be obviously noisier (more hissy). If it turns out that the SM58 recording has obviously less noise than the Rode, then there is probably a problem with the Rode's circuitry.

. . .

These little gun mics for video aren't magic bullets (see wot I did there?). Rode aren't different from the rest in this regard, as far as I can tell. All gun mics, including really pricey ones like Sennheiser, work by cutting out sound from directions you don't want. They cannot make the wanted sounds any better than any other type of microphone will, they just help to get rid of the unwanted stuff at source.

You need to understand this when you use them. The tricks you see on TV cop shows (CSI used to be serial offenders in this!), where they point a gun mic across the road to hear a quiet conversation between two villains, are pure fantasy. It cannot be done with any gun mic I've ever seen.

Likewise, you won't pick up quiet dialogue from a camera position with a gun mic, unless you're in a proper, soundproofed film or TV studio, and even then it won't be good. There's a good reason film recordists use poles or even Fisher booms (years ago I used to drive one on live TV!). The idea is to get the mic as close to the action as practically possible, so that the sound isn't noisy. The "gun" thing is about cutting out the unwanted sound at that point, not making the wanted stuff better (it can't do that, as I said).

Hope that explains a bit.

Simonm.

PS: You *can* make wanted sounds (at distance) louder, with a parabolic reflector, but these are bulky, and difficult to set up and use, and I can't imagine why anyone would bother for normal drama and documentary work. Here's an example: http://www.wildtronics.com/parabolic.html. The bigger the dish, the louder the sound and the more bass it has, too. Great for wildlife, some applications for sports coverage (I saw some being manually tracked on the TV coverage of the French Tennis Open last year), but otherwise a right pain in the bottom to use (I own an old one - I know!).

*The Rode is most probably Electret, which is a permanently-charged capacitor system and quite different to Sennheiser's approach, but I welcome corrections on this and I'm trying not to be down on Rode as such. They do make a bit of fuss about having a "JFET impedance converter" but FET preamps typically go with Electret capsules (which are awkward because natively they're really high impedance). I wouldn't put any store by the reviews though. Rode don't publish enough information to know how well it's likely to work in any given situation, and none of the reviews seem to be objective (or to involve measurement equipment!).

**You can calculate the resistor values - there are Javascript tools on the web for this that make it easy (Google), but you need to know or guess the source and load impedances. For something like a professional mixer or straight from the back of the CD player, assuming a source value of 150 Ohms and a load (camera input) impedance of about 10,000 Ohms (10k Ohms) should get you in the right ballpark.

60dmicrophone

Wow Simonm! Thank you so much for the incredibly thoughtful reply. It is truly appreciated, and I am going to try what you suggested. I just have to gather the supplies first! Thank you again.

simonm

Happy to (try to) help! Please do report back :)

It'll be really interesting to know where the problem lies.

Have fun, and be systematic with the fault-finding. You should find the reason and probably the solution too.

Cheers,

Simonm.