What causes background hiss
 

I have a general question. When I record audio using an Azden wireless system, or with shotgun mics such as the rode NT-2, I always get some level of background hiss. With the shotgun mics, this happens both when I record through a pre-amp to the camera, and when I record to a separate digital audio recorder.

I am wondering what the causes are of the hiss (which, when there is sufficient audio, with effects, music, etc. in the background it is not noticeable but definitely IS noticeable if you need just that track of audio), and what I might be able to do to diminish it.

There are a lot of possible sources for hiss but based upon where and how you experiencing it, the most likely scenario is that the recording and mixing devices you are using have low quality pre-amps.

Other possible issues could be lousy mics (Azden, frankly is not a good mic system) but your Rode is pretty decent quality so I doubt it is the mic if you are hearing when using either of these mics.

To diminish hiss, you need to track down it's source by process of elimination. Is the hiss coming from a particular mic? If the answer is no, you have eliminated that variable. Is it coming from a particular recording device? Is it coming from your mixer? I can't tell from your post if you are even using a mixer.

The other thing to consider is that all recording devices have hiss and noise, but the better quality ones have hiss at a much lower level, there is such a thing as a "noise floor". If you put on headphones and plug them into almost any recorder, even very high quality ones, you will hear some noise/hiss. But if it the level you are hearing it at is messing up your projects, you need to use the process of deductive reasoning and eliminate variables.

Dan

Background hiss is a form of audio noise.

The hiss is usually generated by the internal components of the amplification process itself.

The more a circuit designer is constrained by the need to keep costs low, the more temptation there is to use components that add small amounts of noise at each stage of the audio handling process.

Typically, the quality of a circuit is judged by it's s/n or signal to noise ratio.

In a circuit that's passing a strong signal, a poor to moderate S/N may be acceptable.

For example, if a narrator with a strong voice is speaking into a microphone in a quiet room - the SIGNAL is big, so a modest amount of noise is likely going to go unnoticed.

Take that same microphone, put it outside and try to record a birdsong, and you'll find that the small amplitude of that birdsong requires you to amplify the signal tremendously. And any noise that the signal components add get amplified right along with the sound of the bird.

So like most noise issues in audio, you face tradeoffs. Mics with very low self-noise that can be "turned up" greately without much background noise tend to be expensive because they use high quality components and careful electrical and mechanical design.

Low priced gear like Azden, is much more prone to noise, but if used properly with strong signals and limited transmission distance, they can work just fine.

Like with a lot of pro gear, the point of spending more money on gear is not *necessarily* to get better performance in the sweet spot of operation, but rather to invest in the capability to get better results when conditions ARE NOT ideal.

So for your trained narrator, any of a dozen mics might work just fine.

For that interview with the business woman with the quiet, whispery voice, it's comforting to know that when you have to add a lot of gain, you're turning up the sound of her voice rather than the sound of the mic's electronics.

Hope that helps.

I see that the S/N ratio of a mic is measured in dB. What exactly should I be looking for? I've seen very expensive mics with S/N ratios of 88dB and 18dB. Could you shed some light on this?

Generally, yes. But be warned. There's no SINGLE numerical value that tells you the difference between how two mics perform relative to each other.

Self noise - which is one element of signal to noise ratio - is just one aspect to how a microphone circuitry operates. Low self noise GENERALLY means you can apply more gain (volume) to that mic's ouput signal without the noise generated by the mics circuitry becoming a problem.

HOWEVER, there are plenty of other aspects to mic design that might be MORE important in a particular use.

For example, if I was going out to record something loud like an auto race or a rock band, I'd look for a mic with the capacity to handle a high SPL (sound pressure level) without damage and I wouldn't care a jot about S/N since I'd know that the amplitude of the signals I'd be recording would BURY any minor self noise. On the other end of the spectrum, if I was recording something quiet, I'd look for high SENSITIVITY.

Accuracy is another element of a mics design. Extreme accuracy as you find with so-called "measurement mics" are great to take a pink noise reading of a room's accoustics, but sound pretty dismal when you use them to record music.

Mic A might enhance mid bass - which could help strengthen a weak male voice - while mic B might have a presence peak at the high end which might make a female's voice cut through a music bed better - OR might end up making her voice sound shrill.

These are just a few examples of why there's never been a single microphone made that does well for everything.

I'm sitting 5 feet from a $1200 studio VO mic. It's a wonderful piece of German Engineering. But the idea that I'd take it out on a field video ENG shoot is laughable. In a breeze the same exceptional sensitivity that makes it great for use in my purpose-built acoustic foam lined VO room with precision low airflow AC/Ducting would be expected to get me a crappy location recording since every breath of that breeze would be expected to set it's sensitive diaphram rumbling like thunder.

Sorry, but Sound isn't easy.

It's as simple as that.

Bill,

Thanks for all of the info. I certainly didn't mean to imply that the S/N ratio is the sole factor I am looking for in a mic. I should have phrased the question: If I am looking for a mic with low self-noise, what does the S/N ratio mean exactly and why do many mics that are considered sensitive vary so much in the S/N ratio.

How much does the output impedance affect the noise in the signal, if at all?

Natan, Look at pages 316 and 317 of the new B&H catalog. It lists all the specs, side by side for 27 different mics.

The key to this is remembering that you're looking at a RATIO here. Something compared to something else.

In simplest terms, the HIGHER the signal to noise ratio in decibels - the lower the amount of self noise you'd expect from the mic.

Just as 1/4 is a smaller fraction than 1/2. So given a SIGNAL of 1 of something (in this case decibels) you want the noise factor to be the smallest fraction of that you can get.

So a signal to noise ratio of 10 decibels is OK. But a s/n of 40db is a lot better.

Impedance is a measurement used by circuit designers to denote the resistance of a circuit to an energy source.

In modern impedance bridging circuits - as long as the input fed to a circuit is within the acceptable range of that circuit - and with bridging it almost always IS in modern electronics since they're designed to be happy with input impedances much higher than they'd commonly expect to see. - it should NOT contribute to noise in any appreciable way.

So feeding a low impedence (Z) signal into a high Z input (which is the way almost everything is designed today) shouldn't EVER be the source of noise.