# White Noise (static detecton)

is it also possible to find the baseline of static? because no microphone will have no static at all, everyone has some, more or less depending on the microphone. is it possible to find that in audacity?

I you record “silence” and then look at the spectrogram view, the colouration of the track indicates the background noise.
Are you wanting a numerical value of some kind?

for my project I’m trying to find out how much static, or noise intereference, or whatever a microphone contains using audacity. so I do need a numerical value along with measurement units. if you could help me out about how to approach this on audacity that would be great.

I can give you that number. Very nearly zero. In the case of dynamic or moving coil microphones, it is zero since induction and inductive reactances do not generate noise in the classic sense.

However vast fortunes have been made and lost in designing microphone amplifiers or MicPres.

The raw microphone signal usually hovers around -55dB to -65dB. The amplifier’s job is to boost that up to +4dB and still have a noise floor of at least -60. If you do the arithmetic, that gives you a required amplifier noise floor of -125dB. That’s very close to the noise contributed by the electronic components, ICs, transistors, and resistors, just existing.

I don’t remember the absolute numbers, but I seem to recall a -135dB or -140db or so is the best you can possibly do with a perfect amplifier, perfectly matched to the microphone. The microphone, by the way is irrelevant. You can put a 150 ohm metal film resistor in place of the microphone and go with that – you can do all your noise measurements that way – assuming you have an instrument that will measure down that low.

Anything you do in Audacity is subject to the computer soundcard. Most of the Windows built-in sound cards are terrible and will be far worse than whatever it is you’re trying to measure.

Most people trying to do serious scientific measurements in Audacity run into the design goal which is to edit productions and presentations very well, sometimes at the expense of absolute digital accuracy.

"How come the bit pattern at the output doesn’t exactly match the one I generated?’
“Does the show sound OK? That’s why we’re here.”

Koz

Most decent microphones will have their specifications published in the product documentation.
The two figures that are relevant are SNR (signal to noise ratio) and “Sensitivity” (size of signal for given sound input). If a microphone has low sensitivity (produces a small signal for a given sound pressure level) then it could have a lower noise floor than a more sensitive microphone but still have more noise in practice because the gain needs to be increased due to the low signal level. A microphone that is too sensitive for the job in hand could have a better SNR than a less sensitive microphone but have worse “THD+N” (http://en.wikipedia.org/wiki/Total_harmonic_distortion#THD.2BN) due to higher distortion for high sound pressure levels.

Unfortunately it can be very difficult to compare one manufacturers figures with another as they often use different units for their measurements.

In order to measure the noise floor in a meaningful way you need to decide whether to compare the noise floor to a given “sound pressure level” or to “full scale” (“full scale” will be the maximum signal before a specified amount of distortion). In either case the microphone “self noise” is likely to be almost negligible compared to the self noise of the microphone pre-amp.

so can the spectrogram track view help me out or is there something else I can use?

I think we’ve answered that one - microphones have extremely low self noise. Background hiss in a recording is in the main a product of the microphone pre-amp. Measuring microphone self noise requires specialised and expensive equipment otherwise the noise of the measurement hardware will drown out the microphone noise.

You have still not made it clear exactly what you want to do.
Are you trying to compare one set-up with another?
What is the “project” that you refer to?

This project is for science fair. My science fair hypothesis is the effect of the diameter of a microphone on the amount of static, or noise interference it has. I’ve already bought different microphones based on their diameters. I’ve heard from different people that audacity can help me out with finding the amount of static that a microphone has, like I mentioned earlier, every microphone has a baseline of static, no microphone will not have any static, so I heard that Audacity can help with it. So basically what I’ve been trying to ask is is it possible to find the amount of static with audacity or if not can you help me out with finding another variable which I can use which can be accomplished through audacity, or another computer program, for that matter. I’ve also asked about graphs on audacity because I need them to show my data, if it at all is possible. Your help is greatly appreciated.

The problem is that the noise from the microphone will be much less than the noise from the pre-amp that it is connected to.

The usual way to measure noise is as “rms A-weighted”
There is a plug-in available here that can measure this: "Wave Stats" plug-in
Installation instructions are here: Missing features - Audacity Support

You can measure the noise from the pre-amp by replacing the microphone with a 600 Ohm resistor (the value is not critical - a 1k resistor will do). Make a recording with the recording level turned up and the recording will contain noise from the pre-amp. It is essential that the recording level is exactly the same for each test.

To make the measurement, select a short section of the recording and use the above plug-in.
Avoid the start and end of the recording as there may be clicks at the start/end.

To make measurements with a microphone attached you will need an extremely quiet room to conduct the experiment or the results will be meaningless.

Note that condenser and electret microphones have active electronics built in, so it is not possible to measure the noise level of the mic capsule alone.

Thank you for your help. Are u sure there is no simpler way to measure the noise interference? To me it just doesn’t seem that complicated, such as finding a baseline if static. Aside from that do you have another suggestion for a variable I could measure and if it could be measured through audacity or another program? Also, what is wrong with using the spectogam track you mentioned at first? sorry for my persistence but time isn’t on my side. Thank you for your patience through this.

Bury the microphones in multiple pillows in a quiet room and record them in Audacity. Average out the jumping sound meters. Make the meters really big by grabbing the right-hand edge and pull so you can read the numbers.

http://kozco.com/tech/audacity/Audacity1_record.jpg

or undock the meters and make them even bigger.

http://kozco.com/tech/audacity/AudacityPanelFull.jpg

Don’t pay any attention to the blue waves.

Increase the accuracy of the meters:
Audacity > Edit > Preferences > Interface > Meter Range = -96dB.

Don’t change any settings and cycle the microphones through the system one after the other and read the bouncing sound meter – average out the bounces.

Koz

The more sensitive a microphone the better the signal-to-noise ratio for recording an arbitrary constant sound, ( the noise floor is a fixed overhead ).

The smaller the mass of the microphone diaphragm the more sensitive it is :
as the less massive the microphone element is, the less sound energy is required to move it to create the same signal voltage.

Loudspeakers & earphones can be used as microphones , but their signal sounds more noisy than a proper microphone because their diaphragm is more massive than in a proper microphone and consequently less sensitive, ( you have to shout using an earphone as a microphone ).

So large mass microphone diaphragms are less sensitive , but do not create more noise , it’s just their signal-to-noise ratio is worse because they are insensitive because of their larger mass.

Re: science fair project, maybe make your own microphone … make microphone carbon OR magnet site:youtube.com - Google Zoeken

Thanks for your guys help, I think I have an idea of what I want to do. Quick question though, I want to use my preamp to connect my microphones to my laptop to record with audacity. Do you have any idea of what I need to connect the amp to a computer?

Can you fill out a bit of detail, ideally make an model number for each item, then we can probably give a more useful answer.

I have a Ultratone K450FX 45 watt pa system /keyboard amplifier which I want to be able to plug in my microphones in and then connect it to my laptop in order to record with audacity. Several of my more techy friends have told me that if I can get an adapter I will be able to record and I won’t need the amp. Is this true what they’re saying or should I be using something else?

Most PC and laptop sound cards have a microphone input, but generally the quality is somewhere between poor and atrocious.
A few laptops have a “Line Input” connector, or a combined Microphone/Line input. Many laptops only have a “mic” input.

To connect your amp to the computer you need to connect one of the “Line Out” connectors on the back of the amp to “Line In” connectors.

Even if your laptop has “Line In” I would not recommend using it because (a) it is probably not brilliant anyway, (b) the weight of a decent cable and regular plugging and unplugging could easily damage the connector, which is likely to be directly soldered onto the main circuit board, making it a potentially expensive accident waiting to happen. Personally I would go for a USB sound card.

There are lots of USB sound cards with “Line Input” on the market. I use a cheap Behringer UCA 202. I usually use it with the Line Out on my mixing desk but it should work equally as well with the line out of your amp.

If the Line In is a stereo pair (most line level sound cards are), then because the K450FX is a mono amp, connect the Line out from the amp to the left input of the sound card and set Audacity to record mono. This has a slight downside that if you are monitoring the sound through headphones it will only come through one ear, but the recording will be correct and will play back through both ears. Alternatively you could connect the line output from the amp to both left and right inputs with an appropriate “Y splitter cable” (the splitter cable must be a “shielded” cable. If you try to use a headphone splitter for this it will probably cause a buzz in the recording).

Next problem. If you want this for recording, plug headphones into the amp otherwise the microphone is likely to pick up an echo or howling feedback from the speaker.
If you want to be able to hear previous tracks playing on the computer while you record a new track (overdubbing) you will need to plug your headphones into the USB sound card - so that is 2 pairs of headphones, though you will only be using one pair so the other pair can be from the dime store / pound shop.

An alternative to using your keyboard amp: There are many USB sound cards available that have the 3 pin XLR microphone inputs. These are likely to cost more than a “Line input” device, but may give better sound quality as much of the price goes into the microphone pre-amp.

Out of interest but slightly off topic, what do you think to the K450FX?

And as above for pages and pages, you won’t be measuring the microphones. You will be measuring the preamp noise, but see how it goes. Koz

I have changed my experiment so that I do not need the preamp anymore. After getting advice from friends and stumbling upon a few youtube videos I found this

I feel like this is what I’m looking for. The guy mentions static in it and even shows where it is. Is what this guy saying legitimate? If it is then the only problem I have now is being able to quantify it. Is there a way? I am in desperation mode right now and will be grateful for any last minute help.

That YouTube is just a poor tutorial about how to use Audacity’s noise reduction feature.

As we’ve mentioned for passive* microphones, their sensitivity determines the signal-to-noise ratio. The noise (floor) level is created by the active electronics, (see here), and will be constant for a given set-up. The output of the microphones for a given constant test noise is the only variable, i.e. the microphone sensitivity. Passive microphones do not generate “static” noise.

A quick science project using Audacity and stereo-microphones : measure the speed of sound in air … Using audacity to measure the speed of sound - #11 by Trebor [ the speed varies with temperature ].

[* cf. “active” types which contain powered electronics ]

In the beginning where he shows you the blue waves for static? That’s not static. That’s the rumble and clicking from him handling the microphone. Normal microphone hiss and static doesn’t appear on the blue waves. Good try, though.

Given as far ahead as you are, I would fake it. Do a Hollywood Test. Get a small desk fan and have it blow gently on the microphone. You will get fake blue waves which you can record as the “Noise Signature.” Then change the microphone (or just leave the original one in place) and move the fan so the noise is different. Record that and write your report proving through weeks of work and exhaustive analysis how this microphone has much less noise than that one.

I never saw you before.

Koz