Unexpected Noise

A

azactro

Active member
Hi all,

I am extremely new to audio, but I am having quite a problem getting clean sound out of my microphone system.

I have the Rode NTG-3 and Tascam DR-60D.

I'm not referring to the base noise that the DR-60D produces. I believe it is coming from the NTG-3, which is really surprising me. It sound like a dehumidifier or AC unit, but I am testing this outside.

I'm honestly completely baffled as to why this is happening and would really appreciate some help from someone more experienced with audio. Unfortunately, I don't have anyone to turn to as no one around my town understands anything regarding video or audio production. So, I turn to the wonderful DVXuser community for guidance.
 
Hehe. I was getting to that. Here is the noise I am talking about. Again, sounds like an AC unit or something, but this was recorded outside and it is the same in every direction. I am going to go out in the middle of the desert in a few hours just to make sure.


Please be warned there is a -12dB peak in there a second in as a reference.
https://www.mediafire.com/?kajm9ael6dg7ize
 
azactro said:
Again, sounds like an AC unit or something, but this was recorded outside and it is the same in every direction.
Click to expand...

Sounds like mic self-noise and pre-amp noise to me. If you have the gain cranked all the way up on your DR-60, you should expect exactly this. This is the noise floor, and you can generally only hear it when you don't have an actual signal to compare it to, like your example.

Try recording something real, like a person talking. Get your shotgun within about 60cm of the speaker, pointed at their mouth from overhead, and set the gain on the DR-60 so that you record the dialog with peaks around -12dBFS. When the dialog completes, continue recording for a few seconds. Now listen to the recording at a level that makes the dialog reasonable, and keep listening when the dialog stops. What you should hear in this case is a convincing silence. The noise floor should be sufficiently below the signal level so that you have a relative silence compared to the signal. Not an absolute silence, but a convincing relative silence.

Welcome to analog audio.
 
Can't hear much on my POS laptop's speakers, but a spectrum analysis shows a bump centered at 60Hz, so it could acoustically borne noise... typical of HVAC.
 
This is with the gain set on medium and the knob was turned half way. These levels were chosen to make the speaker get -12dB. I was talking loud too. The shotgun was about 6-8 inches away.

I will try to post a real sample later as well as a sample with absolutely no life around.
 
All practical equipment used in the Real World has noise. One of our primary jobs as audio recordists is to maximize the Signal-to-Noise ratio.
This is accomplished by many techniques. One of them is spending a lot of money on microphones that have low self-noise and good sensitivity.
And mic preamps, mixers, recorders, etc. with good signal-to-noise ratio (SNR)

Another major technique is selecting the optimal microphone for each particular situation, and GETTING AND KEEPING IT IN THE RIGHT PLACE.
This is sometimes done by attaching a microphone to the subject, either hidden or visible. It has become more popular to use tiny, nearly invisible mics very close to the mouth.
Or having an experienced boom operator (monitoring his mic on headphones) holding the microphone on the end of a boom and moving, turning it as necessary for optimal pickup.

Your Rode NTG-3 and Tascam DR-60D are quite decent pieces of gear, not the top of the line, but nothing to be ashamed of.
We would expect the NTG-3 and DR-60D to perform adequately when used properly.

Your sample sounds like completely ordinary and expected "rumble" to me. Unless you are WAY WAY out in the countryside somewhere practically everywhere has "rumble".
Without any acoustic reference, we really have no clue whether this normal and expected level of noise, or whether it is excessive.

Note that many pieces of gear, particularly mic preamps and even microphones have "low-cut" or "high-pass" or LF-filter" switches.
The reason for the popularity of this feature is exactly because of this "rumble".

Unless you are recording pipe organs with 32-foot or 64-foot pipes, or steam engines, or flamenco dancers on a hardwood floor,
it is typically safe to filter out this low-frequency noise, preferably as close to the source as possible, i.e. in the mic, or at the mic preamp.

It is not uncommon to find that the "rumble" or LF noise is so bad that it actually interferes with the proper recording of the DESIRED signal.
Particularly when one is using auto-level or even peak-limiting, the low-frequency noise can influence how the desired signal is recorded.

Also, it sounded to my ears like the level dipped for a period right after the tone, and then came back up.
PLEASE do not tell us that you were using some sort of auto-level here somewhere!

If you want to post some samples of actual recordings, we can help you evaluate the signal-to-noise ratio.
Listening to the noise by itself, without an acoustic reference is of rather limited value.
You could take the very best and most expensive gear on the planet and produce exactly the same sample recording.
 
Richard, thank you very much for that info about low-cut, high-pass, and LF-filters. I will definitely have to try this again to see how much of a difference it makes.

As far as auto-level, I have gone through the settings numerous times and cannot find anything that would automatically set the levels. Of course, I may be overlooking a setting, but I am fairly certain.

I will get an acoustic sample for further analysis so it can be more useful.

Again, thank you.
 
Here are two samples. Unfortunately, they may not be very helpful as it was quite windy. I was attempting to talk quite loud. One thing I noticed is that it does seem like some sort of "auto-tone" is on. Let me know if you come to the same conclusion. Unfortunately, I still do not see any settings that would make it auto.

https://www.mediafire.com/?c7anijiz9833j9k

https://www.mediafire.com/?c7anijiz9833j9k

These are the raw files out of the Tascam. Nothing has been modified.
 
It is very confusing what you are trying to do here.
Are you trying to address some kind of microphone or recorder self-noise here?
Or are you trying to improve your recording technique in noisy locations (like outdoors)?

You seem to have ONE equation and TWO unknowns here.

First, start with your recorder. How noisy it it all by itself?
If you use a regular microphone at a reasonable distance and speak in a normal voice IN A QUIET ENVIRONMENT, then what does the signal-to-noise ratio sound like?

Without changing those "normal recording settings", what does the recorder sound like when you unplug the microphone?
This test will SEPARATE the effect of the microphone and environment from the noise floor of the recorder by itself.
It is possible that you may have a bad recorder that needs to be returned for repair/replacement. But I can't tell from your experiments so far.
Note that the proper way of conducting this experiment is to use a "dummy plug" in place of the microphone. It has pins 2-3 shorted together (or a 150 ohm resistor).
That way you can hear what the mic preamp is doing under "normal" conditions (with a low-impedance source connected).

If you want to test the microphone self-noise, one (perhaps not terribly practical) method you could use is this.
Get up at 3am on the quietest day of the week in your neighborhood.
Take all the bedding in the house and pile it on one bed. Bury the microphone in the middle of the pile.
This will create as close as you can get to an isolation, anechoic chamber so that you reduce the acoustic sound as close to zero as possible.
Then you will be hearing the self-noise from the microphone.
After recording a minute or two of silence, take the mic out and speak into it in a normal manner to establish what the signal level is.
Without that acoustic reference, listening to the noise by itself means nothing.

Note that there are many samples of what the mics should sound like on the Rode website.

Do you have a proper outdoor blimp and furry cover on that NTG-3?
The lack of details about your experiments greatly limits our ability to evaluate the results.
As they say in the credit-card commercials: "Don't leave home without it!"
The foam wind gag that comes with the microphone is suitable only for indoor use.

It still sounds like there is some sort of significant "auto-level" or auto-something going on there.
The noise level REALLY goes up and down. That makes me think the experiment isn't what we think it is.

If, as you say, you are "extremely new to audio" you may have unrealistic expectations of what microphones can do.
To get what you hear on TV and movies took very experienced audio people and good equipment and good production practice.

Frankly, I would much more suspect your recorder than your microphone.
But we haven't really heard any examples that would allow us to do any differential diagnosis here.
 
I may very well have unrealistic expectations of my equipment. I'm using Youtube videos with the same or comparable equipment (Zoom H4N instead of Tascam DR-60D on some) as my comparison though.

I agree about the auto-level. I will make a video showing the settings of the recorder. Please let me know if anything jumps out at you.

Yes, I do have the Rode Blimp, I just wasn't it expecting to be so windy when I went out. I will make another test.

I will try to conduct these tests. Thank you.
 
I highly recommend attacking one thing at a time.
Resolve the equipment noise problem FIRST, and BEFORE you venture outdoors.
Else you well never be able to distinguish between equipment noise and environmental noise.
You can't solve one equation with two (or more) unknowns without remaining unknowns.
 
I went outside hoping to get away from some AC or dehumidifier that cannot be turned off. (I live in an apartment.) You are right though, venturing outside has its own set of problems.

Time to build a fort!
 
Isolation of the equipment is the key to tracking down the source of the noise.

Do you have a camcorder or other recording device that you can connect the NTG-3 to directly? If so, does the noise continue. If not, then that rules out the microphone as being the source of the problem -- assuming that when you connect the mic to your recorder in the same room and conditions, you DO hear the noise. If so, then move to step 2.

Take your microphone and recorder. Get in your car and drive to some remote, quiet place out in the middle of nowhere. Turn off you car's engine and keep the windows rolled up. Make a recording of just the ambient sound inside the car. Do you still hear the noise? If so, then the problem is likely to be in your recorder. If not, then it's likely some background sound in the environment where you were -- HVAC (even a neighbor's), refrigerator, freezer, ice maker, fish tank, etc.

All of this assumes that you use the same microphone CABLE for all of these tests. It's not impossible you have a problematic cable. If you have a long cable run, and keep it wound up tightly, that might introduce a harmonic into the signal. Make sure your cable is stretched out for these experiments.

Test everything, eliminating one factor at a time. Chances are you will find the source of your noise.
 
Noiz2 said:
Try page 46.
of THE MANUAL

Manuals are a good thing.
Click to expand...

Manuals are good when you read them correctly.

From page 46: "The plug-in power function can be turned on when an external mic
that requires it is connected to the 3-4 jack. "

Unfortunately, this only affects Channels 3/4. I am using an XLR input. Thank you though.
 
Richard Crowley said:
Without changing those "normal recording settings", what does the recorder sound like when you unplug the microphone?
This test will SEPARATE the effect of the microphone and environment from the noise floor of the recorder by itself.
Click to expand...

Do you terminate your inputs with something that approximates the impedance of the mic? Or is the input amp looking at 'open', that is 'infinite' impedance?

I've been meaning to do these sorts of tests, but other than just burying the mic, and recording with the Zoom H4, and a couple of USB audio digitizers like the 2 channel Fasttrak, I've not done much in the way of determining the noise floor of my audio capture equipment.
 
Impedance aside, just an open XLR input is capable of picking up environmental (electrical) "noise".
That is why at minimum, it is advised to use a shorting XLR plug to avoid turning the mic input into a receiver.

Some mic preamp circuits perform very differently with a high-impedance (i.e. "open circuit") at the input nodes.
I believe that 150-250 ohms is a standard "dummy source" impedance for proper evaluation and testing mic preamps.

Doing any kind of evaluation or testing of a microphone preamp with an UN-terminated input is not a valid test in anyone's book.

But in this case, just listening to the recorder with the mic plugged in vs. unplugged is a quick-n-dirty first-order differential test for the noise source.

We are dealing with a user with an extremely limited range of test options.
Apparently he will have to travel way out of town just to find a quiet ambient.

My money is on the recorder. That Rode mic has a much better reputation than that recorder.
That recorder is relative new and has a remarkably low price for all those features.
I wonder if they cut corners with the quality of the mic preamps to pack all that into such a low-price box.
 
Last edited:
You must log in or register to reply here.
Back
Top