Noise Gates and Downward Expansion

Hookin up those Audio Boxes without gettin yourself Perplexed.
Post Reply
User avatar
Voodoo Guru
VooDoo Site Admin
Posts: 719
Joined: Wed Feb 23, 2005 8:47 am
Location: Down there on the right...

Noise Gates and Downward Expansion

Post by Voodoo Guru »

Here is a post that we made on another forum that describes our viewpoint on Noise Gates and Downward Expansion along with how to set them up.
Enjoy...


Howdy All,
Yes Marty, We will agree with you on all points.

In our experience, We absolutely abhor a true 'Noise Gate'. They are too abrupt, it's either ON or OFF hence the name 'Gate'. There is just not enough control. Thats why you where experiencing the cut off first syllable Joe...NO CONTROL!!! hehehe

We do like and use a Downward Expander. The description that Marty gave as it being opposite a Vox is good, but a better one is that it's function is the opposite of a soft knee compressor. Instead of compressing the dynamics of the signal when the threshold point is reached on a compressor, an Expander RELEASES the compression when the threshold point is reached. An Expander is exactly the opposite in operation of a Compressor.

The unit used here is the Aphex 622, and it is positioned directly after my Bellari RP520 Mic Preamp. We like this Aphex because it gives you ultimate control of the expander processes. It has controls for Attach, Hold, Release, Ratio, Range, Threshold, and it also has an adjustable Frequency and Bandwidth keying Filter. Correct manipulation of these controls can give you the most transparent transition between full attenuation and full release, then back again. That is the secret of these devices, to set them for the most transparent operation possible so that no one knows that they are even in line. No pops, clicks, abruptness, or cut off syllables (winking at No Control Joe).

How do we set ours? Glad you asked....
We must say that the best thing to do is to correct the noise in the background FIRST! The less you have to use the attenuation in a Downward Expander the better. In fact, not using an expander at all is the best proposition, but like Marty said, not everyone has an anechoic chamber for a studio. The more dB in noise you have in the background, the more 'Range' (attenuation) and 'Ratio' you have to dial in to reduce it, then you have to adjust the other controls for the smoothest transition to full attenuation release when you speak, and then back to full attenuation in your speech pauses. You will find that in order to get the best sound, that you have to find a 'happy medium' in the attenuation, in other words, don't try to suck ALL the noise out, leave some of it in the background. Not much, just a little, that way the expander doesn't have to work so hard and you don't loose so much of your dynamic range in between the syllables when you speak.

The 'Ratio' control works mostly with the 'Range' control, as it determines the amount of compression or attenuation that is applied to the signal when the 'Threshold' point is reached. This can get a bit tricky to explain, so suffice to say that for voice with a small amount of background noise, a 2:1 Ratio is a good starting point. If you have more back ground noise then you might want to try around 3:1, for bad noise 4:1, but usually more then that is too much. This control works in conjunction with the 'Range control.

Well, how do you set the controls for a smooth transition?
OK, so you have now set your 'Range' and 'Ratio' to an acceptable attenuation level. Next is to set the triggering of the Expander. This is done by adjusting the 'Threshold' control. There is usually a light or a meter to show when this is achieved. The Aphex 622 has an 'OPEN' LED and a 'Closed' LED. The Red, Closed LED will be lit until you adjust the 'Threshold' control to where your voice will trigger the expander and light the Green, Open LED.

Once this is done you need to set the 'Attack' control for the best triggering on your first spoken syllable. This is the control that is used to keep from getting the No Control Joe Syndrome artifacts (NCJS, wink wink)! Don't go to fast with this control because it will introduce clicks into the audio mix. We usually turn it as fast as it will go until it starts clicking on a fast 'sss'......'sss'....'sss', then back it off just enough until it will absolutely not click.

Next is the 'Hold' control. Once the 'Attack' and 'Threshold' circuitry allow the Expander to be triggered, the 'Hold' control does just that, it holds the expander open for a set amount of time before it hands control off to the 'Release' circuitry. For Speech, we usually keep this control pretty short in time, usually around .2 to .4 of a second. This depends on your cadence in speech and is different for everyone. This control can only be set by trial and error but it will usually be pretty fast. If its set too fast then the expander will drop out between low level signals, too slow and the expander will hang open too long. This control must be adjusted in conjunction with the 'Release' Control.

The 'Release' control is also self explanatory. After the 'Hold' circuitry is finished, it hands over control to the 'Release' circuitry which allows the expander to decay (Release) back into full attenuation. For our cadence, we usually set this control to around .5 seconds. This control must be adjusted to your voice and cadence and works in conjunction with the 'Hold' control.

Now you are beginning to see how everything works in unison with each other. It is the individual pieces which make up the whole, just like in the entire rack line up, everything must work and compliment the preceding and post-ceding pieces. It is actually like a finely tuned concert and everything has it timing and place.

You must now go back through the controls and fine tune it. You should hear a very transparent and smooth transition when the expander operates. No pops, no clicks, and no NCJS artifacts.

We don't use the Key Frequency/Bandwidth Filter on this unit as we like the flat frequency response keying in lieu of it.

While we're on a writing frenzy, we can also let you in on another trick that's done with our equipment. Allot of people like to use condenser microphones in there studios. The problem with these are that they are VERY sensitive, and if you have a background noise problem to begin with, these mics only make the problem worse. We actually use 2 mic pre amps in our set up. We use a condenser microphone and the next piece in line is the Bellari 520 Mic Preamp, then the Aphex 622 Expander, then the RP533 which has another mic preamp in it. The first Mic Pre is set to give just enough signal to the Aphex 622 to work and still have an acceptable Signal to Noise Ratio. This is a fine line, if you don't give it enough of a signal then it just plain will not work, you won't have enough 'Threshold' control to open the expander, and you will get a horrible background hiss when the expander is at full attenuation. Too much signal and you start to increase the background noise and the Expander has to work harder. The RP533 preamp is then used to bring the signal the rest of the way up to a -3dB level for the equipment down line. This allows us to close talk the mic, which a we prefer, and gives enough signal for everything to be happy. This might not work for you, but we thought we would mention it.

Well, thats what happens here in the Voodoo LABS for Downward Expansion.
There are allot of different Expander units out there, but the above stated procedure should be very similar in there operation.
Ya'll Take Care now, ya hear?!
Voodoo Guru
From Deep in the IDD of the Sub-Harmonix Realm

Post Reply