Need help tonight on gain settings... Channel vs. Master

R

Ryan-Guy

Well-known member
I was just leant a Shure fp33 and a h4n for a shoot tomorrow in a boat/water scene. Hooked up to a 416 in a blimp/furry.

I need help with setting the gain levels and which levels to adjust. Specifically the level vs. master dbs.

1) Where do you start each control and which ones do you start adjusting first?

2) Do you stop at a certain db point on the channel or master and start adjusting the other?

Thanks!
 
From the manual: "The 1 kHz tone oscillator level is set by the Master Level controls. To calibrate other devices, adjust the Master Level controls for a 0 VU response"

Basically you connect the FP33 output into the H4n, set the Master Level to 0, turn on the 1kHz generator, lower the recorders level until it reads -20dBFS.
That gives you 20dB of headroom on the mixer, once set, you should make all your adjustments from the mixer and never have to touch the recorder again.

Couple things to watch our for... never set the level on the H4n below 1.0 (it goes down 0.9, 0.8, ect...) because below 1.0 it's no longer lowering the input gain and instead just applying a compressor to the signal. If that's a problem, you have 2 options...
The H4n has 3 line level inputs. The 2, 1/4" jacks on the bottom (the XLR are hotter, mic inputs). Or the 1/8" stereo input on the back cover. If you have the right cables to make that work, great.
Otherwise you need a padded cable.
 
Amplifiying on Gohanto's response a bit ... the Master output is normally set to unity gain, the "0" mark on the dial, and left there. That should result in tone reading 0VU on the mixer's meters. All adjustments in levels for various sources during recording are carried out with the individual channel faders. So the procedure is to turn down the channel faders, set the master level to unity, send 0VU tone to the recorder, adjust the recorder's level controls to -20dBFS, turn off tone, turn up the channel faders so the mic(s) are at the proper recording level (~0VU) on the mixer's meters. Note that the recorder meter will read higher on speech than it does on tone - this is normal and does NOT mean the master gain should be turned down. 0VU tone will read -20dBFS while 0VU speech will read about -12dBFS, due to the different response characterisitics of a peak-reading meter such as on the recorder to a pure sinewave tone versus a complex waveform.
 
Last edited:
Also be aware that compared to more current mixers, the metering on the FP33 is very slow (VU=virtually useless).

If you're watching the FP33 levels bouncing at a particular level, more than likely at the camera it will show you a hotter signal coming in.

I suggest watching both the FP33 and camera metering at the same time (at least for awhile) so you can see the differences and make adjustments
with that in mind.
 
I listened to my recordings late last night and almost everything was perfect. No noise, I could clearly hear the actors in the water from about 3 feet away from mic, and no boat noise or hiss from the pre-amps! Sounds like they were floating alone in the middle of the pacific, which is what they were supposed to sound like like :beer:

The only issue I had was peaking on the screaming (whispering to screaming in 3 seconds) and luckily those exact parts can be easily fixed with ADR. I couldn't figure out how to (if even possible) make a safety channel on the h4n in stereo. I have a tascam DR-680 coming in on wednesday so hopefully that will give me some more tools to work with on padding(?) separate channels.

Thanks again guys!
 
Remember the key rule of gain staging: raise the gain early in the signal chain. Each gain stage adds noise and multiplies the noise of earlier gain stages. So always try to keep your master at unity (meaning, it does nothing to the signal) and set your channel levels as close to "perfect" as possible. Then, you can leave the channel levels as the reference point, and make small adjustments to the master with the easy ability to go back to unity for the "perfect" level... this helps, e.g., in quickly adapting to a scream in the middle of a scene....

That said, also play around with your setup, and see how low the levels can go before the noise becomes problematic when boosted in post. You may very well be able to set the levels properly for the scream and still get decent audio for the surrounding dialogue. Remember, with digital, levels that are too low degrade gracefully, levels that are too high degrade catastrophically...

If you're just using a single mic, it looks like, in spite of what I said earlier as a general rule, you can pan the input to center at a level that's good for the scream, then use the decoupled master controls to raise one channel to the proper level for the dialog; this should feed the H4N with one signal optimized for dialog and one for the scream... I've never used the FP33 but this seems like the way to go based on the manual...
 
ullanta said:
Remember the key rule of gain staging: raise the gain early in the signal chain. Each gain stage adds noise and multiplies the noise of earlier gain stages. So always try to keep your master at unity (meaning, it does nothing to the signal) and set your channel levels as close to "perfect" as possible. Then, you can leave the channel levels as the reference point, and make small adjustments to the master with the easy ability to go back to unity for the "perfect" level... this helps, e.g., in quickly adapting to a scream in the middle of a scene....

That said, also play around with your setup, and see how low the levels can go before the noise becomes problematic when boosted in post. You may very well be able to set the levels properly for the scream and still get decent audio for the surrounding dialogue. Remember, with digital, levels that are too low degrade gracefully, levels that are too high degrade catastrophically...

If you're just using a single mic, it looks like, in spite of what I said earlier as a general rule, you can pan the input to center at a level that's good for the scream, then use the decoupled master controls to raise one channel to the proper level for the dialog; this should feed the H4N with one signal optimized for dialog and one for the scream... I've never used the FP33 but this seems like the way to go based on the manual...
Click to expand...

This is a little over my head right now. I am just starting with mixers and separate audio. I did try to split the mic in to sperate audio channels on the fp33, but couldn't get it to work. And the meters on the fp33 were not working, they showed no movement at all. I had to base my gains on the h4n. Also my headphone monitors were very quiet until I maxxed them out, then it sound just right. So I thought some of the peaking was due to the headphones hitting their limit, but after reviewing it was really peaking, even at -8db on the h4n (is that normal?)

Thanks again taking the time to help me out! This thread saved the shoot.
 
Sounds like either the mic or the FP33 peaked (or was severely limited), while the level being sent to the H4n is perhaps lower than optimal... you can get distortion by
1) Overloading the mic capsule (sound just too loud)
2) Overloading the mic electronics (sound just too loud)
3) Overloading the preamp input (a pad might help)
4) Overloading preamp output (reducing gain might help)
5) Overloading Master input (reducing channel level can help)
6) Overloading Master output (reducing master level can help)
7) Overloading H4N input...
etc.

Sounds like you're clipping/limiting somewhere BEFORE 7, since the levels recorded by the H4n are still well below peaking. Go through the signal chain in the order above, watching level meters on the Fp33 (though if they don't work...?) if the meters indicate no clipping but you hear clipping, while the master gain is at unity and the channel level is low, then the mic is either distorting itself or overloading the FP33 inputs. Try attenuating the mic with any pad built into the mic or with an in-line attenuator... also try creative booming!

Anyway, I'd suggest you calibrate the mixer so that one set of meters is reliable... either make sure that the mixer will always clip before the H4n does, and watch the mixer meters (which I guess might not work?), or make sure that the mixer never clips before the H4n, and watch the H4n meters.... but make sure the ALWAYS/NEVER condition holds!
 
I would be very skeptical if the meters are not working. What else is not working?

As a super simple guideline, most audio devices like to have their gain someplace between 50% and 75%. Anything over 75% and you should be very alert to preamp noise creeping in. So if meters and everything else is a bit wanky if you start from the recorder end and set things to 50% and try to get nothing over 75% then even with out markings you should be in the ball park. Usually around 75% is unity gain (if it's not marked) or in the case of a preamp the place it was designed to be quietest. You should strive to never turn up a signal after you have turned it down. Actually you don't want to do the reverse either. Anyway so if 75% on one and 50% on the other is a good setting you would want the recorder at 75% and the mixer at 50%.

This is a pretty gross over simplification but it's easy to remember and will keep you in the ballpark. Basically your goal is to not have any gain over 75%.

LOL, I just noticed your sig ;~))) I may have to change mine. So how is your mothers basement? My mothers had a dirt floor so not so good to work out of. Made my day.
 
You must log in or register to reply here.
Back
Top