Help! with basic recording levels

  • Thread starter Thread starter Roger Mac
  • Start date Start date
R

Roger Mac

New member
Hey guys. My angle is mostly for vocals but I assume this applies everywhere.
My main concern is a mild harshness (clipping?) that is being added on the louder vocal parts.

1) I record dry and usually try to aim for the peak on the recording channel to stay just under 0.0 (say, -0.5 to -0.2).
First thing, is this the correct approach or should I try to keep it lower?

2) My pre-amp (FMR RNP) is set at +18 gain. The mixer level (cubase) for that channel is usually set around -6.0 (so the recorded peak stays under 0.0).
Is this correct approach? Or should I drop the pre-amp gain by 6 (to +12) and up the mixer channel setting by 6 (to 0.0). I'm not sure of the proper combo between these two settings

3) I'll add a little compression (thresh= -14, ratio=3:1, gain=5.0) to the vocals when mixing to bring up the lows without pushing the highs over the top (being careful to keep the channel under 0.0). Then after mix-down, I'll add some compression to the whole mix (same reason) to up the overall volume.
Am I on the right track?

much appreciated.
in: Rhode K2 -> FMR RNP -> Lynx Studio L22 sound card -> Cubase
out: Cubase -> Presonus Central Station -> Tannoy &/or AKG headphones
 
describe the sound its makeing alittle better i mean is it like poping or wat , and try using a limiter instead of commpressor limiting usally works good with protools acually i limit every track also add a reverb
 
You're close. Everything is right on except:

The fader in Cubase, however, has NOTHING to do with your input levels. You can't pull the fader down to keep the input from clipping because the fader comes after the input. If you clip the convertors in the Lynx on the way in thats it- its clipped. I know Cubase's input channel has a gain control, but it also can't turn down a signal to keep it from clippng at the convetors. Which sounds like what's happening.

So you have to set your faders in cubase to 0 in order to see the actual level coming in. Then you set the RNP so that the peaks don't clip. Just to save my blood pressure I don't care if the peaks are considerably lower than -.5 to -.2 as long as I'm recording 24 bit. If your input and channel faders are set to 0 then you are looking at the level that's hitting the convertors. If they clip and you can hear it... you're hosed.

That all said, just because the red clip light turns on doesn't mean you've clipped the convertors. Most DAWs light the clip LED when you get 3-6 dB from 0dBFS (full scale). So you have a little room. If you do light the clip LEDs listen before retracking- it might be OK.

Take care,
Chris
 
follow up info

It's like an electronic raspiness on the louder vocals. No popping, more of a background "paper crinkle" sound.
 
Roger Mac said:
It's like an electronic raspiness on the louder vocals. No popping, more of a background "paper crinkle" sound.

That's clipping. You are recording way too hot. There is no need to peg every track during tracking. The software faders do nothing to save you if your converters are being clipped.

Lower your preamp gain by 12 dB, shoot for peaks at -6dB with your software fader at unity (0).
 
Good point

Hey guys.

Damn! Good point (Cubase fader is 'after' input)! Next time i'll set zero across the board and adjust the mic-pre accordingly.
Much thanks.
 
Last edited:
You would be in better shape if you had the peaks at about -5 or so once you put the faders in Cubase back to zero.

You don't win anything by getting the peak as close to zero as you can.
 
Judging from what you described as the sound it is making, that sounds like you may actually be clipping the RNP also. Most digital devices are set so that -18 to -15 should be your average, or nominal input level. There is absolutely no reason to to have your peaks that close to 0. As far as limiters go, I would definately say that it is bad practice to blindly use a limiter or a compressor on every channel. Realistically, if you are constantly getting input levels that are around 0, than you are most likely close to clipping your preamp.
 
levels vs. faders

Ok. So now I have to show how little I (apparently) know. :)

So now I get the issue with the mic-pre and the sound card/converter. So what difference is the fader in Cubase going to make to the recording levels? Or, during recording, does the fader only control the volumn "out" after the track has been recorded into the software?

You expertise is greatly appreciated. Thanks much.
Roger
 
The input faders in Cubase will adjust the recording level (why anyone would want to is beyond me) However, it will not affect the level of signal going into or coming out of the converter. The only way to have any idea how your gain staging is set up is to leave the input faders in cubase at 0.
 
Been years since I tried Cubase. These days I use Sonar and Pro Tools. In Sonar at least, you have a post level setting, and an input trim. So I suppose you COULD adjust input trim instead if Cubase has such a setting, but I wouldn't recommend it. For one thing, it won't help if the clipping is occurring at the pre-amp anyways. It's useful if you're using a tube pre-amp and want a little of that warm soft clipping, you can use the input trim to bring the level back under control as it's being digitized. Other than that, all it's doing it introducing one more step of signal path, and every step will detereorate your signal a little.

As for compressors and limiters, you should certainly not be using one to bring wild peaks under control at the recording phase. In fact, using a limiter at all for the recording phase is generally a bad idea. I'd save that for outboard mixing and/or mastering. It IS a good idea to record using a compressor (sometimes), but only to bring the dynamics under control (soft level vs. loud level), NOT to reduce clipping. If it distorts at the pre-amp and THEN hits your compressor, it's already too late.

For sure, I'd try reducing the gain on the pre-amp first. However, if Cubase is recording the signal obviously well below the digital ceiling, and it STILL sounds distorted, then another possibility is that your pre-amp can't withstand very high SPL. I used to have a cute little 2-channel pre (M-audio audiobuddy) that would crap out with any SPL over 90 db or so. If you use a mic with a pad, that's one way of solving that. Otherwise, you can try backing up off the mic a bit...but unless you're recording in a fairly dead room, this will make the sound uncomfortably live, and it tends not to pick up the high frequencies as well.

For sure, I'd try reducing the gain on the pre-amp first. However, if Cubase is recording the signal obviously well below the digital ceiling, and it STILL sounds distorted
 
(incorrect) reason for my levels

All good info/advice as I expected. :)

What I am trying to do is:

1) first and foremost - get the best possible levels recorded. Looks like I'll accomplish this by backing off on the mic-pre gain.

2) boost the volumn of the vocals to the vocalists headphones as she is singing.

She has an issue hearing herself in the 'phones. She is usually singing louder than her vocals through the headphones (AKG). If she can not hear herself, she is ok. If she can hear herself, she is fantastic! So I'm trying to help out both of us. If the levels are too low, even setting the 'phones at 10 may not be loud enough. (rock/blues/funk). So I'm trying to boost what she hears without overstepping good recording levels.
 
There should be a way to adjust your monitoring levels independently of the recording level.
 
separate volume

The soundcard output goes into a Presonus Central Station. The PCS has separate volume control for the speakers and each (two) headphone.

The track that she sings too is fine. I actually have to lower the track (Cubase mixer) fader to about -7 to match the average vocals volume coming in. Otherwise we'd hardly hear the vocals.

Or are you refering to separate controls within the recording software?
 
Last edited:
What I mean is, you should be able to give her a mix from within cubase that does not affect the recording level. Or by using the monitor mixer that came with your soundcard.
 
I think your on the right track with lowering your level before the converters. At 24 bit, you have a little more to play with. I thought it curious that your software isn't indicating a clipped recording. You could still be putting a hurt on the converters, enough to rough up the sound, just never actually enough to pop the red lights.

I still suspect the microphone-preamp and microphone technique as the cause of the problem. Generally, I get a real nasty sound when I over drive the ADC. You describe what your hearing as harshness. Sounds like analogue to me. You know how a good singer, at point blank range, can expose any weakness in your gear and technique. The ability to handle a hot signal is what separates the tools from the toys.. To ultimately solve this situation, you may have to alter your technique to compensate for the limitations of your equipment. .

poux
 
headphones & amp

Headphones: AKG K271 Studio
Amp - Presonus Central Station, has two headphone jacks each with it's own volume control. Speakers from the PCS also have their own volume control.
 
You can raise the volume output in Cubase without affecting the recorded signal if you need to send more volume out. If you need even more than the software will give naturally, you can cheat and add any plug-in with gain compensation (like most EQs and compressors), and use that to up the signal some more.

If you'd rather not do any of that, Radio Shack actually sells a little battery-powered device that you plug an 1/8" cord into, and then has three 1/8" ports that provide a signal amplified an extra 30% or so.
 
I hate to have to say this, but maybe a Behringer headphone amp is in order here. This would add a little more control to headphone levels, offer more jacks, is pretty cheap, and gets good and loud:D
 
Back
Top