Looking for help to understand singal levels in mics and audio interfaces

[Celdor]

Hello,

I am afraid I need a hand with some understanding signal levels.

I was essentially searching for reviews to couple Focusrite Scarlet 2i4 and Rode Procaster Dynamic Mic. It turns out that Focusrite is not powerful enough for the mic. I found a website The Rode Procaster Levels are Too Low & I Need More Gain! where someone complains about low level of sound using aforementioned configuration. I can follow the excplanation in general but can't figure out a few details. This is a direct citation of the part I don't understand:

The Focusrite Scarlett range, including the 2i4, is capable of producing a gain of 50 dB max compared to the 65 dB max on the Presonus AudioBox. That extra 15 dB would more than solve your problem. If you figure that you're capturing your speech around 74 dB with the mic around 6 inches to a foot away, that 50 db of gain is only going to push the signal to around 0.16 volts. The preamp is meant to boost the signal to line level around 1.2 V. You're still drastically below, but 10 more decibels of gain could push you to 1.6 V, well above line level.

I can't figure out, using this particular configuration, where 0.16 vols comes from and how it happens that number is an output from the interface. I do understand that in order for two items to work, the audio interface should be more powerful, so that output of a "line level" of a signal from the interface should be about 1.2V. I guess this has something to do with different signals levels I just read on Wikil

So, person answering the question says that speech is around 74dB. I take it as a fact. The difference between reference level 94dB and 74 dB is 20 dB. That makes:

20 = 20*log10(x) => x = 10

Therefore, considering speech level, signal this particular mic would provide to interface is `1.6mV / 10 = 0.16mV`. Because Focusrite amplifies a signal by 50dB, that is:

50 = 20*log10(x) => x = 10^2.5 ~ 316.2

the interface's output signal would be just 0.16mV * 316 = 50.6 mV! This is completely different from 0.16V. That's not even close and I don't know what I am missing.

Thanks in advance.

 

[keith.rogers]

I don't know how much you can believe that consumer interfaces actually differ by 15dB in the amount of clean gain they can deliver.

Are you whispering? Yes, the Procaster is on the lower sensitivity end much like the SM7b, but you should be able to get sufficient signal into the DAW and simply add a GAIN plugin to the track. If not, then look at the Cloudlifter CL-1, Fethead, or the sE DM1 recently mentioned here.

 

[ecc83]

That Rode mic is actually pretty much bang on the same sensitivity as the ubiquitous SM57/58 at 1.6mV/Pa and I think a lot of people use the 57 with the 2i2 and other budget AIs, most of which have only 50-55dB of gain*.

Yes, you will need the gain pot well advanced, maybe on the end stop but modern AIs are very quiet these days and that should not be an issue (see *) The problem here is I feel the USE to which the AI is being put. You see, for recording (24 bits) you only need the signal to average around -18dBfs on the DAW meters, about 1/2 way up if there are no dB marks. Even if the level is only -20 to -25dB you can still boost it digitally post tracking and get it close to 0dBfs with no worsening of the noise level. However for PODCASTING you probably want the live, real time level to be hotter? If so you have a problem.

As mentioned you can get inline spook juice powered pre amps but they are a bit pricey. Personally I would consider a capacitor mic with a good 15dB sensitivity boost over a dynamic.

* My NI KA6 has just 55dB and needs the gain flat out for a 57 on acoustic guitar but even so the noise floor is still -70dBfs and most of that is "room" noise even at 2am! I am tho' getting only -20dBfs or so.

Jusfort...you could look for a cheap, maybe S/H wee mixer?

Dave.

 

[rob aylestone]

The trouble is you're apply the maths in an abstract manner without context or accuracy. To make sense of any manufacturers specifications on sensitivity - then it's a voltage figure coupled to a pressure at the diaphragm. Unless you have both - and usually we don't - sensitivity is a kind of 'relative' figure. The notion of extra gain in a preamp also needs coupling to the signal to noise. Some preamps have lots of gain, on the knob, but at a certain point the noise increases non-linearly and becomes intrusive. This point is rarely detailed in the specs. Some preamps can be turned up to maximum gain, and be pretty noise free, but are often criticised as being 'low gain'. Other have loads more gain on paper, and loads more noise.

I have a couple of mics in my collection that are always linked to noise issues because they are 'low output' - which they are. For a given pressure level, their output is low. They work fine for me used as they were intended to be used, close in, mouth almost on the grill. I see the Shure SM7 often moaned about on video forums when mounted on a boom - quite a way from the actor's mouths. The gain required to get a decent level means a very low noise preamp is needed. If you are close to the mic - the only real requirement is decent level and low noise. I don't have the Scarlet, but a friend has one and gain and noise is never an issue with close miking.

The other thing to remember in your calculations is the inverse square law. The voltage at double the distance drops to a quarter of what it was. double that and it's a 16th. So mic outputs can be a huge range - quiet and distant will be microvolts, rather than millivolts. Inside a kick drum? wow! If you have a mic with a pad, 10dB can easily be missed without even noticing. On one of mine the most cut that can be applied is 18dB and while you see the result on the meters, it's not an immense drop in the volume when you engage it.

I'm not certain how you are using the voltage formula?
Voltage gain (dB) = 20×log (Audio output voltage / Audio input voltage). The input and output level component seems a bit wrong, isn't it? You need two accurate voltages to compare. Indeed, with a preamp with outputs at +4dB pro level there will be a difference from one working at the .775V level. The preamp could be boosting the level, then knocking it back a bit in the output stage. All these things impact, yet are rarely detailed in the specs - or even if they are, the measurement is not explained. I'd bet that many do what looks best on paper.

 

[ecc83]

The sensitivity figure for microphones is given as "X mV (or dBV) for one Pascal which is the SI unit of pressure equal to 94dB SPL and, AFAIK that means the pressure AT the diaphragm.

The pressure produced by any other sound, some distance from the diaphragm will obviously be, as Rob said proportional to the pressure of said sound and the distance.
It is quite correct to say that IF you know you are generating a pressure (at the D) 20dB down on 1Pa then the signal voltage will be 20dB lower but usually you don't!

But, Rob again, most folks with a "one and a bit mV" mic get on fine with the majority of AIs these days but only because they are happy to get neg 20 or so in the DAW. If you want to hit -3dBfs in real time you either need a pre amp or shout!

Best thing is to post (attach as a 320k MP3) ten seconds of "Mary had" and tell us where that sat on YOUR DAW meter.

Dave.

 

[Chili]

Yes, you will need the gain pot well advanced, maybe on the end stop
...Dave.

I think this is all that needs to be said. For a dynamic mic, don't be afraid to turn the gain up. All the way, if you need to.

To set the level properly on the 2i2, sing into the mic at a level slightly louder than what you'll record at. Turn the gain up until the LED flashes red, then turn the gain down until it stays a steady green. When you go to record at a normal singing level, don't worry if the LED flashes green or stays steady, it's all good. Just as long as you aren't hitting red.