A couple thoughts for you:

1) Have you checked the manual for the proper procedure to level a mic input? That's the first step. Audio equipment has made me a manual reader where all other fields (except ax handling) have failed.

2) Is the microphone connecting into the mixer via an XLR or 1/4" connector? If the cable is connecting into the mixer with a 1/4" connector then the mixer is expecting a line level signal and you'll not be getting any volume.

3) Have you checked for any switches witch affect the output level of the board (mic vs line. -10db vs +4db) Have you checked to see what level your DAW is looking for? if you got a mismatch that may be your problem.

Good luck,
Adam

Thanks for the advice. Unfortunately, I have done all of that. The manual makes no mention of the issue. I am using XLR. And there is no switch looking for a mic vs line input :(

Typically speaking, do mixers contain an internal mic preamp, or is this something you should usually by separately? From what I've seen, most systems have the microphone connected directly into a mixer, without first going into any sort of amp.

You need to specify levels in dBFS peak as in "the signal peaked at -12dBFS" or dBFS RMS as in "I analyzed the signal and the RMS level measured -20dBFS". Just saying "low" gives us exactly zero useful information.

The instructions for setting gain are on Page 7 of the manual. Even by Behringer's standards the explanation is confusing. The procedure is simple. Solo the channel in question, speak or sing into the mic at performance level and turn that channel's gain knob until the meter shows signal hitting the zero mark without going over too often. Repeat for each channel. Additionally, each channel has a clip indicator LED that warns you when the level is too high. Since the mixer can only record the main mix through the USB connection the main faders may also affect the recording level.

The XLR inputs on your mixer feed microphone preamplifiers. There should be no need for more amplification if everything is set up correctly unless the board is faulty or badly designed (which may be the case).

Thanks for the reply. Although the manual is hard to understand, I do have the gain set up as you described. And, I'm sorry to say, but I'm not used to using any variation of dB units. What I mean by "low" is that, in my DAW, the signal shows up very close to the flat-line, such as in this photo which I took from elsewhere:

In the top track, you can see the levels being recorded (in my DAW, the signal is even "smaller" than what you see here in this photo that I stole from another forum). The bottom track shows what you get after normalization (which is closer to what I want my levels to be in the first place).

The smaller waveform is probably closer to correct. Tracking (recording new tracks) is not the time to try to get maximum volume. You do not want to burn up all your headroom until you have the project mixed down. Then you can finalize your mix to an appropriate volume. The numbers I used above to demonstrate dBFS are about the range you should be shooting for at the tracking stage.

That said, it's possible your levels really are too low. There has to be some sort of meter in your software that indicates your recording peak levels in dBFS.

There should also be a tool for magnifying the waveform. I often do that during editing to better see what I'm doing.

You are always going to normal a vocal track even when recorded on quality equipment. A good target average for peaks is somewhere between -20dB to -12dB. Especially on cheap equipment, it is very easy to overload the inputs of the preamp whether it be stand alone or the in built preamp of a Beh***** mixer. FWIW, the channel fader and the master fader have zero to do with the signal sent to the AD converter. That is controlled solely from the initial gain at the top of the stick.

Here's a question for you, is this device an amplifier?:

Amazon.com: Best Connectivity SD-AUD20040 7.1 Channel USB 2.0 Sound Box: Electronics

I ask because, while I am currently using the USB output of my Behrniger mixer, it does also have RCA outputs. When I choose to go RCA-out to my PC-in, I need to use this device. I connect my RCA-outputs to this device via a RCA-to-TRS cable. This device then connects to my PC via USB. When I go this route, I don't have the low signal level problem I've been talking about. The only conclusion I can come up with is that this thing is an amp. I simply don't like to go this route because (1) it just adds more stuff to carry around with me, (2) I'm stubborn and demand that the USB from the mixer works, because that's partly why I bought it, and (3) the quality isn't as good (I get better quality by recording at the low level and then normalizing).

Thanks again for all the advice so far.

Also, while I agree with your premise that it's good to not use up all of your headroom during the initial record, the image I uploaded was not a perfect representation of my situation (I took the picture from someone else who was having the same problem). Here it is again, slightly Photoshoped to better illustrate my levels in the DAW:

It seems a bit low. Hard numbers in dBFS are needed or it's all just guessing.

The reason you can get more gain NOT using the Behr*** ADC is that you have two more amplification stages prior to the extra USB ADC. At this level of gear however all you are adding is noise. Use one or the other with minimal circuitry in between the mic and the ADC. Normalizing the final result is standard procedure. Of course this is provided the peaks are at least -22dB. As Boulder states, we are just speculating without numbers. Your images mean nothing at all and in fact you may just be getting wrapped around the axle for no reason. A red herring as it were.

jgiannis, post: 392159 wrote: ....
Microphone (Shure SM58S) --> Mixer (Behringer Xenyx X1622USB) --> PC Win 7, recording with GoldWave software (using USB cable from mixer to PC)

I've tried boosting the fader on the microphone channel. The input in my DAW is still low!
I've tried boosting the master fader on the mixer's main output. The input in my DAW is still low!!
I've tried boosting the input levels within the Win 7 interface. The input in my DAW is still low!!!
I've tried boosting the input levels within my DAWs (GoldWave) interface. The input in my DAW is still low!!!!
....

Where have you set the gain pots and the compressors? (At the top of the board, near the XLR input)

Here is the mixer I'm using (you can zoom in to a good degree with these images): [[url=http://[/URL]="http://www.amazon.c…"]Amazon.com: Behringer XENYX X1622USB Premium 16-Input 2/2-Bus Mixer with XENYX Mic Preamps & Compressors: Musical Instruments[/]="http://www.amazon.c…"]Amazon.com: Behringer XENYX X1622USB Premium 16-Input 2/2-Bus Mixer with XENYX Mic Preamps & Compressors: Musical Instruments[/]

My mic (Shure SM58S) is coming into channel 1 via XLR. I set the gain knob (just under the XLR input on channel 1) such that I get the levels in the green-yellow region, and even into the red region when the speech gets intense. The compressor is off (I will be installing an external one soon). All EQ is off (except the "low cut" filter button that's located under the XLR input on channel 1). My AUX is not being used, and I'm panning in the center. My channel 1 fader is up to 0dB. My main mix faders are also up to 0dB. My USB cable goes directly into my PC.

I'm actually surprised to hear that the image I uploaded doesn't bare any relevance to the issue of "strong" or "weak" levels. I'm not saying I don't believe you (I'm sure you know your stuff here), I'm just surprised and perplexed by it. My understanding was always that, theoretically, the ideal recording would obtain the hottest-levels-possible without clipping, and therefore you obtain the strongest signal-to-noise ratio. In my case, where I need to normalize the audio, I was under the impression that I am simply raising the noise floor (which would consist of internal noise). By "internal noise," I mean that I've recorded silence before (hit my record button with my faders all the way down). When I normalize that audio (which appears to be a flat line), I hear the internal system noise (buzzing/hissing sound). Thus, when my waveform appears to be near silent (as in the photo), and I normalize it, I am boosting up that same noise floor. The system noise is always going to be there, so I just want to avoid boosting it. At least that was my understanding. I'm going to try and look up dBFS and other dB scales so I can see where you're coming from. In the meantime, I hope you can try to explain how dBFS is the more accurate measuring tool, and why my logic is wrong (if you do think it's wrong). I hope this doesn't sound cocky. I'm really not trying to be.

jgiannis, post: 392202 wrote: My understanding was always that, theoretically, the ideal recording would obtain the hottest-levels-possible without clipping, and therefore you obtain the strongest signal-to-noise ratio. In my case, where I need to normalize the audio, I was under the impression that I am simply raising the noise floor (which would consist of internal noise).

This is your first error in knowledge. The "max signal prior to clipping" idea is left over from pure analog recording days. If you are not recording to 1/2" tape then you need to forget this idea altogether. Your peaks should NEVER hit the red unless you want the desired resultant effects which you would have to have Remy's experience to utilize anyway. Suffice it to say, for most folks beginners and intermediate and advanced, peak levels while recording should be somewhere between -20dB and -12dB. Normalizing will not change the signal to noise ratio at all in the digital domain. I normalize normally when in Norman to -6dB.

Now, your graphical representation of the sound wave is just that, a representation. It is not the sound wave. Within a modern DAW program that is set to 24 bit recording, there is no possible way an analog signal can clip the digital bus and fill the track FFT visualization. Can't be done. Therefore what you really have to be concerned about is clipping the input and output stages of all the analog devices (or mixer sections). These clip points will vary based upon EACH device in your chain. This management is called gain staging and we had to be aware of it back in analog only days as well since again, each piece of equipment was different unless one custom calibrated everything the same.

Keep the signal from clipping at all points. Don't worry about the graphical representation as long as your dB peaks are between -20dB and -6dB.

Maximizing gain to stay above the noise floor is the realm of gear designed fifty years ago and some cheap modern gear. Using anything decent you will have much more room between the noise floor and clipping.

Same deal with 24 bit digital. Decent analog gear will have a S/N approaching 100dB. Digital audio at 24 bits has over 130dB of S/N. You can run your analog peaks 20dB below digital clipping and the analog noise floor will be no less than 10dB above the noise floor.

It's like driving on a two lane highway. The guardrail is the noise floor and the yellow line is clipping. If you go too far right there will be noise. If you drift too far left you will get clipped. Give yourself some room on both sides with a bit extra on the left since that's more catastrophic.