How Do Techs Magically Pinpoint a Problem Frequency?

Tweet about this on TwitterShare on FacebookShare on Google+
It takes a few attempts.

It takes a few attempts.

So there I was, hunkered down in the sound booth with the congregation rioting around me.  Two instruments were vying for the same dominant frequencies and I could hear an elder yell, “MAKE THIS NIGHTMARE END!”  Sweat was pouring down my face.  “Think man, think,” I told myself.  “You’ve trained for this very type of scenario.”  My hand reached for the channel EQ.  I moved the mid-range sweep knob to 1257 Hz.  Suddenly, confident of my next move, I applied a 6 dB cut to that frequency…and the congregation went wild!

This story seems outrageous but in the mind of some audio techs, it reflects a question I occasionally get via email; “how do techs pinpoint a frequency so easily?”  

There are four ways that techs learn to pinpoint frequencies…but “pinpoint” isn’t the best description.  Let’s look at the four and you’ll see what I mean.

1. It’s what I do every day.

It takes 10,000 hours to master a skill.  Working weekends and maybe a mid-week practice, it would take 24 years of working 8 hours on live mixing each weekend, every weekend.  Professional audio engineers are putting in a lot of more time and thus they have trained their ears to identify frequency areas in relationship to vocals, guitars, drums, etc.  Even with that type of near-every day experience, could they pinpoint a specific frequency?  No.  They would be able to be very very very close in finding the frequency area for their first modification.

2. I trained my ears.

It is possible to get a jump on mastering frequency area identification if you train your ears.  When it comes to this, there have been a number of products which help with this training.  Quiztones is a great one.  Some people have golden ears and it’s easy for them to identify the frequency area they need to change, but for most people, it takes training your ears.

3. I learned the common frequency areas.

Each instrument and vocal has a set of audio frequencies that are known to affect the sound in a certain way.  For example, here are two key vocal frequency ranges and their use:

  • 100 Hz – 300 Hz : Clarity / Thin (Good for cutting)
  • 400 Hz – 1,100 Hz : Honky / Nasal (Good for cutting)

Knowing these frequency ranges, you know the frequency area you should first investigate when you have a problem with your audio channel…or you don’t have a problem but you want to improve the sound, such as add presence to a vocal or an acoustic guitar.  My guide, Audio Essentials for Church Sound, covers all of these frequency areas for the different instruments and vocals.

4. I learned to sweep.

Looking at the previous three points, you’ll notice I didn’t mention how one learns to “pinpoint” a frequency such as in the 1257 Hz in the above story.  It’s because you can’t.  The best you can do is come close on your first attempt and from there, “dial it in.”  Let’s look at how you’d do this.

First of all, it doesn’t matter if you are on an analog or digital board as the concept is the same.  Let’s go with an issue where your vocal is a bit on the nasally side (some vocalists are like that, you work with what you’ve got).  Focus on the below rage:

400 Hz – 1,100 Hz : Honky / Nasal

Go to your mid-range sweep EQ knob and move it to 400 Hz.  Next, set the mid-range cut/boost knob at +6 dB.  You have now applied a 6 dB boost at 400 Hz.  Using the mid-range sweep knob, slowly sweep your frequency center point up until the nasally characteristic jumps out in the mix.  You might find it at 800 Hz or 1,011 Hz or 1, 100 Hz.  Once you find the right point, cut the frequency area to the amount you need.  +6 dB is used because it’s easier to listen for extremes and adjust once you’ve found the point.  The next time you’re mixing that vocalist and they sound nasally, you have a good idea of the frequency area you should give that initial cut before sweeping a bit to make sure you’ve dialed it in.

The Take Away

I encourage you to focus on the last three points.  Start by learning the key frequency areas for instruments and vocals.  Then check out Quiztones and train your ear to recognize those frequency areas.  Finally, learn to sweep.  Even if you don’t use Quiztones or regularly train your ear, if you know the ranges and learn to sweep to find the right spot, you’ll be doing great!

Subscribe to our mailing list

Comments

  1. says

    You can train your ears to be an octave band analyzer in less than an hour with some quality time with a graphic EQ, lower number of bands better to start. Most readers of this post with a windows PC have Windows Media Player which has a built in octave band graphic EQ. Start with your favorite music and a known set of cans and independently raise these bands below. Do this with about 5 different songs that you know well sonically and it will start to imprint the character of the various bands into your brain.

    If you don’t have that on hand, you can use the vowel formant dominant frequencies as a guide.
    1. Oooo sound, long u. If ‘boom’ is what the offending frequency sounds like, it’s almost guaranteed to be under 500 Hz. Start at 500 and sweep down. Can be anywhere between about 200 Hz and 500 Hz.
    2. ‘Oh’ sound, long o. This is usually between 500 hz and 750 Hz. You won’t have to sweep much because by the time you are at 1000 Hz, the vowel formant changes significantly.
    3. ‘Ah’ sound, short o. This is 1000 Hz. Again, you won’t have to sweep very much if you start at 1000 Hz and sweep slowly down then up from 1000, if the offending frequency sounds too much like ‘pop’, ‘box’, ‘knock’ ‘la-la’, etc. you have a hot spot right around 1000 Hz.
    4. short a. If there’s too much ‘smack’, ‘snap’, ‘crack’ etc., the offending frequency is normally between 1500Hz and 3000 Hz. Start sweeping at 2 kHz, sweep down a little then up a little.
    5. ‘Eeee’ sound, long e. Anytime this is over-emphasized, you will notice it in vocals where any long-e sound is way over emphasized. 4 kHz is where to concentrate your efforts, start at 4kHz, sweep down a little, and up a little from there – you’ll find the offending frequency right in that vicinity nine times out of ten.
    6. ‘Shhh’. 6-8 kHz
    8. ‘sssss’ – anywhere above about 10 kHz can cause this.

    • says

      Scott, thanks for those details. I’ve never seen it explained that way with the vowel formations. When I first got into this, I found frequencies that I thought would be “high” were more mid-range.

  2. Quaid says

    When I sing at a place where we set up our portable system, and the mixer is set up offstage, obviously I can’t adjust the channel EQ, or the house eq for what I like. I’m left to trust others, who aren’t as particular as I am, when it comes to knowing what frequency range does what.
    So, in these curcumstances, I have 2 split track recordings on a labtop, which both have the demonstration on one channel, and me singing with it, on the other. The vocal has no EQ applied, and was recorded with the mic I used at the time to sing with. (I’ve obtained another mic since then, so I technically should re-do the vocals.)
    The idea is, I can play the songs, adjust the house EQ and the channel EQ for my track till I sound better, and then transfer those channel strip settings to the channel that the mic is plugged into.
    If there’s a bad frequency, I can find it that way, and adjust the channel EQ/house EQ to correct it.
    Then, as long as the channel settings aren’t changed when I’m onstage, and my voice is in good shape, things will sound like what I think is good, or at least, better than what it’d be without the tuning. (unless it’s a room where I can’t really get it to sound good in, but anyway, the idea is using the tracks to improve the system’s sound)

    • says

      here’s a thought – you can separate the system tuning from the channel EQ.
      If you take time (not during soundcheck) to tune the PA correctly with a measurement mic and DSP then you’ll know the PA will at least reproduce your program material correctly. Once you’re in the room you can tweak it further by listening to known source material & using a sine wave generator to find low end room resonances. Then in theory the PA sounds as good as it can in the room and you could mix with headphones or in-ears (or studio monitors) even if you can’t hear the PA. The biggest problem you’ll have is dynamics changes, but I think your method of using something that you know what it sounds like is a good start.

      • Quaid says

        Yeah, the house EQ should be set separately from the channel EQ. If I decide to add a lot of highs, for example, on the vocal channel of the split track recording, then I might cut the house EQ too much in the higher frequencies, then everything will sound too dark that doesn’t have the highs turned up on it’s respective input channel.

        Once I set the house EQ, then I’d use the recording of “me” to tweak the mixer’s channel settings to improve the sound of “me”. Then I’d put those settings on the channel that my mic is plugged into.
        Looking over what I wrote in the other post, I could have clarified what I meant.
        If I had a digital mixer, I could develop a preset and save it.

  3. Mike says

    I generally use the cut and sweep method instead of the boost and sweep method because in some cases dealing with in ear monitors from FOH, the channel sends to monitors are not pre EQ. A church i used to be on staff at was sued by a musician who claimed to be hit with a spike at 4k through his Aviom mixer while the FOH engineer was sweeping frequencies to remove harshness from the guitar players tone. I am always cautious with the boost technique anytime IEMs are involved. The cut method is harder to use because you do have to have your ears trained for a more subtle change in the sound but is far safer with IEM set ups and potential feedback issues. I have had to train all our sound techs at the church I am on staff at now in the cut and sweep method due to the fact we are all IEM from FOH and channel sends are not pre EQ. Anyway, just another angle on EQing techniques.

  4. Adam says

    Best training ever is spend the money and get an app called Quiztones. It keeps me tuned into what frequencies sound like. Great sound training app!

  5. Jeff Carter says

    Occasionally, the offending frequency is easier to find if I *boost* the mid-range EQ by 6 dB or so and then sweep the frequency until I find the spot where it sounds worst. (I think I first read about this in the church tech blogosphere, or maybe it was Pro Sound Web…)

    Obviously, this technique works best in rehearsal or sound check rather than your service.

    • says

      Jeff, great point. I’ve done it both ways. The more I thought about your comment, the more I thought it best to update the post to go with the BOOST / sweep. Oh, and yes, definitely DURING the soundcheck, not the service.
      :)

Leave a Reply

Notify me of followup comments via e-mail. You can also subscribe without commenting.