1. Register NOW and become part of this fantastic knowledge base forum! This message will go away once you have registered.

Just Moved OTB-Calibration

Discussion in 'Recording' started by mightyeskimo, Sep 11, 2012.

  1. mightyeskimo

    mightyeskimo Active Member

    Ok, so I just moved OTB by building my own 8 channel passive summing mixer (DIY Recording Endless Summer) and I would like some input on calibration. Here is my setup:

    Logic Pro with a MOTU 828mk3 Hybrid

    All 8 analog outputs are being used as 4 pairs of stereo stems (1/2, 3/4, 5/6, 7/8)

    Each stem inputs to the 8 channel passive summing mixer

    Stereo output of the summing mixer into an ART Pro MPA II (2 channel pre)

    Outputs of the pre input into an ART Pro VLA II (2 channel compressor)

    Outputs of the compressor back into the MOTU on inputs 7/8


    What's the best way to calibrate this setup?
     
  2. Boswell

    Boswell Moderator Distinguished Member

    What are you aiming to achieve by calibration? Do you want relative calibration, i.e. to set your passive mixer up so that all the channels have the same attenuation with the faders set to 0dB, or do you want absolute calibration, where the total attenuation through the passive mixer is a specified amount that can be exactly compensated by using gain in your post-summing amplifier?

    While it's helpful to know that all the channels have the same attenuation so that you can swap channels around if necessary without having to adjust trims or faders, in practice it's your ears that should have the final say, and the actual positions of controls end up where your ears say they need to be.
     
  3. mightyeskimo

    mightyeskimo Active Member

    Absolute calibration. This will be my setup for printing a mix so I want to be able to make up for the gain loss plus have the ability to add more gain for a louder overall mix.
     
  4. Boswell

    Boswell Moderator Distinguished Member

    OK, that's pretty easy if you can get your DAW to generate a 1KHz sinewave to give a level of +4dBu out of your 828Mk3. You take that into a channel of your passive mixer set to 0dB on both the channel fader and the master fader, and adjust the gain of the make-up amplifier so you get back to +4dBu again at its output. Alternatively, you could choose to take the output level up to, say, +24dBu (i.e. a 20dB nominal gain increase) if you really want high output levels.
     
  5. mightyeskimo

    mightyeskimo Active Member

    And using a multimeter, I can measure the voltage on the +4dBu output which should be 1.22 volts?
     
  6. Kurt Foster

    Kurt Foster Distinguished Member

    No calibration needed.

    .... it's no different than using any other mixer. Passive mixers typically have 15 dB or so of loss. You should be able to make that up with the a pair of mic pres. Just crank it up until it comes up to the level you want to print the mix.

    -15 digital scale is about 0 Vu analog. Don't compress the 2-bus. Leave some meat on the bone for the ME.
     
  7. Boswell

    Boswell Moderator Distinguished Member

    Yes, as long as you have a good multimeter that will measure true r.m.s. of 1KHz sinewaves.
     
  8. mightyeskimo

    mightyeskimo Active Member

    I have an Innova Equus 3320 auto ranging dmm.
     
  9. Boswell

    Boswell Moderator Distinguished Member

    Hmm, you might have to drop your sinewave frequency to 400 Hz.

    Absolute measurement accuracy is not hugely important, but repeatability is. If you can get a reading of somewhere around 1.22V rms on the interface output, adjust your makeup gain to get the same figure on the pre-amp output. Repeat using the other channels - they won't all read exactly the same, but you can easily see if you have any rogues.
     
  10. mightyeskimo

    mightyeskimo Active Member

    I just searched the manual for my dmm and the word 'true' was not listed. Also, each instance of 'RMS' in the manual had no further details to the extent or accuracy of that measurement. So, as this was the one i could afford, it seems I'm stuck with a dmm that is not ideal. However, I completely understand the concept of repeatability of measurements. As long as I'm careful and perform as many tests as possible in a logical manner I'm sure I can get this system in the ballpark.
     
  11. RemyRAD

    RemyRAD Well-Known Member

    Your digital multimeter is just fine. All of those digital units reflect a high degree of RMS accuracy. These are not volume level meters, and not intended to be used as such. They are designed for single sine wave audio frequencies. And will reflect a correct RMS voltage reading.

    Even other studio items that have been calibrated in the past, were able to be calibrated later, at different levels, i.e. analog tape recorders. In those it would depend on what type of tape, you were utilizing. So calibration levels for analog tape changed radically as tape formulations improved over the years. Whereas their input and output calibration levels were and still are referenced to +4 DBM (the M, indicates a 600 ohm load), which you are not loading into nor have to. Most levels today are calibrated to +4 DBu, which generally indicates loading into 10,000 ohm inputs. And not 600 ohm inputs, as in yesteryear. Either way you're DMM is perfectly fine.

    Your signal chain should work just fine, if you tweak your levels properly. Bottom line is that you are working in the +4 DB 0 VU level realm. Which means peaks will be 12-16 DB beyond +4. So your output 0 VU from your compressor should be at 1.23 V RMS and should calibrate to around -18 on the digital meters in your software. This will ensure that you will have adequate headroom without distortion, for recording. You could also live dangerously and utilize -15 or, -12 as your 0 DB VU, reference level. Because 0 DB on your digital meters indicate the speed of light, which cannot be surpassed. Today in digital as compared to yesteryear analog recordings, we are now dealing with some very different metering. Metering that actually reflects more of what we are recording than the old-fashioned averaging analog VU meters did. We knew there were peaks beyond what we were looking at on VU meters. And we knew what kind of peaks our preamps were able to deliver. Some of this comes from very specific testing procedures, and some of this comes from just plain listening. Because if you don't know what overload sounds like, you shouldn't be recording for anybody. Peak red lights do not indicate that the audio is at its peak. It actually indicates you have just run a red light and will likely cause a crash, a.k.a. horrible clipping distortion. An occasional blink of a red light can and should be expected from time to time on transient peaks, generally coming from drum sets. Otherwise, you don't want to go to the red light district. In some places, it's fun. In other places, you're out of business.

    I like to play 0 DBFS roulette, and I like to land on black.
    Mx. Remy Ann David
     

Share This Page