Old Production Takes From an Old Guy

Using the Matrix for Consistent Record Levels

One of the challenges we face as church sound engineers is to get consistent levels on our recordings. Almost every church I know of records it’s services, either on Hard Disk, CD, Video, Cassettes (the horror!) or a combination of those. Unless you have a dedicated recording mixing position, getting the levels consistent on those recordings is tough.

The challenge is simple; the dynamic range of a typical service can be a good 20-30 dB. In our case, we typically mix our music in the low 90’s; say, 92-93 dBA SPL. Announcements, and teaching however, falls in the high 60’s; figure 65-68 dBA SPL. Some quick math means that we have a good 25 dB difference between the levels of our speakers and our band. Our ears are fine with that, because it makes sense in context. For a recorder, however, 25 dB is a big difference. If I set my input level on my CD-Recorder for good levels during the music set, my sermon will be way too low. On the other hand, if we try to maximize the level for the sermon, the music set will not be pretty.

One way to handle that is to manually adjust the input levels (or change the output level from the board, depending on how you are set up). But that can be tough, especially when going right from a song to a message, which we often do. I could run the signal through a compressor, but applying that much compression to music will suck all the life out of it. There are auto-level devices that do a decent job, but they are expensive and add another piece of equipment to the signal chain.

Having just installed our new DIGiCo SD-8, I think I’ve hit on a good solution using Groups and the Matrix mixes to control and equalize output levels. Here’s what I’ve done:

The first step is to identify my “groups” of levels. In other words, how can I place all the sources on my stage into buckets based on their level. I broke it up this way:

  • Mono Group 1—All Speaking Mics (I call it Spkrs to Mtx)
  • Stereo Group 2—All Vocal Mics (Vox to Mtx)
  • Stereo Group 3—All Instruments (Band to Mtx)
  • Stereo Group 4—All Playback (PB to Mtx)

On the SD-8, it’s easy to send channels to groups; just select the channel and turn on the group send(s). All my channels go to the Master group (which feeds my house PA). Each channel also goes to the appropriate group based on what it is. Unlike a mix bus, a group has no dedicated send control for each channel. It is simply assigned there by a switch. The level a given channel hits a group is based on the input gain and fader position.

Those four groups are routed into four channels of my matrix (actually, it’s 6, I do the same thing with Lobby & Cry Room); Video Rec L&R, CD Rec L&R. Very soon I’ll be using an external 4×8 system processor to split those up, but for now I’m splitting them on the board.

Routing signals into the matrix is easy on the SD-8On the SD-8, I have several places where I can adjust the level of those groups feeding the matrix. I originally set the relative levels on the matrix page, but then decided that was going to be too cumbersome to change on the fly. I decided to bring each group into the matrix at the same level, then adjust the outputs of the groups to equalize the levels.

In practice it looks like this.

Left to right the groups are Speakers, Band, Vocals & PlaybackAs you can see, the speaking mics are the highest, music and vocal mics are lower and playback is in the middle. When the output of the matrix channels hits the recorder, my levels are consistent throughout the entire service. The best thing about this is that the dynamic range of each section is in tact. I do apply some high-threshold, high-ratio compression to the outputs of the matrix sends to make sure we don’t clip the record levels, but it’s set to engage only rarely.

This process has been a great boon to my workflow. I don’t have to remember to adjust my record output any longer, and my levels are spot on all the time. Our video recordings are more useful now (now that the ALC isn’t clamping down on the music all the time), and board mixes translate much better. I’ve also safe’d the group sends and the matrix outputs so they don’t inadvertently get changed in a snapshot.

This set up is easy on the SD-8 as it has a flexible and customizable bus structure. You may have to play around with your console to come up with a similar routing system, but hopefully it gives you some ideas to get started with.

24 Comments

  1. jblasongame@gmail.com

    This is not too different from how we record our services (all of it) What I did was use and auxiliary send and simply sent more of the announcement/pastor mics to the recording to bring their level up and the overall mix level down in our record matrix and then used an external input on the matrix from the aux I’m boosting the pastor mics. On the reordings that’s brought the overall dynamic range of the service from 25dB to about 8dB. i tried fixing the dynamic range a while ago with the high compression thingy. I didn’t like it. I’ll hit my record mix about 4-6dB of gain reduction through the TC Finalizer but that’s it. Any more than that it starts sounding really squishy in a bad way.

  2. jblasongame@gmail.com

    This is not too different from how we record our services (all of it) What I did was use and auxiliary send and simply sent more of the announcement/pastor mics to the recording to bring their level up and the overall mix level down in our record matrix and then used an external input on the matrix from the aux I’m boosting the pastor mics. On the reordings that’s brought the overall dynamic range of the service from 25dB to about 8dB. i tried fixing the dynamic range a while ago with the high compression thingy. I didn’t like it. I’ll hit my record mix about 4-6dB of gain reduction through the TC Finalizer but that’s it. Any more than that it starts sounding really squishy in a bad way.

  3. frizbplaya@hotmail.com

    The matrix is real?

  4. frizbplaya@hotmail.com

    The matrix is real?

  5. justin.langman@gmail.com

    I would do the same, but the problem I find is that the Worship leader will always speak much softer between songs when they are talking to the congregation – praying/offering/chruch news etc. – than when they are singing. So their speaking is always quiet – even after you’ve pushed the fader right up, say 10dB, it will still be 10-15dB less than the music…

  6. justin.langman@gmail.com

    I would do the same, but the problem I find is that the Worship leader will always speak much softer between songs when they are talking to the congregation – praying/offering/chruch news etc. – than when they are singing. So their speaking is always quiet – even after you’ve pushed the fader right up, say 10dB, it will still be 10-15dB less than the music…

  7. mike@churchtecharts.org

    JB-
    That’s certainly another way to do it. I assume your aux sends are post fader so as to track with the changes your making in the mix. The only real downside to the aux send method is that if you need to change a whole group (say the band) at once, it’s a lot of knobs to turn. But if you don’t have groups…

    mike-
    Yes the matrix is real. But don’t tell anyone!

    Justin-
    The system is not 100% fool-proof. One thing you could do is double buss (or double-patch) your worship leader vocals and send the second copy of his vocal to the matrix in another group (but not the main house mix). Then push that fader up when he’s speaking. If you’re mixing on a digital console using snapshots (see Wednesday’s and Friday’s posts), it’s a walk in the park. If not, you can still do it, albeit with a little more effort.

    mike

  8. mike@churchtecharts.org

    JB-
    That’s certainly another way to do it. I assume your aux sends are post fader so as to track with the changes your making in the mix. The only real downside to the aux send method is that if you need to change a whole group (say the band) at once, it’s a lot of knobs to turn. But if you don’t have groups…

    mike-
    Yes the matrix is real. But don’t tell anyone!

    Justin-
    The system is not 100% fool-proof. One thing you could do is double buss (or double-patch) your worship leader vocals and send the second copy of his vocal to the matrix in another group (but not the main house mix). Then push that fader up when he’s speaking. If you’re mixing on a digital console using snapshots (see Wednesday’s and Friday’s posts), it’s a walk in the park. If not, you can still do it, albeit with a little more effort.

    mike

  9. chadgreen@charter.net

    Mike,

    We don’t routinely record our worship music. Is there a copyright issue for music that everybody needs to be aware of?

    Is there a lot of fader riding for a spoken word? I usually set my sermon level and wait for feedback.

    Are you saying that the sermon recording is PostFL for group send?

    Chad

  10. chadgreen@charter.net

    Mike,

    We don’t routinely record our worship music. Is there a copyright issue for music that everybody needs to be aware of?

    Is there a lot of fader riding for a spoken word? I usually set my sermon level and wait for feedback.

    Are you saying that the sermon recording is PostFL for group send?

    Chad

  11. matthew@MeMediaWeb.com

    We have an analogue board, so we use an Aux send (post fader, of course) for our recordings/overflow. It works great most of the time, but can offer its own problems (which you mentioned above).

    Matthew

  12. matthew@MeMediaWeb.com

    We have an analogue board, so we use an Aux send (post fader, of course) for our recordings/overflow. It works great most of the time, but can offer its own problems (which you mentioned above).

    Matthew

  13. tgibson@efccl.org

    Mike,

    A timely post for me. I had been working on a new system for recording and i will be combining your method with my new one. I’ll let you know next week how it works this coming Sunday.

    BTW, I’m on an A&H iLive T112.

    tim

  14. tgibson@efccl.org

    Mike,

    A timely post for me. I had been working on a new system for recording and i will be combining your method with my new one. I’ll let you know next week how it works this coming Sunday.

    BTW, I’m on an A&H iLive T112.

    tim

  15. tgibson@efccl.org

    Mike,

    On these 4 groups feeding the matrices, are you using any dynamic controls, or are the channel dynamics enough?

    Hope that makes sense.

    Tim

  16. tgibson@efccl.org

    Mike,

    On these 4 groups feeding the matrices, are you using any dynamic controls, or are the channel dynamics enough?

    Hope that makes sense.

    Tim

  17. mike@churchtecharts.org

    Chad,
    We record the entire service for archive purposes, and so I can listen to my board mixes afterward. There are copyright issues, but they are easily handled with the appropriate CCLI or Christian Copyright Solutions license. There are rules that need to be followed, but it’s not that hard to do.

    I don’t do much fader riding during the message as I have a compressor inserted in to the pastor’s mic channel. That handles almost all of the level control for the room, and it translates to the group send (which is post-fader).

    If you’re getting feedback, you either have speakers that are not properly set up, improper room EQ or the wrong mic on the pastor. Alternately, the pastor could be turned up too high.

    mike

  18. mike@churchtecharts.org

    Chad,
    We record the entire service for archive purposes, and so I can listen to my board mixes afterward. There are copyright issues, but they are easily handled with the appropriate CCLI or Christian Copyright Solutions license. There are rules that need to be followed, but it’s not that hard to do.

    I don’t do much fader riding during the message as I have a compressor inserted in to the pastor’s mic channel. That handles almost all of the level control for the room, and it translates to the group send (which is post-fader).

    If you’re getting feedback, you either have speakers that are not properly set up, improper room EQ or the wrong mic on the pastor. Alternately, the pastor could be turned up too high.

    mike

  19. mike@churchtecharts.org

    Tim,
    I am not using dynamics on the group sends right now. I compress almost every channel, at least a little bit, so that keeps the dynamic range under control. As I mentioned in the post, I do have some high threshold, high ratio comps set up on the output matrix, just to keep clipping from being an issue. I typically set my CD-R to peak at -6, which leaves me a little room for occasional high peaks. If I start seeing clip lights on the meters, I just back the matrix output down a touch.

    mike

  20. mike@churchtecharts.org

    Tim,
    I am not using dynamics on the group sends right now. I compress almost every channel, at least a little bit, so that keeps the dynamic range under control. As I mentioned in the post, I do have some high threshold, high ratio comps set up on the output matrix, just to keep clipping from being an issue. I typically set my CD-R to peak at -6, which leaves me a little room for occasional high peaks. If I start seeing clip lights on the meters, I just back the matrix output down a touch.

    mike

  21. tgibson@efccl.org

    Mike,

    That’s exactly what I’m doing with the comps.

    2nd question: how are your room mics sent to the recorders?

    tim

  22. tgibson@efccl.org

    Mike,

    That’s exactly what I’m doing with the comps.

    2nd question: how are your room mics sent to the recorders?

    tim

  23. mike@churchtecharts.org

    Tim,
    Right now I’m not running room mics to record. Our room mics are directly routed to our monitor desk until I switch over to our M-48s. At that point, we’ll get the new house mics (DPA 4098HBs) run into my FOH console, and will probably feed them straight into the matrix. Or maybe not… ‘;-)
    mike

  24. mike@churchtecharts.org

    Tim,
    Right now I’m not running room mics to record. Our room mics are directly routed to our monitor desk until I switch over to our M-48s. At that point, we’ll get the new house mics (DPA 4098HBs) run into my FOH console, and will probably feed them straight into the matrix. Or maybe not… ‘;-)
    mike

© 2021 ChurchTechArts

Theme by Anders NorenUp ↑