Recording and mixing an energetic gospel song
by Bruce Bartlett, Copyright 2011
One of the most memorable events in The Blues Brothers movie is the scene where a church congregation dances to the gospel band. The dancers do insanely high flips and cartwheels to this exuberant, joyful music.
I was honored to make a studio recording of similar music played by a top local gospel band, the Mighty Messengers. If your house of worship has a praise band who wants to record a CD, the tips in this article might help.
On the day of the session, we set up a drum kit in the middle of the studio. Surrounding the drummer were two electric guitarists, a bass player, a keyboardist and a singer who sang scratch vocals.
Because the bass, guitars and keys were recorded direct, there was no leakage from the drum kit, so we got a nice, tight drum sound. I miked each part of the kit. Kick was damped with a blanket and miked inside near the hard beater. We recorded the guitars off their effects boxes, so we captured the effects that the musicians were playing through.
I set up a cue mix so the band members could hear each other over headphones. We recorded on an Alesis HD24XL 24-track hard-drive recorder, which is very reliable and sounds great. Most of the songs required only one or two takes — a testament to the professionalism of this well-rehearsed band.
A few days later after mixing the instrument tracks, we overdubbed three background harmony singers (each on a separate mic). Finally we added the lead vocalist.
I copied all the tracks to my computer for mixing with Cakewalk SONAR Producer, a DAW which I like for its smooth workflow, top-quality plugins and 64-bit processing.
Shown below is a typical multitrack screen of the project:
Figure 1. The tracks in the project.
Here's a short sample of the mix of “My Heavenly Father” (written and copyrighted 2010 by Dr. William Jones). If your computer's mp3 player skips, right-click the mp3 file to download it before playing. For the best sound, play the samples through your studio monitor speakers.
Let's break it down. Click on each mp3 file below to hear the soloed tracks without any effects, then with effects.
First, the bass track.
To reduce muddiness and enhance definition in the bass track, I cut 5 dB at 250 Hz and boosted 6 dB at 800 Hz. Then the bass track sounded like this:
Figure 2. EQ used on the bass guitar.
Note that this EQ was done with all the instruments playing in the mix. EQ can sound extreme when you hear a track soloed, but just right when you hear everything at once.
Next, the kick-drum track.
To get a sharp kick attack that punched through the mix, I applied -3 dB at 60 Hz, -6 dB at 400 Hz, and +4 dB at 3 kHz. Here's the result:
Thinning out the lows in the kick ensured that the kick did not compete with the bass guitar for sonic space.
Now the snare track:
I boosted 10 dB at 10 kHz to enhance the high hat leakage into the snare mic. This is extreme but was necessary in this case. I also added reverb with a 27 msec predelay and 1.18 sec reverb time. Finally, I compressed the snare with a 7:1 ratio and 1 msec attack to keep the loudest hits under control.
The drummer did not play his toms on this song. But on other songs, to reduce leakage, I deleted everything in the tom tracks except the tom hits (see Figure 1). Cutting a few dB at 600 Hz helped to clarify the toms.
Next, here's the cymbal track unprocessed. Notice how loud the snare leakage is relative to the cymbal crash.
To reduce the snare leakage into the cymbal mics, I limited the cymbal track severely. The limiting reduced the snare level without much affecting the cymbal hits. This is very unusual processing but it worked. Listen to the cymbal-to-snare ratio with limiting applied.
The guitars sounded great as they were… a little reverb was all that was needed. These two players should do a solo album!
The keyboards also needed only slight reverb.
Moving on to the background vocals, we recorded three singers with three large-diaphragm condenser mics. Here are the raw tracks with the vocals panned half-left, half-right and center:
We applied 3:1 compession and added a low-frequency rolloff to compensate for the mics' proximity effect. I “stacked” the vocal tracks, not by overdubbing more vocals, but by running the vocal tracks through a chorus plug-in. This effect doubled the vocals, making them sound like a choir — especially with a little reverb added. Here are the background vocals with compression, EQ, chorus and reverb:
Finally, here's the unprocessed lead-vocal track. Listen how the word “never” is very loud because it is not compressed:
To tame the loud notes I added 4:1 compression with a 40 msec attack. Figure 3 shows the settings.
Figure 3. Settings for the lead-vocal's compressor.
The track also needed a de-esser to reduce excessive “s” and “sh” sounds. To create a de-esser, I used a multiband compressor plug-in, which was set to limit the 4 kHz-to-20 kHz band with a 2 msec attack time. This knocks down the sibilants only when they occur. De-essing does not dull the sound as a high-frequency rolloff would do.
Click on the mp3 file below to hear the lead vocal with compression, de-essing and reverb. The word “never” is not too loud now, thanks to the compressor.
The completed mix
We're done. Here's the entire mix again without any processing:
And here's the same mix with all the processing just described:
As we said earlier, we recorded the electric guitars playing through their effects stomp boxes, so I didn't need to add any effects to them. Those players knew exactly what was needed. For example, here's another song mix that showcases the slow flanging on the right-panned guitar. It was written and copyrighted 2010 by Dr. William Jones. By the way, this song is in 5/4 and 7/4 time.
Of course, every recording requires its own special mix, so the mix settings given here will not necessarily apply to your recordings. But I hope you enjoyed hearing how a recording of this genre might be recorded and mixed.