A singer’s Logic Pro recording setup, with a Neumann TLM 103 and a Universal Audio Apollo, had been adjusted in so many places that it made more sense to strip it back and set it up again from the start.
A client had recently retired and wanted to record his own singing over backing tracks as a hobby. He’d bought a Neumann TLM 103 condenser mic, a UAD Apollo interface and some of UA’s vintage desk emulation plugins, set it all up on his MacBook with Logic Pro, and made a recording. He wasn’t happy with how it sounded and couldn’t understand why. He’d tried changing things in Console, the volume and the routing, and following tutorials hadn’t got him the sound he wanted. He wanted to spend his time singing rather than engineering.
Starting again
We connected on a WhatsApp voice call and AnyDesk. There was a bit of trickiness at the start getting AnyDesk’s permissions allowed in macOS, which is a normal thing to walk through these days, so I talked him through where to click and then he could share his screen with me. I’d offered to install Audiomovers so I could listen to his audio, but I didn’t need to hear it yet. I could see from his settings that the setup wasn’t optimised.
My approach in a case like this is to strip it back and confirm each part of the signal path is right, rather than pick apart what’s already there. So we deleted his Logic project and opened a fresh one. I showed him how adjusting the faders in Logic to change his volume could put his music out of balance, and suggested he leave them alone and use the volume knob on the Apollo instead. It sits by his right hand, only a little further away than the mouse, and it makes good use of a large hands-on interface.
Clearing out Console
In Console he had various plugins switched on, including some in the aux sends section. I showed him how to strip all of those away, and explained that every active plugin uses the Apollo’s DSP even when it isn’t needed. Once the preamp plugin was removed the gain dropped, so I showed him how to set it by hand with the mic input knob, and how he might need to change it for different kinds of vocal.
- Plugins removed from the Console inputs and the aux sends.
- Mic gain set on the Apollo’s input knob so the meter sat around -12 dB, bouncing in the green.
- Monitoring through Console, with the track armed in Logic using the R button and the I (input monitoring) button off.
We checked he knew what line level was. I showed him the two ways he could monitor his voice, through Console or with Logic’s input monitoring. As he’d be using the Apollo’s DSP effects, monitoring in Console made more sense. We checked the other way worked too, by muting the channel in Console and pressing I in Logic, so he has both, but I gave him the better one to use.
Logic wouldn’t record
Then we did a test recording in Logic and hit a snag. Logic flashed red but didn’t record anything. I suspected it was something to do with punch recording, and after a quick search I was right. The punch button wasn’t showing, so we added it to the control bar with View > Customise Control Bar.
Punch recording was switched on in Logic, but no punch-in or punch-out points had been set.
We switched it off and he recorded a bit of test speech so we could make sure there was no clipping. The signal was a bit low, so I suggested he boost it by another 10 dB, and after that he was happy with the sound of his voice. I also showed him how to change the level of his voice against the backing track.
Reverb in the headphones, effects afterwards
Back in Console, I showed him how to use the sends on his mic channel to feed Aux 1, where we put a reverb. That way he has reverb in his headphones while he sings, and he can change how much of it he hears. After another quick test recording, I showed him how to add effects in Logic once a recording is made. Rather than build a chain from scratch, we pulled up Logic’s Classic Vocal preset. The plugins in it can be swapped for better ones later, but Logic’s presets get you a good sound straight away.
He was delighted, and that was the end of our time. We planned to meet again the following week to build a vocal chain for his particular voice. He’s got professional equipment and it only needed setting up properly. After the session he wrote to say it had made a big difference for him, and that his vocals now sat well with his backing tracks.
When a setup has been changed in lots of places, a fresh project and a step-by-step check of the signal path usually makes more sense than trying to undo every change.
Not sure it’s the same problem? Ask a question first
