How to avoid feedback when having heavy vocal effects during a live performance?

[Hyperwitch]

AUTO-ADMIN: Non-MP3, WAV, OGG, SoundCloud, YouTube, Vimeo, Twitter and Facebook links in this post have been protected automatically. Once the member reaches 5 posts the links will function as normal.

I recently had to perform vocals for an EP I released. Vocals are being radically altered through an FX chain.

Despite that I could perform it without feedback at home in front of my monitoring speakers, once at the rehearsal, I encountered a massive feedback problem (+ that night the usual PA was replaced by a sound engineering student) and I had to cancel all the tracks featuring vocals.

I am very new at performing vocals and I would like to know what I am doing wrong and how to find a way around or compromise to still be as near as possible to the EP vocal's vocal processing while avoiding any feedback issues.

My setup is the following:

Code: Select all (#)

IN: Mic → Scarlett 2i2 (soundcard) connected to MacBookPro/Ableton

OUT: MacBookPro/Ableton -> Scarlett 2i2 (soundcard)

FX on vocals


List of effects on the vocal insert:

Code: Select all (#)

1.utility (raise volume up)
2.vocal control multiband compressor
3.Auto-tune
4.Compressor
5.Reverb (25% wet)
6.Tube amp simulator
7.Compressor

Basically I then simply layer my live vocals on top of the instrumentals and chorus background:



The map of the sound system in the club I performed

The subwoofer was on the floor and the rest of the speakers were suspended 1m50cm above the floor. I had two behind me (my setup was put on the table in the picture), and two others cornered in the space.

[BertKoor]

Contributing factors in feedback:

* used mike. Best are super-cardioid dynamic mikes, such as Shure SM-58, worst are omni-directional large diaphragm condensor mikes. And how you hold the mike contributes.
* speaker placement. Directed away from the mikes please. Good monitors are in front, not behind you, in your face, not your back. In-ear monitor buds are preferred.
* EQ of the mike & PA or monitors as a whole. Find what freqs in that setup are suspect to go resonant: kill them. This is the sound engineers job, not for the artist.
* Overall volume. Can you turn all down? Do you really need to subject the audience to 105dB SPL?

This is all generic. But specific in your setup:
* Compression. Do you really need it? Three!?!? Where without it, the feedback is relatively easy to control for an engineer, compressors add considerable gain in ways an engineer cannot control and thus feedback can go totally out of control.
* Distortion / overdrive. Also this adds gain and thus causes feedback. Often high-pitched like 8-12 kHz while regular acoustic feedback is at 1-5kHz.

[metamorphosis]

As Bertkoor has said, and I would add, use a massive gate before all other effects. Overcompression in a live environment also not a great idea. Dynamic mic, get as close to it as you can. Loud is generally bad anyway, if you want people to be able to appreciate you, don't make them deaf.

[imrae]

Speakers should not be behind you, that is 90% of the problem right there and effects are secondary. I have performed at loud stage volumes (post-rock band) with vocal effects and had no trouble as long as engineering is competent.

Speakers should be in front of you facing audience, with smaller monitors pointing back to you. Another option if possible is to use in-ear monitoring and send nothing to stage monitors.

I was having a bit of trouble with drums opening the vocoder; although I didn’t get a chance to try it the Electrovoice ND96 microphone was recommended to me as having very good isolation and suitable for loud stages. But really you should be able to manage with an SM58.

It is probably a good idea to ease off the compression/distortion gain as well, you should be able to get enough for “flavour” without crushing the dynamics and this will increase the size of your feedback-free sweet spots. It takes some practice to sing into a microphone with consistent volume; you can use the distance from the microphone like riding a volume fade. (There are tonal effects too but mostly they are helpful.) If possible try to book some time in a practice space and get used to performing at stage levels. Consider hearing protection and get used to how that affects singing. I blew my voice out in first couple of gigs getting the hang of this.

A gate won’t help much with feedback unless it’s set up quite cleverly; once you have feedback the gate will stay open...

[metamorphosis]

Speakers should not be behind you, that is 90% of the problem right there and effects are secondary. I have performed at loud stage volumes (post-rock band) with vocal effects and had no trouble as long as engineering is competent.

Speakers should be in front of you facing audience, with smaller monitors pointing back to you. Another option if possible is to use in-ear monitoring and send nothing to stage monitors.

I was having a bit of trouble with drums opening the vocoder; although I didn’t get a chance to try it the Electrovoice ND96 microphone was recommended to me as having very good isolation and suitable for loud stages. But really you should be able to manage with an SM58.

It is probably a good idea to ease off the compression/distortion gain as well, you should be able to get enough for “flavour” without crushing the dynamics and this will increase the size of your feedback-free sweet spots. It takes some practice to sing into a microphone with consistent volume; you can use the distance from the microphone like riding a volume fade. (There are tonal effects too but mostly they are helpful.) If possible try to book some time in a practice space and get used to performing at stage levels. Consider hearing protection and get used to how that affects singing. I blew my voice out in first couple of gigs getting the hang of this.

A gate won’t help much with feedback unless it’s set up quite cleverly; once you have feedback the gate will stay open...

Yeah, that's not correct. You're forgetting that they're compressing the shit out of their signal.
Therefore it will currently be amplifying the tiny amounts of feedback that lead to loud feedback. So yes, use a gate if you're using a compressor. No questions.

[imrae]

It's quite simple.

1. Gate closed, no feedback
2. Singing starts, gate opens
3. Feedback increases to level of singing or louder
4. Singing stops, gate is held open by feedback
5. Feedback continues indefinitely