Compressor usage/setup questions

[Phil_RC_1]

OK, I've done my homework (I think:)), or at least I have read 3 or 4 papers on compressor setup and have a couple of successful setups. But I still seem to be having issues with compressing vocals properly. BTW: This is for live performances.

1) I have a basic Behringer compressor on the mains out, which is setup primarily as a peak limiter simply to protect the amps and speakers. No problem here.

2) I have a DBX 266XL on the snare and kick. Setup with a good amount of compression and a long attack/quick release time to help give more impact. The snare channel is setup a little different then the kick channel, and when I make an adjustment, I have been getting the expected results. no problems.

3) The lead vocal on the other hand is where I'm having problems. I just don't get the results I expect when I make an adjustment. Here again, I have one of the cheaper DBX 266XL compressors. I want his voice to "stand out" from the mix without being overly loud or over-driving the system. The theory or concept of what I'm trying to achieve is similar to how TV commercials are compressed and seem louder then the program even though there peak level isn't really any higher or louder.

Basically, I am compressing the vocals pretty hard, about 4 or 5:1, then setting the threshold around -20 to -30 dB then increasing the gain to compensate for overall level loss. This should increase the "average" signal level of the voice while the peak will not be any higher then the original non-compressed signal (Ideally:D)

When I do this, the channel becomes very sensitive and I get all sorts of feedback issues.

Is my approach just wrong? Would a higher quality compressor help? Any Advice??

Of coarse, the feedback issues could possibly be from something else all together. Since compressor setup is new for me, I'm just trying to make sure it is, or is not in the compressor setup.

Thanx, Phil

[wg_ski]

The theory or concept of what I'm trying to achieve is similar to how TV commercials are compressed and seem louder then the program even though there peak level isn't really any higher or louder.


Multi-band compression. It takes some serious fiddling with to get right, but it's how to get that super-fat "FM radio announcer" voice and still have it be intelligible (retain some dynamics and HF response). Of course, you need a crossover and two (or more) compressor channels.

If you have a friend in the TV or radio biz they'll know how to set it up - to make use of every milliwatt and every percent of modulation index.

[billfitzmaurice]

Read this:
http://www.rane.com/note134.html

Heavy compression won't bring vocals out in a mix, heavily compressing everything else will.

[Phil_RC_1]

Read this:
http://www.rane.com/note134.html

Heavy compression won't bring vocals out in a mix, heavily compressing everything else will.


Good basic info at that link Bill. That's one of the ones I've read before. It's a bit basic and doesn't exactly go into what I'm doing, but thanx.

[Phil_RC_1]

Multi-band compression. It takes some serious fiddling with to get right, but it's how to get that super-fat "FM radio announcer" voice and still have it be intelligible (retain some dynamics and HF response). Of course, you need a crossover and two (or more) compressor channels.

If you have a friend in the TV or radio biz they'll know how to set it up - to make use of every milliwatt and every percent of modulation index.

That makes since but I don't have the equipment to go into that. I would like to find a good paper describing the technique. Maybe it can be done with some low budget Behringer stuff.

For now I guess I'll have to settle for simple compression and riding the gain I suppose.

[wg_ski]

Don't have the equipment? That's hard the believe - a CX2310 and both channels of a dbx 266 isn't that hard to come by. I don't know of any papers, but the technique has come up several times on prosoundweb back when I frequented the place. I've done it for sports announcing, and for Making The Kick Drum Sound Like It Does On The Album. Not difficult in principle, but a lot of knobs to twiddle.

[meep]

I would never compress vocals that hard unless trying to cover up serious problems with the vocalist-- inconsistent voice, poor mic technique. I wouldn't go much beyond 2.7:1 or so, and even then, if the singer is good you shouldn't need much.

Yes, this increases feedback. it also raises your noise floor.

What WILL help them stand out is to get your eq's right. Reserve the freq's needed for vox, for the vocalist. vocalists need that 400-4khz range, especially 900-3k for intelligibility (give or take...). So don't boost highs on other instruments. guitars and keys get mids. bass gets bass. hat gets highs. Few amateurs ever figure this out--- everyone wants to make each instrument sound ideal, but a good mix is a good mix, which means many comprimises are made for the overall best sound.

Also, NEVER BOOST, always cut with your EQ's. any use of eq adds phasing. so if you're going ad phasing, do it will cutting, where it won't muddy up the sound as much.

[Randy L]

Try this....don't max out the instrument inputs. Turn them down so they barely peak at "0". Set the vocal input(s) hotter at ~ +3 or so. This should noticeably bring out the vocals over the instruments...but it greatly depends on how strong a voice the vocalist has too. The final mix level can be adjusted at the processors down the chain to maximize the amp's output levels to the speakers.

I read some time ago that compression raises the signal's RMS output, even though the peaks are trimmed. The higher RMS may be the cause of the feedback. What mics are you using for vocals? Maybe look into a quality super-cardioid for maximum off-axis rejection.

[KevinN]

Basically, I am compressing the vocals pretty hard, about 4 or 5:1, then setting the threshold around -20 to -30 dB then increasing the gain to compensate for overall level loss. This should increase the "average" signal level of the voice while the peak will not be any higher then the original non-compressed signal (Ideally:D)

When I do this, the channel becomes very sensitive and I get all sorts of feedback issues.

I ran into this problem myself when I first started mixing live. It took me a few days to puzzle it out, but it's quite easy to see what is going on when you realize what the compressor is doing without any signal applied - it's not compressing!

So... take the compressor is out of the picture, and set the mic channel up for maximum gain before beedback. Now slowly push the gain up 10-20 dB past that point (representing the full make-up gain of the compressor, which is applied when it fully releases).

Does the resulting humming/howling familiar? It will sound a bit different, as a compressor on the channel would respond to the feedback with its own oscillating attack/release cycle, which will vary the level and frequency of the feedback.

[meep]

actually, the feedback comes because he is raising the pre-compression gain so that he can later squash the signal. Gain up, feedback on.

running +3db on gain for the vocals may or may not work. there is much more in play than *just* volumes. +3db might make it stand out, but it also may take it out from sitting in the mix.

That said, however you should never have to run above 0. cut the interfering noise and frequencies first.

[Phil_RC_1]

Thanx again guys. You've given me a lot to think about and try. I'll be back to it this coming week for 2 weeks so I'll be able to give some of your suggestions a try.

Phil

[Phil_RC_1]

I would never compress vocals that hard unless trying to cover up serious problems with the vocalist-- inconsistent voice, poor mic technique. I wouldn't go much beyond 2.7:1 or so, and even then, if the singer is good you shouldn't need much.

Yes, this increases feedback. it also raises your noise floor.

What WILL help them stand out is to get your eq's right. Reserve the freq's needed for vox, for the vocalist. vocalists need that 400-4khz range, especially 900-3k for intelligibility (give or take...). So don't boost highs on other instruments. guitars and keys get mids. bass gets bass. hat gets highs. Few amateurs ever figure this out--- everyone wants to make each instrument sound ideal, but a good mix is a good mix, which means many comprimises are made for the overall best sound.

Also, NEVER BOOST, always cut with your EQ's. any use of eq adds phasing. so if you're going ad phasing, do it will cutting, where it won't muddy up the sound as much.

Considering the current setup on the console, I think I'll relax the compression a bit then look more into the EQ throughout the board.

When you say "reserve the freq's needed for vox", do you mean to cut those freq's in other instruments?

Thanx again, Phil

[deidrea8]

You should ask the experts at Ingersoll Rand Air (http://www.ingersollrandproducts.com/IS/Category.aspx-am_en-12769) They have a phone line with service and they can break it down to make everything seem easier! Good Luck!