Setting Gain Structure
Home > Home Recording Forum > Live Sound Reinforcement > Setting Gain Structure
Posted on Dec 28, 2008 10:07 pm
Doug Babbitt
Sound as good as you play
Member Since: Dec 23, 2008
I'll tell you what I've been doing, then I have a question:
I solo each channel, so that the level shows on the main meter, fader set at unity, adjusting gain until the level peaks rarely or not at all.
All channels are sent to the main l/r bus and when the main fader is at unity, the level remains well below peak.
The main l/r signals are sent through a pair of mackie powered subs to either 2 or 4 mackie powered mains. I've just started using some compression and EQ between the mixer outputs and the amps in the speakers, but I don't think there's much net boost or cut in the signals.
Finally, to get the right sound levels from the speakers, I generally set the gain on the speaker amps at between -20 and -30 dB.
Just about every discussion I see of gain structure focuses on preserving headroom and maximizing s/n ratio, and I think that's what I'm doing. It's also clear that I'm not at risk of running out of headroom until we get to bigger rooms.
My question is whether the speaker amps are the best place for me to be cutting the signal, or should I go to unity at the speaker amps and cut earlier in the signal chain?
Thanks.
[ Back to Top ]
Dec 28, 2008 10:14 pm I think you are doing it about right. And that's how I would do it. But, it really depends on what your noisiest piece of gear is. If it's your board, turn it down some. If it's the amps, do that same. Sounds like you are on the right track though.
Dec 28, 2008 10:31 pm In live sound, gain structure is subjective to the ambient SPL level of the stage in the room. Sound Reinforcement is just that.
The way you describe is mostly correct. You want to keep your amps at full. And then gain structure is adjusted for the room.
Fader at unity, then turning the gain up until desired SPL level is achieved. If that happens to fall below unity for that channel, then so be it. If it winds up being above. Then some adjustment in EQ is necessary. Remember, EQ is mightier then gain. Keeping the gain at unity, and finding the open holes in the EQ sound will give you excellent headroom.
Running a compressor on the outputs will kill the dynamic range of your PA. When a compressor is in the active range, it moves the entire signal closer to 1k range. Pulling it out of the low and high range. Depending on how you have the attack and release time set, you would hear one kick drum hit nice and strong, then thin and weak on the next. Putting a compressor and gate on the insert of the channel will keep the dynamic range of your PA in check.
Now I know that your thinking right now of how much money that would cost to do something like that. And yes! It does cost. To keep the cost down, you can run through a sub group insert. Yes it will add a small amount of noise to your mix. This is why compressors have a output level knob. Running the group at unity, and adjusting the level of the compressor will help in keeping the noise level down.
You can allow you over all mix to reach a level of +6 to about +9, on the output of the mixer. That should not clip the self powered speakers.
Try this out. See what you think of the overall mix at that point.
Doug BabbittSound as good as you playMember
Since: Dec 23, 2008
Dec 29, 2008 10:43 am Very interesting Rob. Out of a lot of gain discussions I've come across, you're only the second person to say that the amps should be set at full.
The other was a local sound and recording pro who was in the room for one of our shows. I introduced myself between sets and I told him I'd appreciate any input he might have on what I was doing.
First he had me adjust the speaker positions slightly, then he looked at the board and asked me how the gain was set on the powered speakers. With all of the channel faders, channel gain, and the main fader at or near unity he must have known I was cutting the signal at the amps.
He had me set speaker amps at full and just cut back at the main fader. Given your comments, I'm wondering now if he would have pushed the attenuation all the way back to the gain knobs if we had been getting ready for the sound check instead of starting another set in a few minutes.
Can you explain the theory behind the method? My process up till now was based on the idea that you should try to get the strongest clean signal possible at every stage of the sound system, starting at the preamps, which would ultimately result in the best overall signal to noise at the end.
That made some sense, but obviously I wouldn't have posted my question if I was completely sure of myself.
Thanks again.
Dec 29, 2008 11:27 am Sure, the theory behind this type of gain structure is that you want the signal coming in to be as clean and quite (Quite referring to signal to noise.) as possible.
By running the input signal at or just below unity, you can be assured that the signal to noise will be at it's lowest point.
Then by having the input of the amp all the way up on the input, (Remember, the knob is input, not output.) The amp will be able to bring the signal to it's full swing of the output transistors. Translation - More headroom.
So, low or at unity input. Followed by wide open input for the amp, equals more headroom and a cleaner signal path.
Noize2uCzar of MidiAdministrator
Since: Apr 04, 2002
Dec 29, 2008 12:03 pm I will second Rob on the amps setting at full. Although I'll leave the explanation to him as to why. I'll just muff it up if I try and describe it.
Doug BabbittSound as good as you playMember
Since: Dec 23, 2008
Dec 29, 2008 12:52 pm OK, I'll see how that settles in my brain after a show or two.
Let me ask another question and see if I can answer it correctly myself. I had a situation where I was using a pair of powered subs, using the internal crossover to a pair of powered mains. All of the channel gains, channel faders, channel strip EQs, and speaker amp inputs (I'm paying attention) were at unity. When I pushed up the main fader I found the mix to be overwhelmed with low-end boominess.
My immediate "fix" was to turn down the subwoofer input levels by about half. This resolved the boominess, but now I don't get much punch from the subs.
Since then, I've added a 31-band GEQ on the main mix and I'm thinking that the proper solution to the boominess is to turn the powered subs back up and resolve the problem with some careful cutting on the GEQ. As I think about it, turning down my subs had the effect of turning down the entire spectrum below the 120Hz crossover point, but the booming problem likely stemmed from a fairly narrow frequency band interacting with the room.
Am I on the right track? I'd really like to dial-in the bottom-end of the mix. We don't get complaints about the sound of the bass guitar and kick drum, but I'd like to hear compliments like we've gotten for vocals, guitars and keys.
Dec 29, 2008 01:22 pm The way I figure it, the 31 band GEQ on the final mix should be reserved for simply ringing the room out for bad frequencies. Beyond that, if boominess exists, it's likely a room over a sound, tho, dimishing some of the lowest frequencies would help, or, more than that, lower a band at a time and see which frequency is creating the boom...
Doug BabbittSound as good as you playMember
Since: Dec 23, 2008
Dec 29, 2008 01:46 pm I should have mentioned that in addition to the "boominess" I was getting feedback. I'm thinking these 2 problems are really one problem, which can be solved by ringing-out the room through the GEQ before the gig. I'll make sure I setup with some time to spare next time out, so I don't have to rush through the process.
More things seem to be fitting together every time I lay out a problem. With a few more fundamentals under my belt, the sound quality should be noticably better our next time out.
Thanks for testing my conclusions.
Dec 29, 2008 01:55 pm I was kind of expecting to hear you say that, as proper ringing out of a room should help in both instances.
If you are unfamiliar with the phrase "ringing out" read that section here www.homerecordingconnecti...tory&id=426 and you'd be surprised the difference it'll make.
Doug BabbittSound as good as you playMember
Since: Dec 23, 2008
Dec 29, 2008 02:24 pm Good article, maybe I'll ring-out a room at home to get a feel for it.
Dec 29, 2008 02:55 pm It's very easy, should only take a couple of minutes, but it can add headroom to your system and really help control those annoying frequencies that every room has, just in different frequencies.
Dec 29, 2008 07:53 pm dB. Thanks for the article. I must have over looked that one. Essentially, what he is saying though, Doug, is what an RTA(Real Time Analyzer) does. Now, if you can't afford one, 'cause they do cost a little, that way is great and I will try it since I don't have one, but an RTA is much more accurate and quick if you are in a pinch.
Dec 29, 2008 09:11 pm It's true that an RTA is fast and easy. But, yeah! It cost! I bought the Smaartlive 6.0. That set me back $800.00 just for the program. Another $600.00 for the mic, an Earthworks M30. And $550.00 for the Sound Device USBPre.
Much cheaper to bark into a 58 and slowly raise the EQ's faders.
There is one other way to run your subs that will help. That is to put the subs on an Aux. This way you just put what you want in the sub. Giving, yet again, more headroom. Since your not running the whole signal into the sub. And will help reduce feedback.
Doug BabbittSound as good as you playMember
Since: Dec 23, 2008
Dec 29, 2008 09:24 pm Actually, I was trying not to mention the fact that I own an RTA (DBX Driverack PX, which is made especially for powered speaker systems like I use).
Earlier in this thread I mentioned the gig where I imposed upon a local pro to assess my setup. He noticed the PX and asked me a couple of questions about settings. I was clueless and routed the signal around it as we adjusted speaker position and gain. I had read the manual thoroughly and thought I knew what I was doing, but it was clear to me at that point that I didn't have enough basic knowledge to understand what was going on.
Anyway, I pulled the PX and put it on the shelf until I feel more comfortable with the fundamentals. That's when I picked up the GEQ that I'm using now. The PX came with an RTA mic which I may try for ringing out. It also has a variety of crossover options, an auto-feedback system and a parametric EQ, so I'm thinking it may come in handy at some point.
Is it really faster to ring out a room with an RTA? I worry about angering club owners if a try to pink a room with customers in the house.
Another question: I've read elsewhere that some guys, instead of using an open mic in FOH, will open all of the mics on stage, and otherwise follow the same ringing out process. My guess is that this would be the way to go if feedback was expected to be a serious problem during the show. Otherwise, using a mic in a FOH position would be more likely to result in EQ adjustments that improve FOH sound.
Comments?
Dec 29, 2008 09:31 pm Bad bad bad!!!! never open all of the mic's on the stage to try and ring out your monitors or PA!
This could lead to some serious feedback! That could damage your system. Just using a 58 out at FOH will tell you where there is going to issues in the room. Monitors should be rung out one by one the same way.
One other good way to get a clean sound out at FOH, is to work with the monitors first. Then move your way to the FOH.
Once again, this will help you hear what instruments that you will need to run at a hotter level or not.
Doug BabbittSound as good as you playMember
Since: Dec 23, 2008
Dec 29, 2008 09:49 pm Good thing I never tried that method, eh?
I always try to get at least the monitor levels ok before turning up the FOH. I'm sure nobody else ever had to set up under time pressure and was pushed by the band to rush through the sound check so they could roll straight into the first set.
Actually, they seem to get a bit more patient with each gig, so it's not that bad. I think they got so accustomed to sound guys who didn't really care what they sounded like that they thought it was wierd when I asked them to take their time and get things right.
Dec 29, 2008 09:59 pm Yeah right. I never get rushed through sound check.... I don't think that I have ever had time to sit down with a band to tweak to my hearts content.
There where a few times that the band showed up right at doors. So, no sound check at all. That's always fun.
We had an incident where a group showed up 2 hours behind the time that they where suppose to be on stage. 6 guys bum rush the stage all yelling at my monitor engineer to turn it up! When they felt that he was not fast enough, They threw our brand new $3000 a channel wireless mics at him.
Good time, good times!
