PA Help Needed

monquixote

Danny1969 said:

monquixote said:

Sporky;1017393" said:

Do you EQ the system to the room you're in? Rule of thumb for installed audio at least is that correct EQ nets you 6dB more gain before feedback.

We have done that when possible, but usually in a pub the landlord runs out and tells you off for making feedback noises.

Adjusting the EQ to suit a room can help with certain things, like uneven bass response but it's not necessary to get a decent modern PA like you have  loud and clear above a drummer with no feedback issues ...... to be honest any feedback issues with a Lucas and a decent desk would be down to wedge monitors, not FOH. And your not using any wedges!

You haven't said what desk it is ... if it's a QU or similar and your not very familiar with it to work out what's wrong you can just pop your show file onto a USB pen drive and email it to me or any other engineer on here that's familiar with the product and we can see exactly where your going wrong .... no need to drive anywhere

If it's an analog desk then the above point is mute but the fact you mentioned built in dynamics makes me think it's digital 

It's a QU I think. 

I'll see if I can get the config file and send it to you.

maltingsaudio

My first thought when I read it was the compressor, however is it possible to take a shot of the desk settings and compressor settings which would help diagnosis ? 

Roland

How loud is your singer's voice, and what's the mic technique,like? We once had a singer who wouldn't project, and didn't like getting her lips near the mic shield. We had to use too much gain on the channel, and always got feedback. We tried lots of technology, and changed lots of things. Eventually we changed the singer.

benmurray85

don't fall into this trap either. I hate hearing bands say this.....

feedback issues and everyone turns round to look at the guy controlling the PA and he says "its set exactly the same as it was at the last gig!" 

I've had major bust ups in previous bands over this

Sporky

Danny1969 said:

Adjusting the EQ to suit a room can help with certain things, like uneven bass response but it's not necessary to get a decent modern PA like you have  loud and clear above a drummer with no feedback issues ......

You can't get past the physics though; every live sound technician should know the NAG/PAG calculations and microphone polar patterns as well as being able to flatten the PA/room response.

Danny1969

Sporky said:

Danny1969 said:

Adjusting the EQ to suit a room can help with certain things, like uneven bass response but it's not necessary to get a decent modern PA like you have  loud and clear above a drummer with no feedback issues ......

You can't get past the physics though; every live sound technician should know the NAG/PAG calculations and microphone polar patterns as well as being able to flatten the PA/room response.

Well in a conference situation with multiple mics and speakers placed around the room NAG/PAG might be valid but for live music from a single PA stack your gain before feedback is more dependent on the level of vocal going into the mic than anything else. There's many a time I've mixed multi band bashs and some singers literally have twice the PFL gain than others through the same mic without me moving the gain on the desk

I'm not convinced you can compensate much for a rooms acoustics overall ...... if you tame one particular frequency so it's even in one part of the room it can cause a problem in a different part ...... I mean no one use's electronic processing to treat a control room, it's bass trapped and diffused acoustically in material rather than electronically. I used the db Driverack when it came out with white noise sweeps but wasn't convinced. These days I wander around with the iPad and listen in different parts of the room. With pub \ club sound it sometimes has to be a compromise as you generally have 2 stacks in one position and that's your lot. 

With multiple speakers and arrays, especially outside there's all kinds of options though to compensate for ground slopeage, humidity and what not. I have little experience of that, most of the stuff I engineer is indoors and around 200 to 900 peeps 

PolarityMan

For the uninitiated what's NAG/PAG?

monquixote

Roland;1017905" said:

How loud is your singer's voice, and what's the mic technique,like? We once had a singer who wouldn't project, and didn't like getting her lips near the mic shield. We had to use too much gain on the channel, and always got feedback. We tried lots of technology, and changed lots of things. Eventually we changed the singer.

I'm not sure how loud she is but she always belts it out right on mic and she uses a beta 58

Sporky

Danny1969 said:

Well in a conference situation with multiple mics and speakers placed around the room NAG/PAG might be valid but for live music from a single PA stack your gain before feedback is more dependent on the level of vocal going into the mic than anything else.

I've done a fair bit of music system installation too, not just conference stuff. EQing has always improved the level before feedback. Certainly with a typical music PA setup (I'll just say PA for music from now on if that's OK) you've got an advantage in that the speakers are usually in front of the mics, unlike in a lot of conference systems where the speakers are on the wall or in the ceiling... it got to the point where I was being booked to EQ pretty much every system we installed that had mics.

I'm not convinced you can compensate much for a rooms acoustics overall ...... if you tame one particular frequency so it's even in one part of the room it can cause a problem in a different part ......

I've encountered some weird rooms too - one had such a strong response that if you stood on the centreline and spoke it felt like you were mic'ed up, and another with a corner that resonated for 20-odd seconds if you whistled at about 900Hz. But you only need to flatten for where the mics are, feedback-wise.

I mean no one use's electronic processing to treat a control room, it's bass trapped and diffused acoustically in material rather than electronically.

Exactly - the room acoustics are fixed properly, which is a much better approach. I doubt many pubs or small clubs bother with acoustic treatment so a bit of EQ is useful.

PolarityMan said:

For the uninitiated what's NAG/PAG?

Needed Acoustic Gain and Potential Acoustic Gain.

As Danny says, most commonly applied to speech systems but the formulae are relevant for any PA system.

NAG is how much you gain you need to amplify the voice in order that people can hear and understand what's being said/sung.

PAG is how much gain you can apply given the number of open mics and the relative positions of the mics, loudspeakers and audience. The reason I mention it is that one factor in PAG is whether the system is EQ'd or not - if it is you can reasonably assume that you can get another 6dB of gain, which can easily make the difference between being loud enough without feedback and not.

maltingsaudio

Loads of wisdom going on here, the one thing I would add is that all rooms and set ups are different so if your looking for a one "set and forget" setting for the desk its just not going to be possible even if its the same band every night. The key to this is know your gear and know your sound so you can understand when it changes and how to correct it and  always allow an extra hour to problem solve.

With regards to setting up and room EQ, as both Danny and Sporky have said,do pay some heed to them, the simple way to do this is to play a track you know and love through the system and make it sound right by adjusting the output Graphic/available tone controlls not the channel ones. If you play the same track  and EQ it to sound the same  at every gig it will at least give you a starting point. As an exercises keep a note of the setting you use for this and compare them from venue to venue to see how the rooms are changing.     

Sporky

I've always done EQ with pink noise first, but it's really annoying for everyone else in the room. I never got the hang of doing it with music (though I always ran music through afterwards to check the end result wasn't unusable).

There are lots of good automated tools too; there is no shame in getting a technological helping hand, though it is handy to be able to do it by ear.

I should also point out that I'm pretty sure Danny has done this many, many times more than I have, so where I disagree slightly with him it's not that I think he's wrong, just different experiences and approaches.

ICBM

Same here - I've done a fair bit of sound work but not to the level of experience of some here.

My approach has always been the same though - EQ for the room first with all mic channels set flat and at the same level, by very slowly bringing the mains up until you start to hear the start of feedback and pulling down the culprit frequency in turn, until you can't go any higher without it being all at once or random - you don't need to actually make it howl - then once you've done that, back off a dB or two. That's your maximum level. Then set gain and EQ each channel if necessary to make them sound good, and not to fight feedback. Then set the channel levels as necessary to get a good mix. I would never use compression on a mic channel, only limiting (or a compressor set effectively *as* a limiter).

I would almost guarantee there are ways of getting higher maximum volume, but in my experience only at the expense of sound quality, and I *always* put sound quality ahead of maximum volume… which has not always made me entirely popular with bands, although I've never heard an audience complain - most often the opposite.

No numbers or calculations, just ears. The goal is to produce something that sounds like a good recorded track, since everyone out front instinctively knows what that sounds like. Not like a traditional 'band mix' where the instruments are way too loud and the vocals are muddy and inaudible.

ESBlonde

Though it's a few few years since I did this regularly, you do develop the knack of walking into a room and knowing/feeling the acoustic problems before you even load in. Placement of loudspeakers, monitors and microphones makes a significant difference to even the most high tech systems. Then EQ the system to the room. If you have done this right the vocal mics often have little eq movement other than for 'artistic' reasons. If you strap a comp accros the mix or vocal sub group you are asking for trouble. Inserting it on a single chanal is better but then it also imposes itself on the monitors, this can make the singist blow thier voice because when they push harder the voice/monitor doesn't. to get around this split each mic into two chanels, one with the comp for FOH and one natural for the wedge/monitors, this also enables different eq for the foldback if the 'artist' desires it.
Even without big boys toys and fancy systems, the correct placement and system EQ is still the first thing to attend to. Different mics makes and models will each have their own peaks and so present more points to excite feedback than a single type/model for everyone. A single type means you can cut the two worst offenders for feedback and still have a toneful system with good feedback resistance. After cutting more than three peaks you are hacking the life out of the tone and onto a loosing thing. a 31 band Graphic EQ is a bit like using an axe to take out a splinter compared to some of the digital filters available today, but if it's all you've got it can be used.
Try to keep the system sounding natural and it will cut through most things better. For Metal rock and screamers then that may not be true so much but still is a great rule of thumb.
So understand your microphones pickup pattern and response, then use that to your advantage.
Pink noise is good if you have an analyser of some sort, otherwise a slow swept tone from the lowest to the highest audible range tells you where frequencies are reduced or enhanced, the trick is knowing what is causing that peak or trough and taking appropriate action. Just using a graphic can exaserbate the problem sometimes by generating nodes around the source problem.
 

Sporky

ICBM said:

My approach has always been the same though - EQ for the room first with all mic channels set flat and at the same level, by very slowly bringing the mains up until you start to hear the start of feedback and pulling down the culprit frequency in turn, until you can't go any higher without it being all at once or random - you don't need to actually make it howl - then once you've done that, back off a dB or two. That's your maximum level.

I've done it that way and it's a really straightforward and effective approach. I reckon it gets you within a couple of dB (at worst) of the absolute maximum achievable level.

The way I prefer doing it - though as mentioned mostly with installed systems, though I'm not sure there's much difference - is to run pink noise in (while keeping all the mics open) with the system level just below where it starts to ring, then use a fairly sharp parametric with about a 6dB boost and sweep that gently up and down. At the room-and-system's resonant frequencies you'll hear it get much louder, so then I make the filter sharper and sharper until I'm confident of the exact frequency, then knock the level down to a cut - typically about 5dB seems to work OK. Once you've got about 5 or 6 frequencies done you're usually at the point where pushing the level any more results in unpredictable multi-frequency feedback  (as others have said).

Fundamentally it's the same approach, but I'm usually doing this while people are trying to work in the room, and experience suggests that the pink noise and sweep is less annoying to builders, consultants, installers and customers than just pushing the system until it starts to ring a bit.

The advantage of using parametric is that you can be a bit more precise so it has less effect on the sound of the system as a whole.

Danny1969

What I generally find with today's modern quality PA's and digital desks is there's so much clean power available that FOH feedback isn't an issue. As long as you don't do anything stupid in terms of mic \ speaker \ stack placement and don't attempt crazy hi mid boosts, or cane the compressor so it's squashing everything your generally safe. 

If there is any issues it's always monitors. Even with good placement and keeping the monitor in the mics dead zone there's a limit to how hard you can push the monitors before they will feed back. Just about every desk now has a 31 band parametric on every aux send now so you can generally stand there with your iPad on stage and ring out the worse of it but there will come a point where your cutting so much across so many bands your effectively turning it down

Now here's an interesting thing. The more stable and even the monitor amp  \ speaker is across the frequency range the louder you will get it before there's any problems. We have various monitors ranging from HK to Mackie to Thomann own crap and I can get a well designed wedge like an HK Dart a lot louder without any feedback than a cheap monitor.