Need help w/Monitor Setup

[MarkGifford-1]

Hey guys,

 

I'm having some problems getting acceptable monitoring levels w/o feedback.

 

It's always been sort of an issue, but I've just started working with a 6-piece band who all sing, and it's really out of hand.

 

Gear:

 

Mixwiz 16:2, running 2 mixes, which consist of:

 

Mix 1: Ashly Parametric EQ > 1 Ch. QSC 2402 > 3 x EAW 2x10 wedges

 

These wedges were originally 4 ohms, I rewired them to 16 ohms.

 

Mix 2: Ashly Parametric EQ > 1 ch. QSC 2402 > 2 x EV S200 (the better molded plastic boxes)

 

Cabling is all 14G and in perfect working order.

 

Overall, I think that this is a pretty high-quality signal chain, but other systems locally use far worse stuff and seem to sound clearer and more open, w/o the feedback problems.

 

For FOH, I'm using QSC 3002, JBL SR's (12 and baby cheeks) over EV S18 subs, and the sound is always fine - super clear and balanced. Maybe 2-3dB of subtractive EQ for room characteristics

 

Potential problem areas:

 

1. Operator error. I've been doing this for a long time, but there may be something fundamental I'm missing. I'll get everyone's mic up there and open, and ring 'em out, but as soon as they start playing, it'll run away, or I won't be able to get enough gain for various folks (female singer, usually.)

 

2. 6 different mics: Shure wireless SM86, Shure wireless Beta 58, wired Beta 58, Shure PG-48, Audix OM-6, EV-357. Unfortunately, except for the EV (which is loud as hell and doesn't feedback) the band will not switch these, or standardize. I'll do a basic check with the EV-357, and the monitors are so loud it's ridiculous. Add the others in, and it gets squirrelly, fast. Especially the wireless mics, which have always been the bane of my existence...

 

3. The boxes. The EAW's sound kind of strange, could the impedance change be screwing up the way the components interact - there's a HUGE peak in the response around 1.5-2K, that's the real problem area feedback-wise. The EV's sound really good, but there's occasional problems w/them also. I'm presently looking for another pair, as they're very small and really kick ass as wedges. Would having all the same wedges help things?

 

Over the course of various bands I've been in, or worked with, I've had everything from no problems to barely being able to have them on. It would seem that the basic frequency response should be able to be set, and then just tweaked slightly for room variations. I just can't seem to get it there and I'm really frustrated and feel terrible for subjecting the band to this... I've got no problem making some equipment upgrades, the only parameter being size - I've got a full 6-space rack for FOH position, and a full 4-space for the power amps. Can't go any more than that.

 

Thanks in advance, and hopefully I've given enough info for you folks to help.

 

MG

[Rbts]

Whoa, I am hoping to hear what others more experienced than me have to say about this, but I could comment that I think generally, people use some sort of graphic EQ (31 band) to ring out their monitors, as they can carve a more specific troublesome frequency out. I do not know about the ashly unit you are using... maybe it is fine.

 

It is possible that you will need to be more forceful in getting the band to standardize the mics that are being used?

 

Lastly, you talk about setting things up using a EV 357 mic, which is working fine, and then trying to bring the others in and THEN having troubles. Maybe you could work the set up using the MOST OFFENSIVE mics, and gtry to get that set, and THEN bring in the ones that are seemingly more tolerant?

 

Just some thoughts, again, hopefully some of the more experienced helpers will be along any minute now.

[agedhorse]

How are you setting the parametric eq's? Unless you have a LOT of experience, or experience with an analyzer (including knowing when you are seeing uncorrectable data), they can be way more problems then they solve.

 

First, eliminate the parametrics (bypass them). Are your channel eq's flat? Monitor sends are often POST channel eq, so if you are eq'ing the mains, you are also affecting the monitors. If you hear something funny with the EAW's with no eq, then maybe something's wrong with them... perhaps they were "rewired" by someone who did not understand things? Maybe non-stock driver(s) or crossover parts?

[Mr.Talent]

Try turning the stage volume down... Beta58s will feedback before regular 58s. Don't hold the mic with your hand wrapped around the ball.. it doesn't look that cool and sounds even worse.. WTF did you wire a monitor cab for 16Ohms? Aux sends are usually wired post/eq and may have a pre/post switch.. actuall monitor sends are usually pre.

Lose the parametrics and get graphic EQs for your monitor mixes and maybe add a compressor..

Feedback will happen in multiples of the offending frequency, so notching out one won't always cut it- that's what graphics are for.

Analysers are for sissies!

[MarkGifford-1]

 

If you hear something funny with the EAW's with no eq, then maybe something's wrong with them... perhaps they were "rewired" by someone who did not understand things? Maybe non-stock driver(s) or crossover parts?

 

 

Well, I rewired them to 16 ohms so I could hang 3 off one channel of a power amp. I just switched the LF drivers from parallel to series. The problem is, I can't use 'em at 4 ohms, w/o adding another power amp, and I don't have room for one. All components are original. They DO sound a little boxy, but if I can get 'em tweaked in, singers really like them. The singer has a pair of CGM cheap-o's he said he could bring to the next gig, so I'll try them and see how they sound. I'm hoping to locate another set of the EV's, but they're pretty hard to find.

 

Thanks for your help, it's appreciated.

 

MG

[MarkGifford-1]

 

Analysers are for sissies!

 

 

Not gonna go there, I've never used one. I've got pretty good ears and can generally pinpoint offending frequencies quickly and accurately.

 

Thanks,

 

MG

[Mr.Talent]

"1/3 Octave Q, usually no more than 3-6dB of cut. Snoop out offending frequency and just dip it until it's gone. This EQ also has high and low shelving, so I'll take a 2-3dB off below 200hZ, using that. Don't see how that's any different than what a 31-band would do, but I may be ignorant."

 

say you get feedback at 400Hz (kinda typical with our system)- with your parametric.. you find it and dial it out and crank up the mix a little more...what can you do with that parametric when another frequency goes off ? how many sweeps with adjustable Qs do you have?

I don't know about you.. but when 400 goes on my system 800 and/or 200 is not too far behind..

graphics are faster to find that {censored}..

you could be getting frequencies from an instrument combining with certain vocal frequencies at different levels- this could create feedback at a totally different frequency (alot of times in the low-end)- again the graphic eq is better for the job..

compressing the monitor mix by about 3:1 with a fairly high threshhold (only compressing when the singers are wailin) with keep the mix up when you're doin the talky/wispery thing.

I almost never compress vocals individually... I group the vocals and compress the group... my monitors have their own EQ/Comps

Hope that helps..

[Keyrick]

Mark, I know you are knowledgable about this stuff, but I'm going to speculate the the response on the EAW monitors are different than the response on the EVs. And with the parametric, are you setting the EQ's the same for both channels? I could see a problem if you are, once the music starts.

 

And yeah, for monitors, a dual 31 band graphic is really the way to go. Monitors and mics can move to a degree during the show and that will throw off what you rang out with the parametric before the show started. Plus, you can fix the trouble pretty quickly with a 1/3 octave graphic.

 

Just some thoughts.

 

Rick

[MarkGifford-1]

So I'm starting to get the hint...

 

What's a reasonably priced 1RU 31-band that's as good quality as the Ashly stuff?

 

This being where they're made, Ashly stuff turns up used quite a bit, but I'm willing to consider other brands.

 

MG

[tony k]

Just a thought....If he doesn't have much space in the rack....would a dual 15 band graphic work...that way he could eq the EAW's and the EV's separately and only give up one rack space if that's a concern....

 

Never used a 15 band myself but it might work for this app. Though 2 31's would be better of course.

tony

[Craig Vecchione]

 

Well, I rewired them to 16 ohms so I could hang 3 off one channel of a power amp. I just switched the LF drivers from parallel to series. ... They DO sound a little boxy, but if I can get 'em tweaked in, singers really like them.

 

 

 

Hmmm. So the LF drivers present 16 ohms, and originally they were wired for 4 ohms. Assuming a 1st order BW filter....if for example the crossover point was 3kHz, by rewiring the drivers the LPF point will now be at 750Hz, and the LF drivers will no longer 'see' anything above that. The change in impedance put in effect a notch filter into your system. That big chunk of very low output in the midrange would mean big EQ is needed to flatten response, and the huge change in s/n ratio and dynamic response and very likely phase response would make feedback a bitch to eliminate.

 

Get a pair of those speakers back to their original wiring and see if that improves things.

[MarkGifford-1]

 

Hmmm. So the LF drivers present 16 ohms, and originally they were wired for 4 ohms. Assuming a 1st order BW filter....if for example the crossover point was 3kHz, by rewiring the drivers the LPF point will now be at 750Hz, and the LF drivers will no longer 'see' anything above that. The change in impedance put in effect a notch filter into your system. That big chunk of very low output in the midrange would mean big EQ is needed to flatten response, and the huge change in s/n ratio and dynamic response and very likely phase response would make feedback a bitch to eliminate. Get a pair of those speakers back to their original wiring and see if that improves things.

 

 

Thanks, I knew you guys would come through for me. I'll try these suggestions until I get things back under control...

 

MG

[dboomer]

Rewired speakers have shifted your crossover by about two octaves ... you've probably lost a lot of mids. That shouldn't cause feedback ...but maybe whatever you are doing to counter it does.

 

Using all different mics is a problem. If you EQ to take a peak out that is being caused by one of the mics you arre also dipping the frequency response of the other mics which probably don't have any problems right at that spot. The result is you are probably getting a hollow sound from the mics that didn't need to be corrected. Multiply this by the number of different mics you have and I'm sure it sounds pretty odd.

[boseengineer]

Rewiring a passive box is not a good idea. Even if you don't muck up the cross over (depends on how it's done), the frequency response will be way quite a bit off, since the LF portion's sensitivity goes down by 6 dB where as the HF stays the same.

 

You may want to consider wiring them back to original and maybe run all three in series for a total load of 12 Ohms.

 

In my opinion, the best approach to deal with feedback is to take it step-by-step. My elimination figure out which channel and which speaker are the problem. Are all channels equally sensitive or is it always the same offender. Same for the monitors themselves. If it's from mulitple speakers, try it with less monitors. Overlapping monitors usually cost gain before feedback.

[agedhorse]

Absolutely... that "something doesn't sound quite right) with the EAW's is that by rewiring, your LF section x-over frequency INCREASED, the EAW probably has a 2nd order LPF and you now have OVERLAPPING bands by a couple of octaves and the 10's do not have a very smooth response up there.

 

Go back to stock wiring by all means.

 

Are you sure you have 8 ohm drivers in the boxes??? Double-check the specs on the drivers.