What model of guitar were you using? Some types (e.g. the Martin Dreadnoughts) are prone to exactly this effect, and it appears to be caused by a resonant column of air at the soundhole.

The technique that I use in these situations is firstly a hypercardioid mic (Beta 57A) rather than a straight cardioid (SM57), and positioned opposite and perpendicular to the 12th fret, and secondly, care with the blend between pickup and mic: use a blend of mostly microphone in the PA mains and mostly pickup in the foldback. A good pickup (e.g. Fishman) allows you to add volume and body to the FOH mix without detracting too much from the tone picked up by the microphone.

Apart from rolling off the bottom end as expected, no extra EQ should be necessary, at least for feedback avoidance. The pickup needs about a millisec of delay added at the mixer relative to the microphone.

took-the-red-pill, post: 431812, member: 21836 wrote: I thought most feedback issues were the sound bouncing around the room, but outside, with mains in front of the mic, I wouldn't think it would be an issue. Clearly I was wrong.

You're actually not outside, that metal roof is a big reflection surface and some shapes could worsen its effect on sound.

took-the-red-pill, post: 431812, member: 21836 wrote: -Mains at the front of the gazebo, 6' high on stands, maybe 25 feet apart, Yorkville 550 watt powered.
-Mains about 5' in front of the line of the mics/performers.

A lot of bass travels behind the most FOH speaker. 5' is pretty close specially if the musicians are on the same ground level.

But I'm with Boswell, most acoustic guitar needs to be plugged in instead of using a mic and most need a feedback buster to be live friendly.
Something like this :

Thanks for your time and input, gents.

Further experimentation is certainly required. The simplest solution is to just get a piezo, which my other guitar has(a heavenly sounding Fishman Matrix), But dammit I want to figure this out. A guy ought to be able to mic a bloody guitar, as was done in days of yore. I've seen bluegrass bands where they are DI'ed, but they have one 57 at waist level, and for a solo the guy steps out in front of it and lets 'er rip. True, there is no way of knowing if the sound guy actually used it, or just jacked the DI line.

Boswell: Yes, it was a dread.
I always avoid the sound hole, as stated I was at the 14th fret. Maybe not far enough away though. I will try the 12th.
Blending with an on-board pickup was not an option, as this guitar doesn't have one.
Remember, no signal whatsoever in the monitors, so all feedback was from the mains. I can see the sense in your approach though, putting pickup only in monitors does make sense.

PCrecord:
I understand that the roof can have an effect, but it is my understanding that low frequencies go right through thin materials like 28 guage metal roofing, and stuff below 100 ought to do just that. Or at least that's what I would expect. I will try to simulate the scenario on my acreage and see if it still feeds back.

Okay, so next time I'm going to show up early and fart around a bit. I want to crack the code on this one.

I will:

Move FOH much further away from the stage setup.
Try the mic even further from the soundhole, at fret 12, but still close to the guitar.
Rent a 57A, though I want to try my regular 57 first.
Run the channel through my 1/3Oct EQ and ring out that mic.
I will report back my findings.

In the mean time, let us bask in the glow of the greatest live performance of all time, and not a DI in the house.

Cheers
Keith

Any hard surface is reflective. Click your fingers and you hear it. Keep in in mind that while miking at the 14th fret is maybe ok in the studio, the volume level there is much lower than at the sound hole, especially outside where the players play louder. So you turn the gain up to make it heard, and the player moves just a little, the sound hole gets closer, proximity effect starts to work and the bass output ramps up quickly. The same bass leaks around the speakers, and off you go.

The most feedback prone gig I have ever done was outside with a useless person on the mixer who had no idea about eq. Feedback control was simply level. If it honked, turn it down!

took-the-red-pill, post: 431816, member: 21836 wrote: PCrecord:
I understand that the roof can have an effect, but it is my understanding that low frequencies go right through thin materials like 28 guage metal roofing, and stuff below 100 ought to do just that. Or at least that's what I would expect. I will try to simulate the scenario on my acreage and see if it still feeds back.

Yes low frequencies gets through thin and light materials but a part of the sound is reflected. The key point is the distance the source is from it. In your case the source isn't the guitar, it's the back of the speakers. If you would put the FOH further away and/or apart, you would reduce the feedback greatly.
But at no time you speak of monitoring ?? Is it because you have none ?

I've worked with gazebo a few times and they are not easy to work with specially if the roof in some in dome shape.
Sometime, just having a carpet on the floor is a great help (specially it it's a ciment floor)

In extreem situation where the venue is prone to feedbacks, I use the ring out technic. It consist of using a 31 band EQ and lower the frequencies that feeds the most by creating them on purpose (Before the soundcheck) I do that all the time with stage monitors but occasionnally with FOH too.
Even if I don't need to use that technic on FOH, I still EQ it. On a lot of setup, I'd reduce some of the frequencies that are anoying to me. but I'm talking about 10 000 watts setups. I do it by ears every time and even with the same kit on every locations.

The last setup I did was a pre tuned setup. Even then I heard some defects in the hi mids. Also they had put the mixer 100' from the stage on a platform. That platform was resonating at 60hz... I had to get out of behind the desk all the time to go and listen to the mix on several different spots...

Doing live sound is an art. At some point some sacrifice needs to be made for the sake of security and prevent feedbacks.
I'm sure you will find a good recipe that works for you ;)

Thanks for the input gents.

Paulears:Yes, any surface is reflective, and a snap of the fingers will reveal that, but the guys over at the acoustics forum are always harping that low notes with lots of energy simply pass through a thin surface, like metal. So that's what I would expect, but maybe not.

Yeah, I'm guessing it's a cumulative error like what you describe: distance, mic, sound hole, bass coming off back of monitors...etc. Maybe I'll try a different spot on the guitar that has volume but still avoids the sound hole. Any luck behind the bridge?

PCrecord: As stated, I do have monitors, and at this gig I was running the vocal mics and pickup guitar through the monitors with no issues.

I will try your idea of throwing down a carpet or some such material on the floor to quell reflections, as well as the other things I mentioned above.

Cheers mates.

took-the-red-pill, post: 431831, member: 21836 wrote: PCrecord: As stated, I do have monitors, and at this gig I was running the vocal mics and pickup guitar through the monitors with no issues.

Oh sorry... so the feedbacks were comming from the FOH or the monitors ?? As I said monitors can be EQ to avoid feedbacks. Also, since a lot of bass may come from the back of the main speakers, you can reduce the bass in the monitors alot ! ;)

I think this is less of a 'feedback' issue (technically speaking) and more of a resonance problem not uncommon in acoustic guitars. Bear in mind that even though the horns of your Yorkville cabinets are very directional [80˚horizontally x 50˚ vertically] and easily aimed - the lower the frequency, the more omni-directional the speaker as a whole becomes. There is often going to be a lobe of low-mid radiating out the back/bottom/sides of the box.

I'd be curious how much gain you can give the mic before feedback when it's not pointed at the guitar body vs. how much you can give it while aimed at the guitar before it starts to howl. To me, that would be a very telling test result.

I filled in on guitar for a friend a couple weeks ago, and between the smallish room, the bass player's big rig, and my proximity to the corner of the 'stage' area - on certain notes you could feel the wind blowing out of the f-holes of my 335. On the louder songs it was enough to alter the overall tone and even skew the pitch when the body of the guitar would go crazy, hyper-resonant. Relocating myself further from the corner and some tone-notching were the only way I was going to avoid the problem. These guys had nothing but vocals going through their Main speakers on tripods. So although I was pretty close to the back of the one speakers, there should not have been enough low energy coming out the backside of the cabinet to make matters worse. At any much louder outdoor festival stage volume this guitar rig is a thing of beauty, this was a first. The problem on the small club fill-in date was the sum of a standing wave in my corner and an enthusiastic bass player. I can play in my own band in any venue (even a comparable size room) with my MUUUUUCH bigger sound system, with drums, bass, keyboards, guitars and all pumping through the mains, monitors, and side-fills, and never have anything like that happen with the exact same guitar rig.

On a related note: This weekend I'll be doing a sound job outdoors with my PA, for a friend from Nashville who performs as a solo acoustic act these days - nothing but guitar and vocal. His standard tools of the trade are high quality Taylor and Martin acoustic guitars - all with built-in pickups, an LR Baggs Para-DI or two, and one of those feedback buster plugs like PC shows above which I don't think we've ever needed. And this is a guy who likes a lot of monitor volume, and is playing at a volume to fill the grandstands at a fairgrounds type venue. This guy beats the daylights out of a guitar to provide the rhythm section (a la Tommy Emmanuel, etc.), so we can't have the guitar too dead.

On the other hand, I provided PA last month for a young dude with a middlin' acoustic guitar with a cheap pickup, a $2 guitar cable, and a generic DI. I had a ton of gain before feedback and thought he sounded much better with an SM58 a couple feet in front of him. But to be fair, this was a very modest volume church related event though, not a big show. We'd set up a vocal mic for him and when it turned out he wasn't going to sing, made the quick adjustment.

Hats off to you by the way. I like the fact that you're going to experiment and test different combinations of things. I wish more people would take the time to do that!! Good luck!

acoustic guitar, with plenty of gain because of mic position and monitors that are loud enough to be heard? Yep, that's feedback city!

No guitar in monitors. I'll say it again. NO guitar in monitors. So all feedback was coming from mains.
Did I mention there was no guitar in the monitors?

As long as you're certain there was no guitar in the monitors ........... It could be the guitar. I've got an electro-acoustic that has some real spikes in the output. Some notes boom out others don't - but the sound direct from it sounds pretty nice. I don't know if the pickups just make the resonances worse? It'd doesn't feedback that badly, but just sounds boomy and horrible - maybe you have a similar spike, that perhaps corresponds with the PA's? Other than that, maybe just a simple eq problem. With digital desks it's very easy to forget a graphic is patched across a channel, left behind from earlier? Maybe nothing you were doing at all. Probably the monitors though ................