Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Greg Cameron]

Jack Gunn wrote on Thu, 15 January 2009 22:41

Because if that read clip/peak light comes on, that means your tweeter is being sent a SQUARE wave.  That's no longer audio signal... That's tweeter death signal. It can't acoustically or physically recreate a perfect square wave with a magnet and moving coil.  But it sure tries.  Tries until it dies.

It's not the driver trying to track the waveform that's the issue, it's heat caused by the clipped waveform. If clipped signals could in inherently blow speakers, you would see guitar amps killing drivers every time a rock band played.

Jack Gunn wrote on Thu, 15 January 2009 22:41

If your speaker is rated at 100/200/400W in the spec sheet, go for a 400W amp minimum assuming the program material you're playing isn't extremely compressed stuff like dance/techno.  If you're doing live mixing of most any sort, my personal rule of thumb is go with the highest of the three.

That's a recipe for trouble. The chances of exceeding the rms rating of the speaker greatly increase if you can supply 4x the rms rating. If you clip at all with an amp that large, your chances of taking out drivers are much greater than an amp rated at program power. I can't think of any speaker manufacturers that recommend you use amps which match the peak rating. It's most often, if not always, the program rating. JBL and EAW, who know something about speakers, recommend either 1.5 or 2x the rms rating in their white papers, not 4x. I'm not saying you can't do it, but for the average user it's a risky proposition. Several on the LAB have stated that peak power ratings are something that probably should have been left off the spec sheets as they have little useful meaning.

Greg

[Nick Enright]

I would get another amp.

In my experience with monitors its very important to have enough power. You will have less feedback and use less eq to get there, also I have a good feeling that adding a wedge is not going to help cause it wont be loud enough to be heard anywhere on the stage.

[David A. Parker]

I agree. I got into an argument on the LAB about this same topic. When I doubled the power to my monitors, using less quality amps, my feedback problems got drastically better. Whatever the reason, having plenty of power DOES help with feedback. I think it is because an amp loses it's flat response as you push it harder, therefore some frequencies get exaggerated. Having plenty of power allows the monitor to get up to acceptable volume without having to drive the amp very hard.

[Tom Reid]

David A. Parker wrote on Fri, 16 January 2009 13:15

I agree. I got into an argument on the LAB about this same topic. When I doubled the power to my monitors, using less quality amps, my feedback problems got drastically better. Whatever the reason, having plenty of power DOES help with feedback. I think it is because an amp loses it's flat response as you push it harder, therefore some frequencies get exaggerated. Having plenty of power allows the monitor to get up to acceptable volume without having to drive the amp very hard.

Dave, was this bi-amping wedges?
IMHO passive crossovers are not linear through their power range.
I think that's more of a factor than flat response of amps.
And yeah, some smaller amps get real squirrely when they try to work at their full rated output.

[Ryan Lantzy]

David A. Parker wrote on Fri, 16 January 2009 14:15

I agree. I got into an argument on the LAB about this same topic. When I doubled the power to my monitors, using less quality amps, my feedback problems got drastically better. Whatever the reason, having plenty of power DOES help with feedback. I think it is because an amp loses it's flat response as you push it harder, therefore some frequencies get exaggerated. Having plenty of power allows the monitor to get up to acceptable volume without having to drive the amp very hard.

I recall JR addressing this at one point.  I think it has more to do with clipping and that acting like a compressor of sorts.  Typically compression can exaggerate the onset of feedback.

[John Roberts {JR}]

It's simpler than that.. if you don't bring enough rig for the the gig, you will end up having to turn the amp up to a higher gain to get enough (clipped) power output. The bigger amp will get louder (cleanly) at less gain, than the small amp driven into clipping. Feedback is very simply caused by too much gain.

You can make a 5 watt amp feedback... it's the gain not the powah.

JR

[Ryan Lantzy]

John Roberts  {JR} wrote on Fri, 16 January 2009 15:09

It's simpler than that.. if you don't bring enough rig for the the gig, you will end up having to turn the amp up to a higher gain to get enough (clipped) power output. The bigger amp will get louder (cleanly) at less gain, than the small amp driven into clipping. Feedback is very simply caused by too much gain. You can make a 5 watt amp feedback... it's the gain not the powah. JR

Does any one really turn up a small amp to make it louder and let it clip though?  I guess... if the amp is clipping, a bigger amp might help problems with feedback.  That said, if you are having trouble with feedback before the amp clips, a bigger amp will probably not help.

I guess the key is... can you rely on the clip indicators... or do most amps start to clip (by a very small amount) before that.  And does that small amount of clipping exacerbate the onset of feedback.  Or is it only severe clipping that causes feedback.

It would be interesting to test in a controlled environment with an amp that you could adjust the the supply rail voltage.

Edit... dang it... the clipping doesn't cause feedback, like you said, gain does.  I'm having trouble expressing my thoughts here.

Second Edit...

Ok, say you have a small amp with a signal and an open mic.  Say the system starts to ring... then you immediately switch over to a larger amp with more headroom... but the same gain.  Will the system continue to ring?  Answer: yes. (I think).

You will have to reduce the gain on the large amp to remove the oscillation... but it should sound just as loud as before do to the increased swing in signal voltage, and consequently higher RMS voltage.

[Art Welter]

David,

Feedback is caused by gain, the microphone hearing the speaker and reproducing it, not  “because an amp loses it's flat response as you push it harder”.

If you put a hard limiter on your monitors, you can cause them to feed back with under a watt, at a very low SPL, which can be useful for finding rings without blowing ears or drivers.

In the same fashion, if you don’t have enough power for your monitors to get loud enough, you may turn them up to increase the average level causing feedback, and the peaks will be clipped, a basic (nasty) form of limiting. The clipping will be heard more in the HF, where the clipped harmonics are more prevalent and your ears are more sensitive to distortion, the   “exaggerated frequencies” you mentioned.

With more power, the peaks are reproduced without clipping, so the monitor gets louder, the performers don’t have to sing (scream) as hard, until enough power is put in and thermal compression sets in...

Art Welter

[Scott Smith]

David A. Parker wrote on Fri, 16 January 2009 07:52

you can't hurt a speaker by underpowering it, if you could, then every time you turned them on and turned the volume down they would blow! You can only hurt a speaker by overpowering it. Very simply, if you have an amp that is less than the speakers are rated for, and you overdrive the amp, the amp will clip. When an amp clips, the power output is greatly increased. That overpowers the speaker and cooks it...

+1...I am so glad that someone had the guts to state this correctly.

[David A. Parker]

The passive crossovers wouldn't have been the reason, because they were the same in either case, same passive wedges. For whatever reason, I DID have less trouble with feedback after doubling my power. Also, when I got rid of the behringer eq's and replaced them with DBX there was a huge improvement in gain before feedback.