PA system set-up

[Johnno]

Am I right in thinking that if my mixer has a maximum output of +22dBu and my amp has a maximum input sensitivity of +4dBu then I should either connect them through an 18dB attenuator pad or turn down the amp's input sensitivity control by 18dB?

 

The reason I ask is because our drama teacher, despairing of me as I'm not getting where she thinks we should be at the speed she wishes to travel, brought in a soundie to help set up our system and the first thing he did was turn the amp's control to maximum sensitivity "because," he said, "I prefer to set the volume with the sliders". I'd been using it set at 18dB off its maximum mark. (The gear in question is a Soundcraft E12 and a Behringer EP2500).

 

The next thing he did was take away the graphic equaliser and feedback "destroyer" (both Behringers) I'd been using. The EQ had been set flat except for complete roll-off below 100Hz and above 16kHz, and about -9dB at 500Hz, 1kHz and 2kHz, determined as the worst offenders for feedback. The limiter in the EQ had been catching one girl who screeches at full volume a few times during the play. The feedback detector was left to it's own devices with the filters resetting if unused for five minutes. All the mixer channel EQs were flat. The only signal going through the system was children's speech and singing, no deep voices, no music.

 

Once this technology was removed the fellow rapidly found out why I'd been using it and frantically started twiddling channel EQs and mic trims to cope with the feedback-before-gain here, which is quite ferocious (bog-standard 1960s state school hall, lots of reflective surfaces). He ended up with all the EQs turned down at least 3dB. Surely that's pretty much the same as leaving the EQs flat and turning down the trim or the fader by 3dB or so, since - give or take a few Hertz - you are effectively depressing the entire frequency range of each channel? So why not leave the channels flat and use global EQ? Un-necessary added complexity, perhaps? Use the knobs you're given before adding more?

 

Proof of the pudding being in the eating it didn't sound greatly different to me - but my ears are thirty years closer to their expiry date than his and not experienced in mixing sound so maybe he could hear more than I could or knew better what he was listening to - but it sure looked like harder work! That's my trouble, I'm lazy :D

[Simon Lewis]

There are those who will argue either way regarding gain structure (see the recent thread) but many professionals would agree with your thinking... Furthermore, although I'm not a great fan of using Behringer feedback devices, a carefully used graphic is still pretty much the standard way of globally taming frequencies prone to feeding back. Channel eq simply isn't precise enough for this job.

 

I'd sit back, safe in the knowledge that some one has sweated to get to the point you had already tuned the rig to!

 

Simon

[johndenim]

I would be curious as to why he decided to take out the EQ unit?

 

If he had to reset the mixer eq, why not leave 31 bands in so more control is achieved?

I agree with the OP, I would have left in all outboard dsp's, if you don't like the sound from them you can always defeat the processor, take it out completely and you don't have a lot of options.

 

John Denim.

[Ben Langfeld]

If I were you, I'd put it straight back in again. The guy sounds a bit confused, and I'm guessing it's you who has to use the system. Sounds like a lot of toe stepping to me.

[berry120]

That sounds very odd to me. I can understand him perhaps removing the feedback destroyer (I'm not a huge fan of them myself but I admit in some circumstances they do have their uses) but I can't understand why on earth he took out the eq as well?!

 

If he had a reason for doing so, the only thing I can think of was that the eq was somehow being detrimental to the sound quality. It wasn't hissing at all or anything like that was it? Unless he's one of these people that just hates everything Behringer and therefore removes as much as he can of that make...

[Bobbsy]

Ignoring the "toe stepping" for a moment, one thing in your original post jumped out at me: it was where you said the MAXIMUM input to the EP2500 is +4dBu. This didn't ring true because +4dBu is the normal "zero" level for professional audio gear.

 

A quick look at the spec sheet for the EP2500 confirms this. According to Behringer, at in input of +4dBu, the amp will be putting out 8 watts RMS. It's too early on a Sunday morning here to give you exact numbers, but this means that the maximum output of amp will occur at roughly the same time as you mixer hits clipping.

 

As for the question of running amps at max vs. using the attenuation knobs, this was recently a hotly debated topic here in the Blue Room with several pages of replies. There is no right answer to this but it's worth doing some reading on gains structure and various people's ideas. THIS TOPIC might provide some valuable reading.

 

As for the rest of your story, frankly it sounds like you'd thought out the system with one possible exception. I probably wouldn't have left the Behringer FBQ to "it's own devices". They do have a habit of starting to hunt and doing some nasties to your sound. Far better it to let it "ring out the room" then when it's knocked down several offending frequencies, lock it there for the duration.

 

However, for the most part I would use either the GEQ or the FBQ to control feedback (and/or compensate for inadequacies of the room) and use the channel EQ to tweak the sound of individucal mics--for example control the nasal quality you can get in the mids with some lav mics, tame the presence peak on an SM58, that sort of thing). And you're entirely right...if every channel EQ knob is down by 3dB to control feedback, you might as well have just pulled down the fader.

 

Bob

[Shez]

Ignoring the "toe stepping" for a moment, one thing in your original post jumped out at me: it was where you said the MAXIMUM input to the EP2500 is +4dBu. This didn't ring true because +4dBu is the normal "zero" level for professional audio gear.

 

A quick look at the spec sheet for the EP2500 confirms this. According to Behringer, at in input of +4dBu, the amp will be putting out 8 watts RMS.

 

That's odd... The spec sheet I've just looked at suggests that (like the vast majority of other amps out there), the input required to achieve maximum output is +4dBu. Obviously it's higher than that if you turn down the input attenuators, but that figure strikes me as about right if the attenuators are all the way open.

[Simon Lewis]

The voltage gain for this amp is x50 (34dB), and the input sensitivity for maximum output is 1.23V (+4dBu). So far, everything is normal.

However, the text in the Behringer's spec sheet actually says Input Sensitivity (V RMS (@ 8 W) ) - which is probably what Bobbsy read.

 

I suspect that despite getting the symbols correct elsewhere that this should actually read V RMS (@ 8 Ω) and that the Greek Omega has turned into a W when a different font was used. This happens quite a bit in technical articles, and it's unsurprising that copy proofers would get confused when Ω and W are used throughout the sheet anyway.

 

A further area of confusion, is that some amplifiers (d&b seems to spring to mind) have quite insensitive inputs, thus needing a fair push from the desk to work at full output.

 

Simon

[Johnno]

Thanks for your comments.

 

Bobbsy: I've read all the similar threads that Google pulls out.

 

Berry120: No hissing, all working fine.

 

 

I've checked my copy of the spec sheet and it says +4dBu @ 8 ohms. Behringer's website PDF for the EP2500 says 8 Watts, so yes I guess it is a misprint.

 

If +4dBu is the zero for the amp then the output voltage at a clip level of +20dBu would be 388 volts. This seems a tad excessive as an 8 ohm loudspeaker would then be dissipating 18769 watts :P

 

4dBu = 1.22V,

 

EP2500 has voltage gain = 50

 

So Vout = 50 x 1.22 = 61V

 

Power = V^2 / R = 61 x 61 / 8 = 465W @ 8ohms. This agrees with the spec sheet.

 

So +4dBu must be the maximum input at maximum sensitivity and 16dB of attenuation is needed to connect to a device with an output of 20dBu.

 

 

Edit: added a bit.

 

A concurrent post has been automatically merged from this point on.

 

Most professional systems are designed to handle peak levels in the region of +22dBu or thereabouts (+28dBu is very good, while +18dB is not uncommon on budget equipment). Therefore, if the nominal signal level is +4dBu but the maximum peak level is +22dBu, we have 18dB of headroom in the system.

 

Quoted from Sound on Sound website.

 

This seems to imply that it is right to operate my amp at max sensitivity when connected to a device whose max output is 20dBu, but...

 

... the Chuck McGregor article on Setting Gain Structure at livesound.com uses +4dBu as the maximum output to be expected from a PA and says that attenuation would be needed to drive it from a device that outputs +20dBu max.

 

Now I'm confused!

[Simon Lewis]

....This seems to imply that it is right to operate my amp at max sensitivity when connected to a device whose max output is 20dBu, but...

 

... the Chuck McGregor article on Setting Gain Structure at livesound.com uses +4dBu as the maximum output to be expected from a PA and says that attenuation would be needed to drive it from a device that outputs +20dBu max.

 

Now I'm confused!

 

The SOS quote is speaking in terms of the headroom available at the mixer - a nominal 0dBu operating level, with around 20dB headroom. However, Chuck McGregor's article looks at the desk and amplifier, and there's the problem that you identified, - as soon as the desk goes past +4dBu, without attenuation the amp will clip.

[Bobbsy]

Yup, I took the Behringer online spec sheet at face value even though it seemed at odds with the majority of amps. Frankly I wondered if it was a form of speaker protection from Behringer. (As an aside, I've often thought that in a day and age where mixers tend to have 22-ish dB of headroom--which tends to get used--amp inputs should be set up in a similar way. I tend to think that companies like d&b may have it right!). Anyway, given the accurate spec I'd probably set up my gain structure exactly as you do.

 

Bob