FAQ: Output Compression

[RoadRanger]

I use the "No Clip" preset on all my outputs  .

[BlendedMix]

Hi Road...

Thanks for the help in the Airplay thread.  The system I described in the elem school has one cause for concern.  Blowing speakers.  There is the ability to push more than the speakers can handle.  I'm versed in some things about mixing sound, but dealing with limiters is something new to me.  It was never a concern with my acoustic guitar duo playing for small crowds.

Is there an output compression setting that will effectively limit the output to the amp to prevent sending too much to the speakers?  Does the "no clip" setting do that, or is it actually designating that it's allowing the DL to output at full strength?  It seems it could be interpreted either way.    

[WK154]

It would be helpful if you had the list of equipment you are talking about. Name and model# otherwise you'll get nothing but speculation (amp and speakers). If there is any protection for the speakers it would be in the amps. Compression and limiting has little to do with it.

[BlendedMix]

It would be helpful if you had the list of equipment you are talking about. Name and model# otherwise you'll get nothing but speculation (amp and speakers). If there is any protection for the speakers it would be in the amps. Compression and limiting has little to do with it.

The amp for the mains is a QSC CMX300V, 120V.  I read about it, and it does appear to have some speaker protection features included, but I didn't see anything about a limiter.  I just thought I'd check to see if there was any additional ways to prevent users from getting happy with the faders on the iPad. 

I was told that since this was being run as a 70V bridged mono system, I would want to run the amp at full power, and then set my gain structure through the pre-amp and channel faders.  I've done that, but noticed some clipping on the amp that I wasn't expecting at the volume we were playing music through it.

Admittedly, it was "Harlem Shake" and it was a terrible recording, which likely had a lot to do with it.  It just reminded me that it's EXACTLY the type of thing that is going to happen in a school cafetorium, so I thought I'd at least make every precaution necessary to preserve the speakers.

The second amp is a Boeringer BE-EPQ2000.  It just powers the single hanging monitor, which sounds and works surprisingly well, and I'm not concerned about its volume, because it is plenty loud enough on even a moderate channel volume and the amp is more power than necessary. 

The speakers are a Vector Ceiling Array speaker manufactured by Pure Resonance, and they are a 120 Watts Built-in 8 Ohm / 70 Volts.  We have four of them "daisy chained" which is a total of 12 speakers.  This system is mono bridged and running in the 70 volt configuration. 

So that's the rundown.  He also recommended, since these are children's voices, to also have an Aphex Big Bottom and Aural Exciter to "beef up" the sound.  I have to admit, I was very skeptical about that since it is typically used for recording applications, but there is a very noticeable difference with it on or off in the chain, and my guess is that it has something to do with the limited low end response of the Vector ceiling speakers.  Maybe a really good engineer could get the same out of the 31 band EQ on the DL, but I can't. 

Surprisingly, the relatively inexpensive AKG DMS70D with DHT70 mics sound pretty good.  I used my SM58 right next to one, and the difference wasn't what I expected.  We went low cost, because we acknowledge that damage will occur, and they are only $165.00 to replace.  Pleasant surprise thus far...

So that's what I'm up against, and the more I type here and think about it, I'm considering looking into a one time fee for a sound engineer to come tweak the system and get it set up rather than trying to "amateur" my way through it.  I know good sound when I hear it, and I can mix an analog Yamaha powered mixing board and make it sing for two guitarists and vocals, but this thing is getting a little above my pay grade, which is strictly volunteer.   

Once again, thanks for taking the time to help, and it has been invaluable to this point.       

[Greg C.]

Some considerations: compression in the traditional sense can actually cause speaker voice coils to heat up more then a non-compressed signal because it increase the rms voltage value of the signal. Higher rms levels = more heat. So compression is actually not a form of protection at all and can actually promote damage if your compressing the signal and running things full tilt boogie due to the low crest factor of the signal. The only time a compressor can be a relatively protective is when is used as an "rms limiter" where very slow attack and release times are used with a hard limiter ratio to prevent excessive voice coil heating. These types of limiters requires specific non-trivial calculations to avoid waisting system headroom on one side or cooking your voice coils on the other. Standard compressors also don't make good peak/clip limiters because they often don't have a fast enough attack to stop the fast transients from getting though. Most amps have clip limiters in them for basic protection. But they can be punched through on the more basic amps.

Higher end amps with full flown DSP usually have rms limiters as well as peak limiting that work very well. There are outboard speaker processors that have this 2-stage type limiting as well. Neither is cheap. But setup correctly, you can hammer them into oblivion with overdriven signals without damaging your speakers (though it may not sound great). Well made self powered speakers (i.e. not Mackie) will have solid rms and clip limiters as well as protection from driver over-excursion so they hold up to abuse pretty well. Again, they're usually not cheap.

Lastly, keep in mind with distributed systems that you should avoid driving them with a lot of low end frequency content. They're not designed for it. The 70v transformers on the speakers will saturate and distort easily with low frequencies. This also causes problems for the amps which may be why you're clipping more than you think should. You should keep low frequency content below 100Hz to a minimum. A high pass filter around 70Hz is usually a good idea.

[WK154]

When you previously mentioned Cafetorium I was afraid of what I am only too familiar with. A dual purpose room sometimes a gym or cafeteria with lots of echo and no acoustics to speak of. That in and of itself is a nightmare. Tell me otherwise.
"I was told that since this was being run as a 70V bridged mono system, I would want to run the amp at full power, and then set my gain structure through the pre-amp and channel faders. "
A quick look at the specs shows a 256 watt (without insertion loss considered) set of speakers (64 W each). and a QSC amp that is happy driving 600 W at 70V direct (transformer optional). I would say that the advise was bad. Back off the amp. If you have good dynamic material (12 to 14 db crest factor) or a sine wave (steadier signal for setup) available set your DL output at 0 (fader and signal that's ~ 6dBu) and then with earplugs bring up your amp to an acceptable level (OSHA has max requirements not to be exceeded. Your in a school not a nightclub). A SPL meter helps with this. The clip on the QSC is 22 dBu the max output of the DL is 21dBu no clipping on the amp but you will hit the Power limit and transformer core saturation long before this. As Greg pointed out remove the low end with a high pass filter and toss the Aphex and you'll have a truer sound. The spec on the speaker states 35Hz but no tolerance is given and clearly suspect. With these settings you are with lots of headroom and good S/N ratio. Remember your setting the average level not peak. 84 dB SPL is what I shoot for. Once you have the main loudness adjusted then you can play with material eq's, preamps etc. I personally think that the Aphex is nothing but a distortion on sweet young voices unless you want them sounding like the Mackie's Belch King (see DL ad). Acoustic sound treatment would help tremendously but it will be hard to convince the powers to be. Once this is accomplished you could revisit the amp setting and fine adjust. That leaves only the Mic positions and non-existent mic technique to keep you busy riding faders. As to the lone unnamed hanging speaker on a Behringer amp I would say it's probably a better setup than your ceiling speakers. Without info ? What is your DL driving? the Ceiling Spk's or the hanging speaker or both? Back in the stone age we were more concerned about Raptors than Flying threats so our hearing evolved much better in the horizontal plane than the vertical. And so horizontal speaker systems are more efficient for our ears than ceiling ones.

[BlendedMix]

Hi guys...thanks for the info again.

WK...you NAILED the environment.       And yes, the hung monitor with the Europower amp sounds magnificent. 

I could bore you guys to tears with all of the reasons with regard to why ceiling speakers were chosen, but I will submit this.  I took my EMX 5000 1000W Yamaha powered mixer up there with my two S115 speakers to do some testing when this whole thing got started.  It was like one big chamber of horrible in there, and it was primarily because the back wall was reflecting the sound back all around the room making it very muddy and unclear.  Bad news is...terrible place for any sort of entertainment...good news is, I know a great place for an echo chamber if I ever need one.    

Initially, we wanted to treat it acoustically and purchase a sound system.  We had some people take the room dimensions and say that we wouldn't hear any difference unless we spent $5K.  Then some said as little as $2.5 K on the back wall alone would make a huge difference, so start with that and then add more if you need it.  Then when someone actually agreed to come onsite to test it, they submitted a bid for $10K, and insisted that any less than that would potentially just be a waste of money.  So...we were left with a "who to believe" scenario, and decided to start with the sound system first, since it was SOOO bad, and we were working with a limited budget.  They sold everything from raffle tickets to $1 plastic frogs to earn that money for the PTO.  We submitted the sound results to the District with the hope that they may look at the decibel levels and ultimately decide that it needs treatment.  Just last year, they treated their all purpose gym, so we think we may get some traction there.  I digress....badly. 

As shocking as this may seem, the ceiling speakers actually seem to make the clarity of the overall sound much better.  I don't know if that is because the speakers are better with vocal frequencies and don't have much low end, or if since the sound is in the vertical plane and the tiled ceiling is doing a much better job of absorbing some of the sound.  My guess is that it's a combination of both.  Suffice it to say, no reverb is required.   

On my last trip up there to mess with it, which is difficult by the way with the school hours and access, I quickly found that the distortion in the house speakers occurs when too much bass is sent to them.  I already had a HPF on the vocals up to 100 Hz, and I think ultimately I may have to put it on even the music channels, or at a minimum EQ it down significantly, so that I can maintain clarity and not tax the speakers.  Based on the info above, I will likely lower the amp output a little bit as an added measure of protection.

Question for you guys.  I am going to go work more on it early next week.  In lieu of what I have to work with, do you guys think adding a sub would help with the absence of much low end on those open backed ceiling speakers, or would it just serve to muddy things up with that much low frequency in that sound environment?  I was thinking about looking into getting a simple adapter at Radio Shack for an old Velodyne 390W subwoofer that I have that's not in use, and seeing if I could test that from the Aux 6 output just to see what happens.  What do you think? 

 
 

[BlendedMix]

When you previously mentioned Cafetorium I was afraid of what I am only too familiar with. A dual purpose room sometimes a gym or cafeteria with lots of echo and no acoustics to speak of. That in and of itself is a nightmare. Tell me otherwise.
"I was told that since this was being run as a 70V bridged mono system, I would want to run the amp at full power, and then set my gain structure through the pre-amp and channel faders. "
A quick look at the specs shows a 256 watt (without insertion loss considered) set of speakers (64 W each). and a QSC amp that is happy driving 600 W at 70V direct (transformer optional). I would say that the advise was bad. Back off the amp. If you have good dynamic material (12 to 14 db crest factor) or a sine wave (steadier signal for setup) available set your DL output at 0 (fader and signal that's ~ 6dBu) and then with earplugs bring up your amp to an acceptable level (OSHA has max requirements not to be exceeded. Your in a school not a nightclub). A SPL meter helps with this. The clip on the QSC is 22 dBu the max output of the DL is 21dBu no clipping on the amp but you will hit the Power limit and transformer core saturation long before this. As Greg pointed out remove the low end with a high pass filter and toss the Aphex and you'll have a truer sound. The spec on the speaker states 35Hz but no tolerance is given and clearly suspect. With these settings you are with lots of headroom and good S/N ratio. Remember your setting the average level not peak. 84 dB SPL is what I shoot for. Once you have the main loudness adjusted then you can play with material eq's, preamps etc. I personally think that the Aphex is nothing but a distortion on sweet young voices unless you want them sounding like the Mackie's Belch King (see DL ad). Acoustic sound treatment would help tremendously but it will be hard to convince the powers to be. Once this is accomplished you could revisit the amp setting and fine adjust. That leaves only the Mic positions and non-existent mic technique to keep you busy riding faders. As to the lone unnamed hanging speaker on a Behringer amp I would say it's probably a better setup than your ceiling speakers. Without info ? What is your DL driving? the Ceiling Spk's or the hanging speaker or both? Back in the stone age we were more concerned about Raptors than Flying threats so our hearing evolved much better in the horizontal plane than the vertical. And so horizontal speaker systems are more efficient for our ears than ceiling ones.

To clarify, that QSC amp is driving four sets of the 3 speaker arrays at 256W per array right?  Again, I'm a noob at this, but is it not additive?  Is that the reason for the 70V system and the speakers being wired in succession?  Apparently, electricity is not my strong suit either.    

[WK154]

4 speaker sets x 64 watts per set(array) = 256 watts at 70V. The spec is per set not per speaker. Could have been clearer, my bad.

[BlendedMix]

4 speaker sets x 64 watts per set(array) = 256 watts at 70V. The spec is per set not per speaker. Could have been clearer, my bad.

No worries...I didn't even look at what each speaker was rated, and the spec was right there in front of me.