ProSoundWeb Community

Please login or register.

Login with username, password and session length
Advanced search  

Pages: 1 2 [3] 4   Go Down

Author Topic: Stopping non-sound people from blowing speakers (compressor/limiter)...  (Read 12543 times)

Ivan Beaver

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 9533
  • Atlanta GA

Note that that recommendation was for live conventional music (not EDM). Not for playback, especially with how compressed much of recorded music is these day.


Sent from my iPad using Tapatalk HD
I have started setting my "long term" heating limiters at just over 1/4 continuous rated power.  NOT program power, RMS (as some people like to call it) or continuous.

I used to be in the 1/2 power camp-but have found that with some music style and running it into hard limiting can still kill drivers.

I know it sound stupid to run a 4000 watt peak speaker with the limiter set to around 300 watts, but that is the reality these days-with some music styles.

This is what I call the idiot setting-as is "standard".
Logged
A complex question is easily answered by a simple-easy to understand WRONG answer!

Ivan Beaver
Danley Sound Labs

PHYSICS- NOT FADS!

Scott Carneval

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1048
    • Precision Audio

I have started setting my "long term" heating limiters at just over 1/4 continuous rated power.  NOT program power, RMS (as some people like to call it) or continuous.

I used to be in the 1/2 power camp-but have found that with some music style and running it into hard limiting can still kill drivers.

I know it sound stupid to run a 4000 watt peak speaker with the limiter set to around 300 watts, but that is the reality these days-with some music styles.

This is what I call the idiot setting-as is "standard".

What attack and release times are you using?


Sent from my iPhone using Tapatalk
Logged

Scott Carneval

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1048
    • Precision Audio

Our dbx ZonePro offers appropriate ducking features to keep 96% of vocals in check, but just living in the real world here, if an instructor has the microphone set at a given level - so that their "normal" voice appropriately cuts through the music - when they get more excited for small pushes (also known as "yelling"), they're not going to turn down the volume knob first - so we need some headroom built in.  [Somehow I seem to be the only one capable of "sound like" I'm yelling but actually not producing any more volume...so I've given up trying to coach this]

This is what the input compressor is for. Start with 4:1 ratio and adjust to taste. Different people may require different compressor settings, so you're going to have to settle on a compromise here. Set the threshold so that peaks in their speaking are tamed but 'normal' speaking is unaffected.


Quote
I've decided to upgrade the pendants to the Community DP8 which are on the way. 

Can you get your money back? Ten speakers at roughly $400 each is $4k. With a 24' x 31' room with 20' ceilings, you would get MUCH better sound from a single high-quality speaker mounted high. Or a pair if you really 'need to see' two speakers. Recommendations here are JBL SRX812P (self powered, problem solved), Martin CDD10 or CDD12, or Danley SM100 or SM80.


Quote
The Zone Pro has compressors available on mic channels and limiters available on outputs only, so I'm unable to put a limiter on the mic channel individually.  Does it make more sense to add some compression to the mic or just use the limiter?  I'm thinking I may be able to get away with just using the limiter, setting the threshold slightly higher than max music volume (which is itself limited via the input volume control on the ZonePro) and over easy to off, therefore essentially hard limiting to the extent that it will never affect the music and only affect the mic if it exceeds upper limit of the music threshold (so something like over easy off, threshold and overshoot both at 0db and the attack, hold, release settings to auto).

A compressor and a limiter are fundamentally the same thing. A limiter is just a compressor with an extremely high ratio (usually ∞:1). Beyond that, they both perform the exact same function. I believe the 'compressor' on the Zone Pro even allows you to choose ∞:1.

Set the compressor on the mic channel as I mentioned above. To set the limiter, you could run a signal into the amp until it clips, then back down 3db from there. This will limit the amp to roughly half it's available power. I would set the attack as fast as it will allow, and set the release as slow as it will allow. Using a very long release time will prevent the average signal level from increasing. I install lots of sound systems for 'unqualified operators' and [knock on wood] this method hasn't failed me yet.
« Last Edit: July 24, 2016, 05:27:36 PM by Scott Carneval »
Logged

Ivan Beaver

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 9533
  • Atlanta GA

What attack and release times are you using?


Sent from my iPhone using Tapatalk
It depends (as usual) on the particular drivers-size of voice coil etc.

Typically for subs it is 3 seconds, 2 seconds for "bass" speakers, 1 second for mids and 0.5 seconds for HF drivers.

Release is typically 1x attack or a little longer.
Logged
A complex question is easily answered by a simple-easy to understand WRONG answer!

Ivan Beaver
Danley Sound Labs

PHYSICS- NOT FADS!

Scott Carneval

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1048
    • Precision Audio


Release is typically 1x attack or a little longer.

I assume you mean 16x?

Do you ever have issues with the system coming out of alignment when the limiters engage?

Meaning, when the HF limiter is triggered after only 500ms and is attenuated but the mids are not, this could throw the magnitude alignment off, and potentially the phase alignment off as well. But then another 500ms later the mids will also be attenuated, so maybe it's not an issue.

I suppose the 2 second delta of mid to subs limiting would be more apparent. Just thinking out loud...


Sent from my iPhone using Tapatalk
Logged

Ivan Beaver

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 9533
  • Atlanta GA

I assume you mean 16x?

Do you ever have issues with the system coming out of alignment when the limiters engage?

Meaning, when the HF limiter is triggered after only 500ms and is attenuated but the mids are not, this could throw the magnitude alignment off, and potentially the phase alignment off as well. But then another 500ms later the mids will also be attenuated, so maybe it's not an issue.

I suppose the 2 second delta of mid to subs limiting would be more apparent. Just thinking out loud...


Sent from my iPhone using Tapatalk
No.  For "normal" limiters, a release of 15 or 16x is fine.  But for the thermal limiters-with their long attack times, the 1x release seems to work fine.  This is because of the already long attack times

The thermal limiters are for driver protection.  When you are getting that close to possible damage, the last thing I am concerned with is the system coming out of alignment.

Keeping the system working is MUCH more important.

Using the same attack times for all the freq bands does not make sense-from a protection aspect. 

The time for the lows would cause the highs to fail.

Using the one for the highs would cause the lows to limit way to soon.

What some people don't realize is how limiters actually work.

They don't "wait" until the time constant to limit.

The start to limit once the threshold has been exceeded.  It takes up to the time constant to achieve full limiting.

So it is a gradual reduction, not sudden.

And even if it was noticeable, then that could be a "good" thing, as it would give you an audible clue to turn it down.

In fact if you do blind listening, with a moderate amount of limiting, most people cannot tell that it is happening.

It is something that in most cases you really need to listen hard to pick it out-if you didn't see the meters.

The average listener would not be able to tell.
Logged
A complex question is easily answered by a simple-easy to understand WRONG answer!

Ivan Beaver
Danley Sound Labs

PHYSICS- NOT FADS!

Scott Carneval

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1048
    • Precision Audio

No.  For "normal" limiters, a release of 15 or 16x is fine.  But for the thermal limiters-with their long attack times, the 1x release seems to work fine.  This is because of the already long attack times

That makes sense. I didn't think that through before posting. Using 16x on a 3 sec attack would be a bit excessive.

Quote
The thermal limiters are for driver protection.  When you are getting that close to possible damage, the last thing I am concerned with is the system coming out of alignment.

Keeping the system working is MUCH more important.

I agree. But if your threshold is 1/4 continuous power it seems like the system would be in limiting an awful lot, especially the HF.

I don't have much experience using true 'continuous' limiters as we almost exclusively use the Ashly ne8800 processors which only have single-stage limiting We usually set them with a very short attack but a very long release. They become more or less 'brick wall' limiters but the long release time prevents the RMS signal from being raised.





Sent from my iPhone using Tapatalk
Logged

Ivan Beaver

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 9533
  • Atlanta GA

That makes sense. I didn't think that through before posting. Using 16x on a 3 sec attack would be a bit excessive.

I agree. But if your threshold is 1/4 continuous power it seems like the system would be in limiting an awful lot, especially the HF.

I don't have much experience using true 'continuous' limiters as we almost exclusively use the Ashly ne8800 processors which only have single-stage limiting We usually set them with a very short attack but a very long release. They become more or less 'brick wall' limiters but the long release time prevents the RMS signal from being raised.





Sent from my iPhone using Tapatalk
If you were to look at a typical signal that has been high passed above 1Khz, you will see that the average power is WELL below the peaks.  The peaks at those freq are very short in duration.

So 1/2 second may seem like a short time-but it is pretty long when considering the freq involved.

Of course it is best to use several limiters-with different attack times and levels.

If I only have 1 available, I generally limit at the continuous level.  With an attack time of the time constant of the lowest freq in the intended passband. 

So for 1K that is 1ms, for 100hz 10ms. In "extreme" situations that attack time should probably be shorter.  Release around 16x or so.

Of course this can present a problem for HF devices in a passive box.

There is A LOT of room available for each particular situation.

So the saying "it depends" REALLY comes into play.

But it is FAR better to limit early and faster than normal-JUST to be safe.

A couple of extra dB for a short period of time IS NOT WORTH the system going down for the rest of the show.  At least in my opinion.
Logged
A complex question is easily answered by a simple-easy to understand WRONG answer!

Ivan Beaver
Danley Sound Labs

PHYSICS- NOT FADS!

Tim McCulloch

  • SR Forums
  • Hero Member
  • *
  • Offline Offline
  • Posts: 23729
  • Wichita, Kansas USA

That makes sense. I didn't think that through before posting. Using 16x on a 3 sec attack would be a bit excessive.

I agree. But if your threshold is 1/4 continuous power it seems like the system would be in limiting an awful lot, especially the HF.

I don't have much experience using true 'continuous' limiters as we almost exclusively use the Ashly ne8800 processors which only have single-stage limiting We usually set them with a very short attack but a very long release. They become more or less 'brick wall' limiters but the long release time prevents the RMS signal from being raised.





Sent from my iPhone using Tapatalk

Anyone bent on driving a system over the edge - for whatever reason - will find a way to do so and eventually system limiters will not be enough (or those limiters were already crippling the system).  Ivan has addressed this before.

My findings about long term power limiting are also in line with Ivan's.  We've yet to have a thermal failure in any pass band when using I-Tech or I-Tech HD and the sophisticated limiting they have and basing the power limits on 1/4 to 1/3 of "continuous rating".  For peak voltage limiting I use the fastest attack time the device allows for HF...
« Last Edit: July 25, 2016, 12:21:57 AM by Tim McCulloch »
Logged
"If you're passing on your way, from Palm Springs to L.A., Give a wave to good ol' Dave, Say hello to progress and goodbye to the Moonlight Motor Inn." - Steve Spurgin, Moonlight Motor Inn

Justin Goodman

  • Jr. Member
  • **
  • Offline Offline
  • Posts: 81

This is what the input compressor is for. Start with 4:1 ratio and adjust to taste. Different people may require different compressor settings, so you're going to have to settle on a compromise here. Set the threshold so that peaks in their speaking are tamed but 'normal' speaking is unaffected.

Set the compressor on the mic channel as I mentioned above. To set the limiter, you could run a signal into the amp until it clips, then back down 3db from there. This will limit the amp to roughly half it's available power. I would set the attack as fast as it will allow, and set the release as slow as it will allow. Using a very long release time will prevent the average signal level from increasing. I install lots of sound systems for 'unqualified operators' and [knock on wood] this method hasn't failed me yet.

Logistically how does one test the threshold on a mic channel?  Are you saying to yell into a mic with trial and error until I find the threshold at which the amps clip, then lower the threshold 3db from there? 

RE: Pendants vs. higher powered traditional PA speakers, we decided to use pendants for a few reasons. 

One is that we wanted even sound all around the room ... so either more speakers closer to people or larger speakers farther away -- but we did not want an environment where the front row was significantly louder than the 3rd row.  The problem with the larger speaker at greater distance approach is that we're in a busy shopping center with neighbors and it would require more sound abatement--which is expensive in construction. 

Additionally, the bar joist on the exposed ceiling was able to hang pendants as is--another cost saving in construction.

Our installer is also willing to use the 67's on another job (not at full refund, but a helpful amount), so we won't have to write them off totally.  The DP8's we got for ~$285/ea all in, so $2900 or so and we get to use the amps, wiring, etc that we already have in place.  I'm comfortable running cable, but not 20' up, so again this is a cost saving.  Jumping on a 10' ladder and quickly swapping speakers OTOH, I don't mind doing. 
Logged

ProSoundWeb Community


Pages: 1 2 [3] 4   Go Up
 



Site Hosted By Ashdown Technologies, Inc.

Page created in 0.045 seconds with 25 queries.