Print

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

[Edd Lewington]

New to the forums, I'm an installer more used to pulling cable and building racks etc. I'm just starting to try and teach myself programming and system set up. I'm hoping someone can help with this problem.

I'm practising by writing a system file for a BSS soundweb Blu 50. The end client wants to utliise the Ambient Noise Compensator:

https://audioarchitect.harmanpro.com/aa_help/Soundweb_London/Signal_Processing_Objects/Dynamics/Ambient_Noise_Compensator_Non_Gap.htm

This processing object will automatically adjust program material (background music) to sit at a prescribed level above the ambient/background noise level. In this instance it's for a restaurant, through some research I can see the average level of conversation is about 65db (I assume this to be SPL or an A weighted measurement). The plan being that the program material will sit 3 - 6db above this. The ambient noise level will be measured using a condensor microphone and an ambient input channel on the compensator processing object.

The threshold of the processing is measured in Dbu which I understand is a reference of voltage. What I don't understand is how to set the threshold limit using the 65db SPL as the reference point for adjustment of the program material.

Am I right in thinking I need to convert DB SPL to DBu? If so how do I go about doing this?

Thanks to anyone who can heklp me with this problem!

[Mac Kerr]

New to the forums

Thanks to anyone who can heklp me with this problem!

Please go to your profile and change your name to your real full name as required by the posting rules.

Mac
admin

[Edd Lewington]

Sorry didn't realise the rules, thanks for unlocking

[John Roberts {JR}]

New to the forums, I'm an installer more used to pulling cable and building racks etc. I'm just starting to try and teach myself programming and system set up. I'm hoping someone can help with this problem.

I'm practising by writing a system file for a BSS soundweb Blu 50. The end client wants to utliise the Ambient Noise Compensator:

https://audioarchitect.harmanpro.com/aa_help/Soundweb_London/Signal_Processing_Objects/Dynamics/Ambient_Noise_Compensator_Non_Gap.htm

This processing object will automatically adjust program material (background music) to sit at a prescribed level above the ambient/background noise level. In this instance it's for a restaurant, through some research I can see the average level of conversation is about 65db (I assume this to be SPL or an A weighted measurement). The plan being that the program material will sit 3 - 6db above this. The ambient noise level will be measured using a condensor microphone and an ambient input channel on the compensator processing object.

The threshold of the processing is measured in Dbu which I understand is a reference of voltage. What I don't understand is how to set the threshold limit using the 65db SPL as the reference point for adjustment of the program material.

Am I right in thinking I need to convert DB SPL to DBu? If so how do I go about doing this?

Thanks to anyone who can heklp me with this problem!

not exactly apples and oranges, but all dB so can be related.

I would try a seat of the pants calibration where you play x dBu of noise through your system and see how much dB SPL you measure. Raise the level up until you get a clean measurement. You should be able to work backwards from there to establish how many dBu correlate to how many dB SPL. 

JR

[Steve-White]

Ambient noise compensator?  Interesting idea.  The whole house music system at my place evolved over a 15+ year period.  Tried a minimalist approach, then a tapped autoformer type zone controller which is just in/out switches on the autoformer taps.  In the end there's a 15" powered sub at the far end of the hallway, a 12" sub in my shop, bedrooms, bathrooms and kitchen went from wall hang bookshelf on adjustable mounts to in-wall.  I still use bookshelf in shop and patio on adjustable mounts.  Took lots of amp channels, so I used a couple of HT 5ch amps and a 2ch for the patio zone so amp can be switched off at night.

Now the control part of the equation.  I ended up using a pair of dbx Driverack PA's and set crossover frequency to 120hz.  Shop sub covers the patio area and hall sub the bedrooms and bathrooms along with the main sub in the den in the middle of the house.

Went through the system with pink noise at 75 db and adjusted levels to be even at the listening areas.  I've done a good deal of club work, some of which served food, lunch time paging, cocktail hour with a band in one area and such.

That will be something that the algorithm would need to be adjustable almost to the point of being parametric.  Not throwing cold water or naysayin' on you.  Ambient levels vary greatly in a club/restaurant depending upon capacity and time as you are probably already aware of.

Maintaining the perceived level of background against the ambient level sounds like a very interesting project.  I would guess the system would move ~10db, maybe even more with the final result being "how it sounds" which will probably be more subjective when compared against the background level.

Loudspeaker placement will be important, I would venture to guess nothing will be near-field.  That was something I learned during the evolution of the house system.  The shop and patio speakers are near-field monitors.  4 x speakers ceiling mounted in the shop and two in the patio directly over the table seats.

How to pick ambient out from the background program.  Mics for picking up the ambient white noise levels in the room - placement, how many, etc.

You have a very interesting project.  A bit more complex than the music system in the Grand Cherokee that senses speed and adjusts audio level according to road noise -vs- speed.

Keeping the perceived level of the background music relative to ambient white noise will be of interest to me.  I don't think it will track in a linear fashion.

[Geert Friedhof]

You can't convert this easily with math.
Things you would need to know include, but not limited to:

-sensitivity microphone
-distance soundsource to mic
-input gain soundweb
-input sensitivity amplifier
-gain amplifier
-sensitivity speakers
-distance listener to speaker

Like JR said: just install it and calibrate.

Question: Would this not create a feedback loop where the speakervolume is raised, causing the mic to register a higher ambient level, causing it to raise the volume etc?

[Luke Geis]

I think you are overthinking it. The measurement of the microphone is converted to an electrical level in the processor. Whether it's 100db or 2db it doesn't matter. If your mic produces X input, and you want the speaker to go up +6db, you can convert dbu to db with simple math. In your case, all you really care about is that the added output signal is equal to +6db ( or whatever your spec is ). My hope would be that the unit has a comparator in it. It should be able to hear the music in the microphone signal, compare it against the "ambient " non-uniform sound and see that the music is +6db higher than the ambient or not and adjust. It has to be a PID-type circuit where it constantly tries to fix what it hears against what it knows until things = 1. The volume adjustment is biasing out the known source so that it can compare the unknown data. When the known data is the same as the non-uniform data, the PID is complete. Conversely, if you wanted the music to be -6db under the ambient, it would bias the known data UP so that when it is equal to the non-uniform data, you get -6db.

The dbu signal sent out vs the sensitivity of the amplifier will be where the math is required. This can be done by calibrating or by doing some math. Or simply using an SPL meter.

[David Sturzenbecher]

Unless you can fully isolate the microphone from the speakers you need a processor with an AEC algorithm (not a BLU-50) to cancel out the perceived increased ambient noise that the speaker put into the room. If you don’t, the system could possibly “run away” on its own... turning itself up louder and louder due to its own signal.  This practice is a very common in commercial AV, and BSS has written up tutorials on how to do it properly with their other processors

[Russell Ault]

{...} If you don’t, the system could possibly “run away” on its own... turning itself up louder and louder due to its own signal.  {...}

I always chuckle to myself whenever I find out that a tool that is supposed to make our lives easier has a "malicious compliance" mode.

I'll add this one to my list of "unorthodox types of feedback to look out for" just below "resonant frequency of unmuted cello strings".

-Russ

[Chris Grimshaw]

Unless you can fully isolate the microphone from the speakers you need a processor with an AEC algorithm (not a BLU-50) to cancel out the perceived increased ambient noise that the speaker put into the room. If you don’t, the system could possibly “run away” on its own... turning itself up louder and louder due to its own signal.  This practice is a very common in commercial AV, and BSS has written up tutorials on how to do it properly with their other processors

That was my first thought.

I think there'd also be some level of audience involvement in that feedback loop: music gets louder, so punters talk louder, so the processor pushes things further. In the end, the SPLs end up much too high for a restaurant setting. I think I'd just put a couple of volume controls somewhere and let the floor manager set levels.


Chris

[ProSoundWeb Community]