ProSoundWeb Community

Please login or register.

Login with username, password and session length
Advanced search  

Pages: [1] 2 3 ... 6   Go Down

Author Topic: gain staging a digital rig, DSP, and powered speakers  (Read 24405 times)

Mark McFarlane

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1865
  • Middle East
    • Arkose Records
gain staging a digital rig, DSP, and powered speakers
« on: September 23, 2013, 10:11:22 am »

I'm rethinking/redoing my gain setup for my rig. This will be a multi-part post, I'll ask some more questions after early ones are answered.

The Gear:

01V96 all outputs (L, R, Omni) at +24 dBU max output
connected via ADAT to >

DEQ830 8 channel EQ (ADAT in and out)
connected via ADAT to >

ADA 8000 outputs at +16 dBU
connected analog to

Sabine 4802 DSP, outputs at +18 (manual) or +20 (engineer's specs)
connected analog to >

various RCF TT and NX powered speakers
with built in 'soft' (yellow/orange light)
the TT's have RMS (red light) limiters.
Docs say occasional yellow is OK, continuous red will result in damage.

My first goal is to get the Sabine 4802 input meters to match the output meters of the 01V96.  Doing it 'by eye', +5dBu on the DSP inputs seems to give a decent match.  The specs indicate +6 should match the ADA to the 01V96 better,... perhaps a discussion point: slightly lower than spec'd output on the 01V96, or higher on the ADA8000, incorrect personal observation,...

My second goal is to get the speakers to tickle the soft limiter when the console (and DSP) are at 0 DBFS.

First Question: Are these reasonable goals, any problems so far?

There are various options for achieving goal number 2 (simultaneous clipping), such as adjusting the built-in amp sensitivities versus adjusting the output gains on the console and DSP, and for example, maxing out the amp input sensitivities. 

Let's postpone the output sensitivity discussion until the first two goals are qualified.
« Last Edit: September 27, 2013, 09:42:17 am by Mark McFarlane »
Logged
Mark McFarlane
ARKOSERECORDS
Turn down what's too loud.

Scott Wagner

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1020
  • Richmond, VA
Re: gain staging a digital rig, DSP, and powered speakers
« Reply #1 on: September 23, 2013, 12:43:28 pm »

I'm rethinking/redoing my gain setup for my rig. This will be a multi-part post, I'll ask some more questions after early ones are answered.

The Gear:

01V96
L, R, OMNIs, all at +24 dBU max output
connected via adat to >
DEQ830 8 channel EQ
connected via ADAT to >
ADA 8000
Outputs at +16 dBU
connected analog to
Sabine 4802 DSP,
Outputs at +18 (manual) or +20 (engineer's specs)
connected to >
various RCF TT and NX powered speakers with built in 'soft' (yellow/orange light) and the TT's have RMS (red light) limiters. Docs say occasional yellow is OK, continuous red will result in damage.

My first goal is to get the Sabine 4802 input meters to match the output meters of the 01V96.  Doing it 'by eye', +5dBu on the DSP inputs seems to give a decent match.  The specs say +4 should work better,... perhaps a discussion point: slightly lower than spec'd output, incorrect personal observation,...

My second goal is to get the speakers to tickle the soft limiter when the console (and DSP) are at 0 DBFS.

First Question: Are these reasonable goals, any problems so far?

There are various options for achieving goal number 2 (simultaneous clipping), such as adjusting the built-in amp sensitivities versus adjusting the output gains on the console and DSP, and for example, maxing out the amp input sensitivities. 

Let's postpone the output sensitivity discussion until the first two goals are qualified.
One point before we begin this exercise: I've never been concerned with getting the output devices (amplifiers and speakers) to clip when the rest of the system does.  I've always (in the analog world) made sure that all devices in the signal chain prior to the output devices did, then set the output devices to give the desired level (situationally dependent) with a reasonable amount of headroom built in.  My goals for this exercise were to increase S/N ratio of the system as a whole.  I'm not convinced this is really an issue in the digital world (assuming reasonable levels).  Thoughts?
Logged
Scott Wagner
Big Nickel Audio

Mark McFarlane

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1865
  • Middle East
    • Arkose Records
Re: gain staging a digital rig, DSP, and powered speakers
« Reply #2 on: September 24, 2013, 10:10:28 am »

Continuing on,...

third goal:  Be able to set the sensitivity on my 9 different powered speakers (mains, mons, subs) 'one time' so they don't limit regardless of whether they are plugged into the 01V96 (+24dBu), the ADA8000 (+16 dBu), or the Sabine DSP (+18 dBu).

Actually I have a few other consoles I use for small gigs that I need to look up...

This can be accomplised by using output attenuation on the 01V96 (-8) and Sabine 4802 (-2) to match the fixed gain of the ADA8000.

Crazy idea?
Logged
Mark McFarlane
ARKOSERECORDS
Turn down what's too loud.

Art Welter

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1775
Re: gain staging a digital rig, DSP, and powered speakers
« Reply #3 on: September 24, 2013, 03:58:14 pm »


My second goal is to get the speakers to tickle the soft limiter when the console (and DSP) are at 0 DBFS.
connected to various RCF TT and NX powered speakers with built in 'soft' (yellow/orange light) and the TT's have RMS (red light) limiters. Docs say occasional yellow is OK, continuous red will result in damage.


First Question: Are these reasonable goals, any problems so far?

At 0 DBFS your console has no headroom (if 0 DBFS is indeed 0 DBFS), so anything above that level will be distorted, and a little digital clipping goes a long ways.

Clipping increases HF content which tends to fry tweeters, the last thing I'd do is cut the speaker input to the point that digital clipping may occur without using the full available output range of the speaker.

I'd set the speakers inputs to hit the red lights at about -6 on the console (or downstream devices), and use -6 as the "don't go much louder than this" place to stop pushing things.

Logged

Mac Kerr

  • Old enough to know better
  • Administrator
  • Hero Member
  • *****
  • Offline Offline
  • Posts: 6880
  • Audio Plumber
Re: gain staging a digital rig, DSP, and powered speakers
« Reply #4 on: September 24, 2013, 04:28:07 pm »

At 0 DBFS your console has no headroom (if 0 DBFS is indeed 0 DBFS), so anything above that level will be distorted, and a little digital clipping goes a long ways.

Clipping increases HF content which tends to fry tweeters, the last thing I'd do is cut the speaker input to the point that digital clipping may occur without using the full available output range of the speaker.

I'd set the speakers inputs to hit the red lights at about -6 on the console (or downstream devices), and use -6 as the "don't go much louder than this" place to stop pushing things.

I would be even more conservative, since on most pro consoles have an analog output of +24dBu at 0dBfs (not all, DM2000 is +18dBu). That would make -20dBfs eqivalent to +4dBu, the old analog standard for 0Vu. While it has become common to run everything at higher levels than we used to, it is OK to set a more conservative standard. Headroom is a good thing.

On the pretty big widely seen event I am mixing on right this second I treat -12dBfs as my maximum intentional level.

Mac
Logged

Mark McFarlane

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1865
  • Middle East
    • Arkose Records
Re: gain staging a digital rig, DSP, and powered speakers
« Reply #5 on: September 27, 2013, 09:18:58 am »

I would be even more conservative, since on most pro consoles have an analog output of +24dBu at 0dBfs (not all, DM2000 is +18dBu). That would make -20dBfs eqivalent to +4dBu, the old analog standard for 0Vu. While it has become common to run everything at higher levels than we used to, it is OK to set a more conservative standard. Headroom is a good thing.

On the pretty big widely seen event I am mixing on right this second I treat -12dBfs as my maximum intentional level.

Mac

Note, I never said I would be running the console at 0 DBFS, I just proposed that 0 DBFS on the console would be the point where the speakers just started going into their first stage of limiting.

From the perspective of speaker preservation, keeping the speakers out of limiting, even at 0 DBFS, might be considered the more conservative approach, yet I don't think that is what you meant Mac, or is it?
« Last Edit: September 27, 2013, 09:33:14 am by Mark McFarlane »
Logged
Mark McFarlane
ARKOSERECORDS
Turn down what's too loud.

Mark McFarlane

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1865
  • Middle East
    • Arkose Records
Re: gain staging a digital rig, DSP, and powered speakers
« Reply #6 on: September 27, 2013, 09:25:33 am »

I'd set the speakers inputs to hit the red lights at about -6 on the console (or downstream devices), and use -6 as the "don't go much louder than this" place to stop pushing things.

If I set the speakers to hit the red lights at -6 DBFS (red=second stage limiting, where RCF says damage WILL occur) , then I am running a much higher risk of speaker damage, particularly when I am not the operator.

At 0 DBFS your console has no headroom (if 0 DBFS is indeed 0 DBFS), so anything above that level will be distorted, and a little digital clipping goes a long ways.

By definition, one can't get anything above 0 DBFS.  Are you worrying about about intersample peaks?  Or worried about not enough rig for the gig?  Remember, I never said anything about running the console at 0 DBFS, I only proposed having the first stage limiters kick in a 0 DBFS.

Art, I think I understand where you are coming from, but I'll let you explain rather than ASSume.
« Last Edit: September 27, 2013, 09:31:17 am by Mark McFarlane »
Logged
Mark McFarlane
ARKOSERECORDS
Turn down what's too loud.

Bob Leonard

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 6807
  • Boston, MA USA
Re: gain staging a digital rig, DSP, and powered speakers
« Reply #7 on: September 27, 2013, 10:37:02 am »

Just a question about the speaker thing. Regardless of speaker type or manufacturer the amplifier sensitivity is not a feature designed for, or which should be used for limiting output. Regardless of the setting (unless off) peak amplifier output can be reached. 0db on the board through to the last piece of hardware should be 0db.

If the system hardware is properly matched then there is no reason to do anything more with amplifier sensitivity then lower the gain of the amp to the point where there is no audible hiss.

0db should then be considered the point where all devices are at 0db, and your speakers are working at their maximum output. If your system drives the amplifiers/speakers to a point where maximum output is exceeded then IMO the output levels should be reduced at the DSP, XOVER, or in the case of a digital desk, at the desk outputs prior to outboard devices. But, that's just me and all I can say is I've always had plenty of headroom, as much sound as I've ever needed, and not blown a speaker in over 20 years.
Logged
BOSTON STRONG........
Proud Vietnam Veteran

I did a gig for Otis Elevator once. Like every job, it had it's ups and downs.

Art Welter

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1775
Re: gain staging a digital rig, DSP, and powered speakers
« Reply #8 on: September 27, 2013, 11:35:42 am »

If I set the speakers to hit the red lights at -6 DBFS (red=second stage limiting, where RCF says damage WILL occur) , then I am running a much higher risk of speaker damage, particularly when I am not the operator.

By definition, one can't get anything above 0 DBFS.  Are you worrying about about intersample peaks?  Or worried about not enough rig for the gig?  Remember, I never said anything about running the console at 0 DBFS, I only proposed having the first stage limiters kick in a 0 DBFS.

Art, I think I understand where you are coming from, but I'll let you explain rather than ASSume.

If the first stage limiters kick in a 0 DBFS, your console will be clipping above that, you are leaving speaker headroom on the table.
Continuous second stage limiting, where RCF says damage WILL occur, occurs because of long term average power, which the second stage limiting apparently is not capable of controlling.
I'd guess that to be around 10-12 dB more average level than the point where the first stage limiters kick in.

By definition, one can't get anything above 0 DBFS, except for an increased duty cycle, the more clipping occurs, the higher the average power becomes. In addition, the higher clipped average contains more HF harmonics, which puts more average power to the tweeters, which have lightweight voice coils which can be burnt fairly quickly.

My experience in looking at digital outputs for live use is mostly with my DEQ 830, the range is pretty compressed at the top of the meter scale, -15 is halfway up the meter.
After looking more closely at the scale, I'd defer to Mac's comment "I treat -12dBfs as my maximum intentional level".

At any rate, I'd probably be more likely to cause damage to a speaker that had been turned down to the point where I had to use  O DBFS just to make a peak hit a first stage limiter, getting the next 6 dB average that is generally available would require gross board clipping.

Of course, I'd just walk up to the speaker and turn it up if I found myself in that situation, I hate amplifying digitally clipped signals.

Logged

George Dougherty

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 724
Re: gain staging a digital rig, DSP, and powered speakers
« Reply #9 on: September 27, 2013, 01:01:46 pm »


If the first stage limiters kick in a 0 DBFS, your console will be clipping above that, you are leaving speaker headroom on the table.
Continuous second stage limiting, where RCF says damage WILL occur, occurs because of long term average power, which the second stage limiting apparently is not capable of controlling.
I'd guess that to be around 10-12 dB more average level than the point where the first stage limiters kick in.

Given the yellow limiting does no damage and the red limiting continuous can cause speaker damage, it seems reasonable that yellow might be at RMS handling or wherever xmax occurs on a typical music signal and red would be only ~3db higher at program handling.
Logged

ProSoundWeb Community

Re: gain staging a digital rig, DSP, and powered speakers
« Reply #9 on: September 27, 2013, 01:01:46 pm »


Pages: [1] 2 3 ... 6   Go Up
 



Page created in 0.05 seconds with 22 queries.