Gain Structure WITHIN the DRPA?

Kyle Abel

I saw this post in another forum and decided to give it a shot. I have previously set my gain structure based on the tips in this forum, and have had no problems. But there's always room for improvement, right? I ran the pink noise as the post instructed, brought it up to 0dB, then adjusted the gain on the crossovers to get 0dB on both the HI's and LO's. The end result was I had to bump the gain on the HI's +2dB, and the LO's +9dB. This seemed excessive and thought I'd run it by the experts...

For a real treat, set your crossover upper and lower bandpass freqs at manufacturers specs, and crossover gains near 0dB for now, with the GEQ turned off, all accessories like subharmonic synth OFF(!) and just push the RTA button with NO mic connected.

Did you catch that? NO MIC CONNECTED.

Just push the RTA button for free pink noise.

Compressor and limiters turned off.

Amplifier inputs turned ALL THE WAY DOWN!!!!!!

Page forward on dbx button, turn up the pink noise volume.

Look at all dbx input and output meters. Input meters will be OFF because there ain't no mic connected.

Are all your output meters hovering around the same level?

Do you see any yellow threshold lights above dbx's input and output meters? (YES? Then turn off ALL compressors and limiters!)

You're full range right? Then dbx's high output LED decibel scale should match the pink noise level shown on your LCD display.

NO? Turn the RTA button off. Go into the crossover gain, and adjust it to make it match the output LEDs, when you check your pink noise level input on LCD again.

Repeat above steps.

If you wanted to bi-amp (2x4) or tri-amp (2x6) you could go back and raise the crossover gains of each band pass the same way.

Basically, no matter what setup you have, when you're adjusting gain structure within dbx, you want to make LED inputs match each of your LED outputs.

Then turn your each amplifier gains down from full blast, in each bandpass, (full range for you) so that when dbx clips, your amps may clip slightly 1st, just before dbx does.

In a nutshell...

You have a free pink noise source in your dbx unit.

And a gain structure to set WITHIN your dbx unit as well.

Get your crossovers and EQ flat, and make all dbx input LEDs match dbx output LEDs.

Then test with a console or pink noise source OTHER than dbx, and test input verses output.

So...does this make sense? Is there a need to set a gain structure within the DRPA?

kpippen

Hi Kyle,

Welcome to the DBX User Group Forum... Follow the gain structure procedure outlined about mid-way down in this post...

http://www.dbxpro.com/forum/viewtopic.php?t=959

Cheers!

Kev

Dra

You still have to go back and set everything with mixer providing PN output to the DR. So seems like extra work.
My other concern would be that if it is internal in the DR then -0- should equal -0-, not +2 or +9. The x-over gains are to match amps / speaker sensitivity, etc. The process you reference has nothing to do with that.
Maybe I'm missing something.

What forum did this come from?

DRA

Kyle Abel

Pro Sound Web:

http://srforums.prosoundweb.com/index.p ... 758/18084/
Message #278201, about a third of the way down.

The only thing that I could come up with as to why the output meters on the LO and HI would not match the input meter is because the crossover is cutting out some of the frequencies for each band. I don't have my settings handy, but I know I had set my crossover point at 100hz. So I was thinking that could be a reason why the output on the LO was so much lower than the HI - it was only getting sub-100hz frequencies.

If I'm reading the post that Kev references, the gain on both bands of the crossover should be -0- as long as you can get your amps to clip (which I can). Agree/disagree?

Dra

What did you amp settings look like with those x-over settings?
Have you tried it?

If I'm reading the post that Kev references, the gain on both bands of the crossover should be -0- as long as you can get your amps to clip (which I can). Agree/disagree?
Agree. Increased gain = increased noise.

I'm interested to hear about your results. Keep us posted.

Gadget, Any incite?

DRA

Kyle Abel

After I saw I was going to have to ramp up the Lo band gain to +9, I stopped and went back to what I had - which was Hi at -2 and Lo at -0-. My amps attenuators were roughly at 11 o'clock on both (QSC PLX1804 and RMX2450 respectively).

I'm going to go back and redo the gain structure with both crossover bands at -0-. I was happy with what I had - quiet, lots of headroom, plenty of volume, but thought I'd give the advice in the post a try.

Thanks for the help guys!

Gadget

Well it's a A/D conversion then digital through the DR then D/A so gain structure within the digital domain is less important to signal to noise than in the analog domain...Now I have always professed that unity gain is more important than any other thing in the signal chain so to me leave the input and crossover gains at near '0' vu as possible and set the analog structure for best signal to noise...The process of gain structure then sets the external devices to all talk the same VOLTAGE language. Then we know the signal to noise, unity gain and all are on the same page... This running the DIGITAL portion of the DR at near clipping is a BAD idea in my opinion... NO CLIPPING IN THE DRIVERACK!
Gadget