The AM Forum

Hola,

I have a Behringer 9024  DSP  6-band  audio compressor/limiter processor. I've used it on and off for years, but always took it off-line in the end.  I'm satisfied with what I am using now, but the nagging idea of multi-band keeps haunting me.   Right now I have no compression in the chain, just EQ > LP filter > peak limiting > rig.

I'm going for the natural, high dynamic range sound and fear by adding the 9024 it will raise my background blower noise and make the audio too busy for my tastes.

Has anyone experience with multiband compressors setting them up to be transparent while adding a benefit to the table?  

I read that multi-band compressors can widen out a signal due to more energy in the higher sidebands.  Have you experienced this?

IE, does a multiband work for you or are you satisfied with a single band compressor/limiter?

T

I have nine multiband b'cast processors and don't any them for voice-only amateur radio use. Multiband processors were designed to address a necessity and voice transmissions are not it.

In fact, even the top of the line digital broadcast processors reduce the number of discrete bands when set to factory format presets such as News, Talk, and Sports for a reason.

I agree with BCG.  I have a Schlockwood 200 (I know, it's a dumb name) that was developed by Jim Wood, founder of Inovonics.  It's mainly aimed at the Part 15 LPAM broadcaster market but a few AMers are using it.  It has a 3-band processor with blinking LEDs to show activity in each band.  I never see any activity in the low and high bands, only in the mid band.  Even Jim admits that 3 bands are overkill for ham radio.  Here's his quote in a recent email exchange.

I still really question the necessity of a multiband processor for voice-only transmission.  When you are working with a complex program waveform, multiband compression or limiting can impart a constant spectral profile to the program.  “Spectral Loading” is a term we used for this, meaning that any part of the spectrum that is low in energy gets pumped-up.  With the human voice this is not an issue, as you can’t voice two tones at once.  Sibilance, fricatives and ‘breathiness’ are as easily accentuated with static equalization, and will sound better as the spectral balance won’t be changing all the time.

Rod

A key takeaway from the Behringer 9024 manual:

"Unlike analog technology with its response speed limitations regarding changes in amplitude, digital signal processing permits differences in amplitude to be identified in advance using a preliminary signal delay. Increasing this preliminary delay also increases the potential for the intelligent control. Even “looking ahead” by only a few samples is sufficient to ensure the intelligent application of dynamic processing – such as limiting, which ensures an absolutely reliable signal ceiling – without clipping."

For a natural sound, the Telos Omnia boxes are amazing.

You can also use the digital plugins for podcast type work.  They impart a minimalist approach.

--Shane
KD6VXI

That Behringer multiband processor was my go-to for AM before I moved 14 years ago. A fantastic piece of gear and very much adjustable. I still have it.
My only complaint was the slight latency when live monitoring during transmission with headphones. The slight delay is annoying when trying to monitor your own signal.

It's been a while but I believe the individual bands are adjustable regarding compression ratio so you can set some at zero compression. These are a heck of a lot cheaper than typical broadcast processors. I'd love to have an Optimod but they are hard to find and expensive.

I take credit for coining the phrase "the poor man's Optimod" on the air 16 years ago..  ;D ;D

Our radio stations were "early adopters" of digital processing and purchased Orban 8200s for our two FMs, replacing the analog processors. This was done late on a Sunday night to minimize disruption (keep reading to see how that worked out).

Overnights were automated and our classic rock morning show was hosted by a friend. All seemed fine until 10am when the midday shifts started. This is when all hell broke loose.

"WHAT THE %%@%@ IS GOING ON?!?!?" screamed the Program Director and midday host on the rock station (at my office door, not on the air). This was his first experience with hearing his voice slightly delayed in his headphones. After my explanation that these were the new processors he continued his firestorm with management and it was agreed to revert back to the analog processors until a solution could be found. We went all-music for a while.

I asked my friend how he was able to get through four hours of that when his boss went off the rails after his first break. He calmly replied that when he noticed the delay, he held his headphones near one ear and set just loud enough so he could hear what was going on. See why we're friends?

Meanwhile on our FM oldies station, the morning show host had sounded like he was intoxicated byyyyy draaaaagiiiiinnng ouuuuut hiiiiiiisssss wooooorrrrddddsss.

Freakin' mess those 8200s.  The solution was to use no-latency analog processors to feed the studio headphones. Fortunately, digital technology has come a long way, but there will always be latency. It's just a matter of how much.    

Great story!  :D

In my experience,  in general,  pop-music broadcasters have always been terrible abusers of compression.... and that damn "aural exciter" !!

That would have made a nice aircheck tape. ;D

Hi Glenn,


Good to hear you back on after so many years!  HO-K on the "poor man's Optimod."

Yep, I had the same feelings about that 9024 delay and always found an excuse to take it off line again and again.  It's been a delay whirl lately - I gots a delay in my SDR (receiver compared to the analog FT-1000D receiver)  and now the 9024 delay.

The 9024 was also increasing my blower noise to a level that sounded like a machine shop and if that's not enuff, I started to hear some RF getting into the audio. I couldn't get it outa there fast enuff and now back to a quieter shack with no compression and just 5DB of peak limiting in no-delay DSP.

Clark, reverbitis is a serious hybrid affliction that hits the "talent" from time to time.

BTW, I'm working on my new Tom Vu rendition of  Michael McDonald's sock-in-the mouth,  "What a Fool Believes."   You might hear it on this weekend before it goes platinum...  ;D


T

Ahhhh! Master Vu!.. your Kung-fu is much more powerful than mine!

I'll probably go back to using an analog processor. I have a DBX 162SL in use in my studio mostly used for recording vocal tracks. I'll retire that to the hamshack someday and make some sort of 2 band processor with it.

What are you using with a "no delay DSP"?. Probably something modern and fast.  ;)

Michael McDonald? ugh!  :P Thought I heard him on slop bucket the other day...  ;D

Three bands is reasonable for voice when set to the right crossover frequencies. I've modified those on my DAP-310 to fit the voice spectrum nicely.

I use a home-brew 5 band processor.  Hopefully, my audio does not sound compressed (or at least not too compressed).  That's the idea with multiband (if it is not carried too far) - the audio can sound rather uncompressed, even though it is.

not using any multiband processing. Just some 3 and EQ and simple one-channel processing