Home Brewed Processing

[radiorob2.0]

Either out of necessity or creativity, what has been your best processing creation? Here is your inspiration:
http://www.w3am.com/audiocha.html

 

[Sgeirk]

Nothing fancy, but for FM, a pair of texars feeding a dolby spectal processor, feeding discrete l/r feeds to the xmtr, with 8100 and card 5 and Hemphill composite clipper....sounded pretty amazing in 1998. We hit nothing really hard and grew loudness carefully along the way.

Thanks for turning us onto that site!

 

[LA_Guy]

A Highly modified DAP

I took a DAP 310, removed the 3 expander boards and peak limiter board, replaced the gain FETS with better ones, doubled the size of the release time caps, re-biased the FETS so they had a (linear) 25 db gain reduction range, replaced the crossover board with a passive homebrew 2 pole 3 way crossover with the phase of the middle band flipped, and took a Straight Wire Audio UA-2 board to replace the input and output transformers with differential inputs and outputs. This fed into a completely rebuilt UREI 1/3 octave graphic equalizer (unbalanced inputs and outputs-no transformers). Followed this with a rebuilt BL-40 (PR&E unit with phase scrambler) Modulimiter that had all its caps replaced, its opamps upgraded and its transformers removed (SWA again). This all went into a BTA 50G Ampliphase. Format was Music of Your Life. STL was a two hop Moseley PCL-505/C run without pre-emphasis.

This station SUNG!!!!

It also had a 5.1 share in the Providence, RI book-as a DAYTIMER! It was croaking stronger full time AMs and also a couple of FMs.

The processor acted like a smooth AGC and dynamic equalizer. We ran 12-15 db of gain reduction if the jock was watching his levels well-more or less if he wasn't. It 'fixed' all the variations of the older music, but in a very tasteful way. The Ampliphase would not supermoduate well, so the BL-40 was run symmetrically, with a few db of RMS and a couple db of peak limiting.

This processing sounded 'smooth as fine glass'. To this day I have not heard an AM sound as good......

But then again, what the Hell do I know about processing?

 

[rrsounds]

Either out of necessity or creativity, what has been your best processing creation? Here is your inspiration:
http://www.w3am.com/audiocha.html

A subject dear to my heart.

I came along at a time when the "less is more" concept was just starting to take hold. We wanted the process, but didn't want all the crap that came along with it. And sometimes a particular box had one function you REALLY liked, but coupled to another function that you HATED.

Which is why we 'home-brewed' our chains.

Notwithstanding, while I can appreciate the usefulness from a control standpoint of all the multiple cascaded processors, the additive nature of typical analog circuitry's distortion just didn't allow slapping together dozens of stages in a row. Not if you wanted real clarity.

Note that I said "typical" analog circuitry. Because it took the likes of people such as Kanner, Blau, etc. to weed through commercially available equipment's weak spots, to bypass or improve unneeded or detrimental circuitry to make these monsters of processing 'sing.'

LA_Guy's description of his homebrew chain is a perfect example: you couldn't just pile all this stuff in series and expect miracles; it took knowledge, careful listening, the ability to understand and interpret drawings and circuit flow, but most of all, a lot of hard work. Oh yeah, a hot iron and a sharp set of cutters.

Our guiding force was the knowledge that, while some effects of analog distortion could be seen in a positive light, most of it simply resulted in a dulling and muddying of the sound, and was definitely negative. It was the enemy, and we sought every means to vanquish it. As has been said, the rewards for such efforts could be awesome.


These days, just about all of the functions of those vintage processors can be duplicated, sometimes even improved upon with DSP, without the undesired and unavoidable additive effects of multiple stages of analog distortion.

Up to now, however, the possibility to do a bit of homebrew rearrangement and design control algorithms digitally has been available only to those who possessed workstations and/or serious coding capability, along with the resources to put together the required collection of DSP hardware and software into a useful package. This assortment of roadblocks has excluded all but the select few...

...Up to now, that is.


Kind Regards,
David

 

[FFoti1]

Home brew you ask...

Back in 1986, sequestered in the bowels of a radio station's engineering shop in Secaucus, NJ...was born The Vigilante multiband limiter. That became the next step to the Flame Throwing airchain of Z-100 (WHTZ-FM, NYC)...AND...it was the inspiration for Cutting Edge Technologies, which is now known as Omnia Audio.

The Vigilante actually started out as an Aphex Dominator, but much of the limiting circuitry was changed, along with my own distortion-controlling clipper/low pass filter, which is what set that box apart from the Aphex model.

Texar Audio Prisms (early versions) + EXR Exciter (later replaced with BBE) + Parametric EQ + Vigilante + Stereo Gen + CP-803. That was the chain up through 1988, when I was last in charge of it. Somewhere along the line, Steve (Studs) Kingston got me to try a very small touch of reverb.

-Frank Foti, Original C.E. of Z-100, NYC.

 

[menotti1]

Good mod laguy.i sure remember the dap 310's .Ran them on several AM's.Talk about LOUD.Then Frank comes along tweaks an Aphex product(Aphex is still blushing ha).Noticed that wink at the end of David's post.GOT SOMETHING
UP YOUR SLEEVE? HUM.....

 

[cgould]

These days, just about all of the functions of those vintage processors can be duplicated, sometimes even improved upon with DSP, without the undesired and unavoidable additive effects of multiple stages of analog distortion.

Up to now, however, the possibility to do a bit of homebrew rearrangement and design control algorithms digitally has been available only to those who possessed workstations and/or serious coding capability, along with the resources to put together the required collection of DSP hardware and software into a useful package. This assortment of roadblocks has excluded all but the select few...

...Up to now, that is.

Yes, everything David says is true!

My favorite chain was my totally hand built four band AGC / Leveler into an 8100 - Xt2 with the Xt2 heavily modified to open up the high end, CP-803, and Dividend composite filter...This was on WENZ 107.9 around 1994 to 1996.

The basics of my AGC has been incorporated in my DSP processing project (AGC's Limiter's, etc.) and the DSP version is something I DREAMED of being able to do in analog, but the complications of doing that would have been impractical. DSP brings those barriers down in a big way!

My soldering iron has been quite cool lately, except to make XLR cables and such, but the keyboard and mouse have been red hot from all the processing R&D work!!! 8)

-C

 

[LA_Guy]

Home brew you ask...

Back in 1986, sequestered in the bowels of a radio station's engineering shop in Secaucus, NJ...was born The Vigilante multiband limiter. That became the next step to the Flame Throwing airchain of Z-100 (WHTZ-FM, NYC)...AND...it was the inspiration for Cutting Edge Technologies, which is now known as Omnia Audio.

The Vigilante actually started out as an Aphex Dominator, but much of the limiting circuitry was changed, along with my own distortion-controlling clipper/low pass filter, which is what set that box apart from the Aphex model.

Texar Audio Prisms (early versions) + EXR Exciter (later replaced with BBE) + Parametric EQ + Vigilante + Stereo Gen + CP-803. That was the chain up through 1988, when I was last in charge of it. Somewhere along the line, Steve (Studs) Kingston got me to try a very small touch of reverb.

-Frank Foti, Original C.E. of Z-100, NYC.

Yep..

I came up with a distortion cancelling clipper too, but wasn't sure if I'd run tnto patent problems.

Basically, I'd clip the audio and then run both the unclipped and clipped audio into an inverting summer (with the clipped audio inverted once more) with the levels properly set, what came out of this summer was the inverted difference between the clipped and unclipped audio (clipping products). I filtered these with a bandpass filter so only the 1-4K stuff was left...then I summed it back into the whole mess. That way, the distortion products were subtracted out. The cancellation wasn't perfect due to the phasing from the bandpass filter-but it still gave about an 8 db cancellation. The actual circuit had a few more refinements, but you get the concept.

It sounded pretty good-but I never marketed it (though it's in a couple of 8100s).

 

[cgould]

Yep..

I came up with a distortion cancelling clipper too, but wasn't sure if I'd run tnto patent problems.

Basically, I'd clip the audio and then run both the unclipped and clipped audio into an inverting summer (with the clipped audio inverted once more) with the levels properly set, what came out of this summer was the inverted difference between the clipped and unclipped audio (clipping products). I filtered these with a bandpass filter so only the 1-4K stuff was left...then I summed it back into the whole mess. That way, the distortion products were subtracted out. The cancellation wasn't perfect due to the phasing from the bandpass filter-but it still gave about an 8 db cancellation. The actual circuit had a few more refinements, but you get the concept.

It sounded pretty good-but I never marketed it (though it's in a couple of 8100s).

How did you manage the "Gibbs Effect" of the main 15 kHz lowpass filter?

Not handling that part can do ya in with lost loudness - or by picking up distortion from having to clip harder downstream..

Just curious...You didn't mention that part..

;D

-Cornelius

 

[wgliradio]

For AM

Modified Audimax (modified so all audio below 125Hz would passed unprocessed). Slow reaction.
CRL SEP-400A (stock)
Mosley TAL210 Limiter
Yamaha balanced stereo reverb
CRL PMC-450 (modified to allow further drive to the limiters before clipping)

This was loud and proud and I have yet to hear an AM sound as good. It had desnity w/o mudding up the audio with almost NO clipping artifacts and it hung the meter. If WNBC has to exist under NRSC-1, it probably would have sounded like this.

For FM, I have an 8100 on the air with some tweaks to card 6 and a Sequel that sounds very good. Years ago I had a High School station on a budget with 2 Mic Mazes, an SEP-800 in front of a Ultramod 2000 and a CP-803. It was fairly loud, it was musical and the entire chain cost $300 used.

 

[LA_Guy]

Yep..

I came up with a distortion cancelling clipper too, but wasn't sure if I'd run tnto patent problems.

Basically, I'd clip the audio and then run both the unclipped and clipped audio into an inverting summer (with the clipped audio inverted once more) with the levels properly set, what came out of this summer was the inverted difference between the clipped and unclipped audio (clipping products). I filtered these with a bandpass filter so only the 1-4K stuff was left...then I summed it back into the whole mess. That way, the distortion products were subtracted out. The cancellation wasn't perfect due to the phasing from the bandpass filter-but it still gave about an 8 db cancellation. The actual circuit had a few more refinements, but you get the concept.

It sounded pretty good-but I never marketed it (though it's in a couple of 8100s).

How did you manage the "Gibbs Effect" of the main 15 kHz lowpass filter?

Not handling that part can do ya in with lost loudness - or by picking up distortion from having to clip harder downstream..

Just curious...You didn't mention that part..

;D

-Cornelius

I did basically the same thing to a composite clipper....after the filter.

By the way, that's how Eric small did things in his clipper-he 'unclipped' the pilot.

 

[sparks794]

My "home brew" wasn't on the air chain - it was for remotes at a station I was at in the late 80's.

For the Marti receiver I used a DBX noise gate (463 or 563 I think) on the output of the receiver before the console and then added a little compression with putting a DBX 163 in front of the noise gate.

In the FM studio I used the same combination on the output of the phone hybrid (TI 101). Made a world of difference.

I used some of that thought process a year ago at a contract station. They put in a new phone system. They don't have seperate studio lines so I used a JK Innkeeper. They said it lacked low end. I purchased a cheep mic processor from BSW and put it inbetween the Innkeeper and the console - made a world of difference.

 

[Dr. Bob]

Re: A Highly modified DAP

This all went into a BTA 50G Ampliphase. Format was Music of Your Life. STL was a two hop Moseley PCL-505/C run without pre-emphasis.

This station SUNG!!!!

To this day I have not heard an AM sound as good......

You left out that the Ampliphase had to be running into a "good" load. Bad load on an Ampliphase equals bad distortion.

Dr. Bob

 

[LA_Guy]

This Ampliphase ran into a six tower array.

Enough towers for you?

 

[Dr. Bob]

I was trying to point out to the uninitiated that if the AM station sounded as good as you claim (and I have no reason to doubt that it sounded great) that the antenna system played a part in the sound.

Dr. Bob

 

[LA_Guy]

I was trying to point out to the uninitiated that if the AM station sounded as good as you claim (and I have no reason to doubt that it sounded great) that the antenna system played a part in the sound.

Dr. Bob

LOL! Quite true, Dr. Bob!

Anything on the job front yet?

 

[W1DAN]

Hi all:

Having worked in radio in the 1970's and 80's, I fell in love with transmission processing. So I built my own stereo processor in the early 1990's. This only landed on the air for ham radio AM and car music mixes onto cassette!

It consisted of a stereo AGC that fed a phase linear tri-band splitter (stole the circuit from the Dominator). This fed a tri-band RMS compressor, then into a separate tri-band peak limiter (totally separate VCA's and detectors). All levels and time constants were adjustable as well as the MF to HF crossover point. All sections were switch bypassable. I used the dbx 2150A VCA's and 2252 detector chips and fit all this in an 1RU box. This processor was not obnoxiously loud (no clipping), but I could easily get 40db G/R and have it sound smooth.

One thing I would need to do if feeding a transmitter is to build a limiter/clipper after all this to grab the overshoots (for ham AM I used a CRL PMC-300A after this unit...worked very well).

I learned alot though and still love studying audio processing. One of these years I'll wake up my Analog Devices ADSP-21065L eval board and build a look ahead limiter......

Thanks for the thread!

73,
Dan
W1DAN

 

[menotti1]

Mr.Foti sure had his way with that dominator.i guess you can say the rest is history.Sounds like you have a pretty good mod yourself.

 

[Andyf101]

Man, you guys are nuts! ;D I mean that, in a good way.

My own home-brew, strictly for making mix recordings, is a Symetrix 422 AGC into an FMR Really Nice Comppressor (in SuperNice mode) followed with the free MBL4 PC based multiband limiter. One of these days I'd like to find a multiband compressor to replace the Really Nice Compressor.

No my chain doesn't hang the meters, and the sound quality is somewhat subpar (OK, OK, VERY subpar), but I do the best I can with what I have. 8) Of course I don't expect my chain would ever see the light-of-day in a real broadcast environment.

Andy