Bill DeFelice, you must be talking about WAVZ or KC101 New Haven. In 1977 when I became PD of WAVZ (1300kHz, 1kw DA-N) we built 3-band component processing that went from the STL receiver to a Pultec EQ with a 10db boost at 8kHz and a lowpass filter at 10kHz. The Pultec output fed the input of a Crown crossover which, in mono mode, gave us bass, midrange and high freq outputs, each of which fed a UREI LA-4 optical compressor. The outputs were summed using, as I recall, 5k resistors on the + and - of each of the LA-4's balanced outputs, twisted and soldered together to feed the balanced input of an AM Volumax peak limiter. Our Harris MW1A transmitter had an integral 3db audio clipper that we deliberately undercalibrated and ran in maximum (3db) setting, which gave us virtually constant modulation peaks of 99% negative and 124% positive without the distortion you'd expect. We put the lowpass filter on the input in order to bandlimit what the system reacted to, so the high band compressor wouldn't reduce gain on cymbals and other high frequency audio that would have been well beyond the bandwidth of the typical AM receivers of the time, many of which were still quite wideband by today's standards. We essentially beat the NRSC to the punch by eight years, although the clipper produced some significant out-of-band emissions. The genius behind most of the design of this system was WAVZ/WKCI chief engineer Tom Osenkowsky. I, as PD, added the concept of using the EQ and filter ahead of the gain reduction, it was my idea to use the transmitter clipper, and I set all the controls on the system components. Tom and I agreed he should maintain control over the modulation since he was the chief operator, and he was always great about letting me squeeze out every legal drop of modulation.
In 1979, after we moved the Hot AC format from WAVZ to WKCI as KC101, Tom and assistant engineer Lloyd Prezant purchased and installed a similar system for the new FM format. It required two Crown crossovers to output three stereo bands and three stereo dBx compressors. We did not use any EQ or filtering for FM, and the outputs of the dBx compressors were summed resistively and fed an Orban Optimod 8000, state of the art at the time, running only 1 to 2db of gain reduction in "Limit Only" mode. Our top secret weapon at the time was the pair of diodes with which Tom retrofitted the baseband input of the Harris exciter, this was the state of the art of composite limiting/clipping at that time. Tom had to haul the 'scope up to the transmitter to adjust the composite input and eliminate any pilot modulation whenever I made an adjustment to the multiband system at the studio, since the Optimod 8000 composite output was far from being a "brick wall" like its successor, the 8100.
The Optimod 8100 was not even on the market yet, until well after I moved on in 1980 to NBC's 97WYNY/New York, where, at my request, chief operating engineer Randy Hoffner, who went on to develop the analog TV stereo audio system adopted by the FCC in 1985, built a 4-band component processing system using the then-new concept of active op amp crossovers and lookahead limiting of the type previously used primarily in the vinyl disc cutting process. That system fed Bob Orban's first Optimod 8100 unit, followed soon after by Optimod 8100/A serial #00001. Eric Small prototyped his Modulation Sciences CP803 composite clipper on a breadboard circuit board at WYNY's Empire State Building transmitter, and used our team's suggestions to finalize the ultimate 'sound' of the box that would soon be found at nearly every FM station in the country.
