The AM Forum

   I've been using a W2IHY EQplus with my Ranger for the past year or so.  I know, it's intended mainly for sideband, but it's been working well for me: compressor and noise gate function seamlessly, and the two-band EQ is effective.
   Now, I just got myself one of REA's excellent modulation monitors, and it's illuminating a whole lot of things that my old monitor scope didn't.  Here's a curiosity: when I activate the W2IHY's equalizer, the modulation goes significantly asymmetric, favoring the negative.  Not the compressor, just the equalizer, and it occurs regardless of the setting of the equalizer.  The compressor by itself actually drives the modulation somewhat asymmetric in the positive direction, which is nice.
   It's probably inherent in the design, and probably far less significant - even unnoticeable - when transmitting single sideband. 

Sounds like the EQ is flipping the polarity compared to the compressor. On my transmitter i have a setup where i can flip the polarity of the mic easily to favor asymmetric audio in either direction, positive or negative.

Voices are not uniformly asymmetric, so adding EQ that either boosts or cuts portions of your vocal spectrum will affect the average asymmetry.

For example, lower vocal components might have greater negative amplitudes where upper vocal components might have greater positive amplitudes. Boosting or cutting either will throw off that relationship.

In the case of a Ranger, the audio driver transformer has a tremendous amount of phase distortion (linearity wasn’t a design concern). Reducing the lower frequencies that are most affected by this component will have a huge affect on asymmetry.

Alas poor Ranger, I knew thee well. The driver transformer phase distortion wasn't noticeable with stock telephonium audio, but when the audio chain was modified for AM Gangsta audio the bottom end got muddy. The solution is get rid of the driver transformer and rework the 12AU7 parallel triodes into one half as the driver and the other half as a phase splitter.

DC offset in the W2IHY EQ section?

I did think that maybe cutting or boosting frequencies might play into the Ranger's design quirks, but the asymmetry remains regardless of the settings of the equalizer.

I haven't checked for DC offset; I'll put my scope on it and check.

The output of the EQplus is balanced - for radios like my Ranger with unbalanced inputs, you just ground the unused side, in this case the negative.  I'm going to try grounding the positive side instead and feeding the negative side into the Ranger.  If that doesn't work, I might try this:
http://www.w3am.com/8poleapf.html

Just flip the polarity on the output of the IHY box. Your problem should be at an end.

  Basically that's what I just did, and resolved the problem.  As I said in my previous post, the IHY's output is balanced; for running into a radio with unbalanced input the wiring instructions have you ground the negative side (the output is taken off a transformer) and run the positive side into the radio.  I reversed that, grounding the positive leg and applying the negative leg to the Ranger.  Now I get a bit of positive asymmetry, which I can live with far more than negative.
  This also proves that the asymmetry is created entirely by the EQ circuit: now when I switch the EQ out, it once again become asymmetrical in the negative direction.
   So, anyone running a W2IHY box on AM, check your symmetry.  All in all, it's not a bad little box; a nice low-dollar alternative for those whose pockets aren't deep enough for the high-end stuff.  I've always received good audio reports using it - now it should be even better.  It'll be on the AM Carrier net tomorrow morning.