I have a Kenwood TM-V7A dual-band (144/440 MHz) mobile ham radio. A long while back, the RJ-45 plug on the end of the mic cord lost its little retention tab. I cut it off and crimped on a new one. I plugged it back into the radio, turned it on, and it immediately started transmitting. Every time I plugged in the mic, it nailed up the PTT switch. Whiskey Tango Foxtrot?
I looked at the old and new connectors, and they LOOKED identical. I put the whole thing on a shelf until I had time to dig deeper into this. Maybe strands of the mic signal shield were poking into two different wires and shorting it.
So tonight, months later, I finally pull the thing down and look it over. Took the mic apart, toned out each wire. Sure enough, the black one doesn't connect. Okay, cut off the cable, crimp on another one, and make damned sure the black wire goes all the way down its hole. Same problem. OMGWTFBBQ?
I spend the evening looking over the PC boards in the mic (it's a DTMF/remote control mic, a Kenwood MC-53DM.) Nothing's corroded, nothing's shorted, nothing's burned out. I tear aprt the 'net looking for a schematic. I find one for the MC-45DM, the non-backlit version of this mic. Okay, there's a chain of transistors that pulls the PTT when you dial a digit. That's what must be nailing it up. I get an ethernet cable and a coupler so I can hook the mic to the radio and test it live. (Tuned to the aircraft band, the radio won't transmit, it just beeps at first contact to tell you so.)
I find some rather odd voltages. And the DTMF emits no tones. It just makes a squishing noise in the speaker when I dial a digit. There's points that SHOULD be at ground, yet there's a voltage difference, of about FOUR volts, between them and the PTT ground. There's two ground connections in the RJ-45, one is called out on the schematic as being the mic wire's shield. Okay, an analog ground. Well, why the hell is there a difference between the analog and digital ground? You separate them for noise, but they should still be 0V relative to each other. Unless the analog ground isn't actually CONNECTED to the place where they finally bond. (In the radio.)
There's a black wire in the cable. There's a black wire attached to the mic's PC boards. Rather, there's what LOOKS like a black wire attached to the PC board. I get no continuity, so I strip the insulation off the mic end... the wire at the RJ-45 end is copper colored. This is not. This is silvery. Like the mic shield. It's got a very thin piece of heat-shrink tubing on it.
Clip the RJ-45, attach another one, using the twisted shield line instead of the black wire. It's REALLY tricky now because I'm starting to get into the coiled part of the cord. And I'm back in business.