> keep the line out for my tuner
You didn't say that.
Note that those tricks of shorting the XLR also short the transformer secondary, thus its primary, and the direct-out signal is severely loaded.
> the LED is in the collector circuit
So? There was 12V emitter-to-collector, now 10V e-to-c, signal levels are hardly 1V. Still tons of headroom.
> The LED is also a handy addition.
IMHO, essential for live performance. There's too many ways signals get lost. Adding another signal-kill, you really need a go/no-go indicator.
And the LED can't really go anywhere else because current is tight.
> when the signal is not muted, the LED is {on}
Matter of taste. I figure ON is ON (un-muted). This also confirms that Phantom is ON. After the first screw-up, Ray will always look for the red light before starting to play. It could be wired "Red = STOP! you are muted".
> I can predict a pop when the operating point changes.
Yes, but you can work it out. Mu of a BJT is near 1,000. We induce a 2V change at the collector. There's 2V/1000 or 2mV change at the emitter. I think the pop/tik will be 20db-40db down from normal signal. If it was a $40 trick, I'd think about it more. Since it is a $3 experiment, try it.
A selected resistor (330-470) or two diodes in the LED-shorting line can reduce this to a few mV.
Nobody objects to "huge peak currents" on a shorted emitter-follower? That was my main fear.