Balanced DI output modification for Boss TU-2 tuner- Sch. and ?'s

[Processaurus]

Here is a modification I wanted to ask for feedback on.  It's to add a balanced output to the Boss TU-2 tuner pedal, for stage use, plugging acoustic guitars or electric basses into a snake and mixing board, without the need for a DI box.  Just a TRS to XLR male adapter or cable would be needed.

Some background- I've been bringing a Boss tuner in my live sound briefcase, to use with acoustic acts that don't have a nice preamp with XLR outs.  Besides being a tuner, it acts as a buffer as well and can drive low impedance inputs, like a mixer. Sometimes people will have passive pickups with no preamp, and those need a hi-Z input to sound their best, a passive DI on its own has much to low of an input impedance (5 Kohms maybe) for directly connecting a magnetic or piezo pickup.  I was reading about impedance balanced outputs, a cheap, easy way to add a balanced output to an unbalanced device, all you need to do is duplicate the output impedance to the "cold"/pin 3/ ring connection, and connect that to ground. By adding resistance to ground, the cold line will pick up the same noise as the hot (regular output), which will be rejected by the balanced input at the mixer at the end of the snake.  The impedance balanced output using a 1/4" TRS output jack is also very forgiving of the ring connection being shorted out to ground by a normal TS guitar cable.  There's some testing that can be done to determine the precise resistance to duplicate the output impedance at a given frequency, but here I just duplicated the series resistor on the output and coupling capacitor.

My questions are about how to protect the output from accidental phantom power.  I spec'd an upgraded coupling capacitor for the output, non polar and rated high enough to withstand accidental +48v.  I was wondering what happens when the transient from the +48v gets back through the capacitor and to the emitter of Q6 and Q7?  I put a protection diode to shunt it off to the power rail, is that necessary?  Is it enough?  The right circuitry?

Also, I couldn't decide if the output impedance should be lower (lower series R34, R32).  My mixer has rather high 5K ohm mic/line inputs, but many traditional mic inputs are 600 ohm.  The 600 ohm would load the signal down, it would be 37% of the tuners output would be seen at the mic input, but, maybe that's good, so it doesn't overload a mic input that has a high minimum gain, like Mackie mixers.  A lot of newer digital mixers can take a line in on the XLR inputs, so no concerns there about being too loud.  What do you think about the tuners output impedance, ok to be rather high with the 1K series resistor?

a suitable 63 volt, non polar coupling capacitor
Transistor datasheet

[Processaurus]

Nice article on this issue with the phantom power popping through the coupling caps. Significant is that the energy can be an order of magnitude higher than static electricity. Recommended simple solution is Schottky diodes to V+, and minimizing the coupling cap value to only be as large as doesn't roll off bass in the intended source. Unfortunately mic inputs on mixers can vary a lot in input impedance- the old standard 600 ohm with input transformers, or many modern ones are 1k to 2k, my mixer is 5k.

https://www.thatcorp.com/datashts/AES5335_48V_Phantom_Menace.pdf

[MrStab]

I expect the diodes to suffice - the output resistance would also reduce the incoming 10mA or so before it gets there.
I use diodes in the same configuration, downstream of 200R resistors, on balanced outputs from a 4580 op-amp. It survives +48V atop a +9V DC Vcc (though you might have to go apologising to your PSU before long, not the pedal). BJT outputs in those.

[Processaurus]

I expect the diodes to suffice - the output resistance would also reduce the incoming 10mA or so before it gets there.
I use diodes in the same configuration, downstream of 200R resistors, on balanced outputs from a 4580 op-amp. It survives +48V atop a +9V DC Vcc (though you might have to go apologising to your PSU before long, not the pedal). BJT outputs in those.

Sounds good. I think I'll do the diodes to the +9v, rather than the +7.7v, which is sort of an improvised analog stuff supply rail in the TU-2.

Do you think 200 ohm resistors would be more appropriate in this case? The high value of the 1K output resistor seems unusual, I'm used to seeing 47 ohms on opamp outputs, and assume lower output impedance = better. Don't have enough background with the BJT output to really conceptualize its drive ability. It can only pull high? And the 10k on the emitter is all that pulls it low?