That's correct, by phantom power standards, there are two 6k8 resistors in series with the positive and negative of the balanced signal.
Depending on the load of the circuit, the voltage will drop, like you already mentioned.
But it's about the situation when there is no load on the phantom power supply.
48V falls within the tolerances of a 50V capacitor.
So it could work for now, but I don't agree with you that it's 100% safe.
It's not great design practice.
b_force, yeah, I'm just chinese worker but did you read that: https://e1dashz.wixsite.com/index/cosmos-apu
I am still a bit confused.
If it's 100% safe, why the above quoted procedure?
Also, what happens when the connection is all of a sudden broken?
I don't know what PSU is inside?
Looking at the input, probably a flyback (maybe with a coupled inductor).
These sometimes have a tendency to overshoot momentarily when there isn't any load.
I want to be sure my device will be working fine, if my other gear decides to fail for whatever reason?
Otherwise I am happy to replace the 50V cap with a 63Vdc one myself.
If it's 100% safe, why the above quoted procedure?
Also, what happens when the connection is all of a sudden broken?
I don't know what PSU is inside?
Looking at the input, probably a flyback (maybe with a coupled inductor).
These sometimes have a tendency to overshoot momentarily when there isn't any load.
I want to be sure my device will be working fine, if my other gear decides to fail for whatever reason?
Otherwise I am happy to replace the 50V cap with a 63Vdc one myself.
I have a better idea, in case I'll go for APU-II I gonna throw out 48V power at all. That feature was suggested to me in our discord or maybe forums, and I regret that implemented that. Nobody needs 48V power in Cosmos APU and no one uses it, as I know. The initial idea was using the switch to On/Off 20kHz LPF but I was stupid enough to replace that with the phantom power.
I actually like the 48V phantom power, I don't see why that would be such a big regret?
Makes it a lot more usable this way, otherwise people need a separate 48V phantom power injector.
I did a quick simulation and if you just put a 100k-150k resistor on the output, that will put at least some load on it all.
While bringing down the voltage about 5-10%.
Another idea is just to put a R-zener-R instead of the 6k8 series resistors.
With a 47V zener, the circuit is also always protected
Just curious, but are adequate 63Vdc capacitors so difficult to find?
Makes it a lot more usable this way, otherwise people need a separate 48V phantom power injector.
I did a quick simulation and if you just put a 100k-150k resistor on the output, that will put at least some load on it all.
While bringing down the voltage about 5-10%.
Another idea is just to put a R-zener-R instead of the 6k8 series resistors.
With a 47V zener, the circuit is also always protected
Just curious, but are adequate 63Vdc capacitors so difficult to find?