Phantom current

[bcarso]

Be wary when making these kinds of measurements of thermocouple potentials due to temperature gradients. They can take a long time to settle down after a big thermal excursion, and just handling the part will usually introduce them. Depends a bit on the meter---some will use a relatively small current to keep the voltage drop reasonably low. That's very useful for measuring things in-circuit since various diode junctions stay more or less off, but it exacerbates the sensitivity to other errors.

I guess to the question of 0.1%---there's that feeling that maybe the manufacturer has given them better care and inspection---but that could be illusory.

 

[Rossi]

[quote author="PRR"]
The audio load can be 10K, or it can be under 200 ohms. [/quote]

That's why manufacturers specify a rated load impedance (usually 1000 ohms). The specified performance (max SPL, frequency response, noise) is only guaranteed for loads no lower than the rated load impedance.

Many newer mic preamps have a variable impedance feature, which often enables users to lower the imput impedance to unreasonable values. The lowest I've seen so far is 150 ohms. On the other hand, the main reason for having such a feature is to alter the mic's performance. So people probably don't want a mic that can handle super low impedances. I've yet to see a mic that sounds better at 150 ohms, though.

There is an interesting AES paper by Martin Schneider on the Neumann website ("Electromagnetic Interference, Microphones & Cables") which includes a graph showing how max. SPL is influenced by the mic preamp's input impedance.

 

[bcarso]

hAHAhaha they actually included the disclaimer/warning.

At UCLA there was a fancy discrete DAC that took up several rack units. One module was full of the hermetically sealed Vishay bulk metal foil R's, and they were in a bath of mineral oil. The beast had a whole circuit board per bit switch. It was used for a deflection circuit associated with an automated measurement system for bubble chamber photographs.

 

[Gus]

Question

What is the highest voltage people have seen on failed phantom supplies(shorted regulators, high line voltage etc..) in consoles and home gear?

I would quess about 80V but I would like ask what people have seen

This is for microphone design I am working on I want to make sure it can survive a failed real world phantom supply.

 

[Guest]

48V Zener is your friend.

 

[Guest]

[quote author="mediatechnology"]Not to nitpick but I think I'd go with a few volts higher than 48 for Vz.

[/quote]

Agree.

56V for consensus, if put them directly on XLR.

Fuse is odd: I can't expect 6K81 resistors shorted.

 

[bcarso]

Unfortunately zeners are all over the place. But the spec for a 500mW one in an old Moto databook says 5uA max at 42.6V for a 56V 1N979B. With the newer series the 1N5263B, this improves to 100nA max at 43V.

Zener current is a point process (so-called Poisson distribution) at these currents with potentially irritating acoustical effects compared to nominal white noise.

It's interesting how hard it is to find material on zener/avalanche noise btw. The otherwise comprehensive discussion in this databook is conspicuously silent on noise. I finally found some material in Ambrozy's book, Electronic Noise, ISBN 0-07-001124-9.

When I want to clamp something for overvoltage protection I tend to bias the zener on slightly and couple to it with a normally reverse-biased regular diode. Of course this requires a voltage source above the zener voltage.

 

[SSLtech]

About 87V on a failed SSL 9000 Phantom supply. -Boy, did that do some damage!!!

Good memory, Wayne!

Keith

 

[Gus]

Thanks I will use 90VDC as the max

 

[Guest]

[quote author="mediatechnology"]

At the XLR is a sure-fire way to protect the mic I agree. But I wonder if the 56V (or whatever V) would have enough reverse leakage current noise (even below zener conduction) to be a noise contributor. Worth looking at.

[/quote]

Use only one 48V Zener shunted by electrolytic with current from power supply through one resistor, and silicone diodes on each XLR input to it. A bridge rectifier (something like DF10M) with ~ leads connected to 1 and 3 pins of XLR, + to Zener, - to signal ground will be what the Doctor Wavebourn prescribed.

When a mic connected a voltage plus positive signal peaks will be for sure much less than 48V.

 

[Guest]

[quote author="bcarso"]Unfortunately zeners are all over the place. But the spec for a 500mW one in an old Moto databook says 5uA max at 42.6V for a 56V 1N979B. With the newer series the 1N5263B, this improves to 100nA max at 43V.

[/quote]

Put 48V Zener between base and emitter of powerful enough transistor, better NPN (safier) (see my prev. answer) as a point to connect + leads from bridges.

PS: Wayne, I've messed up with 2 questions, one from Gus (self - protecting microphone itself).