Although this exercise does not seem to be necessary for the LS-11,
it might be useful for others who are using these capsules with
recorders that have noisier preamps and similarly low input
impedance. I am interested enough to try experimenting with lower
resistance values in the battery box, and a lower battery voltage to
keep the current down where it should be (< 0.6 mA).
Klas, as far as I can tell, my battery box has an output impedance
around 3.4k ohms, not 6.8k. If it were 6.8k, I think I'd be hearing
it in the LS-11 since that would mean a 13 dB drop in voltage. When I
attach any resistor, large or small, in series with the microphone
capsule the voltage is divided, with an approximate value of the
"resistance" of the em172 capsule around 7k ohms. This varies
depending on the voltage applied, but always somewhere in this ballpark.
The EM172 spec sheet says they derived an impedance of 2400 ohms
using a 5v source and a 3900 ohm load resistor. I think that works
out to an internal resistance of about 6380 ohms for the capsule,
which is in the ballpark of what I have been getting from measuring
the voltage division using various resistors and voltages.
I am thinking that 500-600 ohm resistors with a 4.8v battery pack
would supply the voltage and current the capsule needs, and only a
Message: 1.
Subject: 8 dB drop in voltage when it meets the LS-11, compared to the 9 dB
drop it is probably getting now. I'll probably try this at some point
just out of curiosity. Resistors any lower than 500 ohms raise the
current too high or require a voltage too low. The battery box as
currently configured is using a load resistor nearly equal to the
impedance of the capsule, thus dividing the battery voltage in half
and presenting about 4.5 volts to the capsules. 9v batteries are a
handy package, but 4.8 v (4 AA) I think might work better.
If Raimund is right, this won't work, and won't make any audible
difference even if it does work, at least not with the LS-11!
All values are approximate. I do not have precise test equipment and
I don't know if I am doing everything right. I'm basing my thinking
on Thevinin's Theorem (how to simplify a circuit for the purposes of
analysis), which I may not be understanding properly, particularly in
how the various resistors in both parallel and series contribute to
the impedance "seen" at the output. It's been 30 years since I barely
passed college electronics theory! But all the numbers more or less
add up and agree which each other, which gives me some hope that I am
not entirely off base.
One of these days I'll get back to making actual recordings, although
winter wind is about the only game in town right now.
John
John Crockett
Let us live in harmony with Earth
And all creatures
That our lives may be a blessing
On Dec 30, 2012, at 5:52 AM, wrote:
> John, Raimund and all...
>
> An electret like EM172 is not a complete circuit. It has a very high
> open circuit impedance, on the other hand it doesn't work, so it is
> not relevant to talk about "impedance".
> To make it work, a resistor is needed between the drain of the FET
> and + of the battery and then the circuit gets the same (approx.)
> output impedance as the resistor! So John's circuit will have a
> output impedance of approx. 6.8k.
>
> As the input impedance of the LS-11 is only 2k, it means that less
> than 1/4 of the signal level remains.
>
> Raimund:
> 1/ What do we know about the LS-11 input impedance with and without
> plug in power?
> 2/ Do we know that it is a resistor?
>
> Also: When reading the specs of electret microphones, you have to
> have a look at the "test-circuit" diagram. If Primo says that a
> EM-172 has an "output impedance" of 2.4k, it is because they have
> used 2.4k in the test circuit.
>
> Klas.
|