[sdiy] This circuit actually works?

[harrybissell]

Hi Glen...  (inline)

Glen wrote:

> I'm designing a pre-amp to use with my computer's sound card and a dynamic
> microphone. I breadboarded a small test circuit tonight, and I really
> didn't expect it to work, but it does. You can find the slightly simplified
> schematic at this URL:
>
> http://webpages.charter.net/glenweb/diy/pre01.gif
>
> In the schematic, the op-amp power connections and compensation cap are
> omitted. What I'm really interested in showing is the fact that there is
> not the  typical input resistor connected to the op-amp's inverting input.
> I didn't expect the circuit to work like this, but it does!
>
> I should mention that the microphone is a low impedance dynamic type. The
> shield has been tied to one of the signal lines coming from the microphone
> and connected to ground. I know this isn't a "proper" balanced input
> circuit, but it does work.

Why have a 'balanced' input with a single ended source... you'd have to convert
to
differential signal first, then back again.

> I also realize that the output is DC coupled,
> and there is indeed a tiny amount of DC offset present, but my computer's
> sound card doesn't seem to mind at all. (I might experiment next with
> nulling out the offset, but I want to try and avoid a coupling cap for this
> pre-amp.) I also realize the output signal is inverted, but it doesn't
> matter for my purposes.

The sound card VERY likely has the cap you wanted to avoid. You're
probably fine.

>
>
> Without an input resistor, how do I calculate (not measure) the gain of
> this pre-amp?

hmmm... the input is probably part resistive, part inductive. I'll pass on the
equasion (hej Magnus ???).   You probably do not have flat frequency
response here because of the inductive part. A series resistor in the input
would flatten the response and lose some gain.

If you like what you have, its a winner.

You'll probably (if the gain is high enough) start to limit the high frequency
response because of gain/bandwidth product of the opamp.  Less gain would
help.

You might be real lucky... the inductance of the microphone coil will probably
make
a rising response with higher frequency, and the opamp roll-off because of GBW
product will make a falling response... it might just cancel out pretty well
(you could be
lucky).

In that case you win (engineering game over)...  or you don't notice / care
about the
response funnies (engineering game over - you win again)

This circuit could also be considered a current to voltage converter, and the
microphone
'sees' zero ohms impedance.  I think if I had to do the math, I'd probably
consider this to be true
and start there.

H^) harry

>
>
> I realize that I have taken many shortcuts here, but what problems would
> there be with not using a resistor between the mic and the inverting input
> of the op-amp?
>
> How do I calculate (not measure) the gain of this pre-amp?
>
> What input impedance does the microphone see?
>
> I'm interested in your comments.
>
> thanks,
> Glen