AD797 mic preamp output question

[hazel]

Hi everyone!

i'm waiting for postman to bring me this:

http://diy-tubes.com/image/data/manuals/ad797-v1.1-en.pdf

it's from user ungifted user here

taking a look at AD797 datasheet (http://www.analog.com/static/imported-files/data_sheets/AD797.pdf) came across that output resistance is 3 mili Ohms.

My questions are:

- do i miss something? how can it be so low?
- mic preamp works as full balanced so 2 IC amplifies + and - so, do i need output buffer / line driver / balanced driver / transformer? or maybe i can plug IC outputs to Audio Interface for recording purpose?

I know they'll be newbie questions...i'm still learning

 

[PRR]

> maybe i can plug IC outputs to Audio Interface for recording purpose?

That's what it's for.

The number of milli-Ohms is not an important spec.


> - do i miss something?

What are you missing?

> output resistance is 3 mili Ohms. how can it be so low?

Ideally the output impedance is ZERO.

In practice we are happy if it is much less than the load impedance. For typical 10K interfaces, <1K is fine, under 500r is dandy. At 600 Ohms (antique gear) we like either ~~600 Ohms (add a 601 ohm resistor) or <50 Ohms.

3 milliOhms meets all these goals.

However it is not a single number.

The naked, no NFB, opamp has an internal impedance. Typically 50 to 300 Ohms for low frequencies. This is the internal devices and resistors.

Apply NFB, and this resistance is reduced. Roughly by the amount of Gain given-up.

Ah!! RTFM! AD797 datasheet page 4, Output Resistance "Av=1, f=1KHz: 3 m Ohms".

What I was thinking: 110MHz GBP or Fig 16 suggest gain of 10,000 at 10KHz (top of audio-band, kinda-sorta). And the naked output impedance is near 120 Ohms. 120/10,000 = 12 milli Ohms. Maybe 24 mOhms at 20KHz, etc. And 0.12 milli Ohms at 100Hz.

The diy-tubes plan throws another 33 Ohms at the output, but takes audio NFB beyond that. This raises the numbers by 153/120 or 1.27.

That's at gain of 1. You may be using gain of 100. All values are 100 times higher.

That's for *one* side of the two-side output. So double everything.

I get as much as 6 Ohms for Gv=100 and F=20KHz. Even 36 Ohms if you jam the gain control as far as it goes.

This is still "MUCH" less than a heavy 600 Ohm load and much-much-much less than modern 10K loads.

For perspective: 1 foot of #16 wire is 4 milli Ohms.

And this is not the load the preamp will DRIVE. It current-limits. Fig 5 shows full output down to 200 Ohms. Per side, so 400 Ohms push-pull. Yes, it will drive that antique 600 Ohm input without strain.

There's no DC control in that preamp. The AD797 has excellent DC specs but counting on thumb I see as much as a half-Volt of worst-case output DC. Nearly all modern interfaces will ignore that. True 600r transformers may not like it.

 

[hazel]

> maybe i can plug IC outputs to Audio Interface for recording purpose?

That's what it's for.

The number of milli-Ohms is not an important spec.

> - do i miss something?

What are you missing?

Thanks PRR!!!!

Due to my ignorance i thought i would need something like 5532 in buffer/driver configuration to avoid connecting AD797 to output.

In unbalanced mic preamps 5532, 1646 etc are used to balance signal but i didn't figure out how to connect with full balanced preamp.

So i'll just plug AD797 outputs to recording interface

There's no DC control in that preamp. The AD797 has excellent DC specs but counting on thumb I see as much as a half-Volt of worst-case output DC. Nearly all modern interfaces will ignore that. True 600r transformers may not like it.

So i need a DC servo? a couple of caps at the output?

I won't use transformers, like to keep this preamps as simple as possible