# converting line level audio to resistance?

Discussion in 'Electronic Design' started by Mad Scientist Jr, Sep 19, 2005.

I need to take a line level audio signal and convert it to resistance
in the range between 0 Ohm and 1 MOhm. Is this difficult to do for a
beginner in electronics?

2. ### John FieldsGuest

---
Well, zero ohms is the tough part, but even tougher than that is
having to cut and paste and repost or retype and repost earlier
replies because a querying poster (you) didn't crosspost to all the
where you enter where you want your post to be sent, and if you fill
in all the groups where you want it to go the first time you send
it, it'll be sent to all of them automatically and then anyone
send replies to all of the other groups as well. That's called
"crossposting" and makes life a little easier for us all, so why not
try it?

3. ### John WoodgateGuest

Not difficult, just impossible. Now say what you are really trying to
do, and we may be able to help.

4. ### Pooh BearGuest

With a name like Mad Scientist I'm sure it's right up your street
! ;-)

Graham

5. ### Pooh BearGuest

V = I*R

therefore R = V / I

So all you have to do is divide the voltage by the current.

Graham

6. ### Fred BartoliGuest

Just convert all the possible line levels to 1K.

Plain easy, even for you.

7. ### Ol' DufferGuest

It's not entirely clear to me why a responder would want or need
to cut and paste, but with modern software it's arguably easier
than having to dig for a post that mysteriously disappeared because
you marked it "read" in some other newsgroup. Which is why some
folks prefer multiposting to crossposting.

8. ### Walter HarleyGuest

Well, so far you seem to be getting responses that, while perhaps relevant,
might not be helping you solve your problem.

That might be partly because doing what you suggest is a bit unusual, and it
might be that you are approaching whatever broader problem you are trying to
solve in the wrong way. So if you tell us what you're trying to do, that
would help.

But to answer one thing I think you might possibly be asking: if you mean
that you want to create a device that has an audio line level input and an
output that behaves like a resistor whose value varies predictably with the
input voltage, yes it can be done but it is challenging. The hardest parts
are getting it to be linear (or logarithmic or whatever you're aiming at),
and getting it to be fast enough to deal with audio frequencies. (And as
others pointed out, the 0 ohm part is impossible; you might be able to get
10 ohms, or maybe even 1 ohm, but 0 is physically impossible.) Think in
terms of using the input voltage to control the drain-source resistance of
an FET.

There are a lot more aspects of this you'll need to specify in detail in
order to get a more useful answer.

9. ### Fred BartoliGuest

IOW, you're begging for some information and you think others have to suffer
the inconvenience you create just because you don't want to bother being
careful enough?

How selfish you're.

10. ### RenéGuest

A very imperfect realisation of the above is using an audio amp to
drive a lamp rather than a speaker, and have it's light shine on an
old fashioned LDR. (This would sort of transform the audio average
*volume* into sort of resistance)

11. ### John FieldsGuest

---
You seem to have missed the point, which is that If I respond to a
post in, say, seb and then, later, find that identical post in sed
(but not crossposted) I will either have to not respond, go back to
seb, locate and copy my earlier response, then go to sed, find the
OP's post, paste my response into it and, finally, send it, or
retype a separate response.

As far as software goes, I think it's relatively easy to set up a
newsreader so it marks posts read but doesn't kill them. At least I
don't have that problem, and I'm using Agent.

That's an interesting idea - how would you get the circuit to max out
at 1 MOhm and at a minimum, go to as close to 0 ohms as possible?

14. ### DeefooGuest

Now that's a cool idea. Let your stereo do the playing for you while you get

driven, so you would only have to amplify your audio to cover the 0-5V
range. For starters try connecting your audio output directly to the paddle
pin (and gnd to gnd) and see what happens.

Foo,D

I'll just end this subtopic and say I was a dumbass and multiposted by
mistake - I posted and then had a 2nd thought "maybe it would be better
inconvenience.

16. ### Frank BemelmanGuest

Very good, very good indeed

I was about to suggest a fixed 470K resistor and forget about the
audio signal.

17. ### Walter HarleyGuest

I'm not clear whether you want the resistance to be related to the envelope
(average volume) of the signal, or to the instantaneous voltage.

Are you trying to make something like an oscilloscope, to display something
like a Lissajous pattern (ever-changing squiggly thing that shows phase and
frequency relationship between two channels)? Or are you trying to make
something that just moves the paddles depending on how loud the sound is?

If the former, you're probably out of luck; I don't think the Atari itself
will respond to paddle movement at 20kHz.

If the latter, then Rene's light bulb idea is not a bad one. One problem is
that light-dependent resistors (LDRs) with 1MEG dark resistance are usually
pretty slow, taking as much as several seconds of darkness to get all the
way to 1MEG. The resistance will go low in a hurry, but it takes a long
time to recover to high - it's sort of like your eyes when they dark-adapt.

Either way, you probably only need a current or a voltage, not a resistance.
That's considerably easier. And I wouldn't worry about getting the
(effective) resistance any lower than 1k ohms, since the pots in the
joystick are almost certainly linear (as opposed to log-taper): the
difference between 1k and 0 is 0.1% which is less than the pots themselves
can distinguish.

Give Rene's idea a try, as a start. If it doesn't work you can evolve it.

18. ### John FieldsGuest

assume it is quoting the original article and then it doesn't
let us know what you're responding to then we won't have to digging
for it.

That sounds easy, I will need to multimeter it.

2 dangers:

1) blowing up the stereo if the output resistance is not 8 ohms (or
whatever the output expects for headphones) - that's why I originally
said Line Level, this would be a single standard as opposed to
different stereos which need different resistance speakers.

2) blowing up the video game - it sends 5v out and measures how much of
that comes back. I assume the signal from the stereo sends voltage as
well. So what happens when you have the 5V from the game, plus the
volts from the stereo? What if the game can only handle at most the 5v
it is sending out? Therefore, I would rather make a circuit that 1)
controls resistance 2) provides some attenuation so the stereo doesn't
blow up