why or why not would these component values be good in a voltage follower?

[ode2no1]

a friend and i are trying to set this up with a 10k fader ahead of it as a simple buffered volume control. we had initially tested it without the caps or resistors since we'd seen voltage follower schematics using only the chip and two jumpers, but it was distorting. adding in the extra components stopped this...though the values in the schematic we found were much different. the resistors were 1meg and the caps were .47uF and .1uF. after running test tones thru the circuit we came up with these values to eliminate a loss of gain in the lower frequencies...but in all honesty we have no real idea what we're doing. we want this to be as hi-fi as possible. is there any reason why the values we came up with will be less than ideal?

[R.G.]

a friend and i are trying to set this up with a 10k fader ahead of it as a simple buffered volume control.

That'll be fine if what drives the volume control can drive a 10K load. If it's a guitar driving the 10K volume control, it will have severe treble loss because of the loading on the guitar by the control, irrespective of what the buffer does after the pot.

we had initially tested it without the caps or resistors since we'd seen voltage follower schematics using only the chip and two jumpers, but it was distorting.

That's because the opamp must be set so its inputs and outputs are somewhere in the middle of the power supply. Otherwise, it can't swing very well one direction, and it distorts badly, as you found out. R1 and R2 hold the + input in the middle of the power supply, so it can work undistorted.

though the values in the schematic we found were much different. the resistors were 1meg and the caps were .47uF and .1uF. after running test tones thru the circuit we came up with these values to eliminate a loss of gain in the lower frequencies...but in all honesty we have no real idea what we're doing. we want this to be as hi-fi as possible. is there any reason why the values we came up with will be less than ideal?

There is probably no "ideal", only "fits with the rest of whatever is hooked up before and after it".

R1 and R2 have to be driven without much loss by the wiper of the pot. They both load the signal, so the wiper sees them in parallel, for about a 100K load. The opamp + input probably has a high impedance, much higher than 100K, for most but not all opamps. A pot's output impedance is a maximum of 1/4 of its value, or in this case, 2500 ohms, and that can drive the 100K load of R1 and R2 without much loss. 2uF driving the 100K load of R1 and R2 give a low pass rolloff of about 0.8Hz, so you could make do with a 1uF, or even a 0.22uF cap there just fine. In fact a 0.1uF cap would be fine, even for hifi.

The 4.7uF cap on the output may be OK or not, depending on what it drives.

There are about seventeen levels of stuff underlying what I just said, and I'm tired of typing for the day. But you need to know
- what is impedance, both source and load impedances?
- series/parallel calculations of resistors
- what the cutoff frequency of an R-C combination does.
- a little about opamps.

If it were me, I'd
- change the pot to 100K
- take the pot wiper through a 0.22uF cap directly to the + input
- make the values of R1 and R2 be something like 22K to 47K (both the same)
- insert a 220K to 1M resistor between the junction of R1/R2 and the + input
- put a 22uF capacitor from the junction of R1/R2 to ground.

But that might not work any better to the ear than what you have.

[ode2no1]

ok, so this is an old thread but i need a bit more help with this. basically what we're trying to do is run audio from the sound card of a computer into this as a volume control, which will go to outboard gear and then back into the computer. we want the audio as unaffected as possible. we did some tests yesterday and for some reason we're losing about 3dB of volume. in trying a smaller cap for C2 we lost more signal...about 6dB. also the taper is kind of strange. when sliding the fader from all the way down to all the way up the audio comes in a bit abruptly and only increases about 4dB from the midway point to full up.

i just realized though that we're taking the audio from the computer and running it into the fader, then thru the buffer....is that backwards?

thanks so much

[R.G.]

basically what we're trying to do is run audio from the sound card of a computer into this as a volume control, which will go to outboard gear and then back into the computer. we want the audio as unaffected as possible. we did some tests yesterday and for some reason we're losing about 3dB of volume. in trying a smaller cap for C2 we lost more signal...about 6dB.

That's a little odd. -3db and -6db as compared to what? No buffer? No fader on the computer output? Neither?

Some questions come to mind.
- what opamp are you using?
- what are the DC voltages with respect to the 0V ground on the opamp pins?

For a 2uF cap in the input and the two 200K bias resistors, you'd see a bass rolloff of F = 1/ (2*pi*R*C) where R is the 100K paralleled value. This works out to 1.4Hz, so you're pretty hifi in terms of bass rolloff. Actually, you could get by with 0.2uF for a 14Hz rolloff or 0.1uF for a 28Hz rolloff. Still pretty much full audio range in terms of frequency.

If you're losing volume when you attach this, the possibilities fall into two broad classes: (1) the computer audio output is being loaded down by the 10K pot and (2) there is something wrong with the buffer circuit, possibly not being hooked up like you think it is, or other flaw.

also the taper is kind of strange. when sliding the fader from all the way down to all the way up the audio comes in a bit abruptly and only increases about 4dB from the midway point to full up.

This fairly screams that you're using a linear taper fader, not an audio taper one. That's what log/audio tapers are for, to bring volume in volts up slowly, then faster with more rotation/movement. This precompensates for your ear's oddities about volume.  You can "taper" a linear pot if you like. See "The Secret Life of Pots" at geofex.

i just realized though that we're taking the audio from the computer and running it into the fader, then thru the buffer....is that backwards?

Not necessarily backwards. If you find that it's the fader loading down the computer output, then you might want to put the buffer first, then the fader. That works, but is a bit worse in several esoteric (although probably not practical) ways.

[ode2no1]

thanks so much for your help R.G. the pot is connected before the buffer, so the volume loss is as compared to the whole unit bypassed. this was  determined using pink noise as a test, but when we ran audio thru it it did sound fine. sounded exactly like the bypassed signal, though we hadn't realized the slight volume loss yet. when we used a lower value cap like you suggested though and ran the pink noise thru it we saw that the loss went from -3db to -6. we tried a 10uF cap on the input just to see if we could get it to unity but -3db it remained. we used a 9v power supply, but plan on using 15v when it's all done. would this make a difference?

as far as pin voltages, i'm a dummy and forgot to take my DMM with me, so i won't know that for a few days. i also forget the chip and am waiting for a reply from my buddy on that. it's a low noise dual opamp though.

as far as the signal being loaded down by the pot....the output impedance from the sound card is super low so we figured running it into a 10k fader would be fine with the buffer after it. it actually is audio taper. my first thought was also that it had to be linear, but we double checked that when we noticed the weird taper problem. what would the negatives of running the buffer before the pot be? we just don't want the pot to load the signal down when lowering the volume, which is why we decided to go the active route instead of passive.

[ode2no1]

ok the chip is LME4270

[ode2no1]

aside from pin voltages does anyone have any ideas? we're using xlr cables so we have the dual opamp set up with the + going to the input of one half and the - going to the input of the other and the grounds all tied together.

[tempus]

we're using xlr cables so we have the dual opamp set up with the + going to the input of one half and the - going to the input of the other and the grounds all tied together.

I wonder if that's your problem right there. The soundcard out is going to be stereo (left and right). It (unless it's a more pro level card) probably doesn't have a balanced out, and it sounds like you're trying to set up a balanced line here (correct me if I'm wrong).

So where are you taking your measurements from? The outputs of both opamps or just one? Also, you mentioned that your opamp is dual. If that's the case you have yours hooked up wrong (I googled LME4270 and came up empty). The standard daul opamp pinout has the output of Pins 2 and 3 on Pin 1. Pin 6 is the - input for the 2nd opamp.

[ode2no1]

hey thanks for your reply. the sound card is pro level so it has xlr outputs. the chip turned out to be a lme49720, which explains why i couldnt find info on it either haha. so we have a dual ganged fader and set it up to where sound card xlr pin 2 (+) goes to first gang lug 1, lug 2 to pin 3 of the opamp, lug 3 to ground. xlr pin 3 from the sound card (-) goes into the second gang lug 1, lug 2 to pin 5 of the opamp, lug 3 to ground. xlr pin 1 is also to ground. pin 1 of the opamp goes to another xlr, pin 2. pin 7 goes to xlr pin 3. i'm pretty sure that's how it would be set up taking the schematic i posted and translating it to a dual opamp. or....i could have no idea what i'm talking about....which is probably the case. what do you think?

[artifus]

couple of quick thoughts: double check soundcard settings, both hardware and software, for any +4/-10 discrepancies. as you are feeding this from the soundcard, could you not lose the input pot in favour of driving it directly from your daw or soundcard software and instead have an output pot? is this even required - what outboard gear are you trying to interface with?

[tempus]

Just read your post but gotta get ready for work now. I'm pretty sure the volume control is not hooked up properly the way it is now. have you tried testing it without the volume control to make sure you opamp is hooked up right?