Buffered Mult, is this correct?

At first I thought I was on the wrong thread as I incorrectly thought I had seen your RGB Crossfader panel :sweat_smile:
Cheers @Fox for confirming the resistor values. I’ve got to read and learn what “pull down resistor” means.

For anyone like me who hasn’t a clue about electrical circuits.

Maybe it should be called termination resistor in this case. Pull-ups and Pull-downs are terms typically used to digital electronics, but each is kind of ambiguous.

The 100k protects a disconnected input from oscillating but also creates a voltage divider with the signal’s series impedance. They also prevent a wire’s resistance from increasing t infinity, but that really isn’t relevant in this case since most user’s wires are a few feet at best.

1 Like

It will only act as an attenuator in this case. For it to create gain, it would have to be in the feedback path of the op amp.

1 Like

That pot on the output is bad news. If you turn the output all the way down, the output will be ground. If you use stacking cables for example, that would force all signals on the stack to short to ground. Put the attenuator on the input.
100k to ground on the input jack, through 499ohm to the pot. Wiper of pot to the input of opamp. 499 between output of opamp and jack.

5 Likes

Thanks everyone :slight_smile: that pot was a mistake. Is it ok like this?

1 Like

If you add 100nF bypass caps between V+ and GND as well as V- and GND on the LM6172, that looks right to me.

Here’s one I posted in another thread:

2 Likes

This then

1 Like

This is starting to look a lot like a proper buffered mult :slight_smile:

1 Like

if you have an unused opamp stage (like in your schematic above), you need to do this:
image

but I recommend to just add another buffer on your pcb.
in the LZX schematics the output jacks switches are not connected to GND.

1 Like

On the input section the 100K is there as a pull-down resistor but what is the 499r doing after? Adjusting the Voltage to a bandwidth of 1V or stopping or limiting a hot signal into the op-amp?

1 Like

ok what about now?

1 Like

Best not to wire the socket switch pins to ground - that will mean you short the tip of the plug to ground briefly as you insert it. On the output side shorting the op-amp to group (via 499R) when no plug is inserted loads down the op-amp unnecessarily. In general leave the switch pin unconnected on outputs and use a small series resistor on inputs.

4 Likes

A nice trick to add to your circuit is to normalise the second input to the first input (might by incorrectly phrased), that way you’ve six cloned outputs if nothing is plugged into the second input.
I’ve seen this done on other threads by some of the various circuit designers here. A resistor was required but where exactly the first & second inputs connect, I can’t say sadly. I’ll scout for some of the threads & post a link if I get lucky.

@jestern, here’s a couple of screenshots of other circuits found on here that show what Adrian was talking about.

3 Likes

Perfect thanks guys!!!

1 Like

It is to reduce settling time , as per LM6172 datasheet (https://www.ti.com/lit/ds/symlink/lm6172.pdf)

REDUCING SETTLING TIME
The LM6172 has a very fast slew rate that causes overshoot and undershoot. To reduce settling time on
LM6172, a 1kΩ resistor can be placed in series with the input signal to decrease slew rate.

3 Likes

^^^ Bastien’s advice is solid, definitely put those series input resistors back in…

2 Likes

Merci Bastien :+1:t3:
I’ve got to read the datasheet again. I’ll pick up more next time as I’ve learned alot on here since first attempting to take in all the various info there.

1 Like

About daisy chaining using the switch from a jack, best is to take the output of the op amp (before the output resistor) and connect it to the switch pin of the jack.

Not sure if a resistor is needed in between, as it will only be connected internally. Didn’t used such resistors in VU002 (where full wave rectification stages are daisy chained).

Taking the output of the circuit (after the 499R) will work too, but will be a little bit more attenuated (as each input connected will add a 100k resistor to ground), though there is some “headroom” since 499R is quite small compared to 100k (meaning it can be connected to multiple inputs without too much attenuation).

1 Like

Again I agree with Bastien’s comments. But one note of warning - in my Quarterizer module I took the output from each input’s buffer op-amp directly to the switch pin of the next input socket, and it caused problems with ringing on the third and fourth inputs, even when I used a small series resistor. But, since that doesn’t happen on the VU002 it’s likely that some other aspect of my circuit was contributing to the problem. My fix (which probably isn’t appropriate for a buffered mult) was to configure the buffer op-amps for 2x gain (which the design needed anyway) and then use a 100K/100K voltage divider from the buffer/gain op-amp into the next input, like this:

Until you build your whole circuit you probably won’t know for sure what the potential problems might be. I recommend bread-boarding and prototyping all your circuits… sometimes even a simulation won’t reveal the problems you might experience.

3 Likes