I see that in buffered multiple designs too, except that’s reverse? It must work?
Hm, I guess it does. If there’s no out-of-loop resistor then a low impedance downstream module won’t drag the others down, I guess.
Here are some reasons for using 100ohm in the feedback of unity gain buffers, but impedance matching does not work like this here and the other mentioned reasons I don’t understand xD I will have a look at the datasheet of the opamp…
It serves as a current limiter, but that’s not relevant for the internal stages, and in conjunction with a capacitor parallel to the feedback resistor it stabilizes the op amp against a capacitive load, but again that seems irrelevant, and there’s no capacitor anyway…
2 Likes
That sounds really cool! I want to use this with a matrix sequencer, so I probably won’t need the precision voltages, but I am tempted 
1 Like
One could set up a rotary switch with octaves, fifths, whatevers…
2 Likes
Maybe Isolation resistors? I don’t think that we have these kind of loads here, but maybe that is just „how it is done“ with precision circuits?
I hadn’t heard of these LT1014 op amps. I looked up the datasheet, it says:
The LT ®1014 is the first precision quad operational amplifier
which directly upgrades designs in the industry standard
14-pin DIP LM324/LM348/OP-11/4156 pin configuration.
It is no longer necessary to compromise specifications,
while saving board space and cost, as compared to single
operational amplifiers.The LT1014’s low offset voltage of 50µV, drift of 0.3µV/°C,
offset current of 0.15nA, gain of 8 million, common mode
rejection of 117dB and power supply rejection of 120dB
qualify it as four truly precision operational amplifiers.
Oh?
And Mouser says:
1: $7.51
Oh.
2 Likes
I’m sure it would work fine with an
Lm324? Maybe I should try with my Doepfer lol
1 Like
Buried a little further down in the LT1014 datasheet:
Slew Rate 0.2 [min] 0.4 [typ] V/µs
1 Like
Bet none of our oscillators are tuned good enough for that fancy pants op amp to make a difference 
1 Like
I still want one! btw on the Doepfer the 1v is normalled thru all the inputs, not just the pot. So you can get +/- multiple octaves/ use it as a 9 octave switch
1 Like
Looks like you’re laying down a fine foundation for my footswitch keyboard 
2 Likes
Yeah, this is just work in progress, I’ll add the other 1V labels later 
This is the first input with the potentiometer. I was wondering what value the pot is and if it plays a role for the inverterting summers. Does the pot value contribute to the input resistor and has to be accounted for when calculating
gain = -R_feedback / R_input ? Or is it different here, because the pot is a voltage divider?
1 Like
Doepfers is an a50k. That’s a good question tho I would think any ol thing would work but the 100k after the pot is weird isn’t it
1 Like
The op amp analysis doesn’t depend on what’s upstream of the voltage at the input resistor. Whatever it is, whether it got produced by a voltage regulator or a voltage reference or a voltage divider or anything else, it’ll get multiplied by -R_f/R_i. You could do an analysis that includes the voltage divider, where the voltage at the top of the voltage divider gets multiplied by a smaller gain that takes into account not only the 100k but the resistances above and below the wiper, but you’d get the same answer. Since you know the voltage at the wiper it’s simplest just to take it from there and ignore where it came from.
Whatever the pot value is, when the pot’s at its limit the voltage at the wiper will be 0 and V_input. But if the pot resistance is not small compared to the input resistor, the voltage won’t vary linearly with the pot position as you go between those limits. If they’re equal the non linearity will be there, but not very large. So anything from 100k down will be all right for the pot. Of course the lower the value the more current it’ll consume, but 1 V through 10k is only 0.1 mA so who cares?
(Really what matters is what fraction of the current into the voltage divider goes out the wiper. If it’s small, you’re good. With a non inverting amplifier with a single input there would be essentially no current regardless (and no input resistor). Here the voltage at the op amp input is 0 so you do have current through the input resistor, and you need the resistor large enough that most of the current goes through the bottom of the attenuator instead of out the wiper.)
5 Likes
I made this little table to compare the opamps.
So, I think it’s mostly about the offset voltage, which is ~10 worse for tl074, but probably still okay, especially if you might use a quantizer afterwards
| lt1014 | tl074 | lm324 | |
|---|---|---|---|
| offset voltage (uV) | 150 | 1000 | 3000 (2000, on A) |
| drift (uV/°C) | 2 | 2 | 7 |
| slew rate (V/us) | 0.4 (0.2 min) | 25 | 0.5 |
3 Likes
Hi all! I found this thread while troubleshooting my A185-2 which won’t quite calibrate. As part of the troubleshooting process (and just for fun) I mapped out the circuit and thought I ought to share it here as this thread helped me visualize the general layout.
It’s not necessarily all correct, but I think it’s pretty close. I haven’t simulated it yet; this is just laid out in fritzing.
3 Likes