op amp splitter design

Frequent asked about building blocks: gain stages, buffers, clipping configurations, ...
Post Reply
User avatar
heaterman
Information
Posts: 25
Joined: 27 Jul 2017, 08:44
Has thanked: 8 times
Been thanked: 18 times

Post by heaterman »

Hello everbody!

I am working on a 4 way splitter (could also be 2 or 3 way) and i have some really basic questions. I did research but could not find perfect answers...
here is the schematic so far:

Image

So: my qustions are:
1) Power: I have seen small resistors or intuctor in series coming from 9V. Does that help the power filtering?

2) Vref: Resistors (R1 and R2) values for the reference voltage (virtual ground, 4.5V here) I have seen a lot of different values from 10k to 1m. Based on what do we choose this values? Lower values means more current, right? But when is this needed?

3) R3 (1M) alone determines the input impedance along with the input Z of the opamp, is that right?

4) resistors at the output: first i had none (no R10, no R11) that caused very loud switch popping in the next pedal (overdrive, truebypass). The R10 100k made that waaaay better. I have seen values for resistors in this place from 10K to 1K. What effects does a low or high value have?
Then: R11 is for protecting the OP amp, plus it sets the output impedance's... is that one correct? I have seen this on also right after the opamp output before the 1uF cap. What difference does the position make?

A lot of questions.... sorry :|
Any further improvement suggestions, other thoughts? have i overseen some important stuff?
The circuit is working good so far!

Cheers ant thanx!!
David

User avatar
FootRock
Information
Posts: 2
Joined: 29 Dec 2018, 16:04
Been thanked: 3 times

Post by FootRock »

Hi Heaterman
1) Power: I have seen small resistors or intuctor in series coming from 9V. Does that help the power filtering?
yes, that is the intention. any series resistive element (including inductors) between the 9V input and the circuit will reduce noise coming from the 9V input, but beware it will increase noise generated by the circuit itself. this is usually taken care of by capacitors down stream from the resistive element.
2) Vref: Resistors (R1 and R2) values for the reference voltage (virtual ground, 4.5V here) I have seen a lot of different values from 10k to 1m. Based on what do we choose this values? Lower values means more current, right? But when is this needed?
higher currents are correct, but think of this more like impedance. your AC current draw from the reference is taken care of by the cap, but this still has a cut-off frequency like any filter. imagine drawing 1mA (DC) from the reference voltage towards ground. there's now a 1mA difference between the current in the upper resistor and the lower resistor. this means that the reference voltage will be offset by this current according to ohms law. in other words, the output has an impedance.
there's also the length of time it takes to charge your 47uF cap when you first power on the circuit.

there are many other reasons to choose small or large value resistors - thermal noise, energy efficiency, cost, power ratings. 100k is about right.
3) R3 (1M) alone determines the input impedance along with the input Z of the opamp, is that right?
not exactly, you also have the reference output impedance mentioned above. in this case its all negligible (tl072 input is 10^12). you can just say its the 1meg resistor
4) resistors at the output: first i had none (no R10, no R11) that caused very loud switch popping in the next pedal (overdrive, truebypass). The R10 100k made that waaaay better. I have seen values for resistors in this place from 10K to 1K. What effects does a low or high value have?
the output resistor to ground needs to balance pulling the output to 0V (lack of which is what cases the pops) against neither loading the op-amp nor allowing the DC blocking cap to affect the frequency response. a high value means less leakage or bias current will cause the DC operating point to drift away from 0V. a low value means you need a larger cap to keep the high pass corner frequency below the desired pass band.

some numbers: with a 1uF cap and 100k, your 3db frequency is about 1.6Hz. 1k brings this up to about 160Hz, at which point you've started filtering out the low E string on a standard tuned guitar. you'd need to up your cap to about 100u to get back the bass.

bear in mind that this cut-off will be impacted by the following device. if you choose 100k, and load the output with 100k, it would be the same frequency response as if you'd chosen 50k. however if you choose 1k, then loading the output with 100k makes no a jot of difference. (well, okay, it makes 1% difference or something, you end up with 990R instead of 1k)

conclusion? as low as possible without loading the op-amp or requiring ridiculous caps.
Then: R11 is for protecting the OP amp, plus it sets the output impedance's... is that one correct? I have seen this on also right after the op-amp output before the 1uF cap. What difference does the position make?
it adds to the output impedance.

no difference really, in its current position, its in parallel with the 100k to ground, but with it on the other side of the cap its in series. either way, its in series with any load impedance, which is presumably what your trying to protect against. so long as the load impedance is much bigger, it wont have an effect.
other thoughts? have i overseen some important stuff?
you may benefit from better / more decoupling caps on your op-amp. some smaller valued ceramics are usually recommended.

Post Reply