Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

OpAmp Resistor Selection Question

Status
Not open for further replies.
Quick question. Let's say I configure an op amp for a non-inverting gain of 11. G=1+Rf/Rg so Rf=10xRg. What is the difference between selecting 1M and 100K versus 330K and 33K? Why would you choose one versus the other?
 
All other things being equal, larger resistor values cause greater errors relative to the simplified, ideal opamp gain calculation because real opamps have finite (though small) input currents that result in voltage drops across the resistors. Larger resistor values also have greater thermal noise. On the other hand, lower values are harder for the output to drive and will result in higher current from the power supply. In your case I would go with the lower values because they are high enough to easily drive, won't draw much current, and will result in lower errors and noise. But, really, both pairs of resistors should do fine.
 
Last edited:
Thermal noise goes up with higher resistance values. Power consumption goes down. There are lots of trade offs. To high, the input bias current of the OP amp gets in the way. Too low, the Op amp is unable to drive it. Those are absolute limits.
 
There is another consideration, which is very important with wideband op amps, and that is the time constant comprised of the input capacitance and the Thevenin resistance seen by the inverting input. This time constant is an additional pole in the feedback loop, and can cause frequency peaking (ringing on fast edges), or even oscillation.
It can also be an issue on lower bandwidth amplifiers, especially on breadboards with high stray capacitances, or circuits which have unavoidable capacitance to ground.
Here is a sim of a gain-of-10 amplifier where the op amp has a 180MHz GBW (gain-bandwidth) product, and an input capacitance of about 2pF. Note that the sim has no stray PC board capacitance, which is inevitable in the real world.
The resistor values in the sim go from 1k/100 to 10k/1k to 100k/10k to 1Meg/100k as the peaking increases. Notice that the peaking never goes away entirely. This can be remedied by adding a cap across the feedback resistor, but the required value is so tiny that it is difficult to do.
 
.................................
The resistor values in the sim go from 1k/100 to 10k/1k to 100k/10k to 1Meg/100k as the peaking increases. Notice that the peaking never goes away entirely. This can be remedied by adding a cap across the feedback resistor, but the required value is so tiny that it is difficult to do.
Sometimes a tiny capacitance is generated by a "gimmick capacitor" consisting of two short insulated wires twisted together. The amount of capacitance can be adjusted to optimize the compensation by changing the length of the wires and/or the tightness of the twists.
 
Last edited:
Sometimes a tiny capacitance is generated by a "gimmick capacitor" consisting of two short insulated wires twisted together. The amount of capacitance can be adjusted to optimize the compensation by changing the length of the wires and/or the tightness of the twists.
Yes, that's what I was thinking. A true "trimmer" capacitor!

"I cut it off twice, and the peaking is still too high!";)
 
Last edited:
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top