Reason for Resistor Before an Op-Amp Follower?

[Higgy]

I'm designing a circuit for experimental use. I'll make this as quick as possible.

Browsing other circuit designs on the web, I initially noticed "followers" placed between other op-amp app's (adders, integrators, inverters, etc). From what I can tell, putting in followers like this must be a good practice in circuit design (because it provides some kind of "buffer" throughout the circuit?).

That said, why do I also see resistors (with seemingly random values) between adders and followers, integrators and followers, etc. Followers don't require an input resistance to do their job! Thus my confusion! Can't we just do without all those resistors?

To help illustrate my question, I've attached one example I saw. Why did the circuit designer choose to put a 49.9k resistor between the two op-amps? Should I incorporate this practice into my circuit design?

Thanks!

 

[waht]

I've always wondered this myself. You can substitute any resistor value up to a megaohm and it would still work because of its high input impedance, and low output impedance.

 

[Higgy]

I've always wondered this myself. You can substitute any resistor value up to a megaohm and it would still work because of its high input impedance, and low output impedance.

Right, that makes sense. Now my thinking is this: For a circuit with one voltage input and output (and, say, lots of op-amps in the middle, as is my circuit), the followers aren't necessary - except, perhaps, one at the beginning to act as a "buffer". So I think that many of the followers I see in these old circuits might just be redundant "hand-me downs" from modifications done to some older circuit.

If that's the case, then I'm going to consider all the "mystery" resistors as just contributing to the input resistance to whatever op-amp app (adder, integrator, etc) that comes next in the circuit.

Of course, I'd feel more at ease if a bit more was said about the actual implementation of followers into circuits somewhere.

 

[Bob S]

Hello Higgy-
To minimize offsets due to input bias currents in voltage followers, there should be an equal-value resistor in series with both inputs.
Bob S

 

[es1]

The resistor in Higgy's circuit is probably for current limiting, not bias current balancing. If the output of the first opamp can go below ground or exceed vcc on the second then is could bias clamp diodes that protect the inputs. The series resistor limits the current into the clamp. See this datasheet for an example: http://focus.ti.com/docs/prod/folders/print/opa348.html

If the first stage of the opamp is a BJT then people also use resistors so that those transistors see similar impedances (or if there is an initial chopper stage). But if this was the case then the feedback path would not be a short but another 49.9K R. There are good reasons to do this to but it depends on the implementation of the amp.