The question is in the title.

 

C1 can't roll off the closed loop gain to ensure stabiltiy, that will be the job of C2 in conjunction with the external feedback resistor(s) (not shown on your circuit) that establish the closed loop.

 

Actually, this is the Mitch electronics student trainer op-amp.

http://forum.allaboutcircuits.com/threads/discrete-op-amp-electronic-kit-diy.106607/#post-817557

So, I have my doubts:

1) C1 is there to set a high end roll-off frequency. If that frequency limitation exists in open loop, it still exists in closed loop, doesn't it? The closed loop gain merely intersects the line on the gain per frequency graph set by C1, and that is your maximum frequency of operation (unless you add capacitance to the external feedback to limit frequency even more than C1 limits frequency). True?

2) C2 sets a high frequency limit by being in series with R4. I drew in a closed loop gain of 2, but I can't see how C2 affects that or is affected by it.

Please type slowly because I am dead serious about not understanding why C2 has a job that C1 can't do.

 

How about this: 10.6 KHz

1.5k in series with C2 means C2 will not start being effective until about 10.6 KHz.

How does C1 work relative to this? A frequency reduction curve that intersects zero-zero?

 

Does C2 cause a 2nd-order rolloff above 10.6kHz to help insure closed-loop stability at low closed-loop gains?

 

Does C2 cause a 2nd-order rolloff above 10.6kHz to help insure closed-loop stability at low closed-loop gains?

That's what I think. You and AG are people that know this so much better than I do. I am hoping you didn't intend to end that with a question mark.

 

A simulation would likely tell you that.

 

A simulation would likely tell you that.

Well now you have my walnuts in a vise. You know I don't Sim.

 

Actually, C1 *does* increase stability, by rolling off the high-frequency gain. It forms an integrator with Q6 that acts as a slew-rate limiter and single-pole lowpass filter. At very high frequencies, phase shifts caused by junction capacitance in the transistors can add up to enough to cause oscillation. C1 acts as intentionally large base-collector junction capacitance to reduce the gain at those frequencies so it is insufficient to sustain oscillation. The corner frequency is the output impedance of the previous stage and the capacitor. C1 is called a Miller capacitor after the guy who figured this out in 1920. Also called a pole-splitting capacitor.

ak

 

Well now you have my walnuts in a vise. You know I don't Sim.

Well, since my math skills are less than stellar, simulation is the only way I have to answer the question.

 

Well, since my math skills are less than stellar, simulation is the only way I have to answer the question.

Now I feel better about the fact that I have algebra'd my way through electronic designs for the last 50 years.

Algebra isn't the only tool one needs, but being good at it can get you by in a scrape.

 

I managed to get through all my algebra and calculus classes in fairly good order but I never was good at "thinking mathematically" if you know what I mean. So I basically try to solve problems with a minimum of math -- some algebra and perhaps a little calculus if necessary.
I would never want to do a KVL analysis for a multiple loop circuit for example, especially with reactive components. That's an exercise in masochism as far as I'm concerned. And S-plane transfer functions and pole-zero plots are mostly Greek to me.
I work more or less on intuition from experience and let the computer and simulator do the heavy lifting when it comes to determining the circuit performance details.

 

One of the habits/beliefs that limited my education was my need to internalize information. That means something about being able to visualize relationships, be it in electronics, physics, math, or chemistry. For instance, I see the voltage on a collector as its distance from the positive rail, a capacitor looks like a coffee can to me, and I see a zener diode as a spacer like a stack of washers on a bolt. This means that what I'm good at, I'm frankly intuitive, but the penalty is a ceiling on my learning ability. When I'm in my zone, I look like magic. When I hit my limit, I have nothing.

 

I'm a visual guy too, and for what it's worth, back when I was good at calculus, I could "see" what I was doing. It was very intuitive and visual. I could see 3D integrals everywhere in the world around me.

It has passed.

 

Einstein was the same way. His genius was hampered by a fantastic imagination. Sounds contradictory, but in fact it cost him the Nobel prize. He could "see" the red-shift that led to Relativity, and imagine *almost* anything, but not everything. He could not endorse quantum mechanics because it is so counter-intuitive that he couldn't "see" it work. He and Bohr whaled on each other for decades before and after Bohr got the Nobel. Einstein got one, but for the photoelectric effect, not relativity.

ak

 

I'm a visual guy too, and for what it's worth, back when I was good at calculus, I could "see" what I was doing. It was very intuitive and visual. I could see 3D integrals everywhere in the world around me.

It has passed.

When the day came that I got in the shower and the equation for the position of the water drops popped into my head, I realized that I had been soaked in Math and Science for too long. Fortunately, I beat it back down enough that I have to intentionally think of that kind of math when I need it.

ps, Einstein and I both failed algebra the first time we took the class.