Title says it all. These op-amps have extremely high bandwidth and a ridiculously fast slew rate. They can output a fair amount of current as well, but their open loop gain is their weakness (3k compared to 100k for a typical op-amp). I've got about 50 of these things I pulled out of an old, broken video amplifier board. Would like to put them to use.

My original plan was to do an analog sine generator with a decent range, but that devolved into a mess quickly. I was attempting to re-shape a triangle wave into a sine; did not go well (high distortion and a lot of "slanting" on the triangle wave). Something like the attached schem.

Any thoughts would be welcome.

Thanks in advance,
E

 

What are current-feedback amps good for?
You stated it in your second sentence.
You use them when you need an

extremely high bandwidth and a ridiculously fast slew rate.

They also do not have the gain-bandwidth limit of standard op amps, so their operating bandwidth is not greatly affected by the value of the closed-loop gain.

I was attempting to re-shape a triangle wave into a sine; did not go well (high distortion and a lot of "slanting" on the triangle wave). Something like the attached schem.

Did you follow the feedback impedance guidelines for that particular op amp?

 

I was able to find surprisingly little in the way of practical application, other than differential video amplification (almost like they were designed for that specific application, hmmm).

Regarding the function generator project, the "slanting" was able to be fixed using a diode and a series trimpot in the feedback loop to trim the duty cycle of the oscillator. The biggest issue was distortion, which came later in the diode clipping stage. At high frequencies, the circuit would struggle with the peaks of the wave. Now that I think about it, this could be the fault of the diodes.

 

Well, if you have a really large backyard, you might build nice gravity wave detector with a CFB: https://www.analog.com/en/landing-pages/001/ligo.html

A couple of years ago, I was interested in high-speed filters. These are the three most useful documents I found:
http://www.ti.com/lit/an/sloa066/sloa066.pdf?ts=1588487405465
https://www.analog.com/media/en/training-seminars/tutorials/MT-034.pdf
http://www.ti.com/lit/an/slyt081/slyt081.pdf

Note the advice from TI about avoiding capacitance directly in the inverting feedback loop.

 

There was an elantec CFA in my old function generator (max038). It had the role of being an output buffer, it could drive a 50 ohm cable.
The transfer function is different from that of a triangle waveform. Voltage gain is common enough anyway what about function of other waves shapes. The composite video signal holds much detail in the eye of a video technician who is familiar with that waveform.
In order to maintain all that composite at that speed the current mirror must be incredible. Someone told me that if we listen to voice from an amplifier like a radio talk show and if can hear good articulation in voice that we can actually hear distortion. Considering what the composite signal can do the CFA has excellent potential to process audio or other signal for high performance radio receiver that makes use of fine details.