Discrete cfp (op-)amp

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi,
I'm looking for (basic) discrete cfp amp circuits to play with and examine. For example, a schematic of the AD811 (but I don't think it's published?)
Tietze/Schenk have a discussion of a transimpedance opamp which is the starting point for me.

thanks in advance,
Rüdiger
 
You think you can duplicate the function of an op amp by reproducing its internal circuit?
Don't ever think of it! Integrated circuits rely their operation in very well balanced transistors characteristics, with they achieve by making them into a single chip (same dopage) and close proximity (well termal mayching)
No way a discrete duplicate can even function.
 
I don't have any examples, but I'm here to say that it is possible to develop descrete op-amps. How do you think SS power amplifiers are mostly designed? A common three-stage SS push pull amplifier done in discrete topology with negative feedback loop and differential pair is really a power op-amp.

I can't say I've ever tried to duplicate an existing op-amp, you will not be able to easily achieve similar operational characteristics. You can make discrete op-amps if you know how to design these kinds of things though. There are cases where this is useful if you need the circuit to do things the pre-made ICs can't do.

Generally, though, the op-amp in its IC form from major suppliers will outperform anything one can build discretely by leaps and bounds from the standpoints of distortion, noise, offset, speed, etc depending on the kind of op-amp in question.
 
Do you have a copy of The Art of Electronics, but Horowitz and Hill? Especially the second edition?

This shows you more or less what you need to know about all sorts of circuitry. It's one of the few 'bibles' of electronics and everyone in the field should have it on hand.

http://www.ecircuitcenter.com/Circuits/opitov/opitov.htm

The link above shows the basic example of a transimpedance amplifier implemented with a normal opamp.

http://mysite.du.edu/~etuttle/electron/elect21.htm

This link shows a Norton amplifier, which is a true differential transimpedance amplifier.

You could build either type on the breadboard with discrete parts if you are stubborn enough; I know I am since I do so from time to time for fun.
 
Onvinyl said:
Hi,
I'm looking for (basic) discrete cfp amp circuits to play with and examine. For example, a schematic of the AD811 (but I don't think it's published?)
Tietze/Schenk have a discussion of a transimpedance opamp which is the starting point for me.

thanks in advance,
Rüdiger


Top engineers have put a great deal of thought and expertise into an op-amp.
Why not just take one off the shelf and use it yourself ?

My power amp designs use a very simple op-amp type cct which has much fewer transistors than an internal op-amp.
 
i'm working on an amp that basically uses a discrete op amp for the input stage and VAS. the "op amp" will basically be on a separate module, and the driver/ output stage will be on the main board. the use of the discrete op amp allows the amp to be scalable (within the voltage limits of the "op amp") simplifying the amp design. the "op amp" also has (like many op amps) pins for compensation caps, so the compensation can be tailored for various output device junction capacitances. the module can also be used as a standalone high voltage op amp for other applications.

i've also seen that Cirrus/Apex now has a dual "op amp" chip (with a package like an LM3886) that can operate at +/-100V rails. it's actually designed for driving piezo elements, but i'm thinking of trying it out as an audio amp driver.
 
I've set up some opamps in LTSpice. They simulate very well. I can post them tonight CET. AD829, AD797, LT1363 etc. I've not made any of them in real life, just some simple "basic" opamps. I did try CFP on the output stage with oscillation as reslut. With my limited experience I think the typical "diamond buffer" is outstanding as output stage. I guess it's not a coincidence that "all" good opamps use this like the three mentioned above.

I was on a quest for a perfect portable headphone amp and was never satisfied with monolithic opamps. People said a discrete battery powered amp was "impossible", but they're dead wrong.
 
Onvinyl said:
Hi,
I agree to both of you. There is no point in reversing an IC and hoping to outperform it's numbers. I'm looking for examples of transimpedance circuits to try discrete circuits. I *do* have hopes for good sound, anyway....
You have surely something to looking forward to. It's pretty easy to make a minimalistic and a lower power cfb amp.
 
When you wrote cfp, did you mean complementary feedback pair or current feedback something? I've tried some variants on this one. The sound is very good and it seems stable, but I have no oscilloscope.

A typical current feedback opamp
An externally hosted image should be here but it was not working when we last tested it.


Yes Per-Anders, they're all clones, not the real stuff. I know absolutely no electronics. Why be so serious. They simulate excellent, but I'm not sure how they behave IRL. If there will be a group buy for "Diamante", I'm going to try AD829 on that PCB. AD797 and AD829 are among the best monolitic opamps, so I'm curious to try something like that.

OPA827
An externally hosted image should be here but it was not working when we last tested it.

AD797
An externally hosted image should be here but it was not working when we last tested it.

AD829
An externally hosted image should be here but it was not working when we last tested it.

LT1363
An externally hosted image should be here but it was not working when we last tested it.

A complementary diff amp
An externally hosted image should be here but it was not working when we last tested it.

A simple complementary JFET input with DB
An externally hosted image should be here but it was not working when we last tested it.

LT1469
An externally hosted image should be here but it was not working when we last tested it.

A simple sym opamp
An externally hosted image should be here but it was not working when we last tested it.
 
you might find the following interesting..... this is the phono preamp from the Holman preamp. you will see that it IS a discrete opamp it has a FET and a BJT as the diff pair in the input stage, which does give the output a 2V dc offset (which is why the "opamp" has an output coupling cap). i suppose you could use a pair of BJTs or FETs instead. you can do a lot with this circuit. the feedback loop has caps and resistors in it which gives it an RIAA frequency response curve. the RIAA network isn't necessary for this being used as an opamp. the gate of Q1 is the nonimverting input, and the base of Q2 is the inverting input. for phono preamp use Q1 was a device hand selected for low noise. Q8 and Q9 are capacitance multipliers which act as active filters for the supply rails.
 

Attachments

  • aptphono.jpg
    aptphono.jpg
    46.6 KB · Views: 782
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.