TL072's are good, they have FET front end. I get nervous about flipping polarity... Was your PID completely analog?

100% analog. Lately, I've been considering to redesign it but not sure yet.

 

P.S. your Avatar reminds me of my IEEE insurance policy

I'm sure a new avatar is in the future... IEEE is good stuff.

 

100% analog. Lately, I've been considering to redesign it but not sure yet.

I did one earlier this spring... I added some improvements to speed up the overall response. I've seen it work in sim but can't say for certain in circuit. Fun project.

 

I did one earlier this spring... I added some improvements to speed up the overall response. I've seen it work in sim but can't say for certain in circuit. Fun project.

I followed closely your thread about your PID. Even posted once or twice there. Impressive.

 

thanks, it had to be the kids in there were 30 years younger than I am. I had to show them what us older folks can do.

 

Guessing if you know, Ron, that we live in rather different countries...

In some countries parts are at the Electronics Market. I can get almost anything there. Where I live now it is 1000 miles to a parts store. All parts must be shipped.

 

Damn... I've been watching strange parts and I'm so jealous of folks living in Shenzhen and Japan...

 

Damn... I've been watching strange parts and I'm so jealous of folks living in Shenzhen and Japan...

The best and most strange experience on that regard was after signing off from my last vessel in Singapore. With two days waiting for our flight to come back home, I got time to visit the Sim Lim Towers; one of them, fully dedicated to components. Why I bought so few things there, is a mystery to me.

 

I have some because I've used them in the past, but I consider them a specialty item like a transconductance amp.

The CA3140 wasn't even good for that. I traded an house design that used the CA3140 for an LF351, but that needed help with stability. I think I went with an OP41 with an (OP41 & LT1010) buffer. before that, I used the OP17.

 

The old CA3140 is too noisy for audio but its Jfet inputs are maybe the first ones.
I used many similar but newer audio TL072 duals and 10s of thousands of audio TL074 quads.

 

The old CA3140 is too noisy for audio but its Jfet inputs are maybe the first ones.

MOSFET input stage, loaded with a bipolar current mirror.

ak

 

For audio I like TL072, NE5532 and many of the Burr Brown ones, I do need opamps for other uses.

 

Yes I was wrong, the old CA3140 had an input made with Mosfets, not Jfets.
It was invented by RCA who invented Cmos logic, then the old CA series was sold to Harris then to Intersil then to Renesas of Japan.

 

@AnalogKid this opamp is strange and elusive. The input has to be at the threshold or lower for Q9/10 to conduct. So R2-5 keeping drain higher while source near VCC. Theres gotta be a Vin max thats lower than Vgs. The zeners may help by activating a Inversion. It’s not an audio opamp. Even the output is strange.

Not a standard opamp at all. What have you used it for? It’s cool because its so unusual.

 

The big claim for those series of opamps ws the very high input impedance, and I think the very low noise. It was much better at what it did best, at that time.

 

I didn’t realize how old they were.

 

Indeed, the amount of choices is overwhelming.
What I do, is to have three workhorses which are fine for the vast majority of projects:
-LM358, low cost, plain vanilla general purpose, wide supply range.
-TL072, audio and signal conditioning, widely available.
-TLC2272, low supply voltage, R2R.

Of course if I require a specific parameter like ultra low Vos, then I’ll search for chopper amps.

 

OK, so now your first selection step must be to decide what the requirements are, and probably how much you are willing too spend. Then , if they still have selection guides, go online to your preferred brands selection guide. Of course now the package is a big part of the deal. The last opamp that they ordered for the project that I was working on had several variations of pin-out available. Which pin-out was identified by the second letter after the part number. What a royal pain, just consider the possible production logistics problems. And this was a single SMT opamp about 0.1 inch square. Our specific requirements were the automotive temperature range and being very stable. Of course our other requirement was reliable availability, since it would go in a small sensor with 5 sensors on every car built by that company. Reliable availability is not a hobby requirement.

 

Reliable ability, kind of burned us last time, we had to special order 5,000 switches when we only needed about 100 for the test run because the client specified NKK... ugh... then they complained to me about it...