As 2022 draws to a close, let's take a moment to celebrate the 50th anniversary of National Semiconductor's grand slam, the

LM324
LM339
LM358
LM393

when the world became a better place.

ak

 

Still among my favorite parts.

 

when the world became a better place.

. . but with rather more crossover distortion.

 

Didn't the LM3900 arrive about the same time? Has anyone figured out how to use it yet?

 

Back in the late 1970s, I used the LM3900 to make a comparator for a rather high input voltage of aprox 160 VDC.
Since the inputs are currents and sit at essentially ground, all you require is to use a suitable high-valued resistor to make a small input current to flow, and on the other input use another resistor, sized to produce the threshold current being drawn from a reference. Easy Peasy, and works like a charm.

 

Didn't the LM3900 arrive about the same time? Has anyone figured out how to use it yet?

You just bend both knees and jump into it. Just keep the input currents below the max limit. The output is short circuit protected so, pretty easy.

 

Amazing it's been this long, just used an LM358 in a PCB today.

 

You just bend both knees and jump into it. Just keep the input currents below the max limit. The output is short circuit protected so, pretty easy.

and it can do much the same things as can a LM324/LM339 but with less accuracy.

 

Amazing it's been this long, just used an LM358 in a PCB today.

Great for producing 0-10V outputs when you need them, provided that you don't forget the 100Ω build-out resistor (if you do, it oscillates)

 

and it can do much the same things as can a LM324/LM339 but with less accuracy.

Yes, you should definitely use an LM324 or LM339 if you have a circuit designed for one of those devices. And, since this was your comment about Norton Amps,

Has anyone figured out how to use it yet?

I'm guessing there is no need to waste my efforts trying to explain the different things you can do with a Norton amp - especially the LM3900's newer, faster younger brother, the 42-year-old LM359.

 

Yes, you should definitely use an LM324 or LM339 if you have a circuit designed for one of those devices. And, since this was your comment about Norton Amps,

I'm guessing there is no need to waste my efforts trying to explain the different things you can do with a Norton amp - especially the LM3900's newer, faster younger brother, the 42-year-old LM359.

Actually, I'm quite intrigued - I know how to use them to replace ordinary voltage amplifier circuit, but when would be advantageous to use one? When would you order one in specially even if you have a box full of LM324s in strores?
I've been working in electronics for almost 40 years and never used one. Perhaps we need a new thread for Norton amplifiers.

 

If I recall correctly, the LM399 preceded the LM324. I was gearing up to design some circuitry that generated sines and cosines and then did some multiplication - a lot op amps were going to be required and the LM399 was the first quad op amp I had ever seen or hear of. They worked pretty well in that application, though it probably would have been easier with the LM324 had it been around at the time.

I still use the LM324/LM358 on somethings and I have a pile of LM393's when I bought a company's surplus inventory in 1990.

 

Amazing it's been this long, just used an LM358 in a PCB today.

Contrary to the opinion of some, being an old design doesn't make it useless. You just need to know how and when to use it.

 

If I recall correctly, the LM399 preceded the LM324. I was gearing up to design some circuitry that generated sines and cosines and then did some multiplication - a lot op amps were going to be required and the LM399 was the first quad op amp I had ever seen or hear of. They worked pretty well in that application, though it probably would have been easier with the LM324 had it been around at the time.

The LM399 is a temperature-controlled voltage reference. . .

There were other dual and quad opamps at the time, but they did not have the 324's (and the others') magical input stage and TTL-specific output stage.

ak

 

There were other dual and quad opamps at the time, but they did not have the 324's (and the others') magical input stage and TTL-specific output stage.
ak

Did the release of the LM324/339 correspond to the invention of some new process for making PNP transistors on ICs?

 

Actually, I'm quite intrigued - I know how to use them to replace ordinary voltage amplifier circuit, but when would be advantageous to use one? When would you order one in specially even if you have a box full of LM324s in strores?
I've been working in electronics for almost 40 years and never used one. Perhaps we need a new thread for Norton amplifiers.

Our EE used Nortons as high-side current monitoring on some 100V DC motors used to adjust the slope of a vibratory feeder. The vibrations and dust would get into the lead screws and motors would strain. His system allowed monitoring current on the motors and we didn't burn out any more motors.

He also used LM359 Nortons to somehow make a proprietary DCS video signal play with the existing displays in the control room. There are better ways to do it now but, back in the late 1980's, he used the hammer he developed to hit his nail.

 

The LM399 is a temperature-controlled voltage reference. . .

There were other dual and quad opamps at the time, but they did not have the 324's (and the others') magical input stage and TTL-specific output stage.

ak

Good catch, I should have typed LM339. National was very perceptive in designing an opamp that can operate from a power supply for the 7400 integrated circuits.

 

Our EE used Nortons as high-side current monitoring on some 100V DC motors used to adjust the slope of a vibratory feeder. The vibrations and dust would get into the lead screws and motors would strain. His system allowed monitoring current on the motors and we didn't burn out any more motors.

That's smart - cancels out the Vbe variation. At first glance, it seems like it might be quite similar to the high-side current sense amplifiers that work at voltages above their supply rails.

 

Did the release of the LM324/339 correspond to the invention of some new process for making PNP transistors on ICs?

The story I have is that Widlar had the idea for the input stage, but it took a while (2 years - ?) working the the device physics and production groups to come up with a structure that had the gain-bandwidth and transconductance flatness he needed.

ak

 

According to Wikipedia, the LM324 was introduced in 1972 by National Semi. But it doesn't mention whom was the actual "father".
Although I read somewhere that Thomas M Frederiksen was the engineer who actually laid out the LM324's architecture, the reality is that as others have mentioned, it must have been a team effort, with assistance from the device science and process engineers.