Electronic Design

Bob's Mailbox

You answered one question for me last year (“Why are FETs so expensive in India?”), and I now have another. How does a bipolar op-amp-based non-inverting dc amplifier amplify dc signals that are below a 0.6-V bipolar threshold (e.g., an LM358- based non-inverting dc amplifier)?

(Okay, you want a gain of +1.5 or 2 or 3 for a small signal that is barely above ground, such as +0.1 or +0.2 V? And the LM358 uses a small +V supply such as +6 V and ground? /rap)

The op-amp datasheet schematics show no bias on the input Darlingtons. Representative schematics show a bias constant current source for the differential-pairs tail current, but none for the bases of the two input PNP transistors.

(The input PNP’s emitters run on the base current of the differential-pair transistors. If you build up this circuit using ordinary transistors, it will work. Of course, 2N3906s have higher beta, but the lateral PNPs in the LM358 do not have that high beta. So there is always some current to run those emitters on. Further, an extra current is often fed to the emitter of the input transistor—maybe only 1/2 µA, but enough to give that PNP some emitter current to run on. Note that the LM358 schematic is a “simplified” schematic diagram. Transistors are so cheap, we can add in another transistor as needed to make it work a little better. We can add a little more current to the inputs’ emitters. Note: LM358 inputs work down not just to +0.1 V, but also to –0.1 and –0.2 V of VCM. /rap)

I have read all of chapter 7 in Art of Electronics as well as many other books that cover op-amp basics, but this particular question is explained nowhere. How are bipolar op-amp input transistors biased into conduction?

They just run on whatever current they can get from the following (differential) stage and from extra current sources. They don’t need a lot. Even 0.1 µA is plenty. If you open up an LM358H, you can see these current sources. Sometimes they are a part of the main 6-µA current source to the differential pair. Sometimes the transistors in the differential pair have two collectors and the smaller collector goes to its base, and thus to the input emitter. Either way will work. Both ways are used in various different amplifiers of this basic type.

In a recent article, you said that too much beta decreases mu. I can’t find anything on this in my textbooks. Would you explain?

You may have mediocre textbooks. Mu is an alternate statement of (1/Hrb). When the beta goes sky-high, the mu and the voltage gain go low. It is related to the Early effect. Have fun.

As I am going to buy a testbench for electronic circuit testing, I want some feedback about isolated power and the ground plane arrangement. Can you suggest a layout?

You haven’t given me much info on what you are trying to do, so I can not exactly give you feedback. But in general, it is a good idea to put a big slab of copper-clad glass-epoxy material on top of the workbench with the copper side up and ground it to the neutral of the power line as a ground plane. I like to set up at least 12 to 16 power outlets for an ordinary bench with local circuit breakers. Then I would usually ground most of my circuits to that grounded copper. I might put cardboard or insulating glass-epoxy over the ground plane if I need insulation. When I am working with high voltages, I try to keep one hand in a pocket before I turn the line power on. I did a lot of work on high-voltage stuff recently, and I never got “bit.” You also mentioned isolated power. I rarely need isolated power. I do use an isolation transformer on rare occasions. I do sometimes use a line-power “three-wire-to-two-wire” cheater so I can float a scope or sine generator, but I am pretty careful when I do that. I hope these are useful ideas for you.

I need to convert a 5-MHz sine wave to a 5-MHz TTL signal. Is there an IC that would do the job?

Take a look at some comparators. There are lots of slow ones, but you’ll want one with delays less than 15 ns. The LM360/ LM361 is fast enough. But you can pick one that gets along with your existing power supplies such as an LMV7219.

Comments invited! [email protected] —or:
Mail Stop D2597A,
National Semiconductor P.O. Box 58090,
Santa Clara, CA 95052-8090

BOB PEASE obtained a BSEE from MIT in 1961 and is Staff Scientist at
National Semiconductor Corp.,
Santa Clara, Calif.

Hide comments


  • Allowed HTML tags: <em> <strong> <blockquote> <br> <p>

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.