But first, an esaeP's Fable.... Once upon a time, the Lion woke up in an especially powerful mood, so he started out through the jungle. He met a hyena and roared, "Who is the Lord of the Jungle?" The hyena replied, "You are."
The Lion met a wildebeest, and roared, "Who is the Lord of the Jungle?" The wildebeest replied, "You are." Further along, the Lion met a monkey, and again roared, "Who is the Lord of the Jungle?" And the monkey confirmed, "You are."
Then the Lion met an elephant, and he asked the same question. Without a word, the elephant grabbed the Lion, hoisted him over his head, and SLAMMED him to the ground. After a few seconds, the Lion collected his wits, and struggled to his feet, and proceeded to slink down the trail.
But he called back over his shoulder, "Well, you don't have to get sore just because you don't know the answer." End of Fable.
About 6 years ago, I declared myself the Czar of Bandgaps. I stated that anybody at National Semiconductor who was designing a bandgap reference must clear his design with me. Sure enough, since that time, I've been fairly busy in my spare time, checking the circuits and the layouts for new bandgap circuits.
Recently, a reader wrote that he though that Paul Brokaw at Analog Devices was the Czar of Bandgaps. I wrote a nice reply, and said that I had never heard Paul claim to be the Czar of Bandgaps, and Paul had never slammed me to the ground. But if Paul wanted to claim to be the Czar of Bandgaps, well, I wasn't going to argue with Paul. However, if anybody had the right to claim to be the original Czar of Bandgaps, it could have been the late Bob Widlar, since his LM113 and LM109 circuits came first. But Bob Widlar never slammed me to the ground. And he never challenged me about my Czardom-for good reason!
First of all, why would anybody make the claim to be the First Czar of Bandgaps?? I was the first to do so, as nobody before me had ever been stupid enough to make that claim. I only made that claim because my boss and his boss were unhappy because several new bandgap-reference circuits had been observed to not work properly, and the mistakes were the same as old mistakes made years ago. It's always a shame when you make a mistake, but it's especially sad when you repeat an old mistake like that. It's akin to the old saying, "Those who do not remember history are condemned to repeat it."
I figured I was pretty good at noticing and remembering things, and as I was a veteran of over 10 years at National, I had a good chance of doing more good than harm. At the time, I had made a bunch of mistakes in my own circuits, but I had never yet made the same mistake twice. So I made this outrageous bluff, and I promised to be helpful to make sure that no old, foolish, unnecessary mistakes were made, and not too many new ones, either.
So far, I think I've done a pretty good job of catching the mistakes in the bandgap references brought to me-at least half of them. I've encouraged the engineers to think about worst-case design, and good layouts and good testing. Our references and regulators have been coming out with fewer dumb old mistakes, although a some brand new mistakes still creep in (which we keep learning from). In a few months, I will demonstrate a few good bandgap circuits that you can't buy, but that you can make for your general amusement and edification.
Additionally, we decided, after a while, to establish a couple other Czardoms: we have a Czar of Startup Circuits, a Czar of Computer Goofups, and a Czar of Czener Czaps. Now, that would ordinarily be spelled, Czar of Zener Zaps-that is, the expert on trimming of circuits, using fuses and "Zener Zaps." But we decided the alliteration was a little too good to resist. Also, we have a Czarina of Linear Data Sheets. And, oh, I almost forgot, I also declared myself to be the Czar of Proofreading. I am the best Proofreader I know, and I have a very bright eye for typo errors. I've caught typo errors in the telephone book, in the Sears Roebuck catalog, and in the Holy Bible (RSV). But that does not say I'm perfect at catching all of the errors in my own work-I'm NOT perfect when it comes to my writing, because every writer is too close to his own work.
In Dec 27, 1990, I used the word tendency twice in one paragraph of my column, and it was spelled two different ways. The English language is flexible, but not that flexible. Bob Milne graciously conceded that he had put the wrong "tendency" in there, at the last minute, so late that it never had a chance to meet the spell-checker. And when he faxed a copy to me, I didn't spot it. It just goes to show that even a Czar can miss an error if he's too close to it. I regularly have fistfights with the spell-checker in my word processor. And Mr. Milne and Roger Engelke, Electronic Design's chief copy editor, have been known to shake their heads at some of the odd spellings that I think are perfectly correct....
What else have I done as Czar of Bandgaps? I gave a 40-minute Tutorial at the IEEE Bipolar Circuits and Technology Meeting, back in September of 1990, not just about how to do it right, but, more interesting, how to do it wrong. I mean, you can look in any number of data sheets to see a circuit that works, but looking at a circuit that DIDN'T work is more educational. I can't give away every secret so that every engineer can copy any circuit with impunity, but it's not to anybody's good to have everybody making all sorts of dumb old mistakes.
For example, at National, when we engineers have a circuit laid out as well as we can (at least to the best of our knowledge), we invite all our peers and colleagues (engineers and technicians, too) to see if they can detect any errors in the circuit's layout. If they find a significant error, they win a mug of beer, to be paid off later. That's why we call it a Beer check. It's much cheaper for me to give away a few pitchers of beer, with pizza to match, than to have my circuit crash and burn just because of a dumb oversight on my part. As I mentioned above, I can easily overlook a dumb error on my circuit because I'm too close to it. I can let through a mistake in my circuit that I would surely spot if it were in somebody else's circuit. Isn't that true for you, too?
So if you think there are certain classes of errors that cause excessive trouble, well, you too might declare yourself (with your boss's approval) the Lord of the Jungle, or the Czar of Thingamabobs. Just don't be too surprised if somebody doesn't understand and tries to slam you....
Recently I was giving a lecture, back East, and I asked the engineers, "Do you have any Czars in your company?" A week later, I got a nice letter from a senior engineer who realized that his company did have several Czars, experts in a particular niche, who didn't get enough recognition or appreciation. He said he was planning to give them a little extra responsibility and appreciation, as Czars. Maybe I'm on the right tack after all! So, I will ask the same of YOU, whether you have previously recognized them or not: Do YOU have any Czars in your company?
All for now./Comments invited! RAP/Robert A. Pease/Engineer
Address: Mail Stop C2500A, National Semiconductor, P.O. Box 58090, Santa Clara, CA 95052-8090