Electronic Design

What's All This "SMWISICDSI" Stuff, Anyhow?

A long time ago, I went out to dinner with Dave Ludwig, an old friend of mine from Massachusetts. We went to a fancy Italian restaurant. After we had ordered our entrées, the waiter asked, "Would you like a salad?" I declined, but Dave said he would like one. "What would you like in your salad?" Dave said lettuce and tomatoes.

In a few minutes the salad arrived. Yes, it had lettuce and tomatoes. It had three kinds of lettuce, and tomatoes and chick-peas and croutons and onion slices and green onions and Italian cherry peppers and slices of hard-boiled egg, and three kinds of salad dressing on the side. I sat there with barely suppressed astonishment, and Dave just sat there with a quiet, resigned smile. After the waiter departed, Dave explained: There was no point in complaining or griping or hollering, because at the best, the manager or the headwaiter would just come over and say, "Hey, you don't have to eat anything you don't want, and you don't have to pay for anything you don't want, so, what's your complaint? Show Me Where It Says I Can't Do It." And he explained that he was an aficionado of these kinds of stories, which we will of course abbreviate to "SMWISICDI."

He pulled a faded clipping from his pocket. The newspaper story was about the advantages of home ownership. "If you own your own home, you can play the piano at midnight." I thought about it. I said, "Dave, I can't play the piano worth a darn. But this story says that I can play the piano at midnight if I own my own house."

Dave smiled and agreed. "You just think you can't play the piano. But, have you ever tried at midnight?" I had to admit that I had not. But, one of these days, I will try playing at midnight, to see if I'm any better than at any other time. What I suspect is that at 12:00:02, I'll suddenly be a rotten pianist again.

Now I, too, have gotten interested in SMWISICDI stories. In fact, last week I mailed 5 ounces of SMWISICDI stories and clippings to Dave. For example, I clipped out a story about how the Israeli Army solved the problem of what to do when Palestinians throw stones at the soldiers. The Israeli Military invented an automated stone thrower to retaliate. An eye for an eye, a tooth for a tooth - and a rock for a rock? What if the U.S. Army sent out a Request For Quote on developing, manufacturing, and deploying a rock- thrower? What if Jack Anderson got wind of this, and confronted the Joint Chiefs of Staff? They would just tell him, "Show Me Where It Says I Can't Do It."

Last week, Wanda Garrett, our senior applications engineer for amplifiers and regulators, got a phone call from an unhappy customer. He had used one of our ICs to design a switching regulator, and it didn't work well at all. The output had glitches and burps and excessive ripple and noise. The regulation was poor, the loop stability was rotten, and the efficiency wasn't even very good.

After a lot of inquiry, Wanda discovered that this person had built up the switcher on one of those solderless breadboards. OHHH!! Patiently, Wanda explained, that is exactly what you expect when you use one of those solderless beasts. The inductances are awful, the capacitances will cause crosstalk between adjacent buses, and if you try to build a switching-type regulator, of course it will work badly. And the customer replied. "SMWISICDI."

Now, Wanda was a little taken aback. She had to admit, of all the things that we might have told people they should not expect to work, this was one that we didn't specifically warn against. But, to put the shoe on the other foot, she asked, "Where does it say you can do this?" And the customer replied that in the solderless breadboard's promotional brochure, it says, "Ideal for high- frequency and high-speed/low-noise circuits." Wanda observed that was probably not quite true - neither for linear circuits, fast digital circuits, fast ADCs, nor switchers. Then she pointed out to the customer that the LM2575 data sheet does spell out: "As in any switching regulator, layout is very important. Rapidly switching currents associated with wiring inductance generate voltage transients that cause problems. For minimum stray inductance and ground loops, the length of the leads indicated by heavy lines should be kept as short as possible. Single-point grounding or ground-plane construction should be used for best results."

So we really did tell every customer that you need a good layout, right in the data sheet. Of course, if the customer believes that solderless breadboards are really great for high-frequency circuits, then we have a problem - which Wanda was able to resolve and explain. The solderless breadboards do cause many troubles.

First of all, most fast ICs, whether linear or digital, require a good ceramic power-supply bypass capacitor, right close to the IC. But with those long buses inside a solderless breadboard, it's hard to get a bypass capacitor with less than 3 or 4 inches of loop. That won't help a switcher or any other fast circuit. Then, the capacitance between adjacent buses - typically 2 to 4 pF, depending on the size - is going to cause stray coupling that will probably make the circuit unhappy. That's my experience. Furthermore, when you have a switching transistor turning off, its collector or drain can easily slew at 600V/µs or more. If your "catch" diode is spaced more than an inch from the inductor and transistor, the L di/dt can cause dozens of volts of overshoot, which may overstress the switching transistor (exceeding its voltage ratings), not to mention generating some horrible spikes in the air. And those white slabs - they are not Teflon. They aren't polystyrene or polyethylene, either. They are made of nylon, or something similar, so the leakage can be pretty bad on a warm or humid day. Even worse, if you push a whole lot of wires into those little solderless connectors, the little scraps of solder will get scraped off until there's a whole pile of solder scraps hidden inside. Then they can start making intermittent short-circuits between adjacent buses - won't that be fun to troubleshoot.

So Wanda explained all of these reasons not to use a solderless breadboard for making a switcher. She sent the customer a little PC board that was neat and compact to help him get a prototype of the circuit working - one of the LM2575 "Simple Switchers." Note, we normally think that these "Simple Switchers" cannot miss - they're very easy to apply. But, if you try to build it on a solderless breadboard, even a simple circuit can be hurt - ruined - by the strays of a poor layout. Simple, yes. But foolproof and tolerant of a truly bad layout? No.

Then she warned the rest of us Applications Engineers that customers might be having trouble when using solderless breadboards, ultimately complaining "SMWISICDI." I'm just passing on this warning to you readers.

Finally, Wanda said she was going to try to put a disclaimer in our linear databooks and applications handbooks, that the "solderless breadboards" are unsuitable for any applications other than medium-speed, medium-impedance-level, and medium-precision circuits. It may sound silly, but I know that she'll find a way to put in a caution flag where it's appropriate. I mean, Wanda is the Czarina of Linear Data Books. She can put anything she wants in there. Show Me Where It Says She Can't Do It!

Comments invited! / RAP
Robert A. Pease / Engineer

Originally published in Electronic Design, May 9, 1991.

Hide comments

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.
Publish