Electronic Design

What's All This FLOOBYDUST Stuff, Anyhow?-part 5

A JACK-KNIFE is Handiest when at Hand: I have carried a pretty good jack-knife for about 50 years. But it wasn't the same jack-knife, it was about 40 different knives. Obviously, it's easy to lose a jack-knife, or leave it behind. I've done it dozens of times. It is not just annoying because it's expensive to buy the replacement knife; you may have to wander around without any knife for a while until you have a chance to buy one. (You can partly alleviate this by buying 2 or 3 knives when they are on sale....)

Recently, my friend Will Frangos said, "Take a look at my Swiss-Army knife. Have you ever seen one with a blade this narrow?" I agreed; it looked like he had been using it and sharpening it for many years. He explained that he, too, always used to lose his knife, every few months—but he decided to tie it to a cord, a tether tied to his belt, so it would not get lost. Consequently it was going strong after 10 years.

I thought about that, but I didn't like the idea of a simple cord. If it was long enough to let me do useful work, the cord would be too floppy and bulky. I remembered the retractable cable for keys that key shops sell. I bought one. The legend stamped onto its back reads: " 'KEY-BAK', US Patent 2732148. Mfd. & Exported by West Coast Chain Mfg. Co., Box 9088, Ontario Canada. Distributed by CTL Co., Wausau WISC. Assy. in Mexico." This little reel, about 2-in. diameter by 1/2-in. thick, and weighing just 2 ounces, clips on to my belt.

When I need my knife, this cable reels out of the reel, as much as 4 feet. When my work is done, the cable reels back in. In addition, I put in a double strand of black shoelace to link from the knife to the cable. Then, when the cable is all the way up, the knife sits comfortably in the bottom of my pocket. I won't be buying very many knives for a while. If you like to carry a jack-knife, I can recommend this.

I got a copy of that old patent. The original design had a light chain. The newer version with the cable seems nicer. Some key stores sell that "Key-Bak" for about $14. Others sell it for $8. Whatever.

A while back, I had a dream where I was trying to protect a couple of old ladies from a vicious snarling, threatening monkey. I got out my jack- knife, and I confronted the monkey. Soon the monkey made a savvy move, and bit my hand—and grabbed the knife. But not for long, as he did not comprehend the return spring that brought the knife back to me before he could bite the shoelaces. Even in dreams, a jack-knife is a handy gadget....

MEASUREMENT ERROR? NO, a thinking error. A while back, in my Measurement Stuff Column, I said that my thumbnail was 12 milli-inches thick—and thus just 0.45 millimeters, not a whole millimeter as the Trivia Man said. One reader caught me up: "Bob, 12 milli-inches is NOT 0.45 mm, it is 0.3 mm." I went back and checked my math. Obviously, I had divided by 25.4, when I shoulda divided by 39.37. That's pretty dumb!

I usually am careful to start out: "1 meter is 39.37 inches, and 1 cm is 39.37 centi-inches, and 1 mm = 39.37 milli-inches. Then 0.3 x 1 mm would be about 12 milli-inches." IF you start from basic definitions, and use CORRECT dimensional analysis, you can avoid being fooled. I recommend good dimensional analysis. More later.

NOTE: While we all know 5280 feet is a mile, it is also useful (and convenient) to know that a kilometer is about 3280 feet, so when you are working with elevations, you can work with that. A mountain at 3,000 meters is 3 ´3280 feet high, or about 9840 feet. Much more precise and useful than "0.62 mile."

However, it is a surprise to find that 3280 feet and 10 inches is EXACTLY 1 kilometer. Not just approximately, or within a couple ppm, but EXACTLY. (Note, 2.54 cm is NOT exactly an inch. It's exactly 2 ppm shorter than an inch.)

When we did the Column on Measurements, a few months back, Roger Engelke, formerly the Chief Copy Editor at Electronic Design was horrified by my usage of "milli-inch," because he could not find it in any dictionary. But we got it published, anyhow. If he sees "centi-inch," he'll just DIE.

Maybe I should explain about the "ranchette." The "ranchette" is a unit of area defined as 1/2 pico-acre. The ranchette is exactly the size of a round emitter 1.0 milli-inches in radius (well, 0.9983 milli-inch), so it really is handy when dealing with large matched pairs of transistors. NO, I did not invent the "ranchette." I read about that, years ago. Yeah, let's include this and give Roger a hard time. Meanwhile, does anybody know to whom we should give "credit" for inventing the "ranchette?"

LAPTOPS— OKAY, you have seen me swap letters back-and-forth with several guys who talked about the advantages of the Tandy Model 100 or Model 102. And you know my opinion of hog-at-the-trough Pentium computers, fancy window-type formats, and power-hungry color LCD displays. (Heck, I cannot type at 166 MHz, 66 MHz, 66 kHz, or even 66 Hz. How about you?)

So you won't be surprised to hear that I got me a Model 102 from one of my readers who was going to give his to the Salvation Army. I figured that anybody who bought it at the Salvation Army would not know what it's good for. But I would. So I made it a home.

Good news—the keyboard is full-sized, and QUITE adequate and comfortable. NOT chiclets. Good news— the version with the full "32k" has an actual 29 kbytes of memory available, which is usually enough for an all-typing weekend or a 3-day business trip. Excellent news? the battery life (4 alkaline AA cells) is over 16 hours, and I can fill it up with 28k of typing in less than that.

The display is just 8 lines by 40 characters, but I can live with that. Moving around on the display is not very quick—but it's livable.

The little $10 cable and the $20 software that you need to transfer a file into a PC are a piece of cake. Its Word Processing system is quite adequate, and sufficiently user-friendly. Not a bad machine. The Model 102 weighs just 3 lb, slightly less than the older Model 100. It's like new. It will probably last forever.

If you are interested in a good old Model 100 (ballpark of $250) or a clean Model 102 (about $450), contact Richard Hanson. You can e-mail him at [email protected]; or fax to (510) 937-5039; or phone to (510) 932-8956; or write to Club 100, P.O. Box 23438, Pleasant Hill, CA 94523-0438. These computers USED to be cheap, but now as they are getting better appreciated, the prices are going up. What that means is that Mr. Hanson will be able to make enough $ to keep them running for over 5 or 10 years more. I'll cheerfully pay to make sure that happens.

POSTAGE BY THE OUNCE—Recently, I got a letter from a guy asking for the answer to last year's April Fool's question on zenering the emitter of a transistor. He was nice enough to put in a SASE for my convenience. BUT he put two stamps on it. Well, perhaps he could not be POSITIVE my reply would not be over 1 ounce, so he put on two stamps.

Then I looked at his letter to ME. He put two 32-cent stamps on that, TOO—and he just had two pages in it. Hey, I like our postal service pretty well, but I hate to pay them double. I like to put just enough STUFF into my envelope to bring it up to 0.95 ounces, or 2.95 ounces, or whatever. I like to get my money's worth whenever I send a letter. That often requires me to leave out a couple clippings that will be sent later. When I am sending JUNK, I have to weigh it. But when I am just sending letters, I know that 4.8 pages of ordinary photocopy paper, when put in an NSC envelope, weighs just under 1 ounce. If I want to send 10 pages, that's still under 2 ounces, and costs 55 cents. NOTE: A lot of people know that the second ounce (and each succeeding ounce) does not cost 32 cents—it just costs 23 cents. So 7.9 ounces will cost 8 '0.23 + 0.09, or $1.93. This is easily done with six 32-cent stamps plus a 1-cent stamp. It's a lot cheaper than just throwing on EIGHT 32-cent stamps.

(But when you get up to 13 ounces, which sounds like it should cost $3.08, STOP. From 13 ounces to 2 lb, it is a flat rate of $3.00, for First Class mail ("priority mail" = air mail.))

Postage for AIR Mail to overseas locations does cost 60 cents for the first half-ounce, BUT only 40 cents for the second half-ounce. So a full 1.9-ounce letter costs $1.80—NOT $2.40. The post office does not publicize this very much.

Mail to Canada is 46 cents for the first half-ounce, and 52 cents for the first ounce. From there on it is nonlinear: 72 cents for 1.99 ounces, and 95 cents for 2.99. Postage to Mexico is nonlinear, too.

A friend once sent me a letter with 78 cents worth of stamps—enough postage for 2.9 ounces. But the letter did not seem heavier than 2 ounces. I weighed it. It was about 1.95 ounces—and 55 cents shoulda been enough. I called him up and asked why he put on 78 cents. He said he checked it at his company mailroom, and it was just over 2 ounces. I suggested he should get those scales checked for calibration. He did, and they were right on. So was MY scale—right on.

We figured out later he was in a damp, humid climate. By the time his letter got to me, it had gotten enough drier to have fallen from 2.05 ounces to below 1.98 ounces. If you live in a damp place, and if you had a letter that was just 3.02 ounces, and if you only had 78 cents of stamps, you could put the envelope in your toaster-oven at 150 degrees for a couple of minutes. Then it would drop down below 3 ounces, and would go legally for 78 cents instead of $1.01. This is not always a great way to save money, but if you are about out of stamps (and if you have already snipped off the borders of the paper) this is a possible way to get the weight down just a little.

But, in general, as a rule of thumb, if your envelopes are not extremely heavy, you can mail AT LEAST 4 pages, and usually 4-1/2 pages, of ordinary paper, for 32 cents. If you have random stuff, you could check to see if it weighs less than 9 new zinc pennies (pennies made after 1985)—and THAT would be less than an ounce. Okay? If you are not sure, make up an envelope full of stuff that ought to weigh 1 ounce, and take it to the post office, and fiddle with the contents until it weighs just 1 ounce. You can keep that letter and use it for calibration. Calibrate your scales, and save money.

By the way, how often do you readers write a small P.S. on a small piece of paper, fold it over 3 or 4 folds, until it is like a popsicle stick, and poke it into an envelope, along the glued joints, after you sealed it? I do it often.

MOISTURE DETECTOR—Recently, I had a drain clog, and my dishwasher started to pump water all over my kitchen counter and onto the floor. After I cleaned up the puddles, I decided to make a moisture detector. I just happened to have around a couple of 12-V "buzzers." I tried four different buzzers from Radio Shack: Models 273-055A, 273-059, 273-029, and 273-060. These are all adequate as a beeper or noisemaker, and all are priced reasonably at $2 or $3.

I took an old 9-V battery and some old copper-clad, and a little 9-V connector (that I'd salvaged out of the top of a dead 9-V battery). I got an old piece of wire (that was the handle for a Chinese-food take-out basket). I decided that was JUST stiff enough and just springy enough. I soldered it to two separated areas on the copper-clad (where I had sawed with an old hacksaw to insulate them.) The two separate wires I adjusted so they would short out, with normal spring-loading. Then, I put an aspirin between the two wires. When the aspirin got wet, the wires would short out—and the buzzer would buzz.

My wife said, "The aspirin won't dissolve. It takes stomach acid to dissolve the aspirin." I told her I had always heard aspirin was good for this. So I put some water on the aspirin as a test. Sure enough—the beeper began to beep—in about 10 minutes. Not the response time I wanted!! But a half of a sugar cube works better and faster. It collapses and fires in about 6 seconds. The only minor problem I foresee is if I put this in a cellar to detect for leaks, would dampness cause any false alarms? Or would the ants attack it? Still, it could help warn about problems with leaks or water-heater failures.

Thermocouples?? A friend asked me what kind of a thermocouple puts out 700 mV, like the one in his gas stove? Well, we know that does not run at 2000°C, so it must not be any ordinary 40 mV/°C thermocouple. It probably uses some kind of exotic metals or semiconductors. Can anybody tell me what is used in them? I'd like to know. It is obviously not a thermopile or stack of several thermocouples.

SMOG, revisited—I got a lot of information (and a lot of disinformation) about smog. To this day, the State of California (and almost all the "media") do not give a fair or correct definition of what a "gross polluter" is, nor do they talk about the best way to avoid having your car declared one—by getting it adjusted and repaired and tested before you take it in for the official tests. SIGH.

We DID learn that the State of California cannot confiscate your car just because it does not pass a smog test. HOWEVER, in some cities, such as San Jose or Santa Cruz, they have "abatement programs." The local "abatement officer" can "abate" your car right out of your driveway, if it is not registered, and can scrap it, and take the $3000 abatement charge, and sell it to some darned refinery or other polluter. Wonderful....

(Heyyyyy — a car that is not registered, and is not on the road, does not emit any smog at all. How come they think they are cutting smog by junking such a car????)

TIP: If you buy some shoes, and you like them, go back and buy another pair before you forget. If you wear the two pairs of shoes alternately, they will last longer. And, that way, you won't be cross to discover, when they are worn out, that you cannot buy them any more!

A friend brought in his new GPS receiver, and we went out in the parking lot to compare his data to mine. Sometimes his receiver made a 50-meter lurch—and mine didn't. Sometimes my receiver made a 40-meter jump—and his didn't! I surely was not expecting that. I had expected many of the lurches to be the same! Still, both gave acceptable accuracy. (Even if they did both list the elevation at -100 or -180 feet.) And they'll get better in a couple years when the Air Force turns off the noise. But, beware, as it is claimed that some older GPS receivers may experience "millennium" problems as early as August 1999.

BOOK REPORT—I got a good book in the mail yesterday, written by Jim Smith (of Cambridge Management Sciences, formerly The Phoenix Group) about Quality. He debunks a lot of myths about Quality. He explains why certain "Quality" procedures are good, and others are bogus. I am recommending to our librarian to get a copy or two. It is a big book, 520 pages: "Optimizing Quality in Electronics Assembly: A Heretical Approach," by James Smith and Frank Whitehall (McGraw Hill). Costs about $50.

I haven't found too many places that deal with the design of ICs, as it mostly deals with the assembly of ICs. So it should be of great value to our customers, and to us, when we are communicating with our customers.

Good writing. He points out the flaws in "the customer is always right, so give the customer the Quality he asks for." He DISMANTLES the Six-Sigma theories. He sorta likes a lot of the ideas of Juran, Deming, Crosby, and Feigenbaum—but not ALL of their ideas. I was pleased to see that he likes about ALL the ideas of Eliyahu Goldratt (The Goal). Conversely, he does mention Genichi Taguchi and his methods. He then prints some nice excerpts of how I was able to show that Taguchi's "optimal" regulator was not regulating at all (Electronic Design, June 10, 1993, p.85). Anybody who is that realistic about Taguchi, and who is VERY SKEPTICAL of ISO 9000, I gotta like. A lotta good stuff in there. I have not finished reading all of it, but I've read half and skimmed the rest.

All for now. / Comments invited!
RAP / Robert A. Pease / Engineer

Mail Stop D2597A
National Semiconductor
P.O. Box 58090
Santa Clara, CA 95052-8090

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.