Midnight Engineering

November-December, 1990

RESOURCE BIN

number thirteen
Perils and pitfalls of patents and patenting.
ur usual reminder here that the Resource Bin is now a two-way column. You can get tech help, consultant referrals and off-the-wall networking on nearly any electronic, tinaja questing, personal publishing, money machine, or computer topic by calling me at (602) 428-4073 weekdays 8-5 MST. Ive got a free pair of insider secret resources brochures waiting for you when you call or write. This month, I thought wed take a slightly different tack. Instead of my showing you lots of great places to get stuff, I will be showing you the one resource that you should studiously avoid at all costs. Because it is certain to waste your time, energy, money, and sanity. The term mark first came from the carnival midway. Any time a scam operator (the rube in carneyspeak) had significantly lightened a prospects wallet, he would give him a friendly exiting pat on the back. Along with a supporting "Gee Fella, thats too bad." Unmentioned and unbeknownest to the lightenee was the fact that the rube had secretly dipped his hand in a hidden stash of powdered chalk just before the pat on the back. And thus marking a large "X" on the lightenee, clearly identifying him as worthy of special treatment by the next rube on down the line. Eventually, every non-rube who so much as entered the carnival midway area became known as a mark. And were contemptuously treated as such. These days, we no longer have too many marks left. So, you substitute the term inventor instead. Any time an "inventor" context crops up, you are assured of an uneven playing field very much comparable to a carnival midway or a casino floor. A scene which is intended primarily to (A) liberate as much money as possible from the mark, and (B) to keep the status quo exactly where it is.
The foremost reason to studiously avoid any "inventor" context is the totally absurd popular mythology which now surrounds patents and inventing. Nearly all of which is dead wrong. To prove this to yourself, just mention the word "patent" at any party and then observe the ludicrous disinformation heaped upon you. Then challenge them to name one individual anywhere, ever, whom they personally know that, in a small scale context, has shown a net positive cash flow from their patent involvement. A cash flow that was worth the time and effort involved. No, the windshield wiper guy has not collected yet. The Sears wrench dude has wasted his entire lifetime by tilting at windmills. To me, Hyatt looks like a rube. Tesla died a pauper. The patent system drove Armstrong to suicide. And Edison was a ripoff artist who made most of his bag by simple theft, using the most ruthless gaggle of renegade patent attorneys ever assembled anywhere. So much for urban lore. Now, patents might or might not retain at least a marginal utility in a Fortune 500 context. Our concern here

September-October, 1994 Midnight Engineering 29.2

Tech Musings

June, 1997
he folks at IBM have added a new patent repository to the web. With a free searchable master file for all patents in all fields newer than 1971. Plus a $3 per patent hard copy fax service. You can access this site from the Patent Avoidance Library Shelf page of my www.tinaja.com Or reach it directly at http://patent.womplex.ibm.com This new service certainly is fast, convenient, and scads of fun to play with. Text and complete figures are included. As are "forward looking" cross references. The site is not yet in Adobe Acrobat, so the figures remain somewhat grubby looking. And I feel their search engine seems to be a tad on the weak side. But do note that not one patent in 200 ever shows any net positive cash flow. Less than one patent in 1000 is ever "new" enough or "non-obvious" enough that it cannot get busted with a thorough enough search for prior art in obscure enough places. Thus, any patent repository might overwhelmingly end up providing a mind-numbing stash of incompetent failures and total losers. An amazing number of patents just plain do not work. Your effort is infinitely better spent studying the trade journals and all of the web sites where the winners consistently appear. Ferinstance, I was two for two on my first visit here. I looked into two patents and now have a pair of superb new candidates for my patent horror story collection. I first searched on "ac phase control" to find out where it would end up leading me. The first patent from 1992 looked vaguely familiar. Sure enough, I had published the exact same waveforms back in the September 1969 Popular Electronics. On page 30. While studiously ignoring obvious high profile public domain prior art, they did, of course, manage to make their design far more complex and way more expensive than necessary. The second patentocity was more recent. And even sadder. They started with a 1938 construction project you

% Consulting services available on concepts shown. % Two-way recordable comm is **REQUIRED** for these utilities. % Routines excerpted from FINDRMS.PS on www.tinaja.com. /scaleamp sqrt mul def /startang 0 def /stopang 180 def /endang 180 def /res 0.1 def % use a 117 volt ac cycle % set limits % % % step resolution in degrees
% Assume a ONE OHM load. Make a normalized array holding the desired % waveform values. This routine optimizes for ac phase control. % Any values array normalized to peak = 1 can be substituted. /makewaveform {/waveform mark 0 res endang {/ang exch store ang startang gt ang stopang lt and {ang sin} {0} ifelse} for] def } def % do the rms and average calculations on the waveform array. /findrms1 { 0 waveform {add } forall waveform length div /normaverage exch store 0 waveform {dup mul add} forall waveform length div sqrt /normrms exch store normrms normaverage dup 0 eq {pop 0.000001} if div /normratio exch store} def % report the results. /crlf true def % IBM or sanity? /return {(r) print crlf {(n) print} if} def /reportrms {return return (The average normalized waveform value is ) print normaverage 20 string cvs print return return (The rms normalized waveform value is ) print normrms 20 string cvs print (.) print return (The ratio of rms to average is ) print normratio 20 string cvs print (.) print return return (The average scaled waveform value is ) print normaverage scaleamp mul 20 string cvs print (.) print return return (The rms scaled waveform value is ) print normrms scaleamp mul 20 string cvs print (.) print return return return} def /findrms {findrms1 reportrms} def % convenience linker % ========== % % demo - remove or alter before reuse. ==========
(A) Plot the rms voltage versus phase angle for a triac that conducts only a portion of each ac half cycle % % % % % 0.1 degree accuracy start at turnon angle go to the end of half cycle stop at one half ac cycle use a 117 volt ac cycle
/res 0.1 def /startang 0 def /stopang 180 def /endang 180 def /scaleamp sqrt mul def
180 { /startang exch store (The phase angle is ) print startang 20 string cvs print ( degrees.) print return makewaveform findrms} for flush % should return these edited results. %% %% %% %% %% %% The phase angle is 0 degrees. The rms voltage is 117.0. The ratio of rms to average is 1.11072. The phase angle is 10 degrees. The rms voltage is 116.936. The ratio of rms to average is 1.11852. << more stuff here >> %% %% %% %% %% %% The phase angle is 170 degrees. The rms voltage is 3.94639. The ratio of rms to average is 4.88476. The phase angle is 180 degrees. The rms voltage is 0.0. The ratio of rms to average is undefined.

Modeling materials

Your local hobby shop probably has smaller quantities of most building stock. I have found it better to go to the actual sources for wider variety and far better prices. For aluminum, brass, and other metal sheets, rods, and tubes, try K & S Engineering. For styrene sheet stock and similar plastic items, use Evergreen Scale Models. And for "lumber" precisely cut in all of the train gauge and scale dollhouse sizes, Northeastern Scale Models is it. For the larger pieces of flat display, exhibit and modeling materials, your best source is Fomeboards. These folks are strong in foam core plastics and similar base materials for architectural mockups, fancier point-of-purchase signs, trade show panels, and such. Several highly unusual decorative sheeting materials are now sold by Coburn. These can include prismatics, holographics, foils, glow-in-the-darks, metallics, and lots of other stunningly attractive display materials. One of my favorite sources for the traditional art supplies is Dick Blick, while the Polyline people are big on cases, labels, and packages for audio cassettes and VCR video cases. I use Polyline cases for my Introduction to PostScript videos. They also stock hard-to-find VHS spine labels.

LASERWRITER SECRETS

The name is short for Educational Lumber Company. These folks are into exotic hardwoods in a very big way. Especially the weird, the beautiful, or the unusual. Nothing like a piece of wenge or cocobolo to liven up a small electronic enclosure. A free catalog is offered.
The reprints from all Dons Midnight Engineering columns. Includes the case against patents, book on demand publishing, toner secrets, paradigm stalking, insider research, lots more. $18.50

Outwater Plastics

This outfit believes they are in the display fixtures business. They have a wildly mind-boggling assortment of low cost and potentially quite useful hardware for you electronic hackers. Plus all sorts of ways of hanging and showing things. They even now offer Grecian urns for writing odes on. Once again, a fat and free catalog is offered. This one is a real page turner, chock full of "use me" stuff. Stuff that simply cannot be ignored.

Model Railroader

Box 1111 Placentia, CA 92670 (714) 632-7721 An all-ads mail order shopper specifically for hardware hackers, ham radio operators, CB folks, computer users, and satellite pirates. Their low-price ads are attractive for most shoestring technical startups.

PaperPlus

1027 North 7th Street Milwaukee, WI 53233 (414) 272-2060 Besides unusual tools and techniques, this hobby magazine has far and away the finest technical writing and technical illustration of any publication anywhere ever. Use it as a style and layout manual, and hope to someday be able to communicate that well. Should be required reading for any tech writer.

Motion Magazine

300 Oceangate #800 Long Beach, CA 90802 (800) 272-7377 If youve ever tried buying paper from an old line source, you know the hassles. Instead, try these walk-in paper supermarkets now in most states. Especially useful for book-on-demand publishers. Also stocks certificates, bumper sticker stock, acetates and polyesters.
Box 6430 Orange, CA 92613 (714) 974-0200 Free trade journal that covers steppers, servo motors, linear actuators, the power control semiconductors, and general robotics stuff. Pricey products but full of good technical ideas and resources.

Mouser Electronics

2472 Eastman Avenue Ventura, CA 93003 (805) 658-0933 Used to be called Power Conversion and Intelligent Motion. Another free trade journal for the robotics crowd. Covers steppers, servos, motors, linear actuators, and their electronic control components.

Player Piano Company

11433 Woodside Avenue Santee, CA 92071 (800) 346-6873 Electronic distributor with low minimums, low pricing, and extensive stock. Very hacker friendly. Carries semiconductors, ics, relays, resistors, capacitors, inductors, hardware, and all the usual goodies. Largely imports.
Northeastern Scale Models
704 East Douglas Wichita, KS 67202 (316) 263-3241 Well, just because it is there, I guess. Unusual source for very unusual tools, materials, and techniques. Has hobby robotics potential, especially for low pressure pneumatics.

Printers Shopper

PO Drawer 1056 Chula Vista, CA 92012 (800) 854-2911 Not really a shopper, but a monthly mail-order catalog for a major printing equipment tools, materials, inks, and supplies house. Many hundreds of items listed. Their prices are usually better than buying locally.

toner to the pc board. Or else a laser printer modified to print directly onto 1/16th inch copper clad. Ive found that a few seconds of pre-etch helps bunches, as does preheating the board so it does not act as a giant heat sink. A post-transfer bake also helps. Trying to use an ordinary iron is an outright joke. I currently use a Kapton film from Dupont that Ive coated with a high temperature mold release from Miller-Stephenson. A commercial toner transfer product called Meadowlake works for some people some of the time. Fake Kroy Color machines and toners are found at Lazer Products. Two other toner sources are Black Lightning and Don Thompson. Two fine trade journals on printed circuits are Circuits Manufacturing and Electronic Packaging and Production, while your best hacker source for pc boards and etchants is Kepro. A low price, low end printed circuit layout package is included in my PostScript Show and Tell from Synergetics.
EMERGING TECHNOLOGIES NAMES AND NUMBERS
Adobe PostScript 1585 Charleston Road Mountain View, CA 94039 (415) 961-4400 Black Lightning RR 1-87 Depot Road Hartland, VT 05048 (800) BLACK99 C & H Sales Box 5356 Pasadena, CA 91107 (800) 325-9465 Carter Carburetor 9666 Olive Road St. Louis, MO 63132 (314) 997-7400 Clippard Minimatic 7390 Colerain Road Cincinatti, OH 45239 (513) 521-4261 Dialog Information Service 3460 Hillview Avenue Palo Alto, CA 94304 (415) 858-2700 DTM Systems 1611 Headway Circle, B2 Austin, TX 78754 (512) 339-2922 Dupont Kapton 1007 Market Street Wilmington, DE 19898 (302) 774-1000 Edmund Scientific 101 East Gloucester Pike Barrington, NJ 08007 (609) 573-6250 Exair 1250 Century Circle North Cincinnati, OH 45246 (513) 671-3322 Flow International 21440 68th Avenue South Kent, WA 98032 (206) 872-4900 Haskell 100 East Graham Place Burbank, CA 91502 (818) 843-4000 Hygenic Manufacturing 1245 Home Avenue Akron, OH 44310 (216) 633-8460 Jerryco 601 Linden Place Evanston, IL 60202 (312) 475-8440 Kepro 630 Axminister Drive Fenton, MO 63026 (314) 343-1630 Kroy Sign Systems 14555 North Hayden Road Scottsdale, AZ 85260 (800) 521-4997 Lazer Products 12741 East Caley #130 Englewood, CO 80155 (303) 792-5277 MasterCAM 2101 Jericho Turnpike New Hyde Park, NY 11040 (516) 328-3970 Meadowlake 25 Blanchard Drive Northport, NY 11768 (516) 757-3385 Meredith Instrument 6401 North 59th Avenue Glendale, AZ 85301 (602) 934-9387 Miller-Stephenson George Washington Hwy Danbury, CT 06810 (203) 743-4447 MWK Industries 1440 S. College Blvd #3B Anaheim, CA 92806 (800) 356-7714 Phillips 2001 W Blue Heron Blvd Riveria Beach, FL 33404 (407) 881-3200 Player Piano Co 704 East Douglas Wichita, KS 67202 (316) 263-3241 Roland Digital 7200 Dominion Circle Los Angeles, CA 90040 (213) 685-5141 SGS-Thompson 1000 East Bell Road Phoenix, AZ 85022 (602) 867-6259 Sharp Sharp Plaza Mahwah, NJ 07430 (201) 529-8757 Sprague 70 Pembroke Road Concord, NH 03301 (603) 224-1961 Synergetics Box 809 Thatcher, AZ 85552 (602) 428-4073 Technical Insights PO Box 1304 Fort Lee, NJ 07024 (201) 568-4744 Don Thompson 23072 Lake Center #100 El Toro, CA 92630 (714) 855-3838 3-D Systems 26081 Avenue Hall Valencia, CA 91355 (805) 295-5600 Toshiba 1220 Midas Way Sunnyvale, CA 94086 (800) 321-1718 Value Plastics 3350 Eastbrook Drive Fort Collins, CO 80525 (303) 233-8306 Vortec 10125 Carver Road Cincinnati, OH 45242 (800) 441-7475 Whole Earth Review 27 Gate Five Road Sausalito, CA 94964 (415) 332-1716

Emerging Opportunities III
t sure is rewarding for me to see you other Midnight Engineers picking up on and successfully going with some of our previous emerging opportunities. All in your own small scale home-based Money Machines. Several examples here include John Rees who offers a great video on converting car alternators into power stepper motors. And Martin Carbone whose new desktop finishing products include a pair of very low cost scoring machines for boxmaking and bookbinding. Or Frank Miller who has bunches of useful direct toner printed circuit products. Or Kevin Bennet with his easy to do "raised print" laser thermography. That uses nothing but a small desk lamp. Or Stan Griffiths and his fine new book on recycling Tektronix classic oscilloscopes. Or Kirk McLoren who has a new Micro Cogeneration book that shows you how homemade power can actually end up cheaper than utility power. Lets return to the scene of the crime. Heres what I see as the current crop of emerging opportunities. Along with several GEnie PSRT filenames you could go to for more details. Stuff that suddenly has become cheap enough and real enough, yet remains fuzzy enough and undeveloped enough for superb Midnight Engineering potential
All you really have here is an air core transformer. With the core being the distance between your chest and your wrist or handlebars. Plain old near field inductive coupling is all you require for effective comm. But wait. What do we really have here? We have a tiny, lightweight, sealed and waterproof transmitter. With a one year or longer life from its internal lithium cell. That can handle a data rate of zero to 200 Hertz or so. At a retail list price of $22, far less in quantity. Providing a signal that is handily received by a coil and an op-amp or two. Largely unidirectional, except for deep axis nulls. Two leading brands of these devices are Polar and Vetta. More details on their internal workings in HACK68.PS. By the way, a dental X-ray is a dandy way to reverse engineer sealed modules of this type. One big new use I see for short haul telemetry

ISOPOD ENERGY REPORTER

Short Haul Telemetry Micropower radio and infrared transmitters have gotten super small and very cheap. To the point where they can be used for all sorts of data comm over ranges of, say, four to six feet. There are a lot of new possibilities here. I like to call the sum total of these devices short haul telemetry. Ferinstance, there are all kinds of new uses for ordinary TV remote controls. There are antishoplifting tags. And implanted animal monitors. And schemes to get data on or off a rotating shaft. Inventory controls. Security systems. Intelligent data tags. New wireless mice and modems. Car locks. 3-D position sensors. Attitude detectors. A brand new trade journal that addresses these devices is Wireless Design and Development. One low cost and grossly underutilized short haul system is called an EKG heart monitor. This is normally used to optimize aerobic excercise sessions. You have a strap that wraps around your chest. The strap picks up your electrical heartbeats and converts them to transmitted 36 cycle bursts of 5 kHz rf energy. These low frequency waves are then picked up using a nearby wristwatch or bicycle mounted computer display. The big advantage is that they perform reliably during strenuous exercise. Cheap finger or ear-clip infrared units do not. Look Ma, no wires.

May-June, 1997 Midnight Engineering

44.-21

December, 1995
just got a helpline call from an "inventor" trying to "protect" a "new" auto headlight idea. To stop "Detroit" from stealing it. Ive never heard of "Detroit" ever paying any outsider for any untested, undeveloped, or unproven idea. Instead, "Detroit" buys parts from suppliers and bolts them together to make cars. They are in the process of outsourcing much of their product engineering. They are significantly reducing their number of suppliers. And holding them to the tightest of razor thin margins. Uh, strike one. Illumination engineering is one of the very few things that Fortune 500 companies happen to do very well. A multi-skill project team approach is usually required, combined with ray tracing computers, arcane production engineering, and outstanding access to the worlds research base. So, those big boys clearly have an unbeatable home turf advantage here. For strike two. Your really big issue on all future headlights is efficiency. Because of downsizing in general and electric or hybrid cars in particular. Anything less than 100 Lumens per Watt wont hack it. You can bet that tomorrows headlights will most definitely not be based on a heated filament. I got the impression the caller was not a member of the SAE. Nor the IESNA. Nor did he seem to be at all into trade journals or online literacy. He seemed to feel that car headlamp efficiency was "not important." And apparently did not have the slightest idea how woefully inefficient his new design was. For a self-inflicted swing and a miss for strike three.
Superb avaition history book A new RGB to NTSC encoder Lamps and lighting efficiency Emerging ultra-fast computers Product development concepts
But be sure to remember the key insider secret rule for all successful new product development: They must come to you. And never vice versa. Do note that you are not selling an idea. Ideas are worth ten cents a bale in ten bale lots. You are instead now offering a proven, in-demand, and a ready to manufacture product. here you have already completed most of the high risk steps. More on becoming a purveyor of risk reduction in RISKDOWN.PDF on www.tinaja.com And much more in general on idea development in my Blatant Opportunist and my Case Against Patents packages.
Or one that long ago fell off the shelf because of inherent problems. Step two is to ask yourself "Who is it that (A) likes bright headlights, and (B) has their own wallet in their own back pocket? Well, out here on my sand dune, the answer is glaringly obvious: 4WD desert off-roaders. To these folks, a "map" light is 50,000 candelpower. And a "running" light can vaporize troublesome boulders at 75 paces. On low beam. So, firstoff, you would have a few four wheelers critique your design. If it is any good, you then let the local 4WD club beta test it. Once you have your tested and proven product well received, you sell a few at regional meets. Then you publish it in all the offroad mags. Next you seek out one or more of those off road lighting outfits. K.C. Manufacturing is but one of the name brand biggies out here. Competitors include Dick Cepek, Hella, Explorer, and Piaa corp.

have been the traditional method of doing panels for limited and small volume production. Because of the front end expense and time cutting a screen, these work best for a dozen or more identical panels. Three good sources for silk screen materials and supplies include Dick Blick, Advance Process, and Southern Sign Supply. And, of course, your leading source for all printed circuit materials and supplies is Kepro. While youll still find traditional electronic decals offered by some old-line sources, these simply arent worth the time and effort. These are basically a sucker bet guaranteed to give second-rate results that range from unprofessional to attrocious. For totally superb panel artwork, consider linking PostScript to those old annodized aluminum dialplate systems offered by MetalPhoto or Fotofoil. These are somewhat pricey but are ideal for rugged and durable one-up or small quantity panels. They are also useful for museum signs and electronic relay rack panels. Picture an aluminum sheet which has only partially gone through the annodizing process, leaving a brightly colored but a very open and spongy surface. A photo emulsion is then applied. You later expose the emulsion through your custom PostScript artwork. A contact printer or an enlarger can be used. Followed up by a traditional darkroom slopping-in-the-slush. After developing, you can boil the panel in a magic glop that reseals the surface, closing a sapphire (literally!) hard surface and locking your image inside the panel. The results are quite durable. It takes a highly dedicated vandal to harm a Metalphoto panel. While the lettering and images are normally black, a wide range of bright colors are offered. Also the "plain old gray" of traditional annodizing. You might use a reverse technique to give you a black panel having colored or gray lettering. Again, this is utterly trivial with PostScript. There is also a slightly cheaper self-stick vinyl based system. This one used to be called ScotchCal, but has been renamed Dynamark. Picture a white self-stick vinyl with a colored photoglop on it. You contact print the vinyl using strong sunlight or some other u-v source, again through your PostScript artwork. Where present, the light hardens the photoglop against chemical attack. You then use a Webril wipe or other
nonwoven pad to apply a chemical that removes the color from all areas which were not photohardened. The result is white over a color or vice versa, depending on whether youve used normal or reverse artwork. A clear epoxy overcoat gives you reasonable scratch resistance. You can then peel and stick your vinyl over your aluminum or other panel. Once again, if you are very careful, the vinyl can also be used as a punching and drilling guide. Kits are readily available, both in single and assorted colors. Very thin aluminum versions are also offered. There is also a non-stick product called Scotch Color Key offered to the printing industry. This gives you a mylar film having color selectively photoapplied to it. Many dozens of colors are offered. There are lots of prototyping opportunities here.

info and other ads for materials with robotics potential. As do pubs from the SAE library, formerly the Society of Automotive Engineers. For lots of sensor and transducer info, check into Sensors magazine, and Measurement & Control. For the motion control info, your best trade journals include PCIM, Motion, Motion Control, and MotorTechniques. There are a number of vocational education magazines that may get into robotic topics. Typical examples include both School Shop and Industrial Education. There are a dozen more. And dont forget Model Railroader. Their ads do offer a great selection of unusual tools and materials. These folks have been in the robotic business for years. They just do not seem to want to admit it.

Robotic opportunities

OK, so what can be done in hobby robotics today? With the possibility of a reasonable cash return for your time and effort? I see several areas where original thought might come up with several long term robotic solutions.
low pressure pneumatics We saw
Magazines and trade journals
There are a dozen or so robotics magazines. The SRS Encoder is a new labor-of-love newsletter, published by the Seattle Robotics Society, a leading amateur robotics group. There is also a Robotics Experimenter magazine. The rest are industry trade journals or scholarly publications. Ive tried to list most of them in our Names & Numbers sidebar. I was unable to review all of these by column deadline time, so do let me know which ones you find useful. Electronics Now runs an occasional robotic project or tutorial. Many trade journals focus on some other field, but may happen to have great robotics info in them. The two best mechanical trade journals are Machine Design and Design News. The largest industrial supply throwaway mags are New Equipment Digest and Industrial Product Bulletin. Appliance, the Appliance New Product Digest, and the Appliance Manufacturer trade journals sometimes have motor
back in Resource Bin #7 and in my Resource Bin Reprints available from Synergetics how you could buy a low pressure 3-way air valve for only a quarter each. Low pressure air has yet to take off but its got outstanding robotics potential. Aquarium pumps can be used as compressors, and your actuators can be nothing but a balloon, bellows, or some rolling diaphram. You can get much more force much more linearly than you could ever hope to with a solenoid or another electronic solution.

Sensor Trade Journals

As I may have mentioned a time or two before, those free trade journals are the best way to get informed in a big hurry on almost any subject. Start with Ulrichs Periodicals Dictionary on your librarys reference shelf. This gem lists some 150,000 trade journals and other magazines. At any rate, your horses mouth trade journal is Sensors. From Carl Helmers of early Byte magazine fame. Also try Measurement & Control. You can also go to industry specific mags for all sorts of useful stuff. Say Pollution Equipment News for typical environmental sensors, Powder & Bulk

Sensors and Sensing

This month, I thought wed take a look at sensors and sensing. A sensor is any device that converts some other physical attribute into an electrical or electronic signal. Sensors of one sort or another are involved in just about any electronic project. And they sure are one hot topic on our helpline and on PSRT. Usually, there will be two stages involved in any sensor problem. First, youll have to do your actual sensing. This gets done using a transducer of some type. The result is often a very small signal, possibly a few tens of millivolts. Noise is always a problem in any sensing situation. And extreme caution is required to take care of this very weak signal. Next, you will typically have to do some signal conditioning to convert the sensed signal into something useful. Such as a higher current, a pulse train, or a digital word. Signal conditioning tricks include shielding, differential mode sensing, isolation, offsetting, amplifying, temp
NEXT MONTH: Don looks at high frequency techniques and hackable resources.
Solids for level controls, or American Laboratory for chemical sensors. Or even Weight Engineering for coverage on strain gauges.

One Stop Shopping

Your highest profile source for just about any sensor is Omega. These folks offer scads of impressive free catalogs on temperature, level, flow, pH, strain, pressure, data acquisition, and bunches more. Pricey, though. One shirtsleeves source for nearly any industrial sensing instrument or tool is Abbeon Cal.
February 1995 / Nuts & Volts Magazine
Devices sells a TMP01 programmable controller. Range of both devices is -55 to +125 C. One obvious use is as a hot tub controller.

Pressure

Silicon pressure sensors are rapidly becoming low cost commodities. These are basically a "drumhead" etched into bulk silicon. Strain sensors are implanted on the drumhead. As the pressure changes, the drumhead flexes, causing a resistance change. There are two main pressure sensor types. The absolute sensor compares pressure against a perfect vacuum. A differential sensor instead compares a pressure difference thats between two ports. If one of those ports is left open to your ambient air, then you have a variant called a gauge sensor. Temperature compensation and a lot of amplification are often needed when sensing pressure. Some sensors are offered both as raw chips or with signal conditioning and temperature compensation built in. Your two most obvious sources for the pressure sensors are Motorola and Microswitch. But the real action comes down from Sensym, I.C. Sensors, and Novasensor. Sensym has a Solid State Pressure Sensors handbook. This one includes a great slide chart. An oddball use for pressure sensors is in low pressure tire alarm systems. Fleet Specialties is one source. Theres also some purely resistive approaches to pressure sensing. One source is Interlink Electronics. With mice and music apps. For ultra cheap, you can sometimes get by using nothing but the black anti-static foam that chips arrive in. But repeatability and reliability can be big problems here. At the high end, Force Imaging sells subminiature and quite thin sensors that work from 1 to 20,000 PSI. A modification to a pressure sensor will let it measure acceleration. The motion of a mass is sensed. The rate of change of motion of the mass is the mass velocity. The rate of change of the velocity is the acceleration. Except for airbag sensors, these devices are still very expensive. One source is Silicon Designs.

around ten percent. Thus, simply to break even, your venture has got to generate more than $100 per year and do so forever. But your tools and materials wont last forever. If they last for ten years, then your venture must generate more than $270 per year to amortize your initial expense and the time value of money over the effective life. With a five year lifetime, your venture must generate more than $329 per year. Just to break even! If your total cost is less than the returns, then you have an economic loss. The total costs must include all parts, labor, and your time value of money. Plus great heaping bunches of intangibles. Not to mention taxes and inflation. Paying cash does not make much difference. Since theres other things you might be doing with the money that gives better returns. Sadly, any "hacker economy" that
Winegard CA-6060 Yagi, fixed roof mount
Radio Shack 15-1117 coax inline 10 db amplifier Solid earth ground! Lightning / static block 75 coax roof drop Line splitter / two set coupler Denon TU-650-RD FM RBDS tuner Existing cable system Other stereo reveiver

Some Details

Say you want to start a technical venture. You first borrow a thousand dollars for your tools and materials. Today, the actual time value of money of those dollars will be somewhere
Fig. 1 MY TEST SETUP for ultra-fringe FM and RBDS.

Wavelets Update

Theres sure been a lot of interest in wavelets recently. These can be a super performing replacement for the older Fourier analysis techniques that relate time and frequency. Important usage areas include everything from video compression schemes to human vision to seismography. Advanced math is required. The field is maturing and there are now dozens of books available. Ive listed several of the more popular of them in our resource sidebar. The best tutorial paper is probably by Rioul and Vetterli in IEEE Signal Processing Magazine, vol 8, #4, Oct 1991 pp 14-38. Also check that Dec 1993 issue of their IEEE Transactions on Signal Processing. Note that these are two different pubs. Yes, we have wavelet shareware up on www.tinaja.com
Code: 0,0,1,0,1,1,1,1,1,1,1,1,0,1,0,0,0,-1,0,-1,-1,-1,-1,-1,-1,-1,-1,0,-1,0 Shorthand Code: 0x2,1x1,0x1,8x1,0x1,1x1,0x3,-1x1,0x1,-1x8,0x1,-1x1,0x1 The fundamental peak amplitude is 1.0545 times the "1" amplitude. Fundamental: (1.0000) 2nd Harmonic: 0 3rd Harmonic: 0 4th Harmonic: 0 5th Harmonic: 0 6th Harmonic: 0 7th Harmonic: 0.1184 8th Harmonic: 0 9th Harmonic: 0 10th Harmonic: 0 11th Harmonic: 0.2230 12th Harmonic: 0 13th Harmonic: 14th Harmonic: 15th Harmonic: 16th Harmonic: 17th Harmonic: 18th Harmonic: 19th Harmonic: 20th Harmonic: 0.0.0.1301 0

Fig. 3 AN "EVEN MORE MAGIC" power sinewave waveform of 210 bits.
the filter only makes a difference on a few stations. But what a difference. Does it make sense to add a lower quality booster at the antenna ahead of an ultra hot receiver? A modest 10 db boost at your antenna gives your antenna a fixed impedance to work into and makes up for both line and distribution losses. The slightly higher signal level is enough to override the "no stereo on weak signal" feature. Which lets you switch by yourself rather than using the factory preset. Naturally, mono is cleaner for marginal signals. But any preamp also adds to your cross mod and can cause overloading. Which was only observable at half a
channel away from a nearby station. If you use a preamp, be sure to use an inline coax version which is totally shielded. Providing you with only the bare minimum gain needed. Does a fixed antenna make sense? Fixed is cheaper and more rugged. And theres not too much east of me because of this slight rise that some folks call the Continental Divide. But Bee does not get her classic Tucson KUAT very well. Despite its strong reception on the car radio in the driveway. Some sniffing around with an unboosted cheap fringe-butnot-ultra Radio Shack 15-1636 FM antenna led to a big surprise. A large number of FM stations appear to now use vertical or circular polarizations. Their apparent aim is to improve nearby auto reception at the cost of distant coverage. For KUAT, flipping the antenna to the vertical dramatically improved the reception in Thatcher. The ultimate solution is antenna height. Which conquers all. At the top of the two mile high mountain in my front yard, I can take a $4 pocket FM receiver and tune it to 93.3. By pointing the whip antenna in one of three directions, I can get KDKB in Phoenix, KKOB in Alburquerque or a

WAVELET BOOK RESOURCES

C.K. Chui, An Introduction to Wavelets, Academic Press, 1992. C.K. Chui, Wavelets, World Scientific Pub, 1992. C.K. Chui, Wavelets: A Tutorial in Theory & Applications, Academic Press, 1992. C.K. Chui, L. Montefusco & L. Puccio, Wavelets, Theory, Algorithms, & Applications, Academic Press, 1994. J.M. Combes, Wavelets, Spr-Verlag, 1989. I. Daubechies, Ten Lectures on Wavelets, Soc Indus-Appl Math, 1992. G. David, Wavelets & Singular Integrals on Curves & Surfaces, Spr-Verlag, 1992. M. Farge, Wavelets, Fractals & Fourier Transforms, 1993. E. Foufoula-Georgiou, P. Kumar, Wavelets in Geophysics, Academic Press, 1994. Frazier, Wavelets, Mathematics & Applications, CRC Press, 1993. Jawerth, Practical Guide to Wavelets, CRC Press, 1994. G. Kaiser, A Friendly Guide to Wavelets, Birkhauser, 1993. T.H. Koornwinder, Wavelets: An Elementary Treatment of Theory & Application, World Scientific Pub, 1993. W. Light, Advances in Numerical Analysis, V2, OUP, 1992. Y. Meyer, Wavelets: Algorithms & Applications, Soc Indus-Appl Math, 1993. Y. Meyer, Wavelets, Cambridge University Press, 1993. Y. Meyer, Wavelets & Applications, Spr-Verlag, 1992. R.L. Motard & B Joseph, Wavelet Applications in Chemical Engineering, Kluwer Ac, 94. D.E. Newland, An Introduction to Random Vibrations, Spectral & Wavelet Analysis, Halsted Press, 1993. Ruskai, Wavelets & Their Applications, Jones & Bartlett, 1992. L.L. Schumaker & G. Webb, Recent Advances in Wavelet Analysis, Acad. Pr., 1993. G.G. Walter, Wavelets & Other Orthogonal Systems with Applications, CRC Press, 94. G. Wornell, Wavelet-Based Signal Processing with Fractals, P-H, 1994. R.K. Young, Wavelet Theory & Its Applications, Kluwer Ac, 1992.