A Tribute to Bob Pease Pease Porridge —5 (2002-2004) ‘work in an upcoming seminar “What's AN Tle Anslog St anyon fe oaie oi mast oticte onpease wt POPrid ge Wr rickrsins stein Semiconcsctor Cop, Santa Ciara, Cait What's All This Analog Computing Stuff, Anyhow? (Part 1) Researches back in 1961, when Philbrick was doing some of the advanced analog computing stutfin the industry. (Note: for you young kids ‘who weren't around then, fil explain ‘everything later) ‘A that time, Philbrick made analog, computing modules—adders,subtrac: tors, scalers, integrators, and analog smultipliers/dividers—and several Kinds 6f operational amplifiers, made ‘with vacuum tubes, hat ran on 4300 V de. We had fun and were profitable. (We also were expanding into solid- state amplifiers, but wll explain that another day) ‘What were analog computers good for? They were good ata lot ofcomput- ing tasks that modem digital comput- cess can now do, They also could simu- late functions chat Spice as learned to do. OF course, like any modern com- puter and as with Spice, they had defi nite limitations ifthe models were Jousy. In the old days, models were oversimplified—and Lousy. In the modern era, modeling is stil often lousy. No surprises there They could simulatea ot of mechan Jeal ystems and other kinds of physical systems such as thermal and hydraulic systems. That was the whole season for having them in the first place, (I'll expand on thisshordy.) in their day, these systems could simulate alot of megabytes per see- ‘ond—in terms of resolution of several variables. Even better, they could dis- play thei results on 2 big cathode-ray tube (CRT), so users could compare the outpur variables ‘What were the weaknesses? There ‘were dynamic ranges that were hard to ‘compute well with fll accuracy. (The upcoming digital systems were sub- stantially unlimited, in theory, for ‘Synamic range.) They were expensive 160 ERODES ny, 2002 ] first joined George A. Philbrick They were not very easy to maintain of calibrate. This is ue for most analog ‘computers in the industry. Moreover, they were difficult o program. You hhad to know a lot about the system, about voltages, and about scale fac- ors—to make it make sense when you ran the computer, (On agood day these analog comput ing systems were “panceleri." That is, they could run at high speeds, or at slow speeds. Unfortunately, the high- and low-speed results didn’t necessarily match. Tha's nota disaster, but it caus- es aconfidence problem. On a good day, you could turn one knob, or dial, and watch the outputs change until you {got an answer that you liked—or until you got an answer from which you ‘ould lear something. like that par. plan to talk about many aspects of analog computation during the next ‘yea especially operational amplifiers, which we now call “op amps.” But to run the operational amplifiers of 1960, which required #300 V dc, you need a good, stable power supply. I recently {gota good ald +300-V supply, an R-300 from 1968. 1 will talk about it first, and some of the very good features ofits design in the next column. To make a good voltage regulator, you need some good special-purpose op amps. Philbrick is no longer in business. ‘Teledyne acquired Philbrick about 30 years ago, and it the merged with Neus Research Labs (one of Philbrick’s spinoffs). Teledyne Philbrick Nesus ran for sever! years, then merged with Tele- dyne Components, formerly Amelco, in about 1985, Teledyne spun this off into Telcom. Most Philbrick products are now dead, but you can still buy a few fiom Telcom. Long live Philbick!—and the analog ‘computers—and analog concepts— land the op amps that it helped intro- duce to the world! More About Knots: Many people ‘commented on "What's All This Knot Stuff, Anyhow? (Part l)" (eugcrRowic DESIGN, Marck 5, 2001, p. 142). Several readers noted that I hadn't drawn a proper bowline knot, but rathera “Dutch bowline,” which supposedly has some disadvantages. Wel, I stand corrected. Several claimed to have never experienced bowlines coming untied, Irs true that with some kinds fof ropes, knots stay tied. But with oth- fer types or sizes of rope, knots—even good ones—tend to untie, especially ifthey can get slack. ‘The good old 1960 book, Moun- taineering—Freedom of the Hills (p. 117), specifically recommends that half-hitches be used to “back up a bowline...for bowlines are especially prone to working loose." This book has no ISBN number, but its Library of Congress number is 60-8907. It was published by The Mountaineers, Sezi- te, Wash. —RAP Embarrassing Typo Er: In the recent “Equations Sul (Part 3),"(sLecrnosic Desicn, Nou. 5; 2001 p. 94), typed out that the response of the filters Your i= Gai fis pCa(RT+RQ) « p'RICIR2C2] ‘Actually, the C1 in the mide term was typed wrong; it should be C2: Your Miu = Gai 1/1 + pCo(R1+R) + p>RICIRC2} All fornow. / Comments invited! RAP / Robert A. Pease / Engineer mapQ@galaxynsc-com—or- ‘Mail Stop D2597A National Semiconductor PO. Box 58090 Santa Clara, CA:95052-8090,ed pease as 20s Past cone 6 tom MT rridge eoreettarccael pol Semiconductor Corp, Santa Clara, Cal What's All This Ripple Rejection Stuff, Anyhow? (Part 1) needed Jow output ripple on a power supply. His 1000-V out ‘pucrequited low, submilivolt ripple. designed a couple of circuits for him, Details soon. To make sure that | was- nv missing any ticks, [looked up a voltage regulator the Philbrick 390, designed by my old colleague Bruce Seddon about 42 years ago. It was ‘optimized to provide #300 V de at 100 mA (sete figure), oF 300 mA ina larg: er 8-300 power supply when addition- 1 output tubes are paralleled ‘Anybody can design an operational amplifier (op amp) with a gain of 100,000 or 1,000,000 at de. Bu this ‘one needed to swing ts output 80 V de plus 15 V pp at 120 Hz, with asum- ‘mingpoint err of less than 150 uV ec ims. Tha’s 100 dB of gain at around 120 Hz! Forty years ago, most regulator amplifirshad wo-tage amplifies. Bat thisone wed thee dual tiodes foreach supply, «and ~ 300 V. Two honest stages of 12AX? provide a lotof gain (= 100), and the 12AU7 (j= 20) gives a good healthy drive to the grid ofthe ‘output tube (6AS7GA) ‘Additionally, che positive feedback of R132, R133, and Rit provides much more gain—even at 120 Ha, and atde. Although the output pass tube (6AS7GA) has a of only 2, this amp can easily drive the grid to any neces- sary voltage, whether no-load oF full: Jad, low-line or high-line. Also, ithas submillivolt gain ero (or lin, load, orrippe “There are good bypass capacitors, such as at C131 and C134, to filter and bypass the noises for frequencies above 412. That helps keep the outputs noise below 250 uV rms, This amplifier was also optimized for fast bandwidth. The ‘output bypass capacitors (not shown) ‘were 150 BF at 525 VDGW, specified with good low Re 70 LscrRAWC OES Ft, 202 R ecently, | helped a guy who me FARCE PHILBRICK SH0-< REGUUATER ~/146t ‘The 5910 dual regulator amplifier provide excellent ripple rejection, and was builtin a litte boxy subassembly. that old machine still cans well today ‘Thetubes stuck up above the box. with Soon, I'll show how to get low ripple the passive components mounted on voltage by adding an add-on circuit to turret terminals along the insides of 20 existing high-voltage supply. Per- the box. haps using a fast FET op amp running Tought a 30-year-old R-300 that on +6V. ‘was still in very good shape. 1useda I'm looking forward to using that Variac to tura up the line voltage very old R-300 to run a whole bunch of slowly to “form up" the electrolytcs. It tests on various old vacuum-tube seould be harmful to apply full line operational amplifiers. I've been power right away. I fired it up and It waiting to do that for years! We rarely regulated nicely, use vacuum tubes these days, but ‘Afier | dic.a general check-out itwas here's a good example of how tubes time to do noise testing. Tused a series could do some very good work. I'll stack of three 25-W light bulbs to draw have more comments on the ald art 190 mA of load, The ripple voltage on of designing with tubes, and the early the main ("upstream") filter capacitors days of operational ampiitiers, 40 oF rose to 9V p-p. The output ripple-plus- more years ago. noise increased from 100 uV shunt. Points A" and B" are well ballasted and the heat is well spread out, distributed over of the wires. This shunt will get pretty hot when you run it at 22 W, but not hot enough for the solder to melt. Will I have to test it at 144 A? No, I can just build and run 12. through it. The rated fusing current for 22-gauge wire is 4o A, so 12 A isn't bad, Now, we have a basic shunt design that can be made up in under half an hour for less than $1, and you can pick the wire gauge, and the lengths and numbers of conductors, to dissipate almost ANY amount of heat for any number of milliohms. Will the bus A-A' carry anycurrent? Not any significant current, as the symmetry forces the current to mostly flow past the wire. So if that circular bus were made of #24 wire and broken in several places, you wouldn't see any significant difference from 14 gauge. It's just to hold things steady and to help spread out the heat. Now,copper has a tempco of about +3300 ppm/°C. So if you dissipate even 3 W, this shunt will warm up and drift in its r value, perhaps +0.5%. What's a mother to do? More on this later. Don't throw out this design. Now, nichrome wire at 22 gauge has about 1 ©/ft with a low tempco. I could lash up 42 1/2 -in. long nichrome resistors, in parallel, to make a good 1-mQ, shunt. How can I lay out 42 1/2-in, wires? I placed a serpentine between two 1/2-in. by 3-in, strips of double-sided copper-clad, spaced 1/2 in. apart. I soldered up seven wires (using strong HCI flux). Then I arranged six pairs of strips in a hexagon, with heavy bus conductors to share the currents, as above. The actual resistors would be quite small, but the copper-clad would spread the heat out. ‘This compact hexagon would have 3-in. sides and be 6 in. across, but only a couple of inches long—compact, but not too small to spread out the heat. It might not be perfect in precision, but its low tempco (20 ppm/°C) would be better for many cases. Any shunt should be mounted in a box with some vent holes to let out the dissipation. ‘To calibrate these shunts, shove 1 A into the Force leads through a calibrated 1-0 resistor at 1 V, and look for 1000 uV at the Sense leads. Add a few extra conductors to get the resistance down to the right calibration, or snip out a few to get it up there. Next month, we'll use one of these millichm shunts in a wide-range watt-meter. All for now. / Comments invited! RAP / Robert A. Pease / Engineer or: Mail Stop D2597A. National Semiconductor P.O. Box 58090 Santa Clara, CA 95052-8090 “Shunts of various sizes can be obtained from many good vendors, such as ; (727) 541- 5771. A 2290.5 8 gaunt LAYOUT-@ 84) caeaig ye ‘BOB PEASE obtained a BSE rom MIT in 1961 and is Sta Scientist at National Semiconductor Corp, Santa Ca, Gavi RAP’s Multirange Wattmeter cers 1.501 KW, some people might ‘be interested in other ranges. So, ‘we will show here a multirange version, ‘with full calibration (see the deuailed schematic). The basic explanation is also applicable to this version with multiple ranges ‘We'll stat with the 2.0-m@ copper -shuntshown in the April 29 column for the 1.501-KW range (wwwelecdesign om/Aices/Index.cfm?ArticelD=17028 ‘SExtension=pd). Then, Imade a smaller shunt, two strands paralleled of about 12 in, of 22-gauge wire, to make up the 9.0-mQ shunt. Ths gives 150.1-W full scale, with 0.1 W of resolution. Ifyou want a 15.01-W range with 0.01 W of resolution, you can add a 50- mAshunt made of fof #24 wie. For 8 1.501-W range | used a hank of $0 ft of #24 wire. Extra switch positions may be needed. These shunts are all added to the basic Wattmeter circuit. Each resistor should be trimmed within about 3% of nominal to make fina csl- ‘bration easier. Measure twice, then cut carefully! ‘The calibration for different ranges is casily madeby adding suitable switches for each cange. If you have a four-posi- tion, twosdeck rotary switch, you could use that to connect up the shunts and the multiple gain-adjust pots, as shown. Myself, | prefer to use four DPST switches, one to turn on each range. Ijust have to remember to turn offeach unwanted range. ‘Note that most rotary selector switch es aren't rated for 15 A. But i's easy to find 15-A-tated DPST switches, ifyou are going (0 Use a rotary selector switch, don't change ranges when heavy load curtentis (or willbe) owing! Remem- ber, a load of just six 100-W bulbs would draw 25 4 of turn-on transient! ‘That would soon ruin most rotary selec- orswitches A though the basic watumeter cov- Ifyou calibrate this at about 2.8.30 Vigeren = 1.01 21% (Vin 12 V) the calibration will berighton at6 A, about 96high at 1 or 0r3 A, about 196 low for 12 dc, about 2% low for 12-A sine ‘waves, and about 5% low for 15-Asinu- soidal ac loads, or as much a5 20% ot 2084 here ne rectifiers are pling s nificant peak currents atthe sine peaks. Inall cases, this meter computes the true RMS multiplication of the V x1, butte scale factor just ails offa couple percent, for currents above hal-scale. | computed that if you use this meter with a load that draws relatively large pulses of current, such as 0.75 Aon a 10% duty cyte, just atthe peaks of the sine waves it will actually draw 120 W average power Ifyou read this with the ‘meter in the 1500-W scale it will read about 119 wats. Bucin the 150-W'scale, itwill ead about 74.4 W. All multipliers have limitations on duty oyle and pulse widths. $o 1 just showed you where this one has its imi- tations. If you have some equipment that draws a ot of watts during the line voltages peas (because the recifirs are charging up capacitors right atthe peaks), anditseems tobe up above 50% of range tr @ higher range: You wil get less resolution. bus beter accuracy ‘The temperature compensation is not perfect. But at all rated loads, the errors should be wel inside 29% be- ‘ween 0*Cand 50°C. IFivis well calibrated at 10 Va it will end to be off at 80 V ac by about 0.2%, oF at 140 V by about +0.2%. Not o bad. This watumeteris no sui able fr reading the wats in loudspeak- es, because it loses a lot aFaccuray if Vg fess than 60 0780 V. (Ifyou want- ed to add the Howland Current Purnp Using AI to compensate down to 2 V, you could try that. It might work pretty Wwell up to 1 kHz, but haven't built this) You can calibrate the respective ranges with 2.0 A, then 0.20, 0.020, 0.0020, and so on. Of course, you will hhave t readjust the zero offset or allow for some offset error in the lowest range, as there may be a small but noticeable offset in the lowest range. Can'tbe helped, Ifyou want intermediate ranges, such as 5 KW, 500 W, or 50 W, be my guest. Make all the fancy switches and shunts ‘you need. You could even compute 15 kW or more using 0.1-m0 shunt. But just remember, the shunt will dissipate ‘over 2 W a ull scale! PS. The basic wattmeter design is bbased loosely on an old design by Catl Nelson in NSC Application Note ‘AN222. | added the ranges the dc cali- bration, and several ather features. PPS. I haven’s had sime to measure de inductance, but the shunt as laid ‘out in my April 29 column is not going to be very suitable for audio work, not even at 5 ot 1 KHz, Ifyou want 1 ty to use this for an audio wattmeter, send ‘me an e-mail, and tell me your snai mail address, I'l reply by July 1. Maybe ‘ovisted pairs will make a good shunt outta 15 KH2! All for nove. / Comments invited! RAP / Robert A Pease / Engineer rap@galaxyns.com—or: ‘Mail Stop D2597A ‘National Semiconductor PO. Box 58090 Santa Clara, CA 95052-8090 Moy 13.202 Bere OESa | BSEE from MIT POrridge FR ir cserscers stein ‘Semiconductor Corp, Santa Clara, Calf. What's All This Wattmeter Stuff, Anyhow? (aka muttiptication Stutt, Part m1 ast month, | designed a good 1- ‘mQ shunt ("What's All This Shut Stuff, Anyhow?" April 29, p, 94). Now welll use tina 1.5-KW wattmeter ‘This one is optimized for the 80- to 4130-V ac range, and for currents up to 15 A (see the figure). How does it work? When Viv is posi- tive, the current in R1 just flows t© ‘ground via D1. But when Vin goes neg- ative, the current flows through RZ to the matched transistors. Ifthe load cur rent I is zero, the balanced transis- {ors will cause zero output, If there's foad current, the tran- sistors will multiply the I, by VL. act- ing asa balanced modulator. The “product” signal will appear at the ‘output terminals, to be read by 2 floating meter. A 3-1/2-digit DMM ‘with 200.0-mV full scale works well. An analog meter with 100- or 150-mi full scale gives poorer reso- lution, but it gets you started. What shunt shall we use? Previ ‘ously, I said that the copper shunt had a BIG tempco, +3300 ppm°C. But here the copper shunt will well ‘compensate for the -3300 ppm/*C of the go of the transistors. (Nichrome with its low tempco would be wrong.) ‘We won't achieve perfect gain terp- ‘co—but pretty good “This meter resolves 1 W out of 1.501 kW, using a 3-1/2-digit DMM. IF you want high resolution over a wider range, check out the multiange-version schematic in the electronic copy ofthis column at wuuzeleedesign.com. It uses four shunts to cover 150., 15: and 1.5 W ranges. Ifyou ean’t attain info from the Web site, circle Reader Service No. 552.0 acquire it by mail ‘What's wrong with this wattmeter? At large currents, the multiplier cell isn’t very linear. The gq of the transistor pair falls off, according to tanh(Vseys¢/52 mV). As the peak voltage across the (86 LECTOWCOESIN «ey 13, 2002 shunt exceeds 10 mY, the diffamp’s gain falls off by 196 o 2%, even ifthe average current is under 3 or 5A ‘That's why I chose a 1-m® shunt for 15-mV “full scale’ at 15 A. Calibrated at 100 Vac, its gain will be low at 80 V by about ~0.196, or high by +0.1% at 140 V, Not two bad. This meter won't directly read the ‘nie power if the load draws current on only one half of the cycle. But a work- around js easy: plug in the load both forward and reversed, and take the aver- age. Or instead, employ a big DPDT switch co reverse the output wires recommend the LM394CH for the super-matched transistor pair. If that's inconvenient, make up a matched cou- ple of 2N3904s (hetp:/uuw.national. ‘com rapStorvbe html), Pair them with- in 0.3 mVat equal currents and temper- atures, and glue them together. ‘The circuits in here are mostly at low voltage, but a few pans are at high volt- age. You can put tape over the high- voltage components. Ifthe low side of line power is connected to the wide prong, most circuitry should be at low voltage. But you must follow proper safety procedures while working on high-voltage circuits. Keep one hand in your pocket, etc, when you see that ‘neon lamp glowing? What's right about i? Well, i reads rue Vx watts in a true rms way. It rejects reactive currents, such as capaci- tive or inductive currents. For example, a capacitive load, like 2 uP (rated at 200 Vac), will generate 90 VARS (reactive). ‘But this meter reads zero because the integral over one cycle of line voltage is zero. [thas low "burden”—the voltage across the shunt is normally less than 0.039 ofthe line voltage. Calibration: & big advantage of this meter is that you can calibrate iewith de without making or mea- suring lots of watts of precision ac power. First, disconnect the meter from line voltage. Apply +8 V de (used only for calibration) to the 48:V jack and apply 100.6 V deto the Vy at the input line plug ‘Then, adjust the offset so the meter reads 0.0 mY, Now, force 2 A de through a cali- brated 1-0 resistor into the input ‘connector prong and cut ofthe out- put socket. Trim the gain adjust so that the meter reads 2 x 1.01 x (Vin = 06 V) x1. (The "2" makes up for the multiplier working just half the time.) So if you had 2.00 A x (390.6 - 0.6) volts, the meter should read 40.4 mv. Then the meter zeads 20.2 mV fora true 200-W ac load—about 1% high, but right on at 6 A, 196 low for 9 A de, 2% fow for 12 A of sinusoidal load, and about 4% low for 15 A, Next, Ill check itout at PG&E’ labs! All for now. / Comments invited! RAP/ Robert A, Pease / Engineer rap@ galaxp.ns.com—or: Mail Stop D2597A, National Semiconductor PO. Box 58090 Santa Clara, CA 95052-8090porridge Bob’s Mailbox Dear Bob: Tenjoyed your “What's All This Drilling Stuff, Anyhow?" column (e1t<: TRONIC DESIGN, March 4, p. 81). Here's a good 1/4-in. 4000-rpm drill: www. oastaltool,comegi-binjSoftCart.exe/af ‘mily/mai0101-20.htm?Ecoastest. No, at $135, ifs not cheap. Milwaukee builds first rate stuff, and I'm sure this isthe last 1/2-in. drill you or your grandchild will ever need. have an old Blick & Decker that still ‘works, thoagh, and T don't expect to upgrade, as it sees litle use. ! work with ‘metal alot, and the high speed is good for burning up bits. I recently discov- cred the advantages offered by air-pow- ered drills. No, it doesn’t do you any good unless you have an air compres- sor. You can geta pretty cheap, good, very high-speed air-powered drill ‘motor ftom Harbor Freight—no plastic gears. A 2-hp (I know, don’t get me started) compressor to run it will cost around $100 on sale. But i's handy for other things too. Plus, the air-driven aril slight and compact In addition to high speed, air-pow- ered drills have an outstanding safety feature: nearer0 torque when stalled. have a 1/2-in. electric drill that's nearly lethal when the bit grabs, say when drilling into an automobile frame. 1 recently boaght a 1/2-in, pneumatic drill motor (geared, slow) because it fit {nto my tight work place. I found out that it had about as much running torqueas the electric one, Butit was safe to hold with one hand, off balance, because when the bit grabbed, the torque dropped to a very manageable level. My electric drill motor now lives ona shelf gathering dust, (Hmmm Point nicely taken, [rap) Here's an interesting note: Carpen- tets use drills to turn their drill bits. ‘Machinists use drill motors to tum their drills. I think it’s a mauer of what one considers to be a tool ‘vin Ferguson vae-mall 52 mcrROMC DES «Moy 27,208 BOB PEASE obtained a BSEE from MIT in 1961 and is Staff Scientist at National ‘Semiconductor Corp, Santa Clara, Cait. CTO ESN If Uhad ato of holes to dri, an air di- en drill might be a very good way t0 go ‘Buc most years, the amount of drilling that 1 have todo would not just that! —RAP- Dear Bob: Re: 2/a-in. electric drills. You obvi- ously aren't looking hard enough ‘uu. toolerib, amazon.com lists several 1/4-in. electric drills, For example, see sunou.araazan.com/exec/obidos|tg/stores] detail hi/B00005Q7AB/qid=10173444 ‘35/br=1-16)refbr_If hi_18/104-72712 33-2720706 (6, ut who the heck do they think they are—Enron? They think that they can rake a conspiracy to cut ofthe supply, then {force me to pay $1392 A good 1/4-in. dill ‘might cost only 3/4 or 2/3 ofthe price for a 3 8in,drill—not 3x or 5x. 11 hep going to garage sales to try to find another one Fortunately I stil have to that are in good shape! rap) Bob, I agree that the price is pretty high. But your allegation was that no one made them, not hat what was on the market was overpriced! ‘Brian Lenharth vae-mail 1 looked in hardware stores in 25 coun tries and found none. I didn’t check out {jewelry stores RAP Bob: Yourcolumns are priceless! Although Thave sacrificed many coat hangers for tool purposes, my all-time favorite adaptation is the bicycle spoke. Decades ago I scrapped some junker bikes and saved the spokes. They're invaluable for retrieving dropped pieces in cramped engine compart- ‘ments. The hook on the end is perfec, and their stifness is ideal. Now Il have to make some drills Pte Stove via e-mall I must admit, I haven't used bicycle spokes each, other than as bicycle spokes. 1 usualy use coat-hanger wire, or bailing wire. But, there's no hurry. These pieces of rmecal are just lying around, waiting until ‘we need a dril in their size! Note that not all coat hangers are of the same gauge wire!—RAP Dear Mr. Pease: Please forgive my formality, but ! have been reading your work for so long that have raised you to near-deity status, (Obviously you don’t know me very well. ap) Anyway, my guess is that you have had multiple responses regarding the use of various forms of flattened wire, etc,, for drill bits. I thought | would let you know that the common term for bits like those you made is “spoon bits” (see wuw.fine-toets. comjbohr3. him for a picture). Such tools have been employed far quite a while (two or three centuries). Now they're primarily implemented in traditional chair making ‘They work wel, as you have shown, and ate relatively easy to make, Also, have you considered using a small hhand-powered drill rather than an elec- tric 1/4-in, drill? I find myself leaving ry electric drills in their boxes when | have only a few holes to drill V'm not one of the traditionalist wackos who only use hand tools, as my collection of 220-V equipment in my garage proves. But | have found that ‘any hand-powered tools work atleast as quickly for small jobs. Enjoy your columns/books/musingsete ‘Mark Enright via e-mail Thad over 100 holes to drill An electric rill was quite appropriat. My homemade bits aren‘ terribly sharp, and high rpm is suitable fer such dviling RAP Allfor now. / Comments invited! RAP / Robert A. Pease / Engineer rap@ galaxy. nsc.com—or Mail Stop D2597A, National Semiconductor PO. Box 58090 Santa Clara, CA 95052-8090Stuff, Anyhow? (Part 11) said that you'd like Charlie Sporek’s ‘new book about Fairchild and the early days of Silicon Valley. Titled Spina it costs about $28, Full info is available at uununationacom/sporck If ‘you can't justify buying it for yourself, get your librarian wo ba * My book on Troubleshooting Analog Crews, now translated into Russian, can be ound at usd But don't {think that your search will work well if you cant read Russian. [have been told ‘that the translation and interpretation ‘are very good though. ‘*IEEE's March 2002 Spectrom had a ‘good story (p. 50:55) aboutar. images, and litle drawings on masks and chips {hich surcrtoNic DESIGN has also cove ‘ered, Apri, 1999, p. 6411). Prety gooe stuff—a “Silicon Zoo.” The Web pages: pcre magnet fx edufreasindes, hheml are a bit slow to come up, but worth che wait, Can you name any more examples of drawings and art on chips? ‘The oldes the better! Pass ong any si icon chip art to that Web ste *# There are several excellent com. ‘ments on the analog art versus digital stuf? at hup://madvsiece.cornel.edu) quips nl Well worth printing out. Suit- able for faring! '* Several people told me where I can buy a Mikita 1/4-in. electric drill for ‘only $139. [refuse to pay that price. beep checking out garagesales. looked in 25 countries and found no 1/4-in. drills in any hardware stores. But 1 didn’t ook in jewely stores. * Are you interested in my opinion ‘ona conventional driver education book? See wuw.terthooileague.org/ 103dr-hum and compare it to my Web site, un arson com + War o rea ap interview with RAP by fon Carroll of the San Francisco Gironicle? Go to wnnusjgatecongi-bin! aticl.ifitenjchronclefarchive/1999/ (08/23/0D32191,071. 2 CTRONC ESN 1822 + Several people asked me. eering 22 “Sudden Cessation Of Stupiity Stuf” (ezcmonacscs ug 7, 2000, 0.134), ifthe ltde wood strips realy worked and detouced the water from running into ry house, Yes, they worked. It ook a while to get them installed, but the 1/2- in, quarter-round strips worked wel. nailed up some and epoxied others. But iss good idea to keep your ain gut vers mostly cleaned out! * Another embarrassing ypo enor in “Equations Par IT (e1:cTHONIC OFS, (ct 15, 2001, p. 90): In the second equation I said "12 = x 1/(C2)I"— bur itshould be [x pC? | Like I told you. ics hard to get equations perfect! ‘The Web version has been fixed «Several people asked me, ‘Whatever happened to Silicon General?” Back in November of 1993, it turned into *symmetricon.” Thea Linfinity became 2 linear AC subsidiary of Spmmetrcon, So it may be stated that Linfinity isa successor of Silicon General, * Similarly, whatever happened to Teledyne Philbrick? After Teledyne Components turned into Telcom, ‘Microchip bought the company. Some people ask me about how they could geta copy ofa recent col- tumn, Most of my newer columas— about 60 of them—and 26 of my old columns are posted on the Eletronic Design Web site; hup:iwselecdsig. com/index chntAction=Pease. This ist oF «columns i accessible via my Web site as well. Go 10 wnn.national.comrap and lick on "ED Columns. You ean also lick on “Eleewonic Design.” *# Some people ask me about other ld columns that arent on the Web. ra ‘eying to add these to my Web site, but there have been delays. i's worth a ty at "Other Columns." Ihaveadded afew new items, ists trek reports, and Homi- ble Pictures. f there's acoluran that you ‘an‘tfind, tell me which one you need, ‘BOB PEASE obtained BSEE from IIT in 1961 andi Stat Scientist at National ‘Semiconductor Corp, Santa Cara, Calf What's All This Floobydust and tell me your snail-mail address. ‘Otherwise, Tl have to pry it outof yout ‘Robert Klabis and I agree, ifyou invest your money in tax-deferred accounts, the taxes will change—the rates are bad, and may get worse. Wel, ‘wewere par right—they seem to have changed. Maybe they got better. Not ‘much. Meanwhile, the Altemative Min- imum Tax (AMT) seems to be getting, ‘worse, What's All This AMT Staff, Any- how? (Later) ‘A friend reminded me that the slide fastener for clothing was a loser until someone came up with the name *“Zip- per” Similarly, you can save ttle time tying and untying your shoes using the Shoe Zappy: un ThighPowercom. + John Trudel (formerly a columnist with Electronic Design) says his new Web comments on innovation are quite popular, See www. trudelgroup.com/ Ayknow.htm. | suggest taking his sur- vey/djuestionnaire about innovation. * Many people have inquired about where to buy Teledeltos paper. The criginal source in England wil stil sell you a big col for $95, including ship- ping; wun searchingsln.com/senco3. Fm! will gt you to the rightaddress for Sensitised Coatings, or call +44 1553764836. Plus, anew source for resistive paper is online: uuw2.pasco. com{products)scrpts|productsaft2N=PK- 19023. The KIT PK9923 costs $114, but PK9025 paper with centimeter grid sutss $34 for 100 23- by 30-cm sheets Buy your own pea with conductive sil ‘ver ink. This is about 32 kO/square. Alor nov. Comments invited RAP Robert A Pease Engineer rep@galansccom—o ‘Mail Stop D2597A, ‘National Semiconductor PO. Box 58090 Santa Clara, CA 95052-8090BOB PEASE obtained a BSEE rom MIT In 1961 andis Staff Scientist at National Semiconductor Corp, Santa Care, Cali. Bob’s Mailbox Dear Bob: nally got around to reading the Feb. 18 issue of Electronic Design. As always, read your colurnn entirely. The tirade of several readers (and you) on the extremes of audiophiles brings two comments to mind. First. isi possible that youngsters (never trust anyone under 30!) have Detter eats than we more mature folks? (Of eourse, mos younger aucophles have beter ears. No problem. Ihave no doubt ‘tha hey an hear many dings beter thar 1. But can stl prove they can't hear any difference! /rap) You bet itis, unless they've destroyed them by loud music. recall teaching electronics to a group of teachers when | was in grad school. The ‘other teacher suggested that forthe lab ‘experiment on amples, they merely twist the dial ofthe audio generator to 20 kblz, 201d measure there walked into a room that was quite literally painful. Unfortunately, my ability 10 feel that type of pain now i long gone. Second, perhaps this is analogous 10 the clists'exteme of spending $100 to shave 10 more grams off the weight of thebike (Oh abd technopiles oe to do tha, even f the difference s hard to rmaasire But 'm slat he aren of $100 a pound. could easily cut pound of ke ‘weight nod below foro more an $100. But refs to consider it Because om along trek, it’s imposi to prove that the lighar compre wil no ft. fap) While ide several'li bikes, i not the weight, But rather the ride and contosion resistance Which dictates that. A good ste! frame Isessentially ust as ight. (in the old days the exper said T “ad not feel right” I guess they learned how fo ‘make them feel “right.” My frame on my Specialized Recthopper AIFS i kami, but ¥ agree that ic’ not appreciably ighuer than a good stet frame of comparctio strength, dual, and price. Lets no tale about carton fiber! Hey now ho good a 22-1 bike feels compared to 27 Uh 18 and 12.88, compared tothe others, Mine swighs 27.6 Ws and tats just ight fr me because I want the most durable, reliable bike can ge under $500—an gota good one frap) Besides, ve noted that a fewr less beers shaves considerably more ‘weight off thebike for much less money! ‘David W.Kooble vie-mal Yes, but you get more fu pr dollar with beer! Or, in our case, rum and lemonade ‘fer our day's ride—RAP Dear Mr Pease: {realy like your acts, Most of the lime, agree withthe wayyou se things But the Wha all This Diling St ‘Anyhow? fuscrsonie Discn, March 1h) ancl Ios rea as an emorin i {don't know about American els but Eoropean drills always turn to the “right” (Cw). (Yeah, a een by the guy dri, ad thas anther way of saying that the workpiece ses the dil earning CCW as it approaches! /rap) This means the machines turing let (COW) com pated the dil teh. Ifthe dil sal, the machine handle Is pushed oward your right and, and eventually pulled out ofyour if hand ‘This isthe opposite of what you stated in the last paragraph. (You're right. 1 dit tat my argument ses bards. I must have been thinking of some crazy npoine Rut wea stare that ings ror symmetcal Dept the wy Tsai icackaands, the guy agred that he ha beter ck shtng of wis ethan ‘/rap) | think this is a good reason to use ftwith your right hand, Maybe it's.a good idea o starcbuldingdils fore Handed people! “nc Van de Wgaert vaemal Same elec ris are reversible, ad there ave left handed cil shafts. But | dt eed ary RAP Dear Bob: Thave three comments, about your ‘column on drills and drilling: 1. When you need to dill hole fora nail, use one of the nails. They te always porridge the right size, Since there are no flutes, the holes wind up a little undersize— just what you want. (I'm not sure ifthe Iles are as mall as want. Maye {should chuck a mail in there and sandpaper i oun abt jrap) 2, The reason drill motors “Tock on* ‘and wind up yourarm is because ofthat stupid and dangerous locking button, ‘on the side. It sits right under your hand, When the dill Binds, your hand naturally pushes it into the lock posi- tion, so the drill keeps going even after you let go. 1 don’t have lockup prob- lems because I figured that out and cut thelasted things oft. 53, Concerning flimsy drills: have a 14-in. Black & Decker that | paid $5.50, forin 1968. bought the cheapest thing that I could find t0 do a quick contract Job fora local radio station. f Aigured that when it wore out. I'd get a good one. Despite running it so hard that smoke has come out, and ithas literally gotten too hot to hold on to several titmes, it's stil fine. Bearings are good, original grease, original brushes, origi nal switch, and no plastic gears. ‘eaac Wings Vae-mal ‘A good Black & Decker is nota flimsy, cheap dri Ie may be inexpensive, butt has ‘excellent quality and durability. Thank heavens Isl have one! I wish tha every= body could buy one!—RAP Al fornow /-Comments invited! RAD; ober A. Pease / Engineer rape@galay.nsccom—or: Mail Stop D2597A, National Semiconductor P.O. Box 58090 Santa Clara, CA 95052-8090 Nowe: Bob is off wreeking this month ‘This time he's riding his mountain bike ‘over the high pass Thorong La at 27,771 firnorth of Annapurna, Ta find out how the trek went, juse send him a request titled “minirepor.” ire 242002 «ECON EES. 5BOB PEASE obtained a BSEE rom MIT Jn 1961 ands Sta Sctentistat National ‘Semiconductor Crp, Santa Clara, Celt. What's All This Multiplication Stuff, Anyhow? (Part IV) ben I'm out on the road, WM preciso, Trun into guys with good questions, every day. Some I can answer, some I can, and others just take a long while forme to igure outa good answer. One question that I car ied for atleast 70,000 miles was, “1 ‘would ike o have a voltage-contolied ‘pot! similar tothe ‘digitally controlled ‘pot! so the ourput will bea function of normalized contol voltage V, (where Waar =k), and Vour= Yaxk+Vax(1-1)." ‘The ‘wiper’ moves proportionally from Vato Vg thought about this a long time and kept the question in my wallet. I pulled itout and mulled over it many random times. Finally, [had a Eureka! moment and wrote to the guy: “Hey, Tim, you asked me why couldn't Linventan ‘ana- log pot! AfierI thought about this fora Jong time I figured out thatthe reason Js because somebody already invented it. Is called the analog muldplier. “The good news: You can go to Ana log Devices’ Web site (wum.analog. 72 kCIRONC OES uy 202 com) and look up the AD534 analog muliplier. You want to connect the Vy and Vs to the differential input pins at V+) and Vx(-)-Then you aso connect, Vato the V;, cuiput offset pin, As you move Vy up ane down from 010 10.0V, you get the ANALOG pot that you said you wanted “the bad news: These things cost $16 and up (even the cheapest grade, ‘even in 1006) and | don't think you're prepared to pay that price for an audio pol. I'd love 10 desige you an analog ‘multiplier using the LM13600, but that's now obsolete, and we ain't got any more to sll” Jim agreed that twas. neat solution, but much more pricey than he would pay. He plans to uy out the ie analog ‘nulplier butatlow pretty thought about it some more. Iwas mentioning this wick solution the other daytoacolleague Hesaid, “Greatsolu- tion, but the LM13600 isn't completely ‘obsolete because you can sil buy the 1M413700, which isthe samefuncion.” Sas astonished. “They told me that ‘the LM13600 and 14137005 were all discontinued, So here I'l give you the good cook- book circuit for this “analog-controlled pot” using the LM13700 as a multipli- ex/modulator (se the figure). This cir cuit has one additional advantage: If you move the Via litle to far the ana- log multiplier causes the unwanted channel to feed through alittle. In my approach, no feedthrough occurs Ifyou want this “pot" to provide a precision output with no offset, you can Use any suitable op amp to buffer the ‘output. Then you'd keep the 1-42 pots and trim for low output offset. If you're just going to need it for audio, use the Darlington output, Use two 499-2 resistors to replace a 1-K9 pot. No op amps or pots necessary—costing 1/10 the price ofa store-bought multplicr. What's good about this circuit? You ‘an implement itith rch current bias cesand get better bandwidth and noise. Ifyou don't mind a litle extra noise, you can run the whole circuit on less than a milliampere. Stor-bought mut- tiptiers won't do that. The bandwidth will stil be over 12 ki¥z, and you can design ito run on just about any sup- plies, such as 5, 6,8, 0F 13 V. ‘What's nat so good? This circuit isn't perfect in linearity, but is beter tnan 1/29 THD. The noise for a 10-kHlz bandwidth is only about 70.€B down (But the analog multiplier may even be better at some settings.) The crosstalk ftom an unwanted channel is just -85 dB, ‘After [built it, what did I see? Jt worked as designed. Fave fun! All for now. / Comments invited! RAP / Robert A. Pease / Engineer rap@galaxyrse.com—or: ‘Mail Stop D2597A ‘National Semiconductor P.O. Box 58090 Santa Clara, CA:95052-6090