Sallen-Key Filters Find More Limits
Another limitation not mentioned in the article \[on Sallen-Key filters\] is that the filter possesses a frequency response for amplifier-generated noise which is different from the filter characteristic modeled (Dec. 17, 1999, p. 96). In low-noise applications, the high "Q" of the filter magnifies the noise and can make the circuit unusable. Needless to say, this is not usually modeled in Spice.
Replacing the op amp with a single low-noise emitter-follower transistor can improve both the noise performance and the cutoff behaviour described in the article, and is usually cheaper too! (An npn-pnp follower pair can reduce dc offsets when required.)
Wireless Systems International
More Corrections To This Design Idea
Several of your readers have correctly pointed out a significant weakness in my "Universal Off-Line Power Supply Uses Few Components" Idea for Design article published on Dec. 17, 1999 (p. 114). The key issue is the lack of a 100-Ù 1/2-W resistor, which should be placed in series with D1.
Without this resistor, there is a small but finite possibility of destroying D1 during startup. Also, the voltage rating of C1 should be increased to 450 V.
I didn't get any feedback regarding the C1 rating, possibly because the figure shows 120 V ac, and C1 rated at 250 V is more than sufficient. However, the text mentions "Universal" and "maximum of 240 V ac." So the C1 rating should be increased to 450 V.
Maybe Classrooms Won't Go Virtual
Being about a month behind, I just read your Editorial in the Dec. 17, 1999 issue (p. 18). You concluded with the question: "Will web-based learning be the choice in the future, or will live instructors in a classroom setting endure?" My answer is a solid "Yes!"
Teachers can and do make a profound difference. If you ever got turned on to a subject which formerly was uninteresting, it is usually because you personally knew or were touched by a teacher. Such infusion of enthusiasm and love for the subject is not presently conveyed by web-based instruction. Thus, I believe live instructors in a classroom setting WILL endure.
As for web-based instruction, a huge number of people are finding it a viable option for their needs. Many are the non-traditional students, like those already in the workplace who can't leave their job for a full-time, on-campus classroom experience. For such people, it is a very welcome option. I believe it will continue to increase.
Over the many years of America's higher education, a continually higher percentage has gone to college. With the continuation of the traditional classroom, plus the options presented by web-based learning, this trend should continue to the benefit of all concerned.
Brigham Young University
Manufacturers Used To Be Greener
I extend my thanks for an outstanding Editorial in the Jan. 24 issue (p. 44). In one simple sentence, you managed to sum up my No. 1 gripe about the current state of technology. Specifically, you wrote, "We can and should build products that can be disassembled, repaired, recycled, and upgraded...."
I could not agree more! Aside from being a second-year EE student, I've been a ham radio operator since 1977. I've lost count of the number of pieces of equipment that I've been able to save from becoming landfill-fodder simply because they could be easily modified, repaired, or refurbished. Such hardware may not necessarily be the latest or "greatest." (I use the latter term loosely.) But the core excellence of its design gives it a much longer life-span than many manufacturers seem interested in these days.
There are a number of things that electronics manufacturers used to do with their products. I would urge any company reading this to please consider bringing these back:
1. Schematics and service information in product manuals. It used to be that nearly every electronic product purchased, from a simple AM transistor radio to a logic analyzer, came with this information. Granted, not everyone wants or needs it. But all too many manufacturers keep such data a deep, dark secret, especially in the case of computer hardware. I've lost count of the times I've asked for a manual, only to receive either a blank look or an outrageous price quote in return.
I say make available, at a reasonable price, a complete service manual for any given piece of equipment.
2. IC sockets. This is such a simple thing, adding only a few more dollars per unit to the cost of a product, that can save so much time if a chip happens to go bad. With the emphasis on surface-mount components, however, socketed ICs are rare. How about bringing them back, at least for critical components?
3. Parts availability. Electronics manufacturers should stop treating those of us who wish to make our own repairs like lepers, and make replacement parts readily available at a reasonable price. This is especially true of custom parts and assemblies, and of "legacy" equipment.
Granted, it's not always practical to keep a stock of legacy parts around. But at least make such data and/or parts available. If it's sensitive or proprietary, make it available through the signing of a non-disclosure agreement.