Monday, October 27, 2008

Cheesy Products: The Case of the Solar-Powered Lamp

Less than a year ago, I responded to my wife's request to install a light on the stairway leading down from our back deck into the back yard. Her elderly father, who lives with us, goes down those stairs at night when he lets his dog out for the usual reason that dogs like to go out, and even with a flashlight the stairs can be tricky. She mentioned seeing advertisements for solar-powered light-emitting-diode (LED) lamps, so I found one at a local hardware chain store and screwed it to the stairway near the bottom.

From the start, the thing was somewhat of a disappointment. After dusk fell the first night, I was hoping its light would be enough to see the stairs by. But frankly, it reminded me of Mark Twain's candle supplied by a skinflint innkeeper. Twain complained that the candle was so dim he needed a second candle to see the first one by. If you looked carefully out in the back yard after dark, you'd see a dim bluish light hovering somewhere in the blackness, but it served more like a lighted buoy in a channel than as a source of illumination for the steps. Still, it was better than nothing.

Time went on, winter, spring, summer, and somewhere along the line, the lamp quit working. It was getting as much sunlight as it ever did, so I decided to do a post-mortem on the thing. It's 98% plastic, of course, and the works are all in the top. A white LED shines down on a conical reflector in the base that scatters light back up along a cylindrical diffuser behind a translucent white plastic box. Inside the lid I found connections to the solar cell itself (installed in a square opening on the top), a cadmium sulfide photocell, and wires leading to a circuit board. I haven't bothered to trace out the whole thing, but it looks like they used a CMOS-type integrated circuit (IC) to detect the light level with the photocell and switch the 2.4-volt rechargeable battery pack to the LED when it gets dark.

That is all fine and good, but this thing sat out in the weather. Somehow water got on the top (not a startling eventuality), made its way inside, and created a nice little rust spot on the circuit board next to the IC. CMOS ICs are notoriously sensitive to small leakage currents, and the conductive rust likely shorted out something or other, causing the entire apparatus to fail.

Now in the grand scheme of engineering ethics, this is not a big deal. My father-in-law didn't trip in the dark and break his hip, the monetary losses are small (twenty-five bucks or so, if I recall correctly), and after all, the lamp did give us nearly a year of service, such as it was. And I suppose it may have even come with a one-year warranty with which, if I bothered to fill out the paperwork, get a return authorization, and ship it back, I could get a new one. But what would happen then? In a year or so I'd have to do the same thing all over again.

Cheesy bargain-basement products intentionally made to last just long enough but no longer are nothing new. As industrialization during the 1800s made possible the mass production of stamped-metal products, complaints arose about how the market was flooded with goods that barely lasted long enough to take home. But somehow, there is always a market for a thing that's a little cheaper than the next comparable product, even if it doesn't work as well. These products are generally made by anonymous factories in Asian countries (the lamp in question is made in the Peoples' Republic of China, to use the official name), sold under nice-sounding brand names (this unit carries the brand of Hampton Bay, which makes a decent line of ceiling fans), and carried by the Wal-Marts and Lowe's Hardwares of the world. And yes, it takes that kind of a system to deliver goods at the lowest prices possible.

But see what you get: a product whose impermanence is almost guaranteed. The problem I'm speaking of could have been prevented with a better weather seal, but that would have added manufacturing steps, labor costs, maybe some added R&D costs, and the price would have gone up a dollar or two. And with the ruthless international market to deal with, the designer said to heck with it, let's ship it as is.

Of course, in principle I could have spent a few more dollars and gotten a better product, but only if one were available. But I did the easy thing, which was to go to the big-box store, find the cheapest thing that did what I wanted (or at least claimed to), and bought it. Judging by the selection available, that's what most people do. I'm a believer in spending a little more if you know you'll get a better product that will last longer, but such options are not always available. In some areas of consumer electronics, the tendency is to drive toward the bottom of the product lineup, cutting costs while maintaining a minimum of functionality. And we as consumers vote with our money to encourage such behavior.

Was it wrong for that designer to neglect the problem that killed the lamp after less than a year? I can't say unequivocally yes, and yet this situation falls into a kind of gray area of ethics that I personally would not want to spend a lot of time in. Cutting corners and trusting to warranties to get you out of legal trouble does not add to one's reputation, but then brand recognition and reputation is such an ephemeral thing nowadays that I'm not sure anyone worries about it much anymore. If I'm in the market for a ceiling fan any time soon (although in contrast to this lamp, ceiling fans seem to run forever), I might consider Hampton Bay, but not for solar-powered lamps.

As for the backyard stairs, I went out and bought a set of three solar-powered lights that use a pole-mounted solar-powered battery. The new lights are much brighter than the old lamp, and you can actually see the stairs better than the lamps. But the other night my wife told me one of the lights had gone out, and a few days later the whole system died. I've just finished exchanging the solar unit for a new one and tracing out a short somewhere that mysteriously disappeared. It works for now, but we'll see how long it lasts.

Monday, October 20, 2008

Ethics of Career Choice: Nuclear Engineering

Since most engineers are not self-employed, the type of work they do is largely determined by the organizations they work for. This means that one of the most ethically significant decisions engineers make is deciding on what job offer to accept. These days, it may seem like finding any engineering job at all is a challenge, but even in the worst of times you are still free to choose where to look for work. Today I'd like to show how you can begin to work through the ethical implications of a whole field of work, namely, nuclear engineering.

Only a small number of nuclear engineers design bombs, but the fact that the first application of nuclear fission was to kill thousands of Japanese in and around the cities of Hiroshima and Nagasaki in World War II has cast its shadow over the field ever since. There are those who are unalterably opposed to any use of nuclear energy, peaceful or otherwise. They argue that besides the danger of nuclear-weapons proliferation, the problem of nuclear waste hasn't been solved and the danger of a Chernobyl-type accident is too great to allow any further growth of nuclear power. (Chernobyl was the name of a city in the present country of Ukraine, then the USSR, near which a nuclear power plant exploded and caught fire in 1986, spewing tons of radioactive material over the countryside and forcing the evacuation of hundreds of thousands of people.)

On the other hand, both current U. S. Presidential candidates favor at least some expansion of nuclear generating capacity in the U. S. as a way of decreasing the nation's dependence on foreign oil. Nuclear generation releases essentially no greenhouse gases such as CO2, in contrast to the burning of fossil fuels such as oil or coal. Proponents argue that the waste problem is manageable and point to countries such as France that generate most of their electricity with nuclear-powered facilities.

You may have heard someone say that technology is ethically neutral, it's what human beings do with technology that makes for good or bad consequences. Most of the time, that view is at least an oversimplification, but it turns out to be almost exactly true of a particular kind of nuclear-related technology: the gas centrifuge. Naturally occurring uranium does not have a high enough percentage of the isotope U-235 to be useful either in nuclear reactors (which need slightly enriched uranium) or weapons (which need almost pure U-235). It turns out that the most efficient way to increase the fraction of the lighter U-235 isotope in uranium, compared to the heavier U-238 one, is to turn it into a gas by attaching six fluorine atoms to each uranium atom, and send the gas through an extremely high-speed centrifuge in the form of a hollow aluminum cylinder spinning in a vacuum. You need hundreds of these centrifuges to do the job, but they can be as small as only six feet high, and a centrifuge plant uses only about as much energy as a food-processing plant of the same size. The very same plant can be used either for making slightly enriched uranium for peaceful nuclear reactors, or highly enriched uranium for bombs.

This is one reason why Iran's gas-centrifuge facilities are so controversial: Iran's government says it's for peaceful applications, but the International Atomic Energy Agency inspectors know the same facility could be used to make bomb-grade material. It all depends on what the engineers do with it.

We've only scratched the surface of what is clearly a complex and multifaceted issue. One of the world's most infamous parties in nuclear proliferation, a metallurgist named Abdul Qadeer Khan, turned his knowledge of gas centrifuges toward making plans available for black-market buyers such as Iran and Libya. Khan is a bad example of how technical knowledge can be abused, but even that viewpoint might be debated by some of the people in countries that benefited from his expertise.

Perhaps more than most other fields of engineering, nuclear engineering is fraught with ethical questions: What products are made? How safe are the facilities being designed? What are the long-term consequences of use for future generations? Young people who are uncomfortable dealing with these issues may consider other fields instead. But the field needs some good engineers—both in a technical and an ethical sense—if the benefits of the technology are to be realized with a minimum of harm.

Personally, I would like no better outcome concerning America's energy situation than if we built a bunch of safe, uniform, French-style nuclear plants, junked our gas-guzzling nineteenth-century internal-combustion cars, replaced them with electrics charged from the nuclear-powered grid, and got in a position to thumb our collective noses at foreign oil producers. But a lot of things would have to change before that vision is realized, and a lot of nuclear engineers would be involved in the change.

Sources: The September 2008 issue of Physics Today carried an informative article by H.G. Wood, A. Glaser, and R. S. Kemp on the gas centrifuge and its role in nuclear-weapons proliferation, pp. 40-45.

Monday, October 13, 2008

Expert Witnessing Ethically

I have never been an expert witness in a courtroom or legal situation. But I have known engineers who have been. And sooner or later, many engineers and academics who teach engineering may get a call from a legal firm wanting to pay for their services as an expert witness. What are the ethical implications of serving as a paid expert witness? Can you both take money from only one side of a contentious legal battle and still preserve your integrity and objectivity?

Let's look at a few of the issues that might arise. To keep things concrete, so to speak, let's say you are an expert in concrete, and have been called in by the owner of a shopping mall whose sidewalks are cracking up after only two years of use. The owner is suing the contractor who poured the sidewalks. Should you accept the job, and if you do, what are your ethical obligations?

Many professional codes of ethics have a lot to say about this kind of situation. Although electrical engineers don't deal much with concrete, the Institute of Electrical and Electronics Engineers (IEEE) has a code of ethics that is representative of many engineering codes, so we will take it as an example since I'm most familiar with the code of my own professional organization.

The first item in the IEEE code that speaks to our hypothetical question says that an engineer should "undertake technological tasks for others only if qualified by training and experience, or after full disclosure of pertinent limitations." In other words, if you don't know beans about concrete and how it cures and why it would crack, you shouldn't trade on your educational qualifications in an unrelated field just to impress a jury. Some lay persons are awed by anyone who can write "Ph. D." after their name, thinking that it confers indefinite wisdom. Those of us who work around Ph. D.s every day know that except for the narrow specialization that the Ph. D. represents, people with doctorates in engineering tend to be all over the map when it comes to wisdom or judgment. And anyway, what your client is paying you for is specific technical expertise that you claim to have. If you don't have it, you've lied to your client, and anyway, the defendant's lawyer, if he's any good at all, will take you apart with great glee during cross-examination. And being humiliated in front of a crowd of people would be just deserts for claiming expertise you don't have.

Suppose you are well qualified to pass judgment on the matter at hand. The plaintiff or his attorneys will offer you a fee for your services. There is generally nothing wrong with this, because everyone understands that an expert's time is valuable and in the course of ordinary affairs, people have to pay experts for their professional time. Of course, if you feel strongly about a certain matter and want to provide pro bono services (for free), there is nothing to stop you from doing that. However, most expert witnesses are paid for their time and effort, which may be considerable in a complicated technical matter, and this is nothing you should be ashamed of.

On the other hand, if there are other connections involved that would look fishy to outsiders, and you don't disclose them, you run up against two more ethical principles covered in the IEEE code: to "reject bribery in all its forms" and to "avoid real or perceived conflicts of interest wherever possible, and to disclose them. . . ." Say for instance that you're married to the owner's daughter and stand to inherit the shopping center when the owner passes on. Most people would say that there is at least a chance that this fact will influence your professional judgment, since you stand to gain a lot more than just your fee for your testimony in that situation. At the very least, this fact should be made known to everyone, to the defendant as well as your client, the plaintiff.

Well, what if you agree to testify, set a fee, and then find that, contrary to the owner's hopes, the contractor he sued wasn't at fault? Maybe the owner made false claims to the contractor about the nature of the subsoil, and it shifted and cracked what was otherwise perfectly good concrete. All sorts of weird things like that can happen to make a case turn out to be otherwise than what it looked like in the beginning. What do you do then?

There's something you're obliged to do, and then you take whatever consequences arise from it. You have to tell the owner the results of your study, whether they're good news or bad news for him. Obviously, if your testimony isn't going to help your client, he won't be wanting you to testify. Whether or not you get paid depends on the nature of your contract with the client. The fairest thing might be just to write it off as a learning experience and not send a bill, but if you spent several weeks working on the issue, you can't afford to undertake too many projects like that. But what you should not do under any circumstances is fudge the data or your analysis to make things look better for your client than they actually are. Here's where the real temptation comes, and here is where it has to be resisted, whether it loses you your fee or not.

Besides all these matters, I have my personal opinion I'll throw in at this point. Like chili powder in soup, expert witnessing is probably best if done sparingly. The picture conveyed to juries is generally that of a busy professional who does "real work" most of the time and has undertaken to benefit the legal system with the expertise that he or she has acquired elsewhere. But expert witnessing pays well, and some are tempted to turn it into a profitable enterprise that takes up a lot of their time. This doesn't seem to me like a good idea for someone who wishes to keep a professional edge on their technical expertise. While the temptations to bend the truth can be successfully resisted if you play the witness role only once in a while, being objective for only one side in a dispute is always a strain. Most people can support the strain once in a while, but I wouldn't advise making a career out of it.

To sum up, expert witnessing can be a genuine public service that can "improve the understanding of technology, its appropriate application, and potential consequences" (another item in the IEEE code of ethics). But this kind of work is fraught with more than the usual quota of ethical hazards, and it takes judgment and wisdom to negotiate them without slipping up. And not everybody—not even those with a Ph. D.—can do it successfully.

Sources: The IEEE Code of Ethics can be found at

Monday, October 06, 2008

NASA At Fifty: A Modest Proposal

Five decades ago this month, a brand-new agency of the U. S. government called the National Aeronautics and Space Administration went into business. In 1969, only a little more than a decade later, NASA scored the biggest triumph of its short existence by putting men on the moon. While it would not be fair to say it's been downhill ever since, there is general agreement that NASA is now a troubled, conflicted, underfunded, and rudderless organization. As a recent Associated Press retrospective points out, the Space Shuttle is a flying antique that NASA can't afford to keep and can't afford to get rid of. The Shuttle is our only way of getting to the International Space Station, and current plans are that when (or if) the Shuttle retires, we will rely on the Russians until we can come up with a new vehicle on our own. These days, relying on the Russians looks about as smart as relying on the housing market to keep rising.

This is not to deny that NASA has pockets of excellence here and there. But a few pockets don't make a garment, and clearly something needs to be done about NASA. In the spirit of Jonathan Swift's "Modest Proposal," I offer the following suggestions.

One way to find out what NASA is really worth is to have a garage sale. You could have different sales for hardware—things like the Deep Space Network, Shuttle spare parts, the giant Vehicle Assembly Building (VAB) in Florida—and the software—outfits like the Goddard Space Flight Center in Virginia, the Jet Propulsion Laboratory in California, the Marshall Space Flight Center in Alabama, and so on. If this garage sale is anything like ones I've had, we'll have to offer some real bargains. On the other hand, I can see some entrepreneurs who might see possibilities in selling rides on high-G centrifuges and swims in zero-G swimming pools. Rocket-engine firings on test stands will always attract crowds on the Fourth of July. And think how many loft-style condos you could make out of the VAB, once the Florida real estate market comes back.

And here's an idea to make the sale go better. Instead of sending a bunch of dull old highly trained engineers up to the Space Station in the next Shuttle flight, we go around the world and offer free rides to the most popular entertainers in the world, regardless of nationality. I have no idea who these people might be, but you can ask any young Chinese or Russian or Indian, and I'm sure they'll have plenty of suggestions. We send them up there with a couple year's supply of food, and then sit back and say, "Surprise, young people of the world! You've got to build the rocket to get them back!" This will do two things: it will probably move a lot more NASA surplus stuff off the shelves, and it will motivate a lot of young people to get interested in space flight real fast.

That ties in with my next idea: the deregulation of space. It is high time that we let the free market determine what we do out there, rather than a bunch of bureaucrats and politicians. Of course, the first step is advertising and publicity. The drama of rescuing those entertainers will make great reality TV. And of course, everybody wants to travel to places where famous people have been, so space tourism will get a tremendous boost. Tourism means motels, restaurants, and all the other things that go with development. Having your latte at an altitude of 200,000 miles will give a whole new meaning to the word "Starbucks."

Naysayers will object that space travel is expensive, dangerous, and ought not to be approached with the reckless exuberance of a prospector looking for gold in a newly discovered territory. I counter that this is exactly the attitude we want. Every new generation looks around for some object to focus its idealism on. There are people out there who want to travel in space more than anything else, and we ought to get clunky old organizations like NASA out of their way and let them. The free market will determine the size of the effort, whether it's one private-enterprise rocket a year or a weekly space-bus trip from starports around the world. The good pieces of NASA that can contribute will find their places in this new order of the ages, and the rest, well, some things are better off simply coming to an end.

Nothing will ever take away the fact that once upon a time, an organization of people and machines known as NASA put men on the moon. But that was close to forty years ago. Five hundred years ago, Queen Isabella funded Columbus's voyages to the New World. But nobody has tried to keep the Spanish court going ever since simply to send more NiƱas and Pintas and Santa Marias out to do battle with the wind and the waves. If NASA's time has come to fold its doors, let's at least try to get some of our money back in the process. And let's encourage the world at large to do what it really wants to do with space—by putting its money where its mouth is.

Sources: The AP story on NASA's troubled 50th anniversary can be found at Jonathan Swift's "Modest Proposal" for the Irish to solve their overpopulation and poverty problems by eating their children was obviously intended to be ironic, as is the case with my proposals above. Swift's original essay can be read at