solar water purification on the cheap

Not too many weeks ago NPR carried a story from a refugee camp in the Democratic Repubic of Congo. The camp is next to a lake. The refugees were washing their clothes, themselves, and doing whatever else they needed to with the lake water. They were also drinking it. As a result, they were getting sick and dying from cholera. All that water, right there, and they can't drink it because it'll make them sick. That's where this project came from. The goal is to come up with a way to purify water that is cheap, reliable, and could be deployed by aid agencies very quickly.

This is a design for a solar distillation system. It doesn't work very well, yet, but perhaps it will eventually be useful with some changes to boost efficiency. Currently, the design calls for a 2 liter soft drink bottle and a plastic cup. I don't know how available that stuff is in D.R. Congo's refugee camps, but we produce huge numbers of bottles and cups here, at extremely low cost per unit.

I'm including directions here for those who want to try their own experiments and post the results.

Start by cutting the top off the bottle about an inch (or the length of the tip of your thumb -- the dimension isn't critical) from where the cylindrical part of the bottle begins.



Next, you'll need to cut a couple slits in the top piece so it can nestle inside the bottom piece. That way, as condensation forms on the top, it'll drip down to the bottom.



The construction work for the basic still design is now complete. The theory and operation are simple. You put a dark colored cup of contaminated water inside the still and set the thing in the sun. Sunlight warms the water in the cup, causing it to evaporate. The water then condenses on the cooler plastic of the bottle. The distilled water drips down and collects in the bottom. Here's a shot of the basic still in operation:



Finally, in this close-up, you can see the distilled water that's accumulated:



This particular test used simulated contaminated water: I let a cup of Earl Grey tea steep overnight. That way, it both had a chance to get good and strong, and it was at room temperature before starting the test. The color of the tea makes it immediately obvious whether the distillation process is leaving behind the heavier stuff, and the distinctive odor is a good marker for the aeromatics. In this case, I performed an el-cheapo chemical analysis by drinking the clear water condensate. There was no hint of tea flavor or color, and only a slight whiff of Earl Gray. That's a good sign that distillation is working as expected.

Unfortunately, all is not not perfect. In fact, right now the still is too inefficient to be useful. After about 5-6 hours of operation, what you see is all the water that accumulated. So, some improvements are in order.

First of all, you'll notice that I'm using a blue cup, not a black one. Black cups are proving hard to find. If you know of a national chain that sells black plastic cups, please feel free to post a comment pointing me there.

Secondly, notice from the shot of the still in operation that the condensate forms above the rim of the cup. That suggests moving the cup higher (or the top lower) will concentrate the condensate in the top of the bottle, leading to more drips forming and accumulating in the bottom.

A taller, skinnier cup would probably help the situation by giving a higher ratio of surface area to volume, increasing evaporation rate.

Also, I plan to try taping an aluminum foil reflector to the outside of the bottle. That may concentrate more sunshine on the cup, speeding the process. (The fact that the bottle is cylindrical conveniently puts the cup at the focal point.)

I'll post more updates as I have time to run more tests.

Later: There's a follow-up here.

federalism and administrative agencies

One of the interesting things that happened during the 20th century is the rise of the administrative agency. A lot of law-making these days is really rule-making: Congress passes a broad law, which tells an agency "you handle the details". The agency then does just that, passing rules. Those rules act pretty much just like laws, except Congress didn't pass them directly. For example, the FCC decides how many radio stations a given company can own in a given market. The FTC figures out if a pair of large corporations can merge. The FAA decides how old you can be and still fly a 747. The IRS decides where and how you can submit your taxes. All of these things affect our lives just like laws, but they don't go through Congress.

Is that a good thing or a bad thing?

Well, in the Old Days, before the administrative state, the rule-making bodies tended to be the courts, who were doing it through common law. If you thought your neighbor shouldn't be allowed to raise hogs up-slope of you so that the smells and nastiness came down-slope into your back yard, you went to court. There, the judge would look at past cases. If there was already a case on point, the judge could apply it. If not, the judge would look at the closest cases, make a decision, and either tell your neighbor to cut it out, or tell you to suck it up (figuratively, of course). And that decision would get written down and (most likely) applied to the next up-slope hog farming case.

Notice, though, that that if there was no case directly on point, neither you nor your neighbor knew what the law really was until after the trial. And after you'd spent a bunch of time and money. In fact, you might not know even if there was a case directly on point, because the judge could say "that case was from a hundred years ago, times have changed, and we're not going to do it that way any more."

Nowadays, there would probably be an administrative agency that passes a regulation saying whether or not you can hog farm up-slope of your neighbor's back yard. That has a couple of advantages. The first is that you most likely know ahead of time what the rules are (unless you're playing really close to the edge, or the rule's ambiguously worded.) The second is that rules usually go through a public comment period before they come out.

Still, you can't vote a bureaucracy out of office, and CSPAN isn't camped out in front of their deliberations. So there might be an "accountability gap."

Also, it raises an interesting question: why the shift?

Is it that society's gotten more complex as the GDP's increased? Increasing GDP suggests an increasing number of interactions, which may increase the opportunities for conflict beyond what we can handle using common law techniques. Or maybe it's the other way around: maybe using regulations has made the system more efficient, so the GDP's gone up.

Or is it driven by a population increase? For a heirarchical management structure, you need a manager for every N peons, and an uber-manager for every N managers, and an uber-uber-manager for every N uber-managers, and so on. Net upshot is that the number of people in management has to increase as the number of peons does. In fact, the fraction of managers to the total population has to go up. Maybe something similar happens with rules and regulations.

Or is it driven by an increased centralization of power in the federal government? Note that increasing centralization might be good: it could make things work more smoothly since they don't have to cross as many government boundaries. It could also be bad: who's to say voters in Salt Lake City want to regulate and manage things the same as voters in San Francisco?

Darth Vader has a blog

Seems like everyone has a blog these days. Now Darth Vader does, too. And it's nicely done. (Kudos to Pete for passing this one along.)

more on the illusions

Here's an update on the previous post.

First, about passports. According to the International Herald Tribune, courtesy of Google News, it looks like President Bush has taken note of the kerfuffle (really, how often do you get to use that word?) over requiring passports from everyone entering the country, or perhaps it was of Canada's protest. He was apparently "surprised" by the proposed regulation. He is now re-evaluating the requirements of the law he signed.

Also, it looks like the rumors about RFID tags in new passports are true. Bruce Schneier's Cryptogram for 15 April contains a link to the specifications for the new passports:

The technology selected for the electronic passport is the 64 kilobyte contactless integrated circuit chip with an antenna. The electronic chip itself has a very short read distance, approximately four inches. This choice is compatible with standards and recommendations of ICAO. The standards and recommendations are found in ICAO Publication 9303, Machine Readable Travel Documents, Part 1, Machine Readable Passports, Fifth Edition 2003; and in the recommendations found in Technical Reports and an Annex supplementing that publication relating to the technology supporting the use of electronic chips in travel documents.

As for the 4" read range, I'm not clear at the moment about whether or not that's with a standard reader, or if you could get better range with a really good antenna and transceiver rig.

Sorry, no updates about the San Jose case.

where are the illusions when you need them?

Some days, the ugly, scary part of the world has a way of poking its head above day-to-day serenity, just to remind you that it's still there. Maybe it's just that I'm primed for it, having read a bunch of stuff about the development of freedom of speech protection -- or lack of thereof -- from World War I, through the McCarthy era, and up till today. Or maybe it's an overdose of Jared Diamond's book Collapse, a review of which will be coming once I get time to finish it. Or maybe it's just general lack of sleep. In any case, a couple news items jumped out at me within the last couple of hours.

The AOPA has posted this article about a proposed Department of Homeland Security program called the The Western Hemisphere Travel Initiative. Basically, it requires a passport for all U.S. citizens entering the country. No mention of whether your drivers license is good anymore. Besides having implications for college kids making their cross-the-border duty-free liquor runs, it may have some privacy ones as well. I've heard rumors that the government wants to start putting RFID tags into passports, which strongly suggests they'll be able to track, timestamp, and log every citizen crossing the border. Good thing? Bad thing? I guess it depends on how else you think someone might use the information once they have it.

In totally unrelated news, Shaun Martin posted this disturbing article about a San Jose police search that destroyed a Hells Angels chapter house. Now, I'm no big fan of the Hell's Angels, but I'm also not a big fan of the police destroying someone's house, shooting their dogs, etc., based, not on a conviction for a crime, but on a search warrant.

I'm going to go to bed now and pull the covers over my head. Sleep well.