Talk:Projects/WaterPurifier

UV Light Water Purifier Discussion

OLPC Category

• http://wiki.laptop.org/go/Water_and_sanitation

References

• Corbin Cambell Project Proposal
• Project Blog
• XOs for Humanitarian Aid / Disaster Relief

OLPC Responses

Nathan Riddle

The maximum output of the USB port is 1 amp (5 watts at 5 volts). The maximum output of the power module is 1.42 amps (about 16 watts at 12 volts).

Water purifiers are roughly 5 watts for one gallon per minute. If you are just purifying a glass of water, units are available that use a couple of AA batteries (3.5 volts).

A UV lamp is a fluorescent lamp with a quartz "glass" (in case of high efficiency models) and as such is a high voltage discharge tube. Some type of transformer (ballast) is needed to step up the volage to operate the lamp -- these are separate modules or may be built into the lamp base as in the case of compact fluorescent lamps. Voltages are about 150 volts or higher to start the discharge and may drop to 90 volts for operation.

Wires held with tension (clips) can be substituted for sockets (or electrical tape, for example, or wires glued against the contacts).

In the hobby market there are six inch fluorescent lamps. These are operated by 6 volt or 12 volt lantern batteries using a small (1" square) circuit board and draw less than 1/2 amp.

Now, if Christian can supply the description of the lamp purchased from Amazon, maybe a ballast can be found to operate from the 12 volt power supply used for the XO (or maybe from the USB port for low amounts of water). The ideal is to just buy a water purification arrangement that has a ballast that is powered by 5 or 12 volts.

Corbin bought:  http://www.amazon.com/gp/product/B0027ND0VO

In any case, the lowest cost equipment appears to be in aquarium and pond germicidals. A 3 watt UV lamp and a separate electronic ballast (120 volt) can be had for $25.

• http://www.soslightbulbs.com/gtl3-uv-germicidal-lamp.aspx

There is a multitude of example circuits on the internet for electronic ballasts powered by 5 to 12 volts (as well as some in the local electronic parts store).

I would suspect that some type of germicidal lamps are available in country and simple ballasts for low voltage power could be obtained or constructed from components shipped into country.

Drinking water germicidals for 2 gpm can be had for about $200 (U.S).

Some follow-up comments:

First, my preference would be a separate 12 volt battery pack that could power off- the-shelf items with higher wattage (portable fluorescent lanterns).

However, a demonstration item powered from the USB might have value.

If I were starting, I would try using ganged miniature lamps of 1 watt.UV - A lamps and the inverter are available from Frys Electronics ($7 and $10 respectively). Their supplier is JKLlamps.com which does sell UV-C lamps and the inverter for one or two lamps operated from 5 volts(about 200 ma per lamp) The lamps are about 3 mm thick and 25 to 150mm long with simple wire leads. Remember that the maximum power from all USB on XO is 5 watts (and that might include radio port), so a more realistic use might be 2 to 4 watts. Lower power means longer contact times and suitable for glass of water size.

One might inexpensively investigate such lamps by buying a commercial product and adapting it or "reverse engineering" it. One such product that uses such lamps is Germ Guardian for sale on Amazon for $28. It is for travelers to "sanitize" surfaces in hotel rooms and uses 4 AA batteries (6 volts) for power. Simplicity might be wiring USB power to the battery terminals.

The lamp that the project purchased is the replacement lamp for the Sub Mariner aquarium sanitizer (which is powered from 120 volt outlets. If it is to be used, the cheapest approach would be to buy the whole unit ($90) which provides everything.

Also, a VEHO microscope for the XO might be useful for assisting in evaluating the sterilization.

ADDITIONAL:

By looking at specifications from JKLlamps, any lamp wattage may be limited to lamp of about 1.5 watts. The UV-C lamp (11 mm x 90 mm) is rated at 0.72 watts and a 5 volt inverter would draw about 0.4 amps. There is a 5 volt dual lamp inverter that draws .75 amps. (https://www.jkllamps.com)

Isolation of the lamp from the water (to allow cleaning of fouling) requires quartz tube or windows. 25 x 125 mm quartz microscope slides are available that could be used as windows. A thin flow cell (made from two slides) might have an efficiency advantage.

These are just comments about hardware that might applied at a small scale-- no knowledge purifying water (except servicing laboratory water supplies). Still, if the numbers scale down, 1.5 watts corresponds to 1 quart per minute. --ncr

PET / UV-A as a low cost variation

The SODIS discussed by Chris uses PET bottles. PET has a lower cut off of about 320 nm and hence operates in the UV-A region. The UV-A lamp plus 5 volt inverter is available for under $ 20 from Frys Electronics as indicated above. A flow cell (flat) could be made with PET (from soft drink bottles) windows. A spacer between two windows with outer blocks for compression is all that is needed. A reflecting foil behind one window and a slot behind the other window for the UV-A light completes the assembly. A kitchen cutting board could provide nylon material for the blocks. Total cost under $25 . 22 Aug 2011 ncr

Chris Leonard

If you have enough sun to charge your XO, you have enough solar to use plastic soda bottles (ideally PET) on a tin roof in a solar water purification approach, a tried and true method, well-documented. I must believe that more UV radiation comes from the sun (especially in the tropics) than you will generate by converting light to electricity and then back to UV through an intentionally low power system.

I think the benchmark for capacity and cost-effectiveness should be SODIS:

• http://www.appropedia.org/Solar_water_pasteurization
• http://en.wikipedia.org/wiki/Solar_water_disinfection

cjl 12:43, 6 July 2011 (UTC)

That's exactly right, but SODIS is even more effective than that (relatively). SODIS depends on a combination of UV and heat the heat makes the damage to the pathogens more effective, and if it gets hot enough, will kill the pathogens in its own right. Taking into account the heat, SODIS will be even more efficient, relatively. The best way to capture the heat is to lie the clear bottle on a dark surface... (a metal roof of any color will typically get hot anyway - how important it is depends on location, cloud cover, i.e. how hot the sun is at that location. --Chriswaterguy 16:19, 15 July 2011 (UTC)