Check the text at the bottom of the page from N3NDJ, Dave Sarraf. Dave wrote me an excellent email detailing some of the problems with the device in the article and solutions to those problems. He also includes a little bit of history concerning thermo-electric devices.

Thanks Dave! vy 73 de NS8O

PAGE 2

Your web page notes mentioned materials for the wires. Use standard

thermocouple pairs such as Type K (chromel/alumel), Type T

(copper/constantan) or type J (....../???) . Once source is Omega

Industries in Stamford CT. Since you don't care about accuracy buy

extension quality wire rather than thermocouple quality wire becuase it

is less expensive.

I question twisting the wires together as shown in the encyclopedia

article. Every place that the wires touch creates a junction, so

current could flow in a loop between the hot junction and the nearest

twist rather than from the hot junction to the cold junctions at the far

ends of each wire. This would greatly reduce the already low output of

the device. Instead spread the wires so they don't touch. Soldering is

ok for the cold junction. Rosin flux will not be aggressive enough to

clean the chromel and alumel wires. The best way to bond the hot end

would be by spot welding. This creates a reliable joint without the

concern of the candle flame evaporating lead from the solder and

dispersing it into the air. Twisting would work for a quick test but

the wires would soon oxidize and make a bad connection.

Thermionic and thermoelecric converters are rather old devices. They

were studied extensively in the early 1960's but fell from fashion when

fuel cells began to improve. About every 15 years the idea is

resurrected and studied again using new materials or new geometries.

The major attraction of these converters is silent and passive operation

and the ability to utilize waste heat from combustion or other

processes. Their efficiency is not very high but in some situations the

advantages are too great to ignore. They have been used on some

spacecraft to convert decay heat from bricks of radioisotopes to

electricity. With no moving parts this power source is highly

reliable. Several organizations have used them to make electricity in

the field using either waste heat recovered from turbines or directly

heated by some source of external combustion. Such generators are

versatile, reliable, and nearly silent.

Most recently TPV is making inroads. TPV, or thermophotovoltaics, uses

photocells to convert heat to electricity. The source or emitter

operates between 900 and 1700 C. The light emitted by the source

(visible or infrared) falls on cells whose bandgap is tuned to the

incident wavelength. Systems having efficiencies comparable to internal

combustion or power plants have been demonstrated, and a few companies

are planning camp stove or lantern-type generators for the backpacker

industry. A few years ago JX Crystals demonstrated a miniature system

that powered a transistor radio from a candle flame and a larger system

that ran a portable TV set from a propane flame. A company in New

England was working toward a gas home furnace that used some of the heat

to make enough electricty to power pumps and blowers to make the furnace

completely self-powered.

All of this does have some applications to QRP. The encyclopedia

article shows that it is possible to power a radio from a candle, and if

changes were made I would expect that you may even power a transmitter.

A lesson to be learned form the TPV folks is that the efficiency of most

PV cells rises quickly if the light is concentrated. Try adding

reflectors to solar panels to increase their output. They also learned

that efficiency falls with rising temperature, so make sure that the

cells are well-cooled if you do try this. Do this by adding fins to the

back of the panel; a fan would most certainly be counterproductive.

Dave Sarraf

N3NDJ