The inventor of this new system is Lonnie Johnson, who also happens to be the inventor of the “Super Soaker” squirt gun.  Johnson’s new focus on electromechanical systems appears to be underwritten by the success of the water toy.

At the core of Johnson’s Thermoelectric Energy Conversion System (JTEC) is a membrane covered with electrodes.  Under thermomechanical stress (heat), the electrodes manage to force hydrogen atoms through the membrane, stripping the electron off the hydrogen in the process.  Any time you can physically separate an electron from an atom, you have electrical current. If you can do it a lot and continuously, then you have a good chance of producing a lot of usable electrical current.

The cool part about Johnson’s engine is that it uses mechanical energy (heat oscillation) to rip the electron off the hydrogen.  The Discovery article also notes that while the system uses hydrogen, it does not “burn” or consume the hydrogen – it uses a fixed amount of hydrogen circulated in a continuous cycle of electron detachment and reattachment (circulating between anode and cathode and back, presumably).

Discovery notes that current solar panels convert sunlight into electricity at about 30% efficiency.  (In the lab, maybe.  High-end commercial PV panels run more in the 20% efficiency range, and mass-produced cheap panels are lucky reach 10%)  Johnson’s heat engine design could reach an efficiency as high as 60%, if the system can be built to handle higher operating temperatures – about 3 times higher than the current working prototype’s 392 degree F operating temperature.

Well done, Mr Johnson!  I look forward to hearing a lot more about this over the next year or so.

Share/Bookmark

Those are grand claims.  While the sentiment to save energy and reduce environmental impact is well placed, I was skeptical.

So, I dug out my only CRT monitor from the closet and hooked it up to my Kill-A-Watt power meter and my laptop’s external video connector.  For a white screen, I used an Outlook message editor window, maximized.  For a black screen, I used a cmd prompt window, maximized.

Nokia 447x 21 inch CRT:  Black screen:  75 watts.   White screen:  101 watts.

On the surface, these results appear to support the claims and pie-in-the-sky global estimates.

However, there’s a catch.

Look at the results for an LCD monitor:

Samsung 21 inch LCD model 204B:  Black screen:  36 watts.   White screen:  36 watts.

LCD screens use fluorescent (or recently, LED) backlights to illuminate the screen.  The backlights consume the same amount of power regardless of whether the LCD crystals are showing black pixels or white pixels.  If anything, LCDs have to work harder to show black pixels because they are flooded with white light.  In a CRT, black is the default state and the CRT has to work to make a white pixel.

LCD screens have been outselling CRTs for many years now.  Laptops surpassed desktop sales years ago.  I don’t know if there are already more LCD’s in the field than CRTs, but it’s clearly the case that LCDs are growing while CRTs are in decline.  Odds are, you’re reading this text on an LCD screen.

While it’s true that a black pixels consumes less power than a white pixels on a CRT screen, pixel color has no effect on LCD power consumption.  Given that LCDs are a large and growing (and possibly majority) portion of the global monitor population, the power savings claimed by blackle.com is a case of diminishing returns.

Originally published on my MSDN blog.

Share/Bookmark