Turquoise Energy Ltd. News #92
  covering September 2015 (posted  October 4th)
Victoria BC
by Craig Carmichael


www.TurquoiseEnergy.com = www.ElectricCaik.com = www.ElectricHubcap.com = www.ElectricWeel.com

Highlights: Free Energy from Thin Air? - Lambda Rays, or Atmospheric Electrical Charge? (see Month in Brief, Electricity Generation)

Month In Brief (Project Summaries)
- Further peltier module 12 V fridge experiments - ARM Reluctance Motor Runs! - New (used) Batteries NiMH & Lithium iron phosphate - Free energy... where's it from again? - Timing of inventions, a perspective.

In Passing (Miscellaneous topics, editorial comments & opinionated rants)
- Syrian Girl's video: an 'on the spot' perspective on what's happening in Syria - Loose Change - Perpetual War - Arguably somewhat humorous things & "Political Spin".

- In Depth Project Reports -

Electric Transport - Electric Hubcap Motor Systems

* ARM reluctance motor project
* "Transverse flux" SR motor rotor idea

Other "Green" Electric Equipment Projects
* Peltier Module/Thermoelectric Cooler ('TEC') Experiments: supply voltage versus attained cooling - 15 amp Peltier module - better ice tray - internal fan - evacuated tube heat radiator - DC to DC converter
* Water Conservation: Best Shower Nozzle and Video

Electricity Generation
* Electrostatic Perpetual Motion Motor?
* Thomas Henry Moray: not only free energy collector -- he invented semiconductor electronics, in the 1920s... and lots more!
* Free energy from the electrical charge of thin air?

Electricity Storage - Turquoise Battery Project (NiMn, NiNi), etc.
* Honda Insight NiMH Battery - converted to 14.4 volts, 65AH.

No Project Reports on
: Variable Torque Converter Transmission, CNC gardening/farming machine, Electric Weel, battery making, aquaponics, Magnet motor project.



September in Brief

   My motors have evolved from "gosh it turns" in 2008 to BLDC motors with excellent specs but not very well built, to much improved constructions, then in recent months to a new BLDC plan with double the magnet poles, and now to the first of two latest switched reluctance designs of great promise, the "ARM" which first ran on the 29th. My several designs of "regular" BLDC motor controllers over the years all kept blowing up at high currents. The new 'unipolar' controller type should be much more reliable and will run either of the new types of motor, but running the ARM has disclosed a couple of things that need looking at in the controller.
   My mom suggested I should finish the things I've started rather than start more new things. In principle I fully agree. But not for lack of trying I have yet to put an "ultra efficient" add-on electric wheel or even a converted car on the road, and somehow the projects seem to all evolve into improving forms rather than to end.
   A video shows that where once I got the Sprint car to move forward in September 2012, I got it to continue, even upslope, until it hit its 'parked' position end stops, driven by about 1 HP. This one is perhaps the best demonstration that the slipping planetary gear torque converter worked once the car started moving, and that it probably just needs a clutch inline (or a big flywheel on the motor) to be practical. I didn't know I had this, or even remember that one of the trials had been that successful, until I found it on my old camera a while back, and I've now finally posted the video. The actual move is near the end of the video, which I didn't have an easy way to edit. (The format/editing problem is why I don't use that camera any more.)

https://youtu.be/hZ_auUjW17M  -- Slipping Planetary Gear Transmission

   On the second (band having not yet started up for the season) I went to the VEVA Islands electric car club. One member had purchased a 1981 Bradley electric. I'm sure I saw the Bradley in Popular Mechanics magazine way back then. The "gull wing" doors were supposed to be great in crowded parking lots. Another had brought his lightweight velomobile - which has pedals but zips along the street at 50 Km/Hr under electric power. We posed them along with my RX7 and a Mitsubishi iMiEV owned by a retired electrician.





The 1981 Bradley Electric (the only one ever made in Canada) with its proud owner.
The body is in great shape. Is it because it's fiberglass, or because it spent 20 of its 34 years sitting in a garage?

(Of course if I was doing this now, I'd use polypropylene ("landscaping fabric") with resin - It's tough; stronger &
lighter than fiberglass, and nice to work with. A good paint should prevent degradation of the plastic by sunlight.
In fact, that could be a fabulous way to reduce vehicle weight, and thus of course increase driving range!
...Make a whole car body?!? Yikes! Don't get ideas, Craig!)


   In practice the Bradley's interesting gull wing door design made it hard to get in and out of the car, and it's easy to see why it never caught on. The unadjustable semi-reclined bucket seats didn't help either.


   On the third I decided to make the planned solar water heater into a pumped system after all. Other than pump and controls, the logistics are simpler. The pre-heat water tank and a system that drains down when the pump shuts off can't freeze and burst. I ended up buying more pipe for this, but didn't get any actual work done on it.

   An 18.5V "power adapter" lithium battery charger in the Mazda RX7 EV quit. On the trip where I noticed it, the voltage of the whole pack was already down to 12 volts - at my destination point, where I noticed it as I got in the car - and to nothing by the time I got home. They should never be allowed to drop below 14.0 volts, 2.8 volts per (3.2V nominal) cell. The drop was very rapid and luckily it was only short trip, about a mile. It didn't seem to hurt them.
   I soldered a plug back on the 17 volt transformer type adapter I had first bought for them. It had worked well enough, but it eventually brings the voltage up to 22 volts, when the cells should never be allowed to rise above 20. A 15 volt adapter might stay under 20 volts, but then it might charge to 16.5 volts very slowly. This shows the problems with unregulated supplies. OTOH the "17 volts" can be down under 15 when charging a low battery without a series resistor yet without overloading the transformer - a bit of 'constant current' operation, or say a sliding scale of current versus voltage. At one point I had tried to put in a 19 volt zener diode to stop it from going higher, but it got hot very fast. I guess I should buy a new 18.5 V regulated adapter - ug! Or maybe I'll try a "15" or "16" volt transformer type. Or add some diodes to drop the voltage a bit.

   Later the Bradley owner was given, and then gave to me, a Honda Insight hybrid (or hybrid Civic) battery: 120 high rate D cells providing 144 volts in series. I converted its strings of 12 cells to parallel, 14.4 volts, thinking to use it in the Mazda RX7 EV. But my enthusiasm waned: I found out on line that the cells were only 6.5 amp-hours for a total of 65 amp-hours. And it weighed 55 pounds - as heavy as lead-acid. (extra metal and less chemicals, to get the very high rate performance.) It should perform well and last for ages, but with only 65% of the capacity of the other batteries it would likely further limit the already short driving range. Perhaps it could be a 28.8 volt, 33 amp-hour, 'spare' battery for the electric outboard? Then I found another 6 tubes of 14.4 volts on ebay for (by the time I get them) probably around 325$C. That makes over 100 amp-hours.
   Then I got three 12.8v, 40 AH lithium batteries for 600$ locally. It was just too good a deal to pass up. I initially got them for someone else who was to pay me back, but then it looked like he couldn't use them. If I put everything together right, I may not need any golf cart batteries for a 300 AH, 36V vehicle drive system. Instead I can probably use two 300 AH, 12 v (or 14.4 v) batteries of NiMH D cells and one of 12 v worth of lithiums in parallel. That would certainly lighten the car, allow regenerative braking, and of course last longest.
   And apparently I bid on an auction for another 5 tubes of 12 Honda batteries, because on October 2nd I got a notice from e-bay that I had won the bid, for 99¢. At first I thought I must have made some mistake and bid on empty plastic tubes or something. Nope, they were batteries. No one else had bid! The shipping was also 99, but with the decimal point on the other side. Still, it was 5 more tubes, 60 cells, for 1/2 the price of the first 6 and 1/4 the price of new Tenergy cells. From the original 10 free sets I'll be up to 21 - total 252 high rate NiMH D cells and about 500$C.


   I did more peltier cooling experiments. I put a 14 or 15 amp peltier module (62x62mm) into the shallow chest fridge and tried running it at some lower voltages with the lab power supply. Obviously the cold from the peltier wasn't being transferred well into the ice tray, so I soon found and put in a new ice tray, a little shallow but much larger and made of thick (cast?) aluminum instead of stamped from thin pressed sheet metal. It worked much better, but the ice still formed from the cooling bar end, and the inner end never froze over. Then I needed the power supply for other things and I put the fridge back on the solar panels, with the battery charger taking the load at night. It was getting quite cold underneath and around the ice tray, as low as 2.5°c, but it was still over 15° at the far wall near the top. I added a small 12V "computer" fan (with some resistors in-line to keep it to a whisper and a light breeze). It took days for the stuffed-full warm end to cool, but eventually it was all under about 8° or so at the warm end and 5.5° under the ice tray - much better overall cooling. But it was using 130 watts during the day. When I find the time, I should make the microcontroller based control, with (along with various features) a DC to DC converter in it to supply a lower voltage to the peltier for a higher COP. It gets that at night when the battery charger takes the load - and the voltage drops to 11 volts (with the peltier still drawing almost 8 amps), and the fridge temperatures in the morning are no higher than in the day (if not lower) when it's up over 13 volts and 9.6 amps.

   I trimmed and epoxied the steel rings onto the ARM motor coils to make them into "cup electromagnets", and epoxied the coils onto the bottom plate. And I welded the rotor to its hub. Then I designed and made a small circuit board for the optical interrupters that tell the controller where the rotor rotation is at. I had meant to check to make sure there were no shorts of the coil windings to the case before the epoxy was hard - too late! There was one. I didn't manage to break the coil loose with a hammer and a wooden block, but finally I unwrapped a bit of the coil wire and the short vanished. At least the episode showed the coils were very solidly glued onto the bottom plate by the epoxy. Then I did the heavy wiring, and found the fat wires simply wouldn't fit as intended into the case. It was too cramped. I finally unscrewed the cable clamp from the case and let the cable hang out. Every time there was a problem I would set the project aside for a couple of days. I had trouble concentrating on it. It wasn't until near the end of the month I checked out the optics board. It had a problem too - wrong connections from a mirror image defined optical interrupter part. I cut traces and added wires to get it to work, then fixed the design on the computer.


A regular cup magnet. The flux is concentrated in
the gap between the magnet and the outer ring,
providing supermagnet levels of strength in that small gap.


The rotor with overlapping rings that match the stator's "cup electromagnet" rings.
Besides the flowery shape, the other unique feature is in having many rotor "poles", 8, per each 3 phase coils, instead of 2.
This places each ring quite close to the stator electromagnet rings as each phase is activated, for highest torque.
Instead of switching once per electrical revolution, each coil switches 8 times - 16 times per physical revolution.


ARM Motor more or less Assembled

Short video: https://youtu.be/cifPWF1Snr0  -- ARM Motor Running (top cover off)

   On the 29th I hooked it up to the motor controller and ran it with the top off. It started promptly and ran quite nicely in both directions except for the rotor rubbing a bit in one spot (which was worse with the top on), but I won't vouch for performance or efficiency. The optics adjustment was more critical than it should have been owing to the thin little light interrupter tabs on the outside of the rotor. They really should have been wider, and rounded on the outside. There seemed to be a lot of torque ripple, and not all phases had the same strength - again perhaps an optics problem, because it changed after adjustment from one stronger phase to two out of three. If the motor was to be produced the torque ripple might be reduced by adjusting the thickness of the rotor rings, probably by increasing the outer diameters a little. But I'm more keen on trying out the highly promising 'transverse flux' design than making more of this style.

   Running it disclosed some issues with the unipolar motor controller which need to be looked at. Nothing quit or went up in smoke, but the energy return coil got smoking hot. Much of the energy seemed to be going into heating it up! I started thinking that there wasn't enough current flowing in the wires to explain it. Was my simplified controller design wrong? But the pulses to it are very sharp, and I remembered that iron cores are only good for low frequencies, up to a few hundred hertz, not 20 or more kilohertz with a high harmonic content. I may need to make a coil with a ferrite core. And the motor would only go up to a couple of hundred RPM. It seemed to go from zero to drawing 11 amps in a very short space on the rotary control, but turning the control up further, even all the way to the top, had little or no further effect on current and speed. Wasn't this supposed to be a 200 amp controller?

   Towards the end of the month I started looking up 'free energy' devices again. I'm not sure anything else in the sustainable energy field would be quite so valuable. There was an interesting electrostatic "Perpetual Motion Motor" video. I don't think I've seen anything using that sort of motive force before. https://www.youtube.com/watch?v=dGPnxLSgnUI It turned, but the motive force was so slight I couldn't imagine trying to scale it up to get useful power from it.
   Somewhere I found someone who actually seemed to know quite a bit about magnet propulsion, youtube channel "MotionMagnetics". He went into three types of magnet motive force machines in three videos. He mentioned that things were very exacting, and that magnets weren't identical. Small variations in the magnetization of each magnet would make a machine work or not work, probably explaining why no one has successfully produced them for sale after making a working prototype. I heard of an attempt where this happened. He didn't get into the easier to make hybrid types where the machines self-propel around the loop with permanent magnets except at the one inevitable "sticky spot", where an electromagnet or mechanical device assists it past. The electromagnet is said to use less energy than is generated in the rest of the loop.

   But my main target was "lambda ray collectors", which I searched on as "radiant energy receivers". After a week of this puzzle, I finally figured that Tesla, Moray and others probably hadn't captured lambda rays, radiant energy, at all - at least not directly. Most likely they had captured the static electrical charge of the atmosphere. Air has an electric charge that increases rapidly with altitude, amazingly up to 100 volts more positive per meter on a clear day. Lightning discharges clearly demonstrate the existence of airborne ionic energy. But some lesser amount of electrical potential energy must be present in the air when there is no lightning. In fact, people have been electrocuted by the static charge on unconnected power lines, and a long steel cable strung from a helicopter makes a huge spark when it touches ground. Is this what was being harvested, and can it be done again effectively with today's technologies? Moray made special germanium semiconductors (the world's first semiconductor devices) and special vacuum tubes that would be hard to duplicate, but other things might work in their place.
   A passive circuit with antenna, ground, capacitors and diodes will gradually build up a few volts and supply a few milliwatts, but it isn't very effective. You want kilowatts (or at least watts); you want to run your house, but it will only gradually charge a cellphone. (I made one in 2013 - TE News #68. With the antennas I tried it only got up to a volt or so and imperceptible power.)
   It appears that the successful people, in particular Moray, made tuned circuits with a driver - oscillators - and actively pulled the energy, literally, out of thin air with resonant circuits. Instead of getting a few charged ions that randomly drift by the antenna, they are actively pulled and pushed in tune with the oscillations, vastly increasing the interaction of the antenna with the air. There probably are some lambda ray energy receiver types of energy devices, but famous(?) powerful 'free energy' devices like Moray's might well instead be atmospheric charge harvesters. Some puzzling things about these various energy devices started to make more sense to me.

   For the past year and a half I had been playing in the Greater Victoria Concert Band "junior band" with my Supercorder. The conductor liked the sound and it had been well received. It was a small band when I first joined and I was well heard. But this season I tried the "intermediate band" to play some more challenging music. It seemed to me to be going well but on the 30th out of the blue I was asked to leave. Apparently the conductor and one or more of the flute players didn't appreciate the fine tonal and expressive qualities of my instrument (which have earned me many compliments) because it didn't sound the same as a flute. It wasn't quite what their ears were familiar with. Well - their loss!
   I know some like it. One flautist had seemed glad to hear I would be there, and one who was elderly sitting beside me (and perhaps having a hard time sightreading some tricky rhythms) had just the previous week told me "You're a treasure! You're right on and the sound is clear and easy to follow."

   Even in music it seems successful adoption of inventions has a lot to do with timing. Sax was too late to have his creations adopted into orchestras, whose instrumentation had by then become more or less "standardized", but new types of bands were springing into existence, and the saxophone came to be standard fare in some of those, including in the large concert band - pretty much an orchestra of wind and brass instruments with considerable percussion. (...itself perhaps a timely invention of JP Sousa?) I suppose if Sax had invented them today they'd be quite unwelcome there. There would be no parts written for them and the unfamiliar sound would probably be considered crass. If on the other hand he had made them 50 or 100 years earlier than he did, they might well have become mainstay orchestra instruments. (and perhaps clarinets would have been dropped?)
   Likewise, if the recorder had had my improvements by perhaps by 1850 instead of 2006, it just might have been the transverse flute that would be considered "airy" sounding and "parochial", and might have been left "unmodernized".


   The earlier part of the day had gone better. Having borrowed a surplus tunable air core coil and a tuning capacitor the day before, I picked up books on LC oscillator and HF transmitter design for the atmospheric charge energy harvester and (bonus!) an evacuated tube heatsink module that I can try out with the peltier camping cooler. And my vacuum jar for one of the steps in making the ethylene glycol/choline chloride DES battery electrolyte arrived by post.



In Passing
(Miscellaneous topics, editorial comments & opinionated rants)

A Perspective on the Syrian Crisis

   It's hard to form an informed opinion on what's been happening in Syria from the occasional disconnected reports with no background context in the western media, which give no overall view. Until refugees started flooding into Europe recently, virtually no mention was made that over 1/2 the entire population had been driven from their homes by war and bombing, with towns, utilities and industries reduced to rubble over the past 4 or 5 years. Listening to Syrian Girl, who has been occasionally reporting from Syria on the events and politics for some time now, may broaden your perspective, if not change your views considerably.

https://www.youtube.com/watch?v=pHFnvFbThDE - Syrian girl

NO MORE WARS!

Loose Change

   I asked in a coin shop how much old (Canadian) silver quarters were worth. These were made from 80% silver and 20% copper from 1920 to 1967. (The alloy was harder than pure silver and so the coins lasted longer.) I found out that a dollar as coins had .6 troy ounces of silver, regardless of denomination. Thus a silver dollar had .6 ozt, a 50¢ piece .3, a quarter .15, and a dime .06 ozt. (A troy ounce, or Trojan ounce as I like to call it, because somehow it seems sneaky, contrived), is about 31.1 grams, where the "real" Avoirdupois ounce is 28.something grams.)

Virtually unrelated: It has been found that an alloy of silver and germanium makes a "sterling silver" that doesn't tarnish. (The patent has probably expired.) This might have made better coins for circulation!

   The price for any and all of this change was thus denominated as the price per face value dollar, ie per .6 ozt of actual silver content. At that store it was 13.50$. This is a good indication of how far inflation has gone. In 1920 the silver content of the coins would have been pretty much a token amount. Coins were probably also the main form of exchange, since prices of most small items were in pennies. Paper bills were for large purchases. (Even in 1960 kid's bus tickets in Edmonton were 6 for a quarter - 4-1/6¢ each.) By 1967, Canada's 100th birthday, the face value of each coin had become worth less than this "token amount" of silver. Oodles of US and Canadian silver coins were melted down for their silver value. An old quarter or dime in change has been a very rare sight for decades now.
   Today four "nothing special" silver quarters or ten dimes - change for a looney - will cost you 13.50$ or more to buy. So the dollar is now worth only 7% of what it was worth in 1967. And since 2011 the price of silver has been manipulated down with "high frequency trading" of huge, uncoverable "naked shorts" (banks selling silver they don't have on the futures market) to make the dollar look better, dropping from over 40 US$/ozt to under 12. The 2011 high price would indicate today's dollar is under 2.5% of what it was in 1967. In 2011 silver may have been overvalued. It is certainly undervalued today. The best guess is somewhere in the middle, that the dollar is worth 3 or 4% of what it was in 1967.
   Backed by nothing and printed as notes or simply conjured into existence on banks' computer screens ad infinitum, the whole world's fiat currency is headed for zero. It is not necessarily that fiat currency can't work, but that those in control of the supply have always opted to print more, gradually transferring the value of peoples' savings to themselves as new tho decreasingly valuable money, rather than face budgetary restrictions. Every fiat currency in history has ended up being worth zero.
   Silver on the contrary is a tangible asset, whose value may fluctuate in accordance with supply, demand and sometimes market manipulation, but which owing to limited (and presently declining) mining supply and electronic device market demand on top of investment demand, can never fall too much... unless some fantastic new silver deposits are discovered or "cold fusion" can be used to create far more. (That would be fabulous, since it's about 12% more conductive than copper for making motor coils - the most conductive of all elements by size. But the chances seem remote.) It will therefore tend over time to at least stay even with inflation, rising in dollar price as the value of paper money falls. Silver and gold have thus long been viewed as "safe havens" for storage of wealth, especially in volatile times, and this is better understood over much of the world than it is in the west after our long period of relative stability and prosperity.

   As I write, the supplies of silver and gold seem to be running short. Month after month there is new record demand from Asia and from a million small investors who wish to cash in their claim checks on wealth (dollars) for things outside the banking system that won't inflate away and can't be confiscated with "bail-ins" and "capital controls" (where you can only withdraw a very limited amount of your dollars each day), as is starting to happen here and there. The CEO of Sunshine Mint, a main coin blank supplier to the US Mint, said in an interview that shortages so far are owing to production time to stamp out the record demand for coins and bars (they're running 24/7 and have tripled their production capacity since 2007), but he also said that they are now pulling old silver ingots out from the backs of their vaults, and that it takes weeks to get more in instead of a day.
   At some point everybody will catch on and there'll be a run. If you don't go to a coin shop or go online and get some 'soon', when that run starts there won't be any at any affordable price. (with no known date on 'soon' - maybe within weeks but quite possibly months, or even a year or two.)

   Some day there will be no more money. World population will be controlled by common consent (2 to 3 billion?), and everyone will have what they need for a good quality of life with free time to learn and grow. Instead of being permitted to earn a certain amount, people - everyone - will be permitted to have a certain amount of possessions and supplies for their personal use. No one will be permitted to hoard unreasonable amounts of goods which would result in scarcities for others. There won't be a small clique commanding most of the resources, and natural resources will be considered common property. The whole mentality will be more social - less "me first!", more "consideration for all but without neglecting me, too." Since everyone will be permitted to own reasonably similar amounts, and since those amounts are sufficient for all to live a good life, all will agree to this system. Those whose talent and enterprise provide large social benefit will be permitted somewhat more than the average - perhaps 3 or 4 or more times as much as the "poorest" or basic allotment: the 'profit motive' or reward is not to be discarded. But no one will have a hundred, thousands or a million times as much as most others. Everyone will do perhaps 3 or 4 hours work per day to maintain this sustainable social system, and then fill their days with more creative or spiritual pursuits or pastimes. It will probably take a thousand years to get there, but in the million year march of human history... that's 'soon'.

Perpetual War

   Someone made this list of countries the USA government and or its "corporatocracy" and or it's military-industrial-financial complex has struggled with, covertly or overtly interfering in its affairs through the military or various government or "private" agencies, since world war two. I don't vouch for its accuracy - some of the dates given seem arguable. And I don't know what the asterisks are for. (Probably this is just a copy of a list that originally gave some context.) Ten years of peace might transform the face of the USA if it allows that to happen - and perhaps the face of some beleaguered countries where "regime change" keeps getting instituted!

China 1949 to early 1960s
Albania 1949-53
East Germany 1950s
Iran 1953 *
Guatemala 1954 *
Costa Rica mid-1950s
Syria 1956-7
Egypt 1957
Indonesia1957-8
British Guiana 1953-64 *
Iraq 1963 *
North Vietnam 1945-73
Cambodia 1955-70 *
Laos 1958 *, 1959 *, 1960 *
Ecuador 1960-63 *
Congo 1960 *
France 1965
Brazil 1962-64 *
Dominican Republic 1963 *
Cuba 1959 to present
Bolivia 1964 *
Indonesia 1965 *
Ghana 1966 *
Chile 1964-73 *
Greece 1967 *
Costa Rica 1970-71
Bolivia 1971 *
Australia 1973-75 *
Angola 1975, 1980s
Zaire 1975
Portugal 1974-76 *
Jamaica 1976-80 *
Chad 1981-82 *
Grenada 1983 *
South Yemen 1982-84
Fiji 1987 *
Libya 1980s
Nicaragua 1981-90 *
Panama 1989 *
Bulgaria 1990 *
Albania 1991 *
Iraq 1991
Afghanistan 1980s *
Somalia 1993
Yugoslavia 1999-2000 *
Ecuador 2000 *
Afghanistan 2001 *
Venezuela 2002 *
Iraq 2003 *
Haiti 2004 *
Somalia 2007 to present
Honduras 2009
Libya 2011 *
Syria 2012
Ukraine 2014
NO MORE WARS!

Advantages of hard assets for wealth retention when things crash - from 2002 after the 'dot com bubble' collapse (Author unknown)

: If you had bought $1000.00 worth of Nortel stock one year ago, it would
: now be worth $49.00.
:
: With Enron, you would have $16.50 of the original $1,000.00.
:
: With Worldcom, you would have less than $5.00 left.
:
: If you had bought $1,000.00 worth of Lone Star Beer (the beer, not the
: stock) one year ago, drank all the beer, then turned in the cans for the
: 10 cent deposit, you would have $214.00.
:
: Based on the above, my current investment advice is to drink heavily and
: recycle.

Caught Red Handed!

Since the US government has withdrawn the border guards and left the Mexican border wide open, citizen militia have started patrolling the border. One group chased down and stopped an SUV with two men that had come across. It was full of cocaine. The men pulled out CIA employee ID cards. But officials at the Cocaine Importing Agency, the world's largest and oldest drug running group, denied any association with the vehicle and men.

Prime Minister's "Famous" Great-Great Uncle (and black is white and white is black! Received in an e-mail.)

Judy Harper, an amateur genealogy researcher in Northern Ontario, was
doing some personal work on her own family tree. She discovered that
Prime Minister Stephen Harper's great-great uncle, Remus Rudd, was
hanged for horse stealing and train robbery in Winnipeg in 1889. Both
Judy and Stephen Harper share this common ancestor.

The only known photograph of Remus shows him
standing on the gallows at the Manitoba Provincial Jail.



On the back of the picture Judy obtained during her research
is this inscription:

'Remus Rudd horse thief, sent to Stony Mountain Jail 1885,
escaped 1887, robbed the CP AND CN trains six times.

Caught by Mounted Police Force, convicted and hanged in 1889.'


So Judy recently e-mailed Prime Minister Harper for information about their great-great uncle, Remus Rudd.

Believe it or not, Harper's staff sent back the following biographical sketch for her genealogy research:

"Remus Rudd was famous in Ontario during the mid to late 1800s. His
business empire grew to include acquisition of valuable equestrian
assets and intimate dealings with the CP and CN Railways..

Beginning in 1883, he devoted several years of his life to government
service, finally taking leave to resume his dealings with the railroads.

In 1887, he was a key player in a vital investigation run by the Mounted
Police Force. In 1889, Remus passed away during an important civic
function held in his honour when the platform upon which he was standing
collapsed."

NOW That's how it's done, Folks! That's a real POLITICAL SPIN!

(Author unknown. Hmm, I'm sure I've seen that before somewhere... where some "notable"(?) died tragically in some apparently freak accident "when the platform he was standing on collapsed" at some public function.)


I only caught the last half of the reportedly excellent total eclipse of the moon,
but the next night there were some really interesting clouds lit up by the moon.
The photo does them little justice. Below are lights across the bay reflected in the water.



Newsletters Index/Highlights: http://www.TurquoiseEnergy.com/news/index.html

Construction Manuals and information:

- Electric Hubcap Family Motors - Turquoise Motor Controllers
- Preliminary Ni-Mn, Ni-Ni Battery Making book

Products Catalog:
 - Electric Hubcap 7.2 KW BLDC Pancake Motor Kit
 - Electric Caik 4.8 KW BLDC Pancake Motor Kit
 - NiMH Handy Battery Sticks, 12v battery trays
& Dry Cells (cheapest NiMH prices in Victoria BC)
 - LED Light Fixtures
 - CAT 12 volt DC plugs, sockets, wall receptacle plates...

(Will accept BITCOIN digital currency)

...all at:  http://www.TurquoiseEnergy.com/
(orders: e-mail [email protected])



Daily Log
(time accounting, mainly for CRA - SR & ED assessment purposes)


September 1: Worked on August newsletter
2: Finished and posted it.
3: -
4: Started voltage versus cooling experiment in 12V thermoelectric fridge. ARM motor - Epoxied steel rings around coils. (Got 2 chickens. 1 egg first day, 2 the next, 0 for a day then 2 more... Hmm, chickens eat a lot more than fish. Then no more at all after the first dozen. Laying shut off like a switch, both birds! ...Roast chicken next perhaps?)
5: ARM - Trimmed sheet metal rings and sanded coils to a semblance of flat and even. Epoxied them to motor base plate.
6: Touched up epoxy job. Designed circuit board for rotor position sensor optics.
7: Printed out, laminated, etched, drilled and populated circuit board for motor. Made an "L" bracket mount for it. Wire brushed top and base and painted them with urethane spray paint.
8: Mounted and tried to adjust board so optical sensors didn't hit rotor. Rotor wasn't quite flush -- it hit both top and bottom during a rotation. Welded rotor to hub. Two tries. (That didn't help the flushness.)
9: Tried to adjust rotor alignment with minimal success. Made cable for rotor position sensor (triple optical interrupter) board.
10: Populated PCB for LED light and restored first LED grow light to operation. (It still runs too hot!)
11: Made a video about (world's most?) water conserving shower nozzle, the Waltec 10C, which uses less than 1/2 the water of any other "water conserving" shower nozzle. (How about a good shower with 5 or 4 or 3 liters per minute?) It appears to be long out of production. It should be brought back! Or re-invented.
12: Wired the pairs of coils together. Tested and found a coil was shorted to the case. Ouch! (Rats! -- I meant to check for that while the epoxy was still soft!) Installed 14 amp peltier module in thermoelectric fridge. (Also see 15th, 17th, 23rd. Voltage trials went on a few days.)
13: Tested to find which coil of two. Attempted to remove coil. I might have to smash it to pieces. Changed Peltier Module on solar fridge to large size 15 amp unit and started cooling tests.
14: A few Peltier fridge temperature measurements. (Heat transfer to ice tray is poor.)
15: Bought rectangular aluminum pan with thick bottom & sides, hoping it would make a better ice tray for the thermoelectric fridge. Repaired shorted motor coil.
16: -
17: Installed the tray.
18: Did the heavy wiring in the motor - it doesn't fit - can't close the case.
19: Pulled the connector off the case and pulled the wires out a bit. Just fits! With the case closed, the rotor rubs at least a bit somewhere, but I'll try it out as is.
20-26th: Studying "free energy receiver" circuits and writings.
23 (& on): Placed small fan to blow air in Peltier 12 V fridge. Trials continued into October. Major improvement in overall cooling.
27: Tested motor optics. Optical interrupter part had mirror image LED.s. Sigh! Cut and soldered wires to get board working, redesigned part & PCB. At last discerned that the "free energy" as collected by Tesla, Moray and others actually does come out of thin air. Air has increasing charge with altitude, and their oscillating devices could apparently collect that energy far more effectively than the passive "cell phone charger" system with only diodes and capacitors.
28: Designed a tentative schematic for capturing air energy charge.
29: Got reluctance motor running. (YAY!) Filed off a coil ring where the rotor was rubbing. (It would work better if the rotor was dead flat and straight!) Did a revised air energy charge harvester schematic and picked up a tunable coil and tuning capacitor for air charge energy device.
30: Tried a few more things with reluctance motor. Obtained some books on oscillators and radio transmitter design, and (bonus!) a vacuum pipe heat dissipator intended for computer CPU.s, but just the right size for a typical 40x40mm peltier module. (I'll try that with the camping cooler some time, seeing the new peltier in the thermoelectric fridge is 62x62mm.)



Electric Hubcap Motor Systems - Electric Transport

ARM Motor


The ARM (Axial flux switched Reluctance Motor)

   On the 4th I did my best to adjust the steel rings I made and rolled up (now months ago) for the ARM motor coils to turn them into "cup electromagnets", and I epoxied them onto the coils. There was lots of trimming to do before the coils were anything like flat and the right height. Much as I feared, attaining the exacting dimensions for .025" flux gaps is virtually beyond my ability for largely hand-made assembly with epoxied wires and curved sheet metal parts. So the gaps will be larger and hence the specs not as good as might otherwise be expected. There will have to be some major assembly simplifications and quality improvements to the process if this motor is to be produced. (The transverse flux design seems more promising.)
   On the 5th I trimmed the rings with tinsnips and then carefully sanded them, then painted on some more epoxy (a couple of the rings were loose at the end) and let it set. In the evening I numbered the coils and positions and ground out dips in the base plate in places on 3 coils where where flattening them had sand them down to bare wires in spots. That should keep them from shorting to the plate.

   I had decided simply to epoxy the coils onto the steel bottom plate, and then build up the epoxy a layer or two, and maybe a little polypropylene cloth, for strength. This is safer than simply epoxying magnets to a rotor, since the stator doesn't spin - there's no centrifugal forces and it doesn't have to be balanced. The day-and-a-half old epoxy (kept in the freezer to prevent rapid setting) seemed rather thick, so I just painted one more layer on when the first coat had set.

Motor Base with "Cup Electromagnet" Coils

   Next came the optical rotor position sensing system. I thought the 3 optical interrupter sensors had to be placed at 60 or 120° intervals around the outside edge. That's how I did it with the magnet sensors. That precluded putting them on a single circuit board. In fact, the available space was a little cramped and they would have to be mounted at an angle instead of straight on. That would keep them inside the outer case ring so I didn't have to cut holes at 3 exact spots and then line it up accurately, and also let them miss the case assembly bolts using existing holes.
   Then it occurred to me that with 8 overlapping rings per 3 coils, the interrupters could be much closer together -- in any three positions where they lined up in a proper 3-phase spacing to the rings, which were about 42mm spacing, so at 0, 14 and 28mm. Perhaps I could drill one or two new holes and mount an assembly with the three interrupters? When I checked, it looked like that would work.
   I measured about 1.65" from one ring to the next at the outside, so 2/3 of that would be 1.1", with the 1/3 position at .55". (which is, sure enough, 14mm) It looked like the board should be about 1.75" tall to mount with a small "L" bracket from the bottom plate and hold the 3 interrupters. (LED + Phototransistor pairs facing each other in a housing with a 5/16" gap between them - ordered from somewhere on aliexpress.com). Make it 1.4" wide to cover the 1.1" plus connection wires plus margins. There was the basis for a PCB design. Making a PCB seemed the best way to go. I've developed an aversion to "cludged-in" component constructions.
   I designed it on the 6th, then printed, 'laminated' it, etched it, drilled it, and populated it on the 7th. And made an "L" bracket mounting to hold it u to the rotor. I managed to forget to print it mirror image on the first try, but I decided to do a couple of flat panel LED light boards at the same time, and the reversed lettering on them prompted my memory before I etched it. I then proceeded to print it out the right way around... on the back side of the toner transfer paper. I didn't notice until the paper wouldn't release. I patiently soaked it off, worked it up and used it anyway. The board came out with a few paper fibers (which I had tried to rub off) shorting close gaps at the optical interrupter modules, seen (by 60 year old me) only under a magnifying glass. I scraped them away with an exacto knife. There were "micro-pits" in the copper everywhere, but all the connections seemed good.

  
Mirror image board - Motor optics and LED light boards ready to etch


Finished (except for wires/cable) Board


Optics in Motor (Light from behind shows copper traces through PC board)

( EAGLE PCB Files: Schematic , Board , Optical Interrupter "lbr" - corrected but untested as drawn )
(The 5/16" optical interrupters were from somewhere on aliexpress.com )

   When I wound the coils, I left the first wire end short (it tends to get in the way while winding) and the second end long. I thought it would average out. But on consideration, I had got more torque in earlier experiments if the two coils of a phase had opposite polarity. The "cup electromagnets" are supposed to negate this difference, but I expected it'd still work somewhat better with opposite than with like polarities. This meant matching the short wires together, or the longs. Either way I would need to extend the short wires, either to connect the them across inside or to make the external connections with them. Oh well. I decided to use the long ones internally and attach heavier stranded wires to the short ends for the external connections. (This caused problems when the fat external wires, and the clamp holding the cable, simply wouldn't fit in.)

   On the 8th I got out the little MIG welder and welded the rotor to its hub. I'm definitely no pro welder. Someone feared it might break off at high RPM. I hadn't yet put the welder away, so I went out again and added lots more equally ugly weld. The poor carpenter blames his tools, but I suspect the cheap, 120 volt welder is too light for these heavy pieces. But then personally I'm not sure I'd do any better with the old 230V stick welder either. Then I ground off the ugliest bumps with the angle grinder and it didn't look so bad.





First and Second so-called Weldings.
The second one looked somewhat better after some grinding to smooth it.
Is that the trick?

   Then I went to try and true up the run, to eliminate the wobble, by bending the rotor plate just a fraction. It grazed the bottom of the optical interrupters at one point of rotation, and the top 1/2 way around from there. I spent a lot of time without accomplishing much. It would be off at different points each time I tried bending it. The next day I checked again. I found I had bent up an actual bump, a raised area, to add to the problem. I went out to the big vise and pushed - one more time after so many - hoping to bend out the bump. Miraculously I did, and the whole alignment was virtually perfect all the way around. Wow! Quit while ahead!
    On thinking about it, it's possible there was some speck of grit between the two upper rings of the thrust bearing (I added the second ring to increase the clearance), and that that rather than the rotor caused the wobble, with the 'random' bad alignment rotations. (I did casually wipe them off once, but without apparent effect.)

   After a lot of work drilling and filing holes in the "L" bracket and much adjusting, I found that the optics assembly had to be removed to install or remove the rotor, and then needed realignment. And the head of the bottom bolt was inaccessible until the upper one was removed, and the upper one had a nut, hard to reach, that made it hard to loosen and tighten for adjustments. It occurred to me that if the bracket was threaded so the bolts were put in in the other direction it would all be much simpler. Easier to do it now than after further frustrations and possibly breaking something. I replaced the hardware store "L" bracket with a new one made out of 16 gauge nickel-brass -- with holes that were where I wanted them, and threaded. (As I gave it a final sanding, I lost my grip and it shot out the back of the bench belt sander and fell behind the bench, never to be seen again. I finally gave up looking and made another one.)
   With the rotor straight and a bracket that would stay where it was positioned to, the rotor spun true in the gap between the optical elements. But some imp must have come along and bent it again while I wasn't looking, as it ran crooked again later. (Or was it just a fleck dust somewhere... or more likely, a fleck dust the time it ran straight?)

   On the 12th I wired the pairs of coils in series. Then, belatedly, it occurred to me to make sure no coil wires were short circuited to the bottom plate. I had intended to do this while the epoxy wasn't set yet, but had forgotten. Now there was a penalty for that omission: there was indeed a short, and the coils were solidly epoxied to the bottom plate and the pairs were wired together. In order to identify the coil, I ran a 1.0 amp current through the offending pair, and checked the voltages to the plate to see which end was closest to the plate. One end definitely was closest, the coil I'd marked "5". I pried off the outer ring of the coil, and scraped away as much epoxy from around the base as I could. But hammering it via a block of wood wouldn't budge it. It looked like I'd have to smash it to get it off. I set it aside for a couple of days. Perhaps an inspiration might come?
   It did. I thought to unwrap just the outer bottom winding. Maybe that would make take away enough epoxy grip that I could knock it loose? But when I started, I only unwound a couple of inches, then thought to check again. The short was gone! It must have been right there at the start of the coil... or else it had been shorted to the outer ring. (Maybe I hadn't needed to do anything but remove that?) I scraped the stuck epoxy off the ring, mixed a very tiny new batch, and epoxied it back on. I put a heavy weight on it hoping to keep it down below the rotor while it set. By night the epoxy was hard - and there was no short. The rotor spun freely again after I filed off the epoxy and metal ring inevitably sticking up a bit. Well, that was certainly easier than breaking off the coil and making and installing a whole new one! The three days lost were of course fully occupied by other things.'

   At least the episode demonstrated that simply epoxying the coils on held them quite well. And the current test for the short incidentally disclosed that the DC resistance of each phase was about 30 milliohms. That's 1/2 the resistance it would be if the phases were used in "Y" configuration, going across two phases instead of having each phase tied to B+. As I recall, the Electric Hubcap motors of the "standard" BLDC configuration with #11 wire coils were about 67 milliohms. This should result in overall copper losses being quite similar, notwithstanding that only one phase is energized at a time instead of two, and so (all else being equal, which it surely won't be anyway) more current will probably be required to attain the same magnetic force.

   I started studying free energy devices rather intensively (along with getting more batteries and doing various things that needed doing) and it wasn't until the 27th that I hooked up the motor optics to the controller and tested it. It turned out that the LED.s were defined backwards in the part "library", so it didn't work. I cut traces and put in wires to make the connections, then fixed up the part and board designs. Once corrected, the unit seemed to operate just right. (The little tabs I put on the outside of the rotor were quite thin, and I had been afraid one or two of the outputs would be blocked too much or too little and wouldn't change state at the right times. Once the motor was running this did prove to be the case and adjustments were required.)


Motor with top off. I ran it like that, and without the rim, too.

Short Video: https://youtu.be/cifPWF1Snr0 -- ARM Motor Running

   On the 29th I bought some connectors to put on the ends of the heavy wires, hooked it up, and (after 5 tries out of 6 possible ways to connect the 3 wires) the motor ran. It seemed to use too much current to spin with no load, but it got up to about 195 RPM with 10 amps at 30 volts. Soon I disconnected the lab supply which only goes to 10 amps and hooked up 24V of NiMH D cells in two 12V tubes.
   But turning the control up higher seemed to have no effect - only about 12 amps of current was drawn, and the speed didn't rise. There was only a small area on the rotary control where the motor went from zero to this maximum. This is probably mainly a controller issue. I hadn't turned up the BLDC 4:3 motor above about 10 amps because I just ran it on the 10 amp lab supply (and it got up to good RPMs with that), probably explaining why I hadn't run into it before. After a time the coil got smoking hot, and the energy return diodes were also pretty hot. The switching mosfets were only slightly warm, and the motor was hardly warm. Before, I thought that it just needed to be a bigger coil with heavier wire. Now it occurred to me (again, as with the free energy device coil and frequency) that the frequency of the switching, and especially the sharp switch-off pulse spikes, was probably too high for an iron [iron powder] core. Surely it needed to be ferrite or something. Definitely much of the power was going into heating up the coil, not running the motor.

   I found there was a lot of torque ripple by holding the rotor back, allowing it to slowly rotate. In some spots it could be easily held, but it would jump through some others with far higher force. I could see as I was putting things together it wasn't going to achieve even torque, the "phase ON" start position being farther from coil alignment than I expected from the cardboard trials. I still think this can be evened out by adjusting dimensions, even if the weaker areas are only increased at the expense of the maximum torque.
   When I went to check the actual levels of torque, I discovered the shaft I had used had no keyslot to attach the device. So I've left it for now.

   One phase seemed to have substantially more torque than the other two. This was a puzzle since the coils were all the same. Then I loosened a screw on the optics board so I could move it around a bit. Side to side made little difference, but moving it in and out changed things markedly. Moved out, the motor stopped entirely. Moved in, the RPM appeared to about double, still with the same 11 amps current. Of course the middle phase interrupter was slightly closer in than the other two, with the flat circuit board not quite matching the curved rotor. Presumably it was full strength while the other two were getting light prematurely and switching at the wrong times. But I adjusted the optics to a new position and was still getting higher currents and torques for two phases than for the third. The whole thing still has some puzzling features, including that the interrupter frequency shown as a number on the oscilloscope sometimes seemed different at the same RPM (perhaps was a harmonic) - and didn't match the actual waveform seen. Later all three phases seemed to have the same strength.

   Since I've already found a better design I don't plan to make any more of this motors. On the other hand, by running this one I might learn some more about reluctance motors and about my own implementations of them in particular before making the next one. And the motor controller, which I hope to use for any and all of my future transport-size motors, definitely needs at least some "tuning up" in the coil - and the synchronous rectifier drivers, and it definitely doesn't run this motor as well as the BLDC4:3 type, so the interactions need study.



"Transverse Flux" Motor

   A probably better way of using the "hole saw" drum came to me watching a video of a commercial reluctance motor. The outer rim of the drum is only 1/16" thick or so. Instead of attaching the "horseshoe magnet shorting bars" across the outside face, they could be attached on the inside. With the thin metal elsewhere, that would provide enough of a difference that it should work pretty well. Anyway that seemed to be how the commercial one was done.
   That would be a big relief from a design standpoint, because if the 'bars' are on the outside, the diameter for designing the stator would depend on the thickness of the bars. The bar thickness would be critical and at the same time might easily vary owing to welding or soldering variances, and any change in thickness would mean redesigning and rebuilding the stator. On the inside, slight variances won't affect the dimensions and the flux gaps. I can now start designing the stator for the drum's exact outer diameter, without fearing that diameter will need changing at a whim.
   With each bar resting at the drum's bottom, they should be easy to silver solder on - they won't try and shift or fall off. Or maybe I can just epoxy them on?



Other "Green" Electric Equipment Projects

Water Conservation: Best Shower Nozzles and Video

   I have what appears to be a "well kept secret", or rather a piece of useful knowledge that just never spread: shower nozzles capable of delivering a fair spray with under 3 liters of water per minute, and a good stiff spray with 4 or 5. Well, maybe 6. A so-called "water conserving" shower nozzle is considered to be one that uses 9 liters per minute, so this is 1/2 the water usage.
   It happened that I bought a "Waltec" bathtub faucet set in about 1980 or 81 (or even earlier?) in bathroom renovations. The faucets and spout were nothing special, but the shower nozzle I would call "clearly superior".  Instead of having the water come out around the edge and through "watering can" holes here and there, it all comes out sideways behind the adjustment knob from a central point, spraying into curved triangular grooves that spread it out and aim it down, and not all the same (which would form a single ring of water) but with varied ejection angles.
   I soon went out to the same plumbing store and bought another one for the other shower, specifically asking for that same part number, a "Waltec 10C shower nozzle" (IIRC the number), and one was produced from the back of the store. The clerk even seemed familiar with that specific number without looking it up - something someone else had asked for? I've never seen anything like it before or since, in spite of looking at shower nozzles in stores on several occasions. "Better mouse traps" seem to often be invented and then go out of production and are forgotten, for reasons that seemingly have nothing to do with performance. Whenever I shower away from home, such as at a public shower (pool or campground), I'm aghast at how much water gushes out, yet often with a rather feeble spray - such a waste of energy and water! Surely they could do better in showers that are in use all day every day!
   But no. As far as I can find, they haven't been available for decades, and there's nothing else like them. I think the original mold and design needs to be found and returned to use, or it needs to be be reinvented.

   The nozzle originally made a soft spray with a minimum of about 4 liters of water per minute. I added a final touch: a small rubber washer around the adjustment knob, the same diameter as the knob. By closing the adjustment down just a bit more, the start of the already small grooves is closed a bit more by the rubber and the most water conserving operation mentioned above is possible, still with good spray.
   I must add here that some people don't like it and open the adjustment wide to get "typical" higher water usage. That too can be accommodated. But I can hear that they just open the taps wide open to start, and then of course the spray is much too hard when they pull the shower knob. They haven't figured out that a good but water conserving shower requires setting the water initially to a pretty low level. (These same people also don't pay my water or power bills.) It would no doubt help if faucets were made that make it easy to make small adjustments to the flow. (I think those single knob bathtub faucets that don't adjust at all - full blast or nothing - are the acme of a bad idea!) Or perhaps reducing the water supply would make the adjustment range of the taps finer. As it is, when I budge the cold tap just a smidgen (...did I actually turn it, or not?), the temperature goes between too cool and too hot. The hot tap has somewhat better range, if only because my hot water isn't turned up too high.
   Another factor is simply what people are used to and what they expect. If one is expecting a huge gush of water, a strong, finer spray probably seems weird, and people may start adjusting to get something more familiar, more in line with expectations, without critical examination of what might actually be better.

   I started out by deciding to do a video about water conservation quite a while ago, and then to cover the shower nozzles specifically. I still haven't posted the first one, but here is the URL for the shower nozzle one: https://www.youtube.com/watch?v=Qs6iKSch2sI .



Peltier Module / Thermoelectric Cooling Experiments

   As voltage on a Peltier module is increased, so is the current, so the power consumption is related more to the voltage squared than to the linear increase. The fridge has been using a 15V, 8.5 amp rated module, which values yield 128 watts. But the calculations aren't so simple because the current drops as the temperature difference rises between the hot and cold side. The experiment with changing voltages on the 12V solar fridge showed that there was considerably more cooling at 9 volts than at 8, and at 10 volts than 9, slight gain going from 10 volts to 11, just a little from raising that to 12, and apparently none going up from 12 to 13 volts. 14 volts seemed, if anything, slightly worse than 13.
   And it turned out that the voltages at the peltier itself were notably lower than at the power plug, with some resistance in the wires and connections. I had been content to use the volt meter on the power supply, while realizing that there would be a little more drop across the wires and the control relay before the peltier modules. When I returned the supply to the solar panels I belatedly measured the solar voltage at the socket (13.4 volts), and then at the leeds of the peltier module/TEC ('thermo-electric cooler') itself. It was only 11.6 volts - a 1.8 volt drop! That meant that my results were skewed - that the voltages I was comparing were actually 10-15% lower than what the meter said. (Also that I should be using heavier wires... or long, light wires to get to a desired lower voltage!) Another unreliable figure is the interior temperature: the fridge is in use, mostly for canned milk for coffee and mixed nuts and things (plus cool storage for many items that are rarely used). Every time the lid is opened, the temperature rises perhaps 1/2 a degree, and it takes some time to return to the pre-opening figure. Plus if a warm item is placed near the sensor it warms up the vicinity.  I present the table below derived from 2 pages of measured figures over 6 days from Sept. 3rd to 9th, and later (9th, 10th) measured real voltages at the TEC's input leeds for the given supply voltages, and otherwise cleaned up and estimated, with various similar readings averaged out.

When (2015/09/--)
Volts (at actual Peltier
module
connections)
Amps
Power
(TEC actual)
Room°
Fridge°
(Cold end)
Notes
/03 - 9:00 AM
11.8 v (solar, at plug)
(~10.3 v at peltier)
-
-
-
6.0
Initial reading. Then set to 8.0
volts (at power supply!) at 9:30
 - 11:20
7.3 v
3.0 a
22 w
20.0°
8.1°
warming. Ice in ice tray melting.
Set to 9.0 volts after readings.
 - 15:00
8.0 v
3.7 a
30 w
20.0°
6.6°
Ice still melting.
Set to 10.0 volts after readings.
 - 24:00
8.8 v
4.1 a
36 w
21°
6.5°
Melting of ice much slowed.
Set to 11.0v after readings.
/04 - 9:00
9.7 v
4.5 a
44 w
20.5°
6.1°
More ice (still < 1/2 tray)
Set to 12.0v at 18:40 PM
 - 21:30
10.6 v
5.0 a
53 w
21°
6.0°
More ice - headed for 1/2 tray.
Set to 13.0v next day at 15:30.
/05 - 15:30 to
/06 - 9:00
11.6 v
5.4 a
63 w
21.5°
6.3°
A bit more ice. (note > room°)
Set to 14.0v, /08 at 8:45 AM.
/08 - 12:15
12.5 v (est)
5.9 a
74 w
23°
6.5
Temperature not dropping
(but room temp higher)


   Referring to the table, the 8 volt setting (delivering just 7.3v to the TEC, <50% of  its rating) was obviously not doing enough cooling and I raised it a volt to 9 after about 2-1/2 hours. That, delivering 8.0 volts to the module, still had the ice tray gradually melting, and I raised it another volt to 10 (8.8v at peltier) after 3-1/2 hours. That reduced the melting speed to a crawl and I left it there 9 hours, but it was still a warming trend. At 11 volts (9.7 at peltier) more ice started to form. 12 volts was a small improvement, but going to 13 (still only 11.6 at the TEC) seemed to make minimal difference. I did some more readings at 11 volts on the 6th to 8th. The room temperature had risen with warm weather, as high as 24°. The fridge warmed up by a somewhat lesser amount, mostly under about 7.0°.
   On the 8th I tried 14 volts (~12.5v at the module) just for completeness. As expected, it seemed to be little better, if at all. The heatsink and fan work too hard to dissipate 74 watts of input heat energy plus about 40-45 watts of heat transferred from the cold side, so the warm side temperature gets much warmer than room temperature, raising the temperature differential the heat has to be pumped across. Going to the higher levels was just wasting energy and worse, probably was a major contributor - if not the main contributor - to shortening the lifetimes of the modules.

   The next thing to try was to install the big 14 amp peltier module, 62x62mm, that I got from somewhere on AliExpress.com (TEC12715... maybe it's 15 amps?). That could certainly be run at lower voltages (8 to 9v?) and hence in a more efficient range, to presumably deliver the same cooling using lower power. Or (9 to 11v?) substantially more cooling at the same power.
   Perhaps it could even get the whole fridge nice and cold. Then I could put in a fan to blow the air around for even all-round cooling. It might even allow the intended operation mode of turning the fridge mostly off at night (or whenever the 12 volt supply voltage was lower) and doing extra cooling during the day (whenever the supply voltage is higher indicating the batteries are charging), ideally freezing the whole ice tray while the energy was coming straight from the solar panels (or other periodic power source [wind?]).
   I installed it on the 12th. When I started to disassemble the old setup I found it had made lots of ice overnight and the fridge was surprisingly (after the temperatures in the above table) under 5°c. Power would have been at about 11v overnight and then 13.5v during the morning. The only thing that seemed to account for it was less opening of the lid.
   I also found that the switch was hot and also some of the connectors. I got rid of it all - switch, Chinese control unit with relay, and some skinny wires - and left just the plug, cord, fan and connectors for the peltier. If it's plugged in, it's on. (One skinny wire remained.) And I discovered that it was a little harder to change the Peltier module than expected: the copper bar didn't stick out far enough to hold the 62x62mm square. However, I had at one point previously fitted it, and there were threaded holes in the right places to do it. I had to thaw out and remove the ice tray to change it, so I put fresh water in it. And the lid was open a long time. Obviously it would take a day or so to settle in before ice extent and temperature readings would mean much.

   When everything was together I set the power supply to 9.0 volts (at 15:00 PM), and it was using 6.4 amps. At the Peltier leeds it was 8.4v. That made 54 watts actually reaching the TEC, about equivalent in input power to the 12 volt setting on the old unit. I also got out an actual ampmeter and compared it to the power supply's meter. It seems that in this middle range at least, the power supply's meter is pretty close. In the lower ranges (eg, under 3 amps) it reads low by about 10%, and I applied this factor, evidently in error, to the amp readings. I adjusted those and the resulting watt figures in the table - or should I say I un-adjusted the erroneously adjusted readings?

   The bottom and outer end of the copper cooling bar were exposed to room air. With it sticking out farther, there was no hollow space to hold a piece of foam insulation on. They soon frosted up. I cut a sort of an "L" piece and pinned it to other insulation with a couple of nails. Less than satisfactory! And considering the water was still +6°C, it didn't say much for the heat transfer from the bar to the aluminum 'bread pan' ice tray. I think that was a problem the previous time I tried to fit the big TEC and moved the tray bolts to the very end of the bar.

   After a couple of hours, things didn't seem to be cooling in any hurry, so I upped the voltage to 10.0, which brought the current up to 7.0 amps. 70 watts! Like the first time I tried this big module, I got the impression the heat wasn't getting transferred very well. Even at 9 volts, the heatsink seemed too warm. Of course, there was less contact area of the tray to the copper bar. And the clamp wasn't centered on this large peltier module. Later I turned it back down to 9 volts. The temperature seemed to settle out at a little over 6°. After another day I set it back to 8 volts. It very gradually (~36 hours) went up to about 7° and most of the ice melted. Yet the copper bar was covered by thick frost.

  
(L) Reduced end of copper cooling bar inside fridge after moving it out 22mm to accommodate larger peltier module.
(R) with flexible graphite to fill gaps and transfer heat better to the ice tray. (An alternative to silicone heatsink compound.)


   Surely a major problem was the lack of contact surface of the ice tray to the copper bar. The bar was covered with frost well before ice started forming in the tray. If the heat transfer was better, the bar wouldn't get so cold before cooling the tray, or hold so much frost while the tray melted. I started thinking maybe I should solder together a copper ice tray with a thick bottom. Or maybe I could find an aluminum pot with a thick bottom that would transfer cold from the bar better than the thin aluminum bread pan. A thick walled pot shaped like a breadpan would be ideal!
   The next morning (15th) I went to "Value Village" to see if there was any such rectangular pan. There was one, much bigger than my bread pan ideal and shallower, but close enough - easier than trying to make something out of copper. I bought it. On the 16th I put it on. The first benefit proved to be the straight vertical sides of this tray. I could use the original bolt holes in the cold bar instead of the 'extras' I had had to put in almost at the end to hold the sloping bread pan with the larger peltier module. That would provide more heat transfer surface.

   I set the supply to just 8.0v, drawing 5.7 amps - about 45 watts. As it was new tap water and the fridge had been open while I was working, it wasn't initially very cold. After an hour or so (at 13:30 PM) the approximate temperatures read: Cold bar under Peltier: 2°c. Cold bar inside fridge under tray: 4°. Water directly above in tray base: 6°. Water at far end of tray: 12°. Fridge below cooler: 11.3°. Room temperature: 23°. It seemed to be cooling quite rapidly, and in 10 minutes those figures were: 0,4,5,11,11.0 and 23°. The heatsink warmest area was just 26°, and I turned down the fan, which was somewhat annoying. (It's hard to get much accuracy with a temperature probe giving no decimal units and touching a thermocouple to somewhere on a surface with poorer to better contact. The probe that reads to a decimal point takes several minutes to reach temperature.)
   The small temperature differences between the cold bar and the water right above it were encouraging. So was the fact that the water at the far end of the big tray was following them down nicely. But 8v is only 1/2 of 15. I turned it up to 8.5v at 14:15. And 9.0 before bed. It seemed to be on the verge of making ice, but none formed. The next day I needed the power supply to charge some batteries, so I plugged it into the solar power system. It was cloudy and the high load brought the system down to 12.5 volts (instead of 14.0) with about 12 volts delivered to the fridge. But there soon was ice. The module draws enough current that at night the system voltage, fed by a battery charger, droped to about 10.4 volts (delivering a little under 10 volts to the fridge), where it was previously about 11.7. (Later when the inside was cold and everything settled down it stayed around 11° and 8 amps (88 watts) at night.) At night and in heavy overcast the low voltage alarm goes off on the inverter (which I charge the RX7 EV through on sunny days) and I have to shut it off. During the day the current measured 9.6 amps, using 134 nominal watts at 14 volts.
   So the fridge experiments are morphing into designing a DC to DC buck converter (see below) to reduce the supply voltage and hence power to this big peltier, which also, as previously discussed several time, increases the COP and thus has virtually the same cooling power using less supply power.

   It's hard to compare the old and new tray performance directly. The new tray is much larger, so it radiates more, doubtless cooling the whole fridge more - more even cooling. So a higher inside temperature at the floor near the tray doesn't mean it's cooling less overall. But after a few days, with room temperatures around 21-22°, there were generally cooler temperatures through larger portions of the fridge. There was lots of frost buildup on the cooling bar and under the ice tray at the cold end, where there was considerable ice in the tray - eventually about 1/2 the tray. I suspect the end of the copper cold bar within the frost was -3 to -5. Right in the back corner at the bottom under the ice tray it got as low as 2.3°, elsewhere under the tray was 3 to 5, and 4.5 to 6 in the area around the tray at the floor, but rising with altitude toward the top. The temperatures there don't rise as much after opening the lid as before, eg .2° instead of .5°, which seems to indicate that it rises more from stirring of the internal air than from introduction of outside air, and that colder air was now more generally widespread. In the middle half way up I got readings of about 7 to 10°. At the far wall past a bunch of stuff blocking the most of the cold air coming from the cold end, at the top, it was still just "cooler than outside" at 16.5°, but seemed to be down a degree or so from before. Of course, any small fan blowing inside would even out the temperature considerably. I should probably put one in -- along with the DC to DC converter and a programmable control for solar and wind - for intermittent power.

   I pulled a small 'computer' fan from the closet on the evening of the 23rd, and connected the wires to a CAT plug (with a couple of resistors to reduce it from a hurricane to a breeze and a whizzing hornets' nest to a whisper). Then I simply set it in the fridge in a corner blowing air at the cooling tray, and plugged it into the 12V solar power outlet with a short CAT extension cord coming out the edge of the lid. That's not exactly 'installed', but it blows the air around. If it can warm up the cold bar to near zero and melt some of the ice, the cooling will gain COP by working across a smaller temperature differential, and the whole average temperature inside the fridge will be cooler than without it.
   After 24 hours, there was considerably less ice in the tray, and the temperatures near the tray were mostly over 5°. But everywhere else was cooler, with anywhere the middle around 7° instead of up to 10, and the warmest areas towards the far end near the top being around 9 to 11° instead of 15+. It seemed more like a real fridge. With the air blowing across it, the frost on the cooling bar had turned to hard ice, and it probably wasn't as cold.
   There was a huge tub of margarine occupying much of the far end, a big heat mass which I thought would probably cool some more in the next day or two and make for further temperature drop at the warm end. But the next morning I found that the rest of the ice in the tray (which had seemed pretty much constant all day) had melted and the coldest temperature under it was up to 5.3°. And most of the ice was gone off the cooling bar. Apparently this was all using up the accumulated coldness as well as the cooling capacity of the system. then again, I suddenly remembered I had also turned the outside fan down for less noise. Could that have made such a difference to the cooling? I put it up to full. A little ice formed right where the cooling bar meets the ice tray, but no more. The difference from the fan seemed to be there, but small. Some days of trying things disclosed that if the inside fan was set somewhere where it moved just a bit of air, everything could be made to a good compromise. Under the ice tray readings were 6-7°c but only the very far wall was over 10, at about 11°. Even upper areas near the far wall were 8-9°. And about 1/2 the ice tray stayed frozen. This of course is with using the 14 amp peltier at max all day on solar, with the voltage dropping under 11 at night.
   Temperature measurements with the fan at different settings (positions) continued into October. Overall result: The coldest section near the ice tray got a little warmer, but within a week the warmer areas in general became much cooler, with the whole fridge staying between 5 and 8° - a great overall improvement in cooling! A control on the fan would be an asset.

   As usual I have other things to do besides design the MSP430 microcontroller control to make peltier voltage and all programmable with the inside fan coming on and off, and I'll be putting it off for yet another month.

   However, there was one more interesting development. On the 30th I got a heatsink with evacuated tubes just somewhat along the lines of what I had tried to make, sized for a 40x40mm peltier module. Since the fridge's new peltier is too big for it, I'll try it out in the camping cooler. And I have the vacuum pump... maybe with that I can successfully make my own evacuated heat transfer pipes, for which sufficient vacuum has so far eluded me.

 
Evacuated Tubes Heatsink. These carry heat away even faster than solid copper.

The copper plate on the bottom is ideally sized for common 40x40mm Peltier modules.
The water turns to steam (even at room temperature) as the plate warms,
and rises up the tubes in the plate to the radiator fins above.
A small computer fan cools the radiator fins.
The cooled water re-condenses and drips back down the tubes.



DC to DC Converter

   It became more and more evident that various projects needed a DC to DC converter, namely a buck converter to reduce voltage in a more efficient way than a linear voltage regulator. Whereas I hoped only for efficiency of 80+%, I reviewed a paper on DC to DC converters that mentioned as high as 94% could be attained with synchronous rectification. Naturally I would jump at the chance to make the project more complicated -- I mean, to make a better design.

   After spending a good part of a couple of days looking up circuits, making one and starting in on a board design, I thought that if I was firstly doing a converter for the solar fridge, and if the solar fridge needed a microcontroller control, I might as well make the converter with the microcontroller and a minimum of other necessary components. Essentially, it would be the fridge control circuit that I'd already designed on paper but never made, with the addition of a driver, coil and a few sundry extras, voltage feedback to the microcontroller, and having the program rapidly turn the peltier drive on and off to keep it near the desired voltage.



Electricity Generation

Electrostatic Perpetual Motion Motor?

   I saw a video on this by "Nikola Tesla" on youtube. Operation was apparently by electrostatic forces - there were no magnets. Unlike most people, his machine appeared to spin by itself, seemingly wherever and whenever, and he explained how it worked. He made the point that the force was very tiny, and it managed to run because it had tiny "needles" to spin on (very low friction) and the rotor was very well balanced, a precision "scientific instrument". I won't try to explain it all here as he does a pretty good job in the short video -- and I can't say I understand it myself because he did leave out explaining a couple of things that may be well known in some circles, but not by me.
   It looked like a pretty easy thing to build and try out. But it would have to be precisely constructed. And even if it would run reliably, it would obviously take a considerable number of big rotors to generate even a few watts. That would still be farther than I've got so far in this realm.

https://www.youtube.com/watch?v=dGPnxLSgnUI - Perpetual Motion Machine

   Then there was another video by someone else that showed "how it was really done". He had built a crude copy of the original machine. Let's see now... it turns on a #10(?) threaded rod with no taper, no bearings, resting on pieces of wood at each end. That already violated the original poster's 'very low friction' requirement and would probably keep it from turning. Then the rotor wasn't balanced - no no #2. Then their electrical charge "points" were the rounded heads of two pan head machine screws: some "points"! Then to demonstrate that it would eventually stop, he spun it by hand -- in the wrong direction. Well, he seemed to have demonstrated that his build didn't work!
   But then there was an irony: he then hooked up a vandegraf generator and he had his young son (standing on an insulator) touch the ball on top of the generator and put his finger (the other hand) next to the aluminum bits on the rotor, in lieu of the two metal points. The rotor turned! (And in the right direction.) "Well, unless somebody has a better explanation, I think this is how he did it!"
   Wow! Far from "debunking" the idea, they had shown that a sufficient electrostatic charge will in fact turn the rotor -- just like the original guy said! The high charge of the vandegraf generator turned even their high friction, off balance unit, with just a blunt finger for a point.

   The point about the electrostatic points and charges focussing at such points is reminiscent of pointed crystals or pointed pieces of gold or copper apparently used in lambda ray collectors ("free energy devices"), but I'm not sure what the actual connection is.

   One youtube video leads to another and  I ran across someone called motionmagnetics who seemed to have a very good grip on 'perpetual motion' magnet machines. He said there were three types, and he had a video explaining each of them. I was impressed with how precise things needed to be, and how inevitable small differences in the magnetization of each magnet could throw everything off, and how someone could have a working prototype and yet be unable to make a reliable - or even working - production model. Compared to that, the electrostatic machine seemed really simple and easy to do.
   Still, one would think a design where key magnet positions could be minutely adjustable could be worked out. Failing that, a design such as a spiral or helix that is easy to get working except where the spiral starts and ends could be made using an electromagnet, or a solenoid moving a magnet or magnetic shield, to carry it past that position. Motionmagnetics said he wasn't going to cover such hybrid designs.



Electricity Collectors... Energy From Where?

   I also did a little reading on Thomas Henry Moray and his 1925-1929 'radiant energy receiver', hoping to better figure out how one might be made. Vital details for how to make such radiant energy receivers or collectors always seem to be frustratingly missing, whether because the invention was a deception, by lack of understanding of why their own devices worked, energy company or other corporate machinations to prevent the spread of such knowledge, or because of the inventors themselves hiding certain things so they could hopefully make money from their work (a factor in Moray's case), is hard to determine and probably is different in different cases. I suppose if that were not so, everyone would have them by now.

   There was something else as interesting as the receiver: In a trip to Sweden when he was younger, Moray had collected some rock samples containing germanium that had interesting properties. Later he connected silver wires to one and got the world's first semiconductor device, which has been called the solid state "germanium triode". Looking at a picture of his device it appeared to have only two wires, and hence could only have been a diode. In the schematics they appear to be used as diodes, having a little triangle symbol, tho without the bar that later became the standard symbol. And in the literature the "diode effect" of silver welded to germanium was mentioned, and the effect of the alloying that we would now call doping. But so was "multiplication" or "amplification" mentioned, which a diode doesn't normally do. Whatever it was, 25 years before the claimed first transistor, Moray had invented semiconductors! He was unable to get a patent because someone at the patent office couldn't conceive that a cold cathode could emit electrons. And he couldn't get a patent for the radiant energy device, the patent office giving various excuses with each application. He got no credit, but it seems that his germanium work spread to another researcher or two and led to the further development of semiconductor materials, enabling today's electronic and computer world.

   As for the radiant energy receiver, it had an antenna, a ground, tuned circuits with high voltage capacitors, and the germanium diodes. It most likely also had special tubes. It seems it was quite finicky. Putting a hand near a (?)capacitor was enough to get it to start or stop producing kilowatts of energy. Touching the antenna would also stop it. Some capacitor had to be stroked with a magnet for a while (a few minutes?) to get it to start (not always successfully). It could be stopped for a moment or two and would restart, but if left off very long, it had to be restarted by stroking again.

   The article also had a paragraph on the radiant energy presumed to be being captured, suggesting it was incredibly high energy like a billion [electron] volts [per photon] and hence ultra-short wavelengths compared even to gamma rays, that it came preferentially more from some directions than others [aligned with the sun] but was always present from every direction, and that no one really knows where the electromagnetic spectrum ends. Except for the presumed distribution, pretty much a description of the "VHE gamma" rays - lambda rays as I've named them - that were identified in 2007, far beyond gamma rays, at up to ten trillion electron volts per photon and up to 10^27 Hz frequency. (That seemed much closer to the mark than more recent "zero point" or "vacuum" energy ideas, which don't seem to be explicable in any sort of comprehendable terms.) The receiver was believed to be tuned to a sub-harmonic of the original "ludicrously" high frequency. (If I may borrow a term from the movie "Spaceballs".) In fact, I now suspect it was capturing the electrical charge of Earth's own atmosphere.

   On the evening of the 24th I again searched youtube for "radiant energy receiver". This time I eventually found someone called "Tesla Cult" doing some rather interesting experiments, and I saw he had several videos. The first one didn't look very interesting, but he showed some interesting 'noise pulses' coming from the antenna outside on his oscilloscope, the biggest ones from distant thunderstorms... and at the end of the video he showed the schematic of the circuit that had lit a couple of dim LED.s. So I called up another one. They all said "3 years ago", but the first one was in April 2012, the next from June, and the last one was from September. So he had persevered and, finding something about "two ground points for an antenna" in a book on radio from the 1920s, made some advances. There was about .6 volts DC between the two grounds. (This could be because they were different metals, but evidently different points can indeed simply be different.)
   And he noticed he got a spike of energy whenever he connected or disconnected his two ground wires. Ah yes! First, disconnecting big switches on poles used to electrocute linemen when they had tried using DC to transmit power early on, and second, switching electric fields on and off is the way to coax lambda rays to release their energy. As I recall from Wikipedia, it's released as a whole stream of particles [electrons] and anti-particles. (Gamma rays release one particle and one anti-particle.) So he had put in a transistor to oscillate and switch.
   This circuit lit about 10 LED.s and more brightly, and it caught my attention when he said "an amp". Now we were talking watts instead of milliwatts! (An amp * 3 volt LED.s = 3 watts) Okay, it was a pretty crude circuit, and it wasn't Moray's 50000 watts, but it was comprehensible (indeed simple) and so buildable, and so perhaps it was a starting point that could then be much improved by many and various things gleaned from other sources such as Moray, Tesla and others more recent. Again there was a good schematic of his circuit, at the end of the video. The tuning coil/transformer was just wire wrapped in a helix around a 2 or 3" PVC(?) tube. Both ground rod wires came in through the window (which he mentioned had an anodized aluminum frame that seemed to have some antenna effect in his earlier experiments). Full marks for observing all this, and the connect/disconnect spikes that led to the new circuit.


"Earth Radiant Energy Receiver"
by "Tesla Cult"


   Something I had recently realized about my own attempt, quite a while back now, was that I had used around 100 KHz switching, but I had used an iron powder coil core. Iron cores are only good up to a few hundred hertz. I should have either reduced the frequency by 99+% or used a ferrite or air core coil. In Tesla Cult's circuit, the frequency is set by the capacitance of the antenna in conjunction with the inductance of the coil, forming some sort of oscillator circuit.
   The next morning it occurred to me to click on "Tesla Cult" channel specifically. He has done a lot since 2012, with some bizarre looking stuff. The next step from the above was to move the "south ground rod" connection instead to an antenna, for which he used his anodized window frame. It was then lighting about 35 of the same LED.s. He said the frequency was about 100 KHz, but there were oscillations at 500 KHz so that would be a better frequency. He said he'd need a new coil to get that. I thought I should continue viewing his videos, but on looking them over, it seems that after the apparent success of apparently generating at least a few watts this way 'from thin air' he diverted into other things.
   But he was getting energy even without a couple of "the usual" key pieces, which might amplify it greatly. Perhaps I'd switch to his air core coil, keep my own 555 oscillator and MOSFET driver circuit, and resume my own experiments in this field. Many things could be tried just for an antenna, as well as for tuning components and circuits. The coil/transformer can be made by hand, and its pickup point is tunable. A tuning capacitor might be had from an old radio dial. The "subharmonic resonance" with the lambda rays is probably a key. Tuning appeared to be finnicky and also a key to operation of Moray's device.

   Suddenly, after all this time, I had the thought of why my transistor had got hot and burned out in my earlier attempt. On the 27th I looked up the schematic and sure enough -- no flyback diode to shunt a high voltage spike as the coil turned off. It was the same as when the flyback diodes had been left unconnected in the new motor controller, causing those ones to get hot and burn out. And the reason for that was... it wasn't in the original schematic I copied the driver from. Perhaps their inductors were so small they weren't needed? I added one in. The antenna and ground seemed to be a key both in "tesla Cult" and Moray's receivers, so I added an antenna connection terminal separate from the coil. Looking at a circuit of Moray's someone had drawn from memory, I also added a diode on the antenna line. There's always been some elusive DC component to all of this, and the air becomes more and more positively charged with altitude (lightning is when this potential suddenly discharges, ionizing the air).

Another Energy Source: Atmospheric Charge?

   Then on the evening of the 27th it hit me: Tesla's experiments were originally intended to capture electricity from the static charge in the air. Franklin's kite experiments were also aimed at learning more about this. Evidently on a good day, the positive charge of air (shortage of electrons) increases (to our considerable surprise when we first learn of it) by up to around 100 volts per meter of altitude. And a lot can happen with this charge prior to - or without - a lightning strike. More conductive channels such as "lightning leaders" and "return (ground) stroke channels" develop in advance of an actual lightning strike. Can they develop and large charges of electricity be tapped without actually causing lightning? And Tesla certainly made a lot of actual lightning, too! In fact, a read at Wikipedia reveals that there are still a lot of unanswered questions about atmospheric charges and lightning. Is lightning triggered by cosmic rays? (or maybe lambda rays?) Why does it generate X-rays and gamma rays when the voltage is supposedly too small to do so? And yet the overall potential vertically through the atmosphere can be a billion volts - Wikipedia's figure is the same one Moray mentions in his description of his "radiant energy". However, at 60 miles, below the ionosphere, it is around 360000 volts.

   It is only natural to wonder about harvesting this known source of electrical energy. A small device can easily be made with an antenna, a ground, and a few diodes and capacitors, which will gradually build up voltage to several volts from airborne electric charge (or from nearby power lines or radio), and very gradually charge a small battery. This is similar to Tesla's two 1901 patents on the subject. (I made one in September 2013 from a "Tesla" circuit found in various places on the web, shown in TE News #68.) But calculations and experiments reveal that such passive, untuned devices would need gigantic antennas to capture much energy. (However, people have been electrocuted by the charge picked up by unconnected power lines, and a long steel cable hung from a helicopter is said to make a huge spark when the bottom touches the ground.) But if the antenna is very tall (eg, Franklin's kite), the voltage is so high it's hard to utilize it. Tesla, who doubtless used a very high up antenna, noted that the charges would build up until they broke through the dielectric material of most capacitors.

   The stuff I'd now been studying for many days started clicking into place, that it all pertained to capturing atmospheric electric charge and not radiant (or "mysterious") energy per se as everyone seems to suppose:
* It has to be well grounded - the earth has the negative charge.
* Large, perhaps insulated antennas to get the positive air charge. (why would a big antenna be needed to capture picometer(?) wavelength rays?)
* Hi dielectric constant antenna insulators for high capacitance (the opposite of what cable insulation makers try to achieve).
* A diode in the antenna line.
* Tuned circuits, usually at ultrasonic or radio frequencies (LF or MF).
* Spark gaps or spark plugs that suddenly discharge or transmit high voltages obtained.
* More energy during the day than at night and early morning -- following the times of day atmospheric charge is greater and lesser, and vaguely aligned somehow with the sun. (It would seem the sun's radiant energy is the major source of the atmospheric charge, so in a sense it's an indirect harvesting of radiant energy. But it still works at night.)
* "Tesla Cult's" antenna noise spikes seen on the oscilloscope from distant thunderstorms. Just how far did lightning energy travel through the atmosphere, and how much energy of a dynamic nature was still there in the absence of actual lightning and storms?

   Some of these techniques starting with Moray's seemed to be inducing high voltage transient fields into the atmosphere around the antenna, like a radio transmitter. This tuned electromagnetic excitation of the air might well cause far more interaction between the antenna and the atmosphere than just a passive antenna and circuit. As the antenna voltage swings negative during its oscillations, it should act on the charge in the air as if the antenna was at a much higher altitude, pulling positive ions to the high dielectric insulated antenna as a capacitance effect. As it swings positive, they should be repelled away again. So it's 'stirring' the atmospheric ions instead of just waiting for them to wander by. And perhaps some of the atmospheric ion channeling effects which are precursors to lightning ("lightning leaders", "return stroke channels" - see Wikipedia, "Lightning") may be generated near the antenna. This could greatly increase the volume of air interacting. The frequency of resonance with the air may depend on the antenna, especially its length, or it may be a property of the air itself. (Obviously I don't have this all figured out, but surely there must be some dynamic atmospheric effects occurring.) Charge from the air would be transferred with each cycle to the circuit, which can flow to ground through an electrical load, powering that load.
   Cyclically discharging an elevated antenna to ground - or lower - would bring the antenna repeatedly down to the voltage from which it would start to recharge fastest, providing the most current. If the voltage was then taken up high, perhaps even above the ambient charge at the antenna altitude, it might in some way "reset" the air around the antenna and prod it into faster charge transfer.

   Ultra high energy, 'ludicrous frequency' (eg 10 trillion EV, 10^27 Hz) lambda rays even beyond where Moray described are now known to exist, and are probably the source of energy for some "radiant energy receiver" devices with no discernible antennas. But grounded devices with antennas placed well up off the ground such as Moray's and "Tesla Cult's", are more likely getting their energy from the atmospheric charge, but collecting it far more effectively with their oscillating circuits than with "Aerial Battery" passive DC circuits. Electricity is literally being pulled out of thin air. or at least regular air.
   (When Tesla lit banks of light bulbs in a field with such devices, it would seem he was transmitting a concentrated ion stream from special ion transmitting tubes located elsewhere - the source of the energy, of most of the charged ions, was the wireless transmitter of radiant electricity, of which he spoke. This appears have been much confused with "free energy" from ions naturally in the air. And his primitive, passive 1901 DC ion capturing patents seem to get confused with his separate 'bifilar coils' patent and with later ideas and writings such as Moray's powerful (1925-1929) energy capture device. Tesla spoke to a newspaper in 1933 about unlimited free energy that was surely soon coming to mankind, an expectation perhaps based as much on Moray's work as his own.
   Tesla's own last patent seems to have been in 1928, unrelated to energy capture. He did do some later work on a particle beam weapon, the original "death ray", and on what may have been the original idea for HAARP. He hid some of his later designs realizing their potential for great harm as well as for good. The US government seized all his designs on his death in 1943 and later wouldn't admit to having them. Reagan's "star wars" program as well as HAARP may have been mainly based on these designs. But I digress.)

   The air has the energy! Lightning is the spectacular discharge of the air's energy. When there's no lightning, which has far more energy than can be handled anyway, some lesser amount of energy must still be there. It appears Moray and others figured out how to tap it. If I'm right Moray was off base thinking he was capturing ultra ultra short wave radiant energy -- his antenna height and size indicate he was surely tapping atmospheric charge energy.
   So my working hypothesis for a potential 'atmospheric energy harvester' experimental project is that the energy comes from the known atmospheric charge, amplified and made accessible by means of a tuned oscillator circuit, with of course the antenna, ground, and other suitable circuit features. (The antenna is ideally about 65 feet up and at least 80 feet long according to Moray - but an elevated insulated plate might work fine too.)
   IF I can make it work, it could supplant most other energy generating projects. Self-turning magnet machines, solar PV, solar hot water, wind turbines, and even ocean wave power and floating river hydro would probably lose much of their attraction. The main drawback would seem to be that an installation with a big antenna and a ground wouldn't really be portable (ie, for electric cars. But it might work with ships and large vehicles like trains?) And I suspect that units located too near each other would probably lose power - that the power available in an area is in fact finite, and they would compete to harvest all that's available in the vicinity.

  On the 28th I drew up a schematic. On the 29th (after getting the switched reluctance motor running!), I went to Microsec R & D and (by invitation) got a big air core coil and three tuning capacitors from their "museum" section. The coil looked like about what I had intended to make, and it had a solderable point for a tap on every second turn, of which there were 32, with an inductance of about 25µH. It was wound on a glazed ceramic insulator - a good improvement over a plastic pipe that could potentially melt. The possible uses for the tuning capacitors were rather speculative, since I didn't draw a capacitor in the circuit. I figured the antenna was the capacitor. But I might need to do some tuning that a tuning capacitor could help with. The largest one (physically) went from about 2 to 95 nanoF. The plates were over 1/16" apart, so it should take quite a high voltage. The next day I returned the other two and picked up books on design of LC oscillators and radio transmitters. (and an evacuated tube heatsink for peltier module experiments, that happened to be sitting there.)

On October 2nd I found a site that seemed to give some real background on the energy issue and things happening from the 1880s to the 1930s. It seems to be a chapter on Moray from a book called Lost Science by Gerry Vassilatos from 1999.

http://customers.hbci.com/~wenonah/history/moray.htm

   It appears that patents for "static-field" "aerial batteries" "fill the archives", and that some of them were used to power machinery. In a most interesting paragraph is:

"While investigating the output of his device, he [Moray] discovered a feature of the natural static energy, which had somehow been overlooked by other aerial battery designers. The electrostatic power had a flimmering, pulsating quality to it. He learned of this "static pulsation" while listening through headphones, which were connected to telephone wires. The static came in a single, potent surge. This first "wave" subsided, with numerous "back surges" following. Soon thereafter, the process repeated itself. The static surges came "like ocean waves". Indeed, with the volume of "white noise" which they produced, they sounded like ocean waves! "

   Herein we recognize the AC or dynamic component of the air charge, and perhaps why an oscillator might harvest much more energy than a "static-field" receiver, and with a much smaller antenna. And we understand that Moray was indeed the first to discover it. Farther down we discover that the ground side of the circuit also has some unexpected characteristics and energies. Specific points in dry ground have energy characteristics not shared by other nearby points or by wet ground.
   In fact, the article contains a wealth of natural energy related information, some of which has been very nearly forgotten - such as some early telegraph stations that worked fine with utterly dead batteries via the ground energy. I read through it that day, neglecting getting this newsletter out. And now I'm not sure. Moray may have been harvesting lambda rays rather than atmospheric charge. I'll have to go through it carefully again before I start trying anything!



Electricity Storage - Turquoise Battery Project Etc.

Honda Insight NiMH Battery

Charging the rewired 14.4 volt, 65 AH "Orange Bamboo" NiMH high rate battery

   On about the 15th somebody gave me a Honda Insight hybrid battery. It consisted of 120 nickel-metal hydride "D" cells. 10 sets of 12 cells were spot welded together, six cells going to the 'back' and the other six returning to the 'front', where the external connections were made. They were all nicely spaced by molded plastic spacers, with air space between each of the twenty runs of six, ensuring good cooling and that nothing would short circuit if things got hot and melted the plastic cell sleeves, which were single sleeves of 'heat shrink' over all six cells. There were three temperature sensors, buried under three of the sleeves, and some other "PTC" temperature sensors that ran right along the sides of each and every "stick". It was electrically divided into two 60 cell, 72 volt batteries, which probably got merged into 144 volts when connected. A cover of metal pieces and extruded styrene(?) foam covered the battery. Surprisingly it weighed 55 pounds, about as heavy as a lead-acid battery of similar nominal rating. It might outperform it by 30% or so and it would surely (at least when new) long outlive it.
   With the battery were the Insight hybrid control parts - the motor controller and charging system. I didn't know what to do with these except to salvage some very nice heatsinks. I wasn't willing to put in the time to learn the details. (If the motor had been included it might have been a different story. Even so, I have my own motors.)


   Nothing seemed to be known about the car the parts came from. It would have been impractical to try breaking up the welded sets of twelve, which were all fully discharged - some were down to 9 volts, and they'd have been there for an unknown length of time. That's below the "one volt per cell" minimum where corrosion starts occurring, but I decided to treat them all as "good" for the purpose of turning it into a single 14.4 volt battery. It would need a special charger for that voltage... but if used in the RX7, it would be up to 142.4 nominal volts, almost the intended voltage of 144. If the cells were 10 amp-hours that would make another 100 amp-hour battery for the RX7-EV, but I expected they were probably about 7 or 8 amp-hour, very high rate cells, capable of zillions of charges and discharges including sudden very fast charging from regenerative braking and steep discharges during acceleration. A check on line indicated they were probably only 6.5AH. A company said they had superior replacements - higher rate and 8.0AH (...and 2100$US) owing to improvements in manufacturing since the Insights (2003-2006?) were made. Being very high rate would help them last longer in the RX7 than the same 65 amp-hour rating in a regular rate NiMH (or lead-acid) battery, but that capacity seems substantially lower. (On line there are various instructions for rebuilding the Insight battery with regular NiMH "D" cells. I wonder how that works out?)

   On the 17th I used up the short copper connection bars it had, and then made my own #14 AWG short connection wires for the rest, to wire all the tube sets in parallel for a 14.4 volt battery. With all the air spaces it was physically large and might be hard to fit, but surely it could go in somewhere?
    Later I ordered 6 more sticks of 12 batteries for 24$US each (or "12 sticks of 6" for 12$) off e-bay. ("4 sold in the last hour"... I looked again and it said: "5 sold in the past hour"... and I took the rest.) It seemed I happened to look at exactly the right time and that the seller sold them all in an hour. Then on October 2nd I discovered in my e-mail that before I saw that deal, I had bid on an e-bay auction for a set of 5 more tubes and won - for 99¢US! Now I'll have 2 whole Honda Insights worth and a tube to spare.
   I only vaguely remember seeing the auction, and thinking ya, that'll be way over 100$ before bidding ends, and probably I bid about that amount just in case it wasn't. But no one else bid!
   I had to have both sets sent to my phone number "c/o Clipper Cargo" in Seattle WA because the sellers wouldn't ship to Canada. The Victoria Clipper jet catamaran then brings them across the strait to Victoria on its regular passenger route for 27$US and I have to clear them through customs myself. (What do people farther from the border do?)
   So shipping will make it total about 170$C. Plus any taxes customs decides to charge me... on 99¢. (That's 99¢, +99+27 $US shipping and the humungus rise of the US$ versus all other currencies as nervous investors seek the "safest" asset. But all fiat currencies are inflating away.)

   Later I got 12 lithium iron phosphate cells (36v, 40AH) for 600$ in town, which was another great deal. I could now power the Chevy Sprint at 36 volts without buying any golf cart batteries, just lots of NiMH D cells and 12v 320AH of paralleled lithiums - if I ever get it running.



http://www.TurquoiseEnergy.com
Victoria BC Canada