Turquoise Energy Ltd. News #111
  covering April, May, June, July and August 2017 (posted September 15th 2017)
  Haida Gwaii, BC
by Craig Carmichael


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

Month In Brief (Project Summaries & short project reports)
- House Move - Project Selections - Electric(?) Ground Effect Vehicle - Lambda Ray Converter - Upgraded Electric Caik Outboard Test ...& the next "Hubcap" type motor improvement - A Nickel-Air Battery Note - LED Lighting - Suzuki Swift: Burned an EV (Yow!) - Miles Electric Truck Chain Drive

In Passing (Miscellaneous topics, editorial comments & opinionated rants)
- Haida Gwaii (AKA Queen Charlotte Islands) - The Move - A Trip Back to Victoria - Antigravity in a Dream? - Time of day - Sigh, Chem Trails Again! - Possible Health Ideas: Cyst and maybe Mole Removal or Shrinking? - Easier Shelling of Peas and Beans?


- In Tedious Depth Project Reports -

Electric Transport - Electric Hubcap Motor Systems
* Electric Bixel Ground Effect Vehicle: - Ducted Fan etc. - The "paper airplane" delta wing design - back to Catamaran design

Other "Green" Electric Equipment Projects (no reports)

Electricity Generation
* New Water Flow Turbine Design Thoughts
* Ocean Wave Power Thoughts
* Short Space Ray/Lambda Ray/VHE Ray Converter: - Circuit Board is working - Programming - How it Works: A bit of speculation

Electricity Storage - Turquoise Battery Project (NiMn, NiNi, O2-Ni), etc. (no reports)



April... May... June... July... and August in Brief

   I was very busy all spring and summer with the move to Haida Gwaii and setting up home and shop there, and there is still a lot to do. I wrote a lot about that, then decided it was mostly of interest to friends and relatives, and here I include just a bare outline, in the "In Passing" section. I found it was harder to get internet up here than I had expected, and without internet at home I couldn't post newsletters. It has been a challenge getting all my stuff moved, and I was without the computer I usually write the newsletters and edit the pictures on. I was missing parts and tools for working on almost any green energy project. I typed article text on a combo-tablet that had a keyboard. Meanwhile the months rolled on. I wrote haphazardly on two different machines, and it became a hodge-podge that took considerable editing to merge into organized texts. And there were dozens of pictures, taken over 4 months.
   Luckily there is cell phone service in my area. (Some areas don't even have that.) Late in May I got a new cell phone, one that I could use as a "WIFI hot spot" and get on the internet. (The price for using that very much is steep, but at least I could get on line from home occasionally.) Late in July I got my trailer which had been stored in Cache Creek and which had much of my computer equipment in it including the one I do the newsletters on, with the subscription list. Now I could upload and edit all those pictures I'd taken - but with so much to do I didn't get around to it. In mid August I bought a "WIFI Range Extender" which also had an ethernet connector so I could connect that old desk computer via WIFI. (Similar WIFI items I had ordered from China in June or July finally arrived on September 11th.) So I finally wrapped up this TE News, #111, which had become more than plenty long with a lot of loose ends. But it didn't work. While the physical connections are there, these WIFI devices expect the LAN to be on the ethernet and the computer(s) to be on the WIFI instead of the other way around.
   I saved a lengthy and revealing In Passing article about World War Two, based on the works of Victor Suvarov, for next issue. (and you thought you pretty much knew what that war was about? So did I. Hah!)

   In these months I couldn't do much with green energy projects except study and plan. I first considered the tidal power project at Delkatla estuary, and the Differential Variable Transmission, and a couple of other projects. But there seemed to be no local interest in actually doing the Delkatla project - at least none I managed to get in touch with. Finally I decided to concentrate on two projects that seemed to me to be more valuable than any of the others.

Electric(?) Ground Effect Vehicle

   Ground effect vehicles seem to me to be the best way to open up transportation to islands, between islands, and along inaccessible coastlines, especially mountainous ones cut by fjords and inlets where road building is circuitous or impractical. Today's options are usually airlines (costly) and ferries (slow). The subject has become of great interest to me since I now live on an island rather remote from the main stream of human activities, which would become much more accessible with such a craft.
   The ground effect vessel flies over the water at very low altitude with the speed of an aircraft, with less special 'pilot' training and 1/3 the energy cost, a "sea bus" enabling many local or intermediate distance routes to be traveled quickly and economically. In scale it could be anything from a personal "sea car" to a ferry carrying vehicles.

Mock-up of the delta wing shape - imagine the
center as a solid hull instead of just an outline.
But the catamaran shape seems more practical.
   My idea of the best form for the ground effect vehicle metamorphosed from a catamaran with optional stubby outer wings to a "paper airplane" delta shape... and, on reviewing some videos, back again. The paper airplane flew well and the delta wing shape seemed "cool", but it seems two outer hulls with a main wing between them traps air at the sides as well as the back, so the air scooped in the front under the wing has nowhere to go except to lift the vehicle out of the water. So this rather "boxy" looking shape is almost inevitably going to give the best performance, stability and fuel economy. And it would be easier to dock and to embark and disembark. But one radio control modeler said when he added stubby wings to the outside of the floats, it worked better. And that's the configuration Bixel seemed to use and showed in his patent drawings. I tend to think simply making the catamaran wider, even square viewed from above, could accomplish the same thing.
   Unexpectedly I have a complete electric drive system that will run at 10 or 15 kilowatts, with 30 KW briefly for take-off, and (while realizing it won't cover long cross-ocean routes) I now hope to make the prototype manned craft electric instead of gasoline. But before that will come the electric radio controlled ("RC") 1/4 scale model.


   In order to maximize thrust for take-off using the least amount of power, I'll be using a ducted fan propeller, as they have more static thrust per horsepower. And perhaps the unit can be mounted at the front with the air aimed or further ducted to blow under the wing instead of over it. That should increase lift for take-off at a lower speed. For the model I found a 5" plastic ducted fan at "Hobby King". I don't know where I'll get a 20" or larger one for the full size craft. The "turbofan", the common jet aircraft engine, is a ducted fan propeller turned by a turbine.
   I didn't get the radio control parts from storage until the start of September, and without knowing how it would all fit together I held off building more than a "mock-up", and that was of the delta wing design.




In the catamaran design, air scooped in under the front of the wing has nowhere to
escape out the back and sides except by lifting the craft out of the water.


   At the start of September I conceived of the ducted fan being at the front and blowing air under the wing instead of over it for still more lift and lower take-off speed. Others have done that with models, but it looks more practical with a small diameter ducted fan than with a big propeller.


VHE/Lambda Ray Converter

   The other especially useful project was one item that did arrive in the shipping container in April before I did. It was a cardboard box labeled "Lambda", which had the parts and pieces related to the project. It seemed to be one project I should be able to work on - at least, once I got my computers back. If I could get it to work, it might replace any and all other means for generating electricity that I might work on. And it could even make battery storage and hence better batteries much less important. It would be relatively easy to duplicate units and I could do it without help and interest from others, which seemed to be lacking.
   So the on again off again converter project was on again as of late July when I got my computers back, with the control board schematics and the software development system for writing MSP430 microcontroller software. This time I hope to get as far as a fairly well built coils/antenna unit in a "bread pan" steel box, a working microcontroller based coil pulse control board, and trying out what might be a viable software strategy for turning the rays into electricity via those circuits.
   In August I got the circuit board that I made a year and a half ago working (at least, the microcontroller ran - not without some initial trouble and confusion), and I made much progress on writing the software. Sometime earlier I realized I had the IRF7307 predriver chip connected wrong and I fixed the design for the present circuit board. But going back to the earlier experimenters' schematics, which I had copied the much of the driver circuit from, I found they all used the same circuit and they all had it wired wrong, identically! My own original mistake was in copying the original mistake! Everyone had simply copied the first person's cludge job assuming it had been properly designed, and so they were all working under a considerable handicap with poor and uncertain drive to the power mosfets and hence to the coils. This may help explain why they got unpredictable results and no successful converters. In August I finally realized it was a poor choice of chips to start with - there are lots of more suitable ones.
   Recognizing these rather basic electronics flaws was actually inspiring. Notwithstanding Mark's safety warnings and emphasis on the difficulties, it gives me good hope that, using the flexibility and precision control gained by using a microcontroller instead of simple discrete circuits and oscillators, it just might be easier than it seems - or even "pretty simple" - to convert VHE rays into electricity.
   I seem to have cludged my board to where it'll probably work, but owing to the poor choice predriver chips (even when connected right) compounded by power supply issues, I'm already wanting version 2 of the board to improve the power mosfet drive while also getting the MSP running within its specified voltage. I picked another mosfet gate driver chip and ordered a few from Digikey. Meanwhile it is most helpful that the MSP hasn't blown up yet at 5.2 volts, when its "absolute maximum" rating is 4.1 volts.

   On reading one experimenter's report I realized it would be just as necessary to varnish or epoxy all the coil wires solidly into place on this unit as it is on an electric motor, or the wires would vibrate until they shorted or broke.


   I also started to dimly see an important aspect of the operation. When the control coils pulse their high voltage pulses and the sudden change in voltage triggers the lambda rays to release their VHE electromagnetic energies, those energies don't simply appear in the "collector coil". Instead, they are transformed into far lower frequency radio (and or other wavelengths) energies in the areas around the coils. These are in turn induced by electromagnetism into the collector coil in the same way as into the secondary of a transformer or perhaps more akin to lighting an unconnected light bulb near a radio transmitter antenna. Here we may begin to appreciate why there may be strange glows or electric arcs around the unit, how people can get RF or radiation burns to their hands from them, and how Mark's earlier transistorized versions didn't work because there was too much electrical interference in the air, getting into the delicate circuits.

   In some ways, I find lambda ray conversion to be a boring project. There's no exciting spinning parts and not much machining or fabricating to do. There's a fair bit of microcontroller programming, but I've already done more programming than anyone should in a single lifetime, and nowadays I can hardly stomach sitting down at a computer to do it. Of course, a bit of success could go a long way to making it more interesting!


   While those were to be the two main projects for the time being, other things deserved putting a little effort into on the way, too.


Upgraded Electric Caik Outboard Test (Continued from TE News #104 & 105)
...& the next "Hubcap" type motor improvement


   One thing I'd been wanting to do for a long time was to test the Electric Caik outboard motor, repaired and with its rotor upgraded for higher RPM.s almost a year ago. But somehow I'd never managed to get the boat in the water. On September 6th, with my 14 foot aluminum boat and all finally at my new abode, I did some repairs to the trailer (maybe lucky the boat made it up here without falling off!), mounted the outboard and equipment in the boat with eight 100 amp-hour lithium batteries (about 25 volts), and by the next afternoon, the 7th, I turned it on. Everything worked properly on the first try, so I hitched it up and drove down to the boat launch at Queen Charlotte. Aside from stupid little things like leaving the drain plug out of the boat when I launched it, and not bringing a bailing bucket along, the trip went smoothly.

  The outboard was rather noisy instead of virtually silent, which experience says indicates downward pressure from the motor onto the drive shaft. (Why?) But not loud like a gas outboard! There was a fair crosswind but I seemed to have navigation - ie, I could steer anywhere without the wind blowing me back - at only about 10 amps. Running at 25 volts instead of 16 or 18 helps. I ran it at different amperages (as redd on the cheap Chinese shunt meter - not very steady readings and probably just a little lower than the actual current) and got the following results:

Amps - RPM
20 - 1530
30 - 1800
40 - 2000
50 - 2100
60 - 2250

   At 50 amps the breaker blew after a bit. (25v*50A=1250 watts.) Well, it's a 50 amp breaker, and maybe it was really 55(?) amps or so. Of course, it took me by surprise "Groan! Something's fried!", but only for a moment before I realized what it was. Reading 60 amps it blew pretty soon. With the improved rotor I wasn't worried about rotor magnets potentially flying off when going to higher RPM.s, but about 60 amps and 2250 RPM was 'full throttle'. I was a bit surprised by this as I thought it would go higher. But the limits are set by the shunt wire in the controller. If I changed that, it could do more... if it didn't blow the controller or overheat the motor.
   Later, I found the control, which I contrived to mount inside the arm of the outboard when I first did it in order to use the original twist grip 'throttle', was acting up. It may not have been reaching the end of the travel of the potentiometer and hence full speed. But now I'm not sure the twist grip in the arm is actually the best place for a speed control. It's arguably best for starting a gas motor and for maneuvering, but in steady travel you're always turning it by accident and it's hard to keep a steady speed unless it's at maximum, and even there you're trying to twist it extra so as not to inadvertently slow down. A control one can set and leave would probably be better, especially when it'll often be run at part speed to conserve battery power, and I think I'll convert it to that setup.
   The boat didn't get up on a plane, but as there wasn't too much weight in it, it probably wasn't too much short of a soft sort of plane. From driving lessons when I was 18: 60 MPH is 88 feet per second - close enough to 90. So 6 MPH is 9 FPS. So 15 FPS, my best very rough estimate at 60 amps and 2250 motor RPM, is about 10 MPH. So assuming proportional RPM to speed: 2100 RPM is 9.3 MPH, 2000 = 8.8, 1800 = 8, and 1530 = 6.8. Note that the speed at 1/3 power (20 amps) was 2/3 of the speed at 60 amps. One sees how with a displacement hull, power consumption increases dramatically to gain just a bit more boat speed.

   I guess I must have run it longer and harder than on previous occasions. I forgot to connect a meter to read the temperature, but afterward the motor was certainly hot. I wouldn't want it much if any hotter. I was rather disappointed after it had seemed to stay so cool on previous tests with similar currents. I guess the "continuous duty" rating would be somewhere around 1200 watts or 2 horsepower, and nowhere near my evidently overrated thoughts of 3000 or more. But I'm not surprised, having in the meantime seen how fast the Electric Hubcap motor coils could get hot in the Sprint car when supplied with up to 150 amps (*36 V = 5400 W) from the Kelly BLDC motor controller. The heads of the (metal) coil clamping bolts (electromagnetically heated by the spinning rotor magnets, with no electrical or good thermal connection to anything else) were especially hot, to no benefit. I thought I had replaced those with nylon bolts... but I guess that was in the next, unipolar, version of the Caik motor. There might be some gain from replacing them, and also by taking the hood off the outboard to get better air circulation to the motor - the hood must certainly trap the hot air.

Improving the motors

   In most motors, the copper wires are close enough to the iron that some heat is carried off through the iron, and heat radiates off everywhere. With my largely plastic motors, only air cools the coil wires. Now I'm thinking of modifying the mold design for all my "hubcap" type motors to get air flow over more surfaces of the coil wires - exposing the bottom (and if possible the top) of the donut as well as the outside edge. The problem with exposing the top is that the flux gap has to be larger, and it may become too large. I've been making my motors occasionally since 2008 and probably achieved excellent efficiency about 2012 (IIRC), and it's amazing that I keep on discovering very significant ways in which they can be improved.
   And perhaps an appropriately ducted "computer fan" that comes on whenever the motor is on, could be used to improve air flow. The magnets acting as centrifugal fan blades seem to move air nicely, but perhaps it needs to be flowing faster. A lot of electric car motors that pack much power into a small size are liquid cooled. Partly because of presumably high efficiency making less heat I don't think I need to go to that extreme, but it's evident they would handle more power with better cooling.

   I may add a couple more battery cells for about 31-32 volts and try the boat again some time with this boost. OTOH, at least at 24 volts I haven't yet blown my motor controller or burned out the motor.

   I didn't think to take any pictures or video until I was back at the launch ramp.



A Nickel-Air Battery Note

   The desire for lighter weight batteries for an electric version of the low flying "sea craft" brought my thoughts back to the potential for nickel-air batteries to fulfill that sort of role. A metal negative electrode is light and compact compared with the oxide/hydroxide based positive. Hence, using air to replace the heavy electrode is an attractive idea that could potentially drop 3/4 of the weight of a cell. Attempts to produce rechargable zinc-air cells do not so far seem to have resulted in practical products. Seemingly few have tried any other metal.
   I've probably mentioned this before, but I'll bring it up again: why would nickel-air be better than zinc-zir or most other metals that might be tried with air?

1. The reaction voltage of nickel to nickel hydroxide (-.72 V in alkali) is less than that of water to hydrogen (about -.83 V in alkali). The nickel won't spontaneously oxidize in solution in the presence of oxygen as do cadmium, iron, metal hydride and zinc. (hence its well known corrosion resistance). Oxygen reaching the metal electrode is hard to avoid in a cell that can't be sealed and where oxygen is deliberately admitted for the air electrode.

2. Theoretical amp-hours aside, nickel has been determined by other researchers to have the highest effective amp-hours per weight in aqueous solution of most anything.

  What special problems are there with making a nickel-air battery?

1. As everyone knows, metallic nickel, alone of all metals, will simply not oxidize in pH 14 alkaline solution. Non-corroding nickel or nickel plated current collectors for the positive electrode made making alkaline batteries simple compared to salt battery chemistries. But it precluded using metallic nickel as an active electrode element. For that, a salt electrolyte must be used instead. The reactions are nevertheless alkaline, but at a reduced pH, usually settling in at 12 to 13.

2. A chloride salt can't be used. KCl worked great with Ni-Mn with the manganese negative electrode. But for reasons I don't understand (and I have no formal training in chemistry) any cell I made with a metallic nickel electrode underwent continuous self-discharge in chloride solution and wouldn't hold a charge at all. Potassium sulfate seemed to work much better. Oxalic acid also seemed to work. To use an acid electrolyte one must choose an acid which won't dissolve the metal or its reaction products. Neither nickel, nickel oxide, nor nickel oxalate dissolves in oxalic acid. This is about the only common acid which will work. Most acids will dissolve most lighter molecular weight metals and so can only work with heavy ones like lead.

New House - LED Lighting

   I brought what I thought was a lot of LED "light bulbs" to my new abode, and three 4-foot, plug-in "fluorescent" style LED fixtures. I immediately replaced the incandescent and compact fluorescent "bulbs" in the house, buying more from the local building supply (at a substantially higher but not outrageous price) when I ran out.
   I took the four fluorescent lights out of the workshop. In two I put in receptacles and hung up two of the plug-in LED "fixtures". For the other two I installed simple screw-in sockets and used the "light bulbs". In one of these I put a "Y" and used two "bulbs" to light the darkest corner. Now instead of 272 watts of "Super Saver" 34 watt fluorescent tubes, it has under 100 watts of LED lights and is much brighter and more pleasant. The growling transformer hum is gone and the room is quiet. One thing I would note is that some of the 4' LED fixtures flicker at 60 or 120 Hz. That seems needless in this day and age and with LED technology - cheapskates! How can you find that out before you buy them instead of after? Some seem to flicker somewhat but only dim-to-bright rather than on-off like old fluorescents.
   I also managed to change the fluorescent tube fixture in the laundry room, but those in the kitchen and dining area are recessed into the ceiling and are going to be a greater challenge. When installed they were doubtless considered "deluxe", stylish, and "energy saving". Now the elaborate settings just made it hard to change them out for something better.
   As an added benefit to LED lights, "LED" is much shorter to type than "incandescent" or "fluorescent". Those long words will mostly fade out of common use with their lighting types.

Suzuki Swift: Burned an EV (Yow!)

   Embarrassing to admit. With the Swift not having quite enough range to get to town and back with a good safety margin, I tried a battery experiment of adding 80 NiMH batteries, in series, to the lithium ones. I thought that if I just made sure there were enough NiMH cells that they couldn't be overcharged by the regular chargers in the car, I could do it safely. But it seemed that however many cells I put in the string (I was up to 83), the charger would up the anti and put out a higher voltage. I didn't get much chance to figure out what was going on before a day when I wasn't feeling well (a tooth) and I had a long nap. Even before the nap the meter reading, another all-time high voltage, should have triggered an alarm bell in my head. And regardless I should have come back soon to check.

   Instead, the batteries were being badly overcharged and the next thought of the car was 3 hours later when the horn started blaring. The car had caught fire and the interior was a blazing mess. I put it out with the nearby garden hose through the open garage door and through the broken windshield, and shortly Tom brought the fire extinguisher, which sped the process up. The garage door above the car was charred. Thank God, the horn going off saved the house! The firewall stopped the fire so all the electric drive components under the hood, as well as the lithium batteries in their metal box under the back seat, were still fine, along with all the heavy cables which were run along the floor. I had all the components to do another electric car conversion except the meters that were mounted above the dash, which had melted.
   The loss of the Swift has already cost hundreds of dollars in extra gasoline. Someone was interested in bringing Nissan Leaf electric cars up to Haida Gwaii, which sounded great. Now he hasn't answered his phone in recent weeks. But now another person, one who lives here, is (evidently) a great mechanic, and is wealthy, wants to bring some up and to build a garage and convert cars and boats to electric here! Even better! If it pans out, the electric drive from the Swift is going to be used for the ground effect vehicle for "driving" over the water. And if it pans out, perhaps we'll be making motors, controllers and differential variable transmissions.

   One can overcharge lithium batteries too. I hooked some of the ones I got from an old Toyota Prius (which I initially thought were NiMH) to a solar panel. If all 5 cells were good, they charged up okay. But if one or two were shorted, the others charged to much too high a voltage. I guess that makes gas, which swelled the cells up.

Miles Electric Truck Chain Drive

Using the Miles electric truck... to haul branches from a cut-down tree
   The low speed Miles electric truck has been pretty useless around here so far, with the driveway going straight onto a high-speed highway. I didn't bother to renew the insurance for the road. Quite some issues back I mentioned ways I had devised to attach a motor to a vehicle wheel with a chain or toothed belt drive. I now thought about doing so with the Miles. The idea would be that the extra efficiency gained by going so directly from the motor to the wheel, bypassing the lossy transmission and drive train parts (which I would simply remove), I could use a lower reduction ratio, speeding up the truck and increasing its range by perhaps 50%, with the same motor speeds and powers.
   I'm ordering a 120 tooth and 12 tooth sprocket set for a 10 times speed reduction, from Rebel Gears. 120 teeth, with #40 chain, was the biggest that would fit on the back wheel. This is probably a change from about 15 times reduction in the original drive train. (I'll try to take the end off the motor so I can see it, and determine the actual ratio before disassembly.) That should take it from about 40 max to 60 max KmPH and similarly increase the range. It would still be more useful in town than on the highway, but it should prove the point.
   If the mechanical end of the project looked tough but doable, the wiring for the motor moved to a new position in front of the rear right wheel seemed like just as big a problem. I doubt it would be wise to lengthen the three heavy wires from the motor controller to the motor, so the controller will have to be moved from between the two seats into the back of the truck. It has a plug with 34 pins with wires coming to it from various sources, and no extra length of wire for any. I expect the only practical solution will be to cut the whole cable and splice in another 6 feet of wire so it can reach, for every wire in the cable.
   Anyway, I'll order the sprockets, and then tackle this project some time when I have some time - or perhaps hire someone to do some of it.



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



Haida Gwaii (AKA Queen Charlotte Islands) - BC, Canada

   The whole archipelago of Haida Gwaii, situated on the outer edge of the continental shelf of western North America, is about the size of the island of Hawaii, which is home to some 100,000 people. Haida Gwaii has under 5000 except in summer, when there is an influx of tourists. Trout, and sea food from shellfish and crabs to halibut and salmon, abound. There are no wolves, cougars or grizzly bears, and the large black bears aren't usually aggressive. Small Sika deer, imported a century ago, abound. With no natural enemies they are considered pests and no one minds people putting one or even a few in their freezer. I planted a garden, and it seems unlikely anyone will go hungry here if the regular delivery chain is interrupted. But the weather this spring (apparently unusually cold and windy) meant that many things would only grow well in a greenhouse, and this may still be true most years.

   One can see that if the average temperature was just a few degrees warmer, people would flock here to live. Cold winters are one thing, but summers where one might wear a sweater much of the time are considered desirable by few. Enter global warming. According to locals, last summer this was Canada's warmest spot more than once. However interesting and unusual, this occurrence was never mentioned in the news. But this summer has been exceptionally cold and wet. At least we have had no forest fires, hurricanes, floods or notable earthquakes.
   Most of the people live on the largest land mass, Graham island. The next largest, Moresby island, has just a village, Sandspit, around an airport not far from the ferry to Graham island. The airport is there, on the "wrong" island, because the large sand spit was a simple place to build an airport.

The Move

   The move to the house I bought on Graham Island didn't go smoothly. Ignoring everything, I grossly overloaded the 20 foot "sea can" shipping container and the truck couldn't budge it. Bill and I had to hurriedly take out 1/2 the entire volume and 40% of the weight and put it in the garage, which I had cleared of hardwood lumber only the day before. We hurriedly shifted the rest of the weight to balance the container, and the truck got the container off just in time to make the April barge to Masset on Graham island. I then had to get two "U-Pak" storage containers and store the remaining 8000 pounds of stuff. Of course, there was no particular order to what was taken and what was left behind. Bill refused to tow my 14' aluminum boat behind my Toyota Echo, so I had to leave it stored at a friend's. I told him he could sell it if he wanted.
   Someone with a flat deck truck was supposed to pick up my electric Miles truck, converted Suzuki Swift EV, and the Sprint project car and take them to Vancouver to put on the barge. (I had them all filled to the roof with stuff, too.) He procrastinated until they missed the April barge, and indeed almost until I had to be gone, but in the last couple of days he got them all, and they came up on the May barge. It was just as well I didn't have to deal with them immediately.
   On May 1st, days late, Bill and I headed out, he in the Echo and I in the Dodge Caravan minivan towing a big utility trailer, a converted tent trailer, with most of my remaining furniture including my bed and sofa. Most of my clothes and personal stuff was in the Caravan. There was a problem with the fuel pump assemblies on these years of Caravans, and I ran out of gas 2 miles short of the first gas station on the highway on the mainland, with the gauge still saying 1/8 of a tank remained. To make a long story short, this caused gunk in the pump line to come loose and then the pump sucked air if the tank wasn't 5/8 full or better. This problem wasn't soon understood, and the resulting confusion it caused led to storing the trailer at a storage facility in Cache Creek, and later abandoning the van 40 Km south of Quesnel and continuing with two of us in the Echo to make our ferry reservation to Haida Gwaii on the 4th. This meant I ended up buying clothes and furniture to make do (at unexpectedly high cost for the furniture - double the 1400$ advertised price after shipping et al), and huge expenses retrieving the two vehicles later on separate trips. In the meantime, I sorely missed the items they contained. And all summer I missed the many tools and items that were left behind in the U-Pak containers.

   There was certainly no lack of things to do around the new house. One of my priorities was to have four huge trees near the house cut down, and I spend weeks cutting up the branches, burning the bits and stacking the substantial ones for firewood, which I'll need this winter. It was like moving from the deep dark woods into an open field. And many other things needed doing: planting the fruit trees, unpacking, outfitting the workshop better, getting decent water from the well or from the rain, fixing the dishwasher...
   There are stores, but not many or large, and being out in the country it's a 25 Km trip to Queen Charlotte, or 80 to Masset where a somewhat different selection of good may be had. It should be a great place for an electric car with sufficient range to go to either destination, or even just to QC. But so far there's just one Nissan Leaf (and my Miles truck) on the whole island, as far as I know.


   Here are some pictures of the house and yard.


The south face of the house.


One of the trees being cut down, branch by branch
owing to limited space to fell it, on the north side.


Around the front, with the highway, my strip of trees
along the waterfront, beach and ocean behind me.


The beach below my waterfront, looking south.
The rocks and sand seem to be rearranged daily with waves and tide -
sometimes gravel, then sand, then rocks, at any given point.


looking north. Off to the right, about one day a month it is clear
enough to see some mountaintops of the BC coast on the horizon.



Looking up from the water's edge at low tide.
The highway and the house are hidden by the trees.


The big livingroom from just inside the bay windows,
(before most of the furniture arrived).



The 24' x 24' Workshop
I have much reorganizing to do.

A Trip Back to Victoria

   A couple of plans to get the tons of equipment and parts I'd left behind in two 7'x9'x7' U-Pak wooden storage containers in Victoria delivered, didn't seem to pan out. Finally I took an airplane to Vancouver and the ferry to Victoria (August 27th), rented a 20' U-Haul truck, transferred everything from the U-Paks into it with the help of a friend, attached my boat trailer behind, and brought everything up, arriving September 2nd. I drove the length of Vancouver island from Victoria to Port Hardy (500 Km), took the ferry from there to Prince Rupert (16 hours), and then the one from Prince Rupert to Skidegate on Haida Gwaii (6 hours). No more long hauls (1700 Km) through the BC interior for me - the previous one ended up costing me perhaps 6000$ when I ran into trouble with the minivan and then had to retrieve it and the utility trailer with much of my furniture separately from different areas. (Counting the extra furniture I consequently ordered in June and the pickup truck I bought to tow the trailer, more like 12000$.) This trip cost me about 5000$ - a bargain! But it cost as much to take the boat on the ferries as I originally paid to buy the boat in 1986. After I unloaded the U-Haul truck, I sent it back "hossled" by BC Ferries to Prince Rupert (meaning driven on and off the ferry by BC Ferries personnel so I didn't have to go myself) and let U-Haul retrieve it from there.


Antigravity in a Dream?

   Probably most or many people have seen a demo of a gyroscope spinning at an angle on top of a rod or post, seemingly defying gravity by not falling off the post, instead precessing around and around it as it spins. Of course, one surmises that the downward force on the post is equal to the weight of the gyroscope regardless of its motions or lack thereof.
   And perhaps many have once held a spinning object and discovered that it strongly resists being turned in a desired direction and instead tries to turn at right angles to the force applied.
   If the precession of the gyroscope is forced to slow, I presume it will start to drop until the axis is horizontal and it falls off. Alternatively, if it were forced to speed up, would it not rise up and spin more upright?
   Some of the same properties may be seen in a pendulum, which also has an uncentered pull on its axis of 'rotation'. (What happens when a pendulum is forced to swing, or to rotate, faster or slower than gravity would normally have it do? What happens if it's pushed at right angles to its usual motion?)
   Is there some unsuspected use for such properties to counteract the downward pull of gravity? I have heard rumors of such a thing, but I have been unable to form any concept of how it might actually work.

   Early on the morning of May 18th I had a dream. I was in some sort of electronics products development place. Some stuff happened that seems little relevant. Then my attention was drawn to a rectangular-ish object perhaps 18" tall and a foot square, which slowly levitated off the floor a couple of feet, moved over a couple of feet, and came down again, all in a smooth arc.
   I said, "You've solved the secret of antigravity! Wow, that's a first! Or at least, the first time I've ever actually seen such a thing." (other than rumors, I meant.)
   I said, "You call it the Pendulum, but surely it has lots of flywheels and gyroscopes in it?" (The idea of what it "must have had" may have been merely my own preconceived notions of what such a device "must have".) There was no verbal response and there was what may have been a nod - or was it just a slight smile, perhaps condescending? (Often the words of a dream bear little relation to the scene and are the important part. In this case they seemed be co-ordinate. But how did I know it was called "The Pendulum"? That term was out of the blue. One suspects that if the dream may be taken at face value, a pendulum or pendulums is or are a key component if not the the key component.)
   A small block of painted wood(?), like a child's plaything, was levitated and set square on top of another similar one, with a peg and a hole aligning them. [What was he meaning of this?... It's child's play?!?] Then the original machine rose up and then moved sideways horizontally owing to the antigravity vector being applied a little off from the vertical axis, going across just above and then settling on a counter top. There I awoke.

   The dream seemed to come out of the blue. I had had few prior thoughts about antigravity other than being sure it exists. My only somewhat recent thought was that if it were discovered and antigravity devices created in the near future, that would make the ground effect craft obsolete, and therefore not worth developing. (...not to mention all aircraft and hovercraft - maybe even commercial ships)
   In listening to the monthly ALTA web bot "predictive linguistics" report summaries on youtube, Clif High's predictions of a new field of "electro-gravitics" in the coming decades had set that thought off in my mind.

   Is it possible that somewhere in those right-angle acting gyroscopic forces, perhaps acting in some timed pendulum fashion, downward pressure can be converted to sideways and thence to up? I had long since put those sorts of ideas in the "too hard" basket. Would it take the same energy to lift an object up a foot that it does when it's lifted from the ground, but somehow lifting itself by its own bootstraps? Logic tells me that all the inertial forces must add up to zero as seen external to the unit overall; that there would have to be something more involved in order for it to work. Well, it was just a dream. And maybe there was more to the machine than I surmised. Then again logic can miss things, and there's the fact that something free to turn can pivot on any axis gyroscopically without an external fulcrum... and the seemingly gravity defying act of the gyroscope on top of a post, turning downward force into a constant velocity horizontal rotary motion.

   In embarking on a new creation to solve a problem, one must decide what project is most worth doing. Is it worth striving to create some untested, perhaps nebulous device which will probably consume a long development time and carrying a high risk of failure, or would it be better to choose something whose principles are better known, seeing a clear path from start to finish with a good prospect for success if it is dilligently pursued?
   In the infinitely variable torque converter, I saw no sure and clear path, but was reasonably confident there must be some practical way to do it. So I chose that path over conventional gearing. Even so it took over 7 years to come up with a clear vision that can surely be built as a practical and reliable design, and even now a prototype hasn't been completed. But the Differential Torque Converter may change transmission designs forever, which making something easily envisioned with fixed ratios couldn't do.
   For the subject of fast over-water transportation, so far my thought is that barring an unexpected inspiration, the ground effect vehicle is the path that can be built using known and certain operating principles with existing materials and products. But the gyroscopes and pendulums have been set spinning in my head!


Time of day

   Growing up, I had always wondered, even before we started using "daylight savings time", why the hours of sunrise and sunset weren't symmetrical. If the sun sets (eg) at 7 PM (5 hours before midnight), shouldn't it rise at 5 AM? Why was it more like 6 AM? In Haida Gwaii this clock skew is highly pronounced.
   In conjunction with my move, I found a minerological map of BC that my brother had purchased around 1980. Among other things, it had each degree of latitude and longitude clearly marked. I had noted before that each Canadian western province was a timezone apart. Alberta, for example, runs from 120 degrees west to 105 degrees: exactly one time zone wide. I thought it was a clever arrangement by design.
   But this time I realized that the "prime meridian" time zone doesn't run from 0 to 15 degrees. It runs from -7.5 to + 7.5 degrees. That means that the 120 degree line isn't the edge of the "mountain standard time" zone, it's the center of the "pacific standard time" zone. The whole western half of Alberta ought to be on Pacific time! This explained the sunset-sunrise discrepancy in my home town of Edmonton, properly at the edge of a time zone and not in the middle.
   Likewise, the whole of northwestern BC (west of 127.5 dergees) ought to be a whole time zone west of "PST". My new location on Haida Gwaii (132 degrees west) is just 12 minutes of time east of the center of this next time zone - nearly an hour off from PST. It would seem we are more on "DST" in the winter than almost half the continent is in the summer, and that this effect becomes extreme when DST comes into effect (now bizarrely for almost 8 months of the year). To tell the real time of day to within the 1/2 an hour time zone standard, the clocks on Haida Gwaii need to be turned back two whole hours. Thus it's not proper to speak of the hours of 12 and 1 during the day as "PM" (post meridian), when they occur before noon ("AM", ante meridian). Noon is at almost 2 of the clock. But speaking of them as "12 AM" and "1 AM", or of the night hours as "12 PM" and "1 PM" is bound to cause confusion.

   The net effect of all this compared to original "local time" is that the clock has become less and less a consistent, reliable indication of the time of day, and more and more a cause for confusion. More and more it is an arbitrary shifter of diurnal cycles of sleep and activity. When one hears of farmers of old rising to start work at 6 AM, here that same time of day would show on the clock as 8 AM. That doesn't sound nearly as early, but the sun is in the same place. And when a "night owl" retires at "midnight" (ie 12 PM) in June, does it really mean what people think of, when it only got dark out an hour before?

   DST is based and has been extended further and further through the year on the premise that it "saves energy", tho it seems that the only actual evidence ever collected about that is that it slightly increases energy consumption. It has also been assumed that most people work daylight hours, and would like to have more of them left to enjoy at leisure after they finish their workday. My most productive time for getting things done outdoors is afternoon, and I find that this period is curtailed. I need to quit earlier to prepare for scheduled evening engagements, which are almost always indoors anyway. So for me it is "daylight losing time", if anything - an hour of sunshine cut from my afternoon. As for the daytime employees, they had to get up an hour earlier to start work, and so are unlikely to have much energy left to do interesting things afterward. The real way to give them that extra hour is to shorten the work day, in accordance with all the labor saving devices we've developed since the 8 hour day became some sort of standard.
   Standard time zones were created by the railroads to help co-ordinate activities of related areas. To a point they are valuable, but when longitudinal boundaries are too much exaggerated the "time of day" in such areas starts to have a different meaning than generally accepted and expected usages. Contrary to what many elsewhere might surmise, people are not lazy "slugabeds" on Haida Gwaii to open businesses at 10 AM, when that's really 8:12 AM by the sun.


Sigh, Chem Trails Again!

   In spring and early summer, the aerosol spraying seemed even worse here than around Victoria. Whenever there was blue sky it was fiercely attacked by jets and hazed over, from horizon to horizon. Even the little sunlight we would have had was always dimmed; there was never a bright sunny day. The moon and stars at night were similarly washed out, only the brightest being sometimes visible. The first time I saw a real night sky here was mid August. I was feeling personally very cheated out of much fine weather by this insanity. And it seemed futile to mount the big solar panels on the roof only to average 300 watts that varies sharply by the minute all day in the thin, patchy, drifting haze, instead of 1000 watts steady in bright sunlight.
   Thankfully, as summer wore on the program seemed to have been considerably if not drastically curtailed - at least in this area.

   In his monthly "ALTA WebBot" reports this spring, Clif High was predicting unproductive crop yields this summer simply from insufficient sunlight to the fields. With seven and a half billion people to feed, this is a very serious prospect. It seems logical that with the immense scope of the program, it is getting to be somewhat like trying to grow crops in the shade, which generally isn't very productive. I don't know the main result, but I've heard that with the droughts - and perhaps the lack of sunlight - ranchers will be killing off their cattle herds again as there's now insufficient hay to feed them over this winter.

   And aerosol spraying in the evening or at night seems to make no sense whatsoever, even with the faulty idea that reflecting away sunlight will cool the ground. What effect can they possibly think they are attaining besides blanketing the atmosphere and preventing normal night time cooling?


Possible Health Ideas: Cyst and maybe Mole Removal or Shrinking?

   I've long had a cyst on my arm. Doctor's advice was to leave it alone. By a decade ago it had grown much larger and was getting irritated every time I rubbed or scratched my arm, or accidentally brushed against it. At some point I got some prescription hydrocortosone cream for a spider bite. There was lots left over. I started applying it to the cyst daily. When that ran out I switched to "over the counter" hydrocortosone cream in a tube. It seemed to work too. The cyst shrank over the course of weeks and months from being a growing concern to perhaps smaller than ever in my memory.

   There are many types and forms of moles and mole-like skin blemishes, and again for most of these the best medical advice has been to leave them alone. But we always think in terms of trying to "kill" or "get rid of" the apparent problem, and the things we try are along those lines. What if instead of that, we think in terms of "making the skin more healthy"?, that the cure is not to kill but to heal? So instead of applying some acid or freezing to try to do harm to the mole, we apply healing hydrocortosone cream to improve the healh and resilience of the damaged skin?
   I recently bought a new tube of hydrocortisone cream and am trying it on the cyst again and on some moles and the like. The cyst has shrunk further. It's barely visible now, tho a lump can still be felt. Perhaps I can get rid of it entirely. What will happen to the other things is entirely an experiment.
   The instructions on the cream say to apply it sparingly and to discontinue after a week. However, no reasons were given for these injunctions. Even Wikipedia was unenlightening. Is there a valid and convincing reason?... or is it just that it was conceived for bug bites and other short term conditions and no one has done any studies on effects of long term use? And so in fear of "malpractice" accusations, they will not commit themselves to recommending, even by omission of contrary (but equally unstudied) instructions, that it just might be beneficial in situations requiring longer term application?
   I certainly use it sparingly, but the idea of removing moles and lesions is obviously a long term prospect, as was shrinking the cyst and now maybe eliminating it. Long term use doesn't seem to have done any harm around the cyst. The only apparent effect was that the cyst has shrunk to a fraction of what it was when I started. It's hard to imagine anything disasterous like moles flaring up and becoming cancerous from this, which they certainly might from doing something harsh.
   So far after 2 or 3 months there have been no observable results except two shrinking cysts (now doing one on my leg too), and I've cut back application from daily to occasionally. If there are any eventual results, I'll write again. One hopes that at some point someone will figure out and create something that will work along these lines.


Easier Shelling of Peas and Beans?

   When we buy frozen peas or beans in a store in a big bag, we give little thought to how they are shelled and processed. When we grow them in the garden, it becomes apparent that we will not be producing enough peas to sell shelled in bags by shelling them by hand. So I did a search on line and found a couple of patents for pea deshellers. One of them rolled the peas back and forth, squeezing them a bit, with the bottom of the roller being a screen that the peas would fall through once they were free of the shell. Well, that might be a bit complex for the small gardener to put together.
   My present thought is to find a rigid screen, perhaps a plastic one such as those sometimes used in fluorescent lighting, and roll the pea pods over it by hand. If that still proves too cumbersome, an advance might be a textured flat plate to go over the top, so one would be rubbing the pea shells between the plate and the screen by pushing it back and forth, again with the peas falling through the screen, which would be dumped as necessary to get rid of the pods. The texture would be one that would grab the pods so they roll instead of slipping.

   This was just ideas. I didn't get to try out much because I only got a few peas and no beans this summer. When I finally got the peas, I tried simply rolling them gently under my hand on a breadboard. That did at least crack open the pods, and it did seem easier than prying them apart and didn't harm the peas. It worked better with 'banana' shaped pods than with rather straight ones.

Misnomer?

   Why is the leading brand of rat and mouse traps called "Victor"? Shouldn't it be called "Victim"?



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

(Will accept BITCOIN digital currency)

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




Electric Transport


Ground Effect Vehicle

   As I thought about how travel times to places I might otherwise rarely or never visit again would be sliced by a ground effect craft, I started becoming more enthusiastic (or at least less reluctant) about the idea of trying to make a "full size" manned craft - of course after building radio controlled models to test out configurations. Haida Gwaii to Prince Rupert in an hour or two is only one potential route along the coast. Victoria is either prohibitive airfare for all but the rarest occasions, or else several days travel with all costs for two costly ferries plus gasoline and vehicle wear and tear - perhaps more costly in dollars plus a high cost of time. With a ground effect craft at 160 Km/Hr it becomes less than 6 hours - a single day's travel and obviously much less fuel than by car. (That can be divided into less than 3 hours from Lawnhill, Haida Gwaii, to Port Hardy at the north end of Vancouver Island, plus 1-1/2 hours from Port Hardy to Comox, and 1-1/2 hours from Comox to Victoria.) Vancouver BC, Seattle Washington, and Anchorage Alaska could likewise become single day trips (one way) for lower cost and without the increasing complications and delays of air travel.
   There are of course various complications. Some of them stem from the fact that no facilities and rules or conventions are prepared for such craft and no one will be expecting them. Much would be gained by establishing ferry routes with terminals for larger craft, but that can't happen until smaller ones are built. Only then people will start saying "How come a few people can flit about the Earth freely while the rest of us are stuck with these cumbersome old "slow boat to China" relic sea transportation systems?"

   To that end, I gave a little thought to what such a project might entail. First, what should the basic configuration or form of the craft be? Before the Bixel design, the reverse delta wing was said to be the most stable. Why was that, and did it still apply to symmetrical wing profile types?
   Let's see... as the angle of attack increases (slowing down, or climbing), the cambered wing's center of lift shifts forward. But if the front of the wing is higher, it gets less ground effect, so the center of lift should move back. With the reverse delta, there should also be less lift overall, because a smaller area of wing is getting the same ground effect. Well, I'm not sure how much sense that makes as far as creating a design.
   With the flat or symmetrical wing, as the angle of attack increases, the center of lift already shifts backward. The rise of the front WRT ground effect can only add to this inherent stability. Why would the reverse delta help?
   A reverse delta would cause inconveniences in the design. I wanted a long rear edge of the wings, that would be awash along their entire length when floating in the water. Their buoyancy would stabilize a craft and keep it from tipping sideways. This would allow for a single hull rather than a catamaran arrangement. And the narrow gap that would open up at the back of the wings as the craft increased speed would (I expect) give the greatest ground effect lift during takeoff. That should give it the lowest takeoff speed using the least amount of power. (Craft taking off from water need substantially more power than those using a runway on land. And if ground effect craft need so little power to stay airborne, the only thing they need high power for is takeoff.) More buoyancy to the rear should allow the engine and propeller to be mounted behind the cockpit instead of in front. Finally, if a wingtip should hit the water during a turn while flying, wingtips at the rear would have the least leverage to try and make the plane pivot and spin, and that more easily counteracted by the rudder.
   On June 7th I made a paper airplane - a forward delta shape. With side rudders, and 'elevators' pointing up a little, it flew great. I decided the reverse delta was an unproductive approach, at least with the Bixel flat or symmetrical wing. The paper plane would fly with a small weight (I used a cup hook, screwed throu the paper) in positions mid to rear.

   For rough estimations I started out thinking that a typical small plane has a wingspan of perhaps 20 to 30 feet averaging maybe 5 feet wide - maybe 100 to 150 square feet. If the ground effect gave 3 times the lift, that would be reduced to 33 to 50 square feet! But the flat wing has less lift per area than the chambered wing, so that might be raised to, say, 50 to 75 square feet. If the "paper airplane" delta shaped wing was just 10 feet wide and 15 feet long it would have 75 square feet of wing area, or somewhat less (55-65?) after subtracting for the fuselage.
   It seems to me 10' is also the maximum width for trailering something on the highway without having an accompanying "wide load" warning vehicle. Being able to trailer it would doubtless be a tremendous advantage - instead of renting hangar space at an airport or moorage at a marina, tow it home and put it in the garage, where it will also be easiest to maintain and service.

   All along I've been thinking of speeds like 175 Km/H to go long distances quickly. But the thought of such high speed just over the water is rather scary, especially if one has no radar, which I'd probably have trouble fitting in such a small frame. But why would I care to go so fast in a prototype? I think I'd be more comfortable at 80 to 120 Km/H, about the speeds one drives on highways. There's more time to react, and the water, if hit, would be less hard. If the boat becomes airborne at 65-75 Km/H that should work out well, and it would need less power. (Of course it will be a great thrill if it actually flys at all! But that's why I'm doing the RC model first, to make sure the basic design... flys.) Higher speeds may be found desirable later when the design becomes more refined and operation is more familiar.

   Another aspect is the propulsion system. Taking off from water in a float plane with a petroleum engine is a power hungry affair. There might be stronger ground effects from having the rear of the wings right down at the water, but still some considerable force to get the speed up will be necessary. At first I thought a couple of hundred horsepower or more might well be useful. A big aircraft type propeller seemed inevitable, perhaps 5 feet in diameter.
Then I thought a ducted fan might be a better way to go. This gives up to double the thrust per horsepower under high load conditions - like taking off from water. Or of course it can be the same thrust from a smaller diameter. Instead of propellers of 8" to 10" diameter for the model, later I got a 5" ducted fan. That would translate out to just a 20" diameter for the full size craft. (Could I actually find such a thing to buy? The most common ducted fan type right now is the "turbofan" - the jet engine - where the multi-blade propeller in the duct is turned by a turbine.) Another benefit of ducted for the ground effect craft is that the smaller the diameter, the closer to the top of the wing it can be, making the thrust more in-line.

   In the paper airplane, bending the rear edge 'elevators'/'flaps' up or down was necessary to trim for changed weight distribution when weights were added. So rear elevators are required. The back of the wings seems like a good place for them, forming part of the wing area, rather than in a higher-up separate elevator. I originally thought ailerons would be unnecessary, but later it occurred to me that propeller torque (ducted fan or not) - or an unbalanced payload - might tend to want to make the craft tip sideways in straight, level flight. If the right and left elevators were separate, making them "elevons", any such tendency could be corrected. The elevators are essentially the same. Only the controls and linkages get more complicated. I bought one too few radio control servos to do elevons, so I'll see what happens in the model without separating the left and right elevators.

   Then there's the hull or hulls. The models on youtube were all catamarans And I thought it should have thin hulls for low resistance in water. But if the flat/symmetrical wings are down at the water (at the trailing edge) and buoyant (filled with foam), a single hull could be very thin and the craft would still be stable, not tippy.
   Then too I read in the 1980s about small (even 12' long) sailing catamarans with flat planing hulls that could outrace larger traditional design cats. (I had been disturbed to read that, having just barely completed making a boat with traditional hulls myself at the time!) With a "paper airplane" shape and buoyant wings, a single hull with a flat bottom - and a flat top - about 2' wide and maybe 18' long, seemed to commend itself. Considered without the wings, such a square cross section boat hull shape should get up on a plane quickly with relatively little horsepower. It could also be aligned with the wings and be part of the lifting body. (The pilot/passenger window canopy was going to louse that up a bit, but maybe not badly.) With the boat up on a plane (or "on the step" in aircraft parlance), the wings (touching the water at the back) would be at the right angle for take-off, and every bit more speed would reduce the contact of the boat and wing-rears with the water, which I hope will mean it will lift off gently with a minimum of power; much less than most seaplanes.

   For materials I think the frame might be a quality aircraft spruce such as was used in early aircraft. I hope to use a very large timber,
a 4" x 6" beam 10 feet long, as a strong cross spar. (One doesn't want it to fold up in flight!) If I can buy or rent a bandsaw mill, there is very good sitka spruce on Haida Gwaii, not excluding the four large trees I've had removed from my own yard, with many more on the "back four" acres.
   Back to the 1/4 scale, to start to turn the "mock-up" into an RC model, I cut two triangles of 1" foam, 12" wide by 38" long for the solid wing pieces. A 1" dowel, ripped in half, can sub for the leading edge components.

   For surface material I thought I would definitely depart from the old doped canvas or indeed any type of fabric. Perhaps lexan/polycarbonate plastic might make a great, smooth, tough skin. Otherwise some other tough plastic.
   Then I thought of epoxied polypropylene... as a last resort? Hmm, that was quite tough, and would be the thinnest and lightest! Lightness counts, so I decided on that: A single layer of PP-epoxy skin over extruded styrene foam, akin to foam sandwich boat construction. That should be pretty strong! I could use 2 or 3 layers of PP cloth on the hull bottoms so it could better be run up onto a sandy beach after landing. Well, I guess that would be a modernized version of doped canvas after all. But with the foam underneath holding it flat.

    The wing(s)... The foam could fill the wing solid, 4" thick, with a semicircle front edge. (With gaps or holes where wires or cables need to run through?) The rear would have the elevator or elevons tapering from the 4" thickness down to a fine rear edge. The PP-epoxy skin would wrap around the whole wing.

Another Design Revolution (Ack!)

   On August 16th I went to town to the library to get on line and order some parts at Digikey. Afterward I went to youtube, and a new radio controlled ("RC") ground effect craft video was there. This one was a rectangular shape with tall but very narrow "hulls" on the outside and a wing between them. When it moved forward, air was trapped under the wing by the two hulls at the sides and by the rear of the wing. So it would fly with three sides almost touching the water (or the ground in the video) and only the front was open, to scoop the air in. Surely this must provide the greatest lift, with no lift-losing air vortexes at the wing tips! And it looked like the most wing area per overall size.
   When something did touch ground it seemed to be of little consequence. Other than not turning very well and often flying somewhat diagonally like a hovercraft, I decided it seemed to be a better plan. It didn't look very glamorous - something like an overgrown snow sled - but it was simple and should lift off the water at the lowest speed with the least amount of power. Some of Bixel's drawings showed this sort of shape but with stubby wings added to the outside, and some of the RC videos had rectangular units with stubby wings. On one it said they had been added later and that they improved the performance considerably.

   So, not being very far along yet anyway, I reverted to the rectangular shaped catamaran form as it seemed to be the most practical. Without stubby outside wings it would be a rectangle with buoyant hulls at the sides - easiest to dock, embark and disembark. If outer wings are being added to improve lateral stability, as seemed to be the case for at least one model, perhaps the whole catamaran form should just be more square instead of rectangular, or even rectangular but more wide than long.


   I got the rest of my stuff from Victoria at the start of September, including the radio control gear. None of the screw holes and fittings on the motor fit those of the ducted fan propeller. Sigh, I'll have to cludje it in there somehow! Anyway now I can start to actually build.

   As I was finishing up this article I suddenly wondered if the ducted fan propulsion unit could be under the front edge of the wing instead of on top of the wing? Blowing air directly under the wing should considerably improve the lift for take-off at a lower speed. But that would put it virtually in the corrosive salt water. Perhaps it could be mounted at the front and placed in-line with the wing, with half the air going over and half under? Or mounted in front of and above the wing, angled or ducted to blow it underneath? It could pivot to blow down at an angle for take-off and then be turned straight once up. Potentially, the wing could even have hinged flaps to cover its front edges down to the water, making it a hovercraft when moving slowly. The flaps would fold down to vertical as the air lifted the craft out of the water, giving it a hovercraft "skirt" on all sides, then fold back flat against the underside of the wing by air pressure as it picked up speed. That would surely take the least power of anything to get off the water, lifting out even before gaining much speed! Such flaps would be an added complication, but they would make it almost certain to work. But would such an operating mode be necessary or desirable? More reasons to build the RC scale model to try things out safely without too much effort and expense!



Electricity Generation


   I moved up to Haida Gwaii with the idea in mind that I could probably enlist others in creating tidal flow plants and wind power, and perhaps wave power. However, the interest level for actually doing anything didn't seem very high and those already doing tidal power didn't seem even to want to visit. If I was going to do a project all by myself, which one would be best to pursue? After some time thinking about the tidal flow and perhaps ocean wave power, by mid July if not earlier I decided the lambda ray converter was the project to work on. Before that, I was thinking about the tidal and wave power.


New Water Flow Turbine Design Thoughts

   One aspect was the site for a first project. The rivers of Haida Gwaii near populated areas seem to flow too slowly to consider, the land being pretty flat. The channel to Masset Inlet fills and empties a small inland sea with the tides - much too ambitious a project. As I had surmised from the satellite images, the Delkatla slough off to one side of the channel seemed suitable if not ideal for a small project. All the water flowed in and out under the road bridge with the tides, to fill and empty a very large nature sanctuary reservoir. There were buildings nearby to connect the output to the power lines. Stopping for a look whenever I went by, eventually I had examined it at several points of tide. It had sufficient depth for some sort of shallow turbine even at low tide, in a channel at least 15 feet wide. Of course I would use the improved Piggott alternator on the end of a "spiral staircase" turbine shaft, with solid core foam sandwich catamaran hull construction for floatation. Then the question became details of the turbine design, and how big to make the unit. It could well be 15' wide and 30' long, but if it was too ambitious, it might not get finished and installed.


This almost empty estuary (much more estuary around corner far left!) fills to the brim and empties
again every 12 hours, with all the water flowing in and out under one narrow bridge.


Under the bridge at the lowest tide.
There's still quite a bit of water.


At a higher tide, flowing outward (to left).

   On the 15th of May, I came up with a new design idea, again using the "spiral staircase" as a model to tap the flow up and down the stream as well as laterally. This time, the horizontal axle would be out of the water, and each "stair" would be a paddle sticking out from it. For river hydro, each "paddle" could be an ideal twisted propeller "wing" shape for maximum lift to drag ratio. (...or is flat a better lift to drag ratio per Bixel, using slightly larger blades?)
   In the case of the intended installation, with bidirectional tidal flow and a unit that won't pivot endwise as the current reverses, that "propeller" shape wouldn't be possible. But perhaps one could just buy canoe paddles or oars, making the project simpler. But a flat twisted paddle shape should work either direction. Brackets would attach the "handles" to the shaft, all in the "spiral staircase" pattern. This design has the advantage of potentially capturing a lot of power, and of being highly configurable in the field in several ways:

1. The height of the shaft above the water would control how deep the paddles would dig into the water.
2. The length of the blades would control the speed of the shaft, the "gear ratio". Short blades would make for a higher spin rate. Long ones would utilize more width of water. It should be possible to mount them so they can be "telescoped" in and out to some extent for adjustment.
3. For example, at 45 degrees twist the paddle would try to move sideways (thus turning the shaft) at the same speed the water was flowing. At 60 degrees it would be around half the speed, and at 30 degrees it would be double. Most ideally (but I think out of the question for an initial unit), a variable pitch would be computer controlled - or perhaps centrifugally controlled - to provide constant RPM and voltage at different flow rates.
4. The number of blades on the shaft could also be adjusted - more blades, longer "staircase", more torque, if the flow has the power in it.

   Being able, within limits, to adjust the speed of the shaft will hopefully mean the imroved Piggott generator can be mounted directly to the shaft, instead of having to be geared up or down.

   A neighbor said he had a 1.25" by 12' stainless steel shaft, from a sailboat propeller, that sounded like an ideal axle. If any other people had been interested, I might have started construction. As it was, I decided to pursue another power generation idea instead.


Ocean Wave Power Thoughts

   Once I had moved into my new abode, the ocean view was right out the window. Furthermore, it seemed I owned the whole waterfront across the highway, a narrow strip from a little north of the house and all along 1.5 Km to the south. Somehow, when the area was subdivided, the BC highways department (how did it get to be their decision?) wouldn't allow the beach side to be associated with the adjacent properties across from them and title was retained with the original property - mine. Solutions to problems that had occurred to me before started to appear.
   For the first prolem, WHERE?, the solution was now obvious: I had my own waterfront for a base.
   The second problem was how to deploy the oscillating wave power buoys. I left my 14' aluminum boat behind in Victoria, with prospects of getting it back doubtful. (I got it - in September) Since it was a test, they didn't, at least initially, have to produce continuous power. Being a flat, sandy beach, I started thinking that I could simply wait for the lowest tides, and then go out to the water's edge and plant the buoys there. Maybe use hip waders to get a little deeper. As the tide rose, they would be immersed and start working.
   Part of the beach was rocky, and the problem of how to anchor the floats was solved when Tom suggested I just use the biggest rocks that could be reached at low tide as anchors. That sounded ideal except that the floats would then be banging up and down on the rocks at some points of waves and tides. I would rather they were in a sandy area. But then I thought one could tie them to sturdy bags, and fill the bags with sand on the spot. Then tie the bag closed. Far more weight for anchoring could be shoveled into a bag than could be readily carried down a long beach by hand as an anchor weight.
   Then, how to connect to the power grid once things were working? This tied into another idea, which Tom came up with. I had purchased a 26' travel trailer for guests. Tom and his family came up to Haida Gwaii two weeks before me and stayed in it, and had found the only RV camping with electrical plug-ins was at Port Clements, 40 Km from the house and 65 Km from the ferry landing. That was over 1/2 way up the island and there was nothing any farther south. Tom said people were paying 40$/night for RV sites with plug-ins. Let's see... 4 RV pads at 40$ would be 160$/night for July and August, say 50 days, 8000$/year. In a couple of years that should pay for putting a power station on the ocean side of the highway! IF permission could be obtained. I would of course make the installation big enough to accommodate net metering "for some solar or wind power"... and of course the wave power equipment as well. But perhaps it might be possible to have a conduit pipe run under the highway? (must investigate.)
   A new problem came up... Where I had hoped for a rocky outcrop overlooking deeper water, the beach was flat and wide, and it was going to take a lot of wire to reach the ocean, even just the low tide shoreline. 3 buoys, 6 wires. Or they could be connected together in parallel, but then they would need a wire 3 times as fat, gaining nothing. I've being calling this metal "copper bullion" it's so expensive. (and it's in the same column in the periodic table as silver and gold.)
   I finally figured I could put them in series. One wire to the first buoy, from there to the second, to the third and back to shore. The problem with that is that the voltage is 3 times higher, eg, 120 volts instead of 40. I was trying to stick to safe levels. Here the "deploy at low tide" testing model comes back into play: when the tide is low enough, the buoys will be resting on the sand and producing no power. Only service them when they aren't making power.
   Another possible saving also occurred to me: The water itself can be used as the ground connection. BC Hydro does that for major power connections. Instead of running the second wire, just dangle it in the ocean, and use the ground rod in the power station to complete the circuit. Withall we're down from 6 wires to one - a huge cost saving.

   All this of course presupposes buoys, turbines and generators that are working. They all need to be built. Even for testing I think the buoys need to be at least 2 meters tall, preferably 2.5 m. I expect even a single full size Piggot alternator with two rotors, with a matching Tesla turbine, would be overkill for the likely output. Perhaps I should think about smaller alternator models. or bigger buoys. 3 m or more would be a better working height buoy, and if fairly wide in diameter they might manage to productively employ a full size turbine-alternator pair.

   Further observation of the waves showed that they stayed under a meter tall even in pretty strong winds. Apparently the spit of Sandspit continues far north just under the water, making a shallow zone, or perhaps two of them, about a mile offshore. The ferry has to take a circuitous route to avoid potentially running aground. One sees whitecaps out there at the marker buoys and at the shore when there are none in between. I should check further north, at Tlell, to see if they are bigger there. This seemed a bit disappointing. OTOH, one can make smaller test unit buoys for smaller waves. But there are logs tossed much too high up on the beach than the waves and tides I've seen so far could account for.

   Even once things are proven and the buoys are in full time operation in deeper water, for series connections there's still the dangerous voltage problem. But each buoy could be isolated (cutting all the power from that string of buoys) with a double pole switch on each one. Or bypassed with a DPDT switch. If the buoy is to be physically removed for servicing, the one on each side will have to be turned off and the string will be out of service. A series string might contain more buoys, perhaps even 6 (240VDC peak) or more, if the safety shutoffs are installed and precautions are carefully followed.


VHE/Lambda Ray Converter

   The box of components for the lambda/VHE ray converter/collector/electricity generator had arrived in the shipping container in May, but the computers had got stuck in Cache Creek until Tom and family went and brought up the trailer I had cached there. It arrived on July 28th. (That left only two "U-Pak" containers with 8000 pounds of my stuff, and a few miscellaneous items - mainly my 14' aluminum boat - still in Victoria.) I still just had occasional internet access, but once I had the old iMac set up, I could at least read my own old newsletters, stored on the hard drive, and it also had the circuit board and schematics for the lambda ray device, done in EAGLE PCB, and the manuals for the MSP430 series microcontrollers. The vital netbook with the MSP430 microcontroller programming software and system was also on the trailer. Now I had the necessities to proceed with this particular project - and almost no others.

   My initial idea was to have all three coils oscillating each at its own frequency, thinking of Mark's harmonic 'chord'. At the end of July, it occurred to me that perhaps only one coil should be driven at a time. One would "multiplex" them, driving them each in turn in a 1-2-3-1-2-3 sequence reminiscent of a three phase motor. Perhaps that would get the currents running around the collector ring?
   All three would be timed the same, in a sequence, adjusting the times on and off of all three as the output voltage rose and fell around a desired value, generally raising and lowering the frequency. The overall frequency would be determined by adjusting both the pulse width and the time until the next coil was "fired". According to Steven Mark, 5 KHz might be an overall "round the ring" frequency to start from. But Mark also mentioned "striking a chord" of three notes, including the "second harmonic". Various things may have to be tried. But perhaps a variety of different pulsing strategies and frequencies can all work? Of course, I'll probably settle on the first one that does and never try any others!
   At first I was thinking there was probably a "DC Kick" energy burst at both switch on and switch off time. But it probably only occurs with the high voltage spike as the coils are switched off. So the pulse width probably only affects the energy of that single spike.

   Anyway, I set up the iMac on the 29th and read over my own TE News reports on July 30th and 31st, and especially after going over #94 and #95 (the last time I had worked on it, a year and a half ago) I started to remember what I had done and what was to be done next. It seemed the first thing would be to hook up the display-controller to my lambda board and try to program it to display and read buttons properly - the same as on the "Launchpad" MSP430 development board but using the pins I had assigned on this board to connect it. That would tell if the board, the microcontroller and the display interface worked properly. The output voltage and wire temperature sensing would be next, and then the coil drivers would be tried, without and finally with the coils. THEN, if all worked well and the gate drive voltage wasn't too low to turn the power mosfets fully on, it would be time to put it in the box and try electricity generating strategies.

   On August 1st BC Hydro was doing some work down the road. I considered asking the workers what people might think about the idea of "shooting" a conduit pipe under the highway to get electricity to the beach side of the highway, to provide power from my house for a little "Lawnhill Beach RV Park" with 4 or 5 sites. As I watched while they replaced a power pole, it dawned on me that if I had VHE ray power working, there would be no need to run a cable. Power could be generated on site - probably as DC with inverters to provide 120 VAC. If I worked at it fairly diligently, and if it was as simple as Steven Mark seemed to suggest and no serious complications arose in the implementation, I could probably have it working this winter for use next summer. That was probably as soon as I'd be able to get a cable under the road anyway -- and with a lot less hassle, explaining and site plan drawings, begging for permissions, permit fees and actual contract costs for the work. I walked back home without talking to them.

The Lambda Board - Initial Tests (no, didn't even get to trying lambda ray conversion here)

   I worked on the software and port pin allocations for the Display LED/Buttons interface and sundry things for a few days. On the 9th I was ready to transfer the microcontroller from the "Launchpad", Texas Instrument's development board, to the "Lambda Ray Converter" board I made in 2015.
   When I finally powered up the board, the display flashed and the IRF7307 pre-driver mosfets got hot. One of them fell off the circuit board: its solder had melted. I checked the pinouts and racked my brain. It looked right. But where did the heat come from? I powered the board up again. One IRF7307 (two remained) got hot and the other didn't. I looked at a couple of the sample TPU driver circuits. The drivers were identical to each other. The two transistors of the IRF7307s seemed to be wired backward (N channel FET on top to Vdd; P channel on bottom to ground - ???) and thus it seemed they would only work "accidentally", with probably very slow "drifting" switching speeds. I thought the IRF7307s were there to improve the switching speed! OTOH... my way obviously wasn't working right. I checked some CMOS device datasheets that showed the internal schematics. In each one, the mosfets were all the same way around as I had done it. No other circuit had them the way around the TPU experimenters had done it. I checked the pinouts of the chip from the IRF7307 datasheet against the EAGLE library part, they seemed to match. Nothing mirror image...
What, then? CMOS Latchup? That too seemed unlikely in this day and age. And why did the TPU experimenters use this strange two-transistor driver chip with its own awkward drive requirements? There were other suspicious points to the TPU circuit drawings: why was a power mosfet rated to turn fully on with 7 or more volts on the gate being driven with only 6 volts? And why were there no flyback diodes on the coils? Was that a special feature of the design or a glaring omission? Every look at the circuit raised more questions than answers and I wondered: Had the first person, without knowing much about power MOSFET design, just cludged together something until it seemed to work for him, and then the others merely copied what had been done? OTOH I thought I knew what I was doing, and I could hardly say my own board was working! There seemed to be more to the IRF7307 than met the eye.
   I looked over the board and noticed the two main filter capacitors were in backward. That couldn't be helping! I reversed them and tried again, with two fingers on the remaining IRF7307.s to feel if they heated up. The board didn't work any better, and one of the tiny chips heated up and in about one second left a little square burned spot on my fingertip. The other seemed to stay completely cool.
   At that time I decided to eliminate them entirely and drive the power mosfets directly from the microcontroller with the 5 volt logic. With the chips gone (or maybe even with them) the microcontroller put the "HELO" message on the 4 digit display. But only sometimes. It seemed to flash between "HELO" and "8.8.8.8." so that it looked like both were going on continually. I suspected a repeated "RESET" function was happening. I found MSP pin 16 was (among other things) "RESET*". I managed to dig out a stray 10 KΩ resistor from a box to pull pin 16 to Vcc. (Even my resistors were in a box still in Victoria. Sigh!) I plugged in the power and this time it worked. The microcontroller hadn't been damaged. For me, that seemed like a big step - from running the microcontroller on a development board all made ready-to-go by Texas Instruments, to running on my own circuit board. Except for the gate driver problem, the power supply capacitors orientation, and assuming the microcontroller's *RESET pin would have had an internal pull-up, I seemed to have got everything right, or at least close enough!

   On August 13th I somewhat idly looked at the circuit board with a magnifying glass. There was a solder bridge short circuit by one of the IRF7307 pads. Did I do that when I was unsoldering them, or was it there when I was testing the board? I wasn't sure it explained the problem... but I decided to put one 7307 back in and turn it on again. It seemed fine. I put another one on and it was still fine. I put the third one back and it was still fine! Well, I assembled that circuit board a year and a half ago, and I had never got around to testing it. Perhaps I hadn't inspected those miniscule chips and their close spaced leeds after I soldered them on. My eyes certainly aren't good enough to spot flaws at that scale without a magnifying glass. Anyway, it seemed there was nothing wrong with the design after all. So that seemed to solve everything, and I would try driving the coil drivers with 5 volts instead of 6.

Voltage Issues

   For a day I thought all was well. But as I looked up specs or the MSP430G2553 to set up more things in the program, I ran across the fact that Vcc was supposed to be around 2.2 to 3.3 volts rather than the old standard of 5.0 volts, with the "absolute maximum" being 4.1 volts. I checked on the development board and sure enough it was regulated down to 3.55 volts from the 5 volt USB power. I was glad it had run without blowing up when I fed it 5 volts on the lambda board, and continued to do so in further tests with the aberrant gate drivers being soldered back on.
   When I was first checking out the MSP430 series microcontollers, I remember reading that if Vcc was above 3.3 volts, it would be regulated down to that value. Thus I thought applying 5 volts was no issue. But perhaps that was for some other specific member of the MSP430 family whose datasheets I happened to be reading, or to a development board, and didn't apply to the MSP430G2553 that I eventually chose. I note that when I start the debugger, there is a note: "Warning: Device does not support power profiling." Maybe that means it doesn't have the flexible supply?
   My board had a place for a regulator, but apparently it now needed a 3.3 volt regulator that I didn't have even in a box in Victoria.

   This created another headache besides just the microcontroller power: The output pins would (and did) also have this lowered voltage. Power mosfets generally should have 10 volts on their gates, and are quite underdriven at 5 volts. At 3.3 would they even turn on? The graphs in the IRF840A datasheets didn't even start until 4.5 volts, and they looked a lot better at 7 volts than at 5.

   In preference to immediately scrapping the prototype board, should I just run the MSP at 5 volts? It did work - so far. Better, maybe I could cut the trace, and put in two diodes to drop (only) the MSP's supply to 3.8 volts. Then everything would be within specs... Except, according to the datasheets, with the IRF7307s having a 5 volt supply, a 3.8 volt input would turn off the high side transistor (ever). The lowest the logic high voltage could be was Vcc - .7v , so the MSP couldn't be less than about 4.4 volts: one diode drop. Still, that was closer to its maximum 4.1 volt rating than 5 volts and so less likely to suddenly or eventually fry it, and since it did work at 5 volts, I decided on that plan. And if 5 volts was too low for the coils to initiate VHE ray conversions, the power mosfets could be given an external supply of any voltage, like 9, 12 or 15 volts, or even higher - 24?, 48?.

   In my initial ignorance, and trusting others must have known what they were doing, I copied the circuit with the IRF7307 that a whole batch of experimenters had apparently used around 10 years ago... without realizing at the time that they seemed to have had it wired wrong. When I designed the "TPU driver" second board after my initial experiments I corrected this design mistake (notwithstanding any missing or shorted soldering initially causing trouble with the assembly). But it's a poor choice for microcontroller based control. A chip for that should take 'normal' logic input levels, but the outputs need to be a higher voltage for switching
the power mosfet gates - at least 7 volts, it would seem, for the IRF840, and in general 9 or 10 volts or more. I decided to look for an alternate chip. I had some IRS2003 hi-low mosfet half bridge drivers from motor controllers that should work. I could just ignore the "hi" driver and use the "low". Seemed a bit silly. For a low side only N-channel driver, I also found I already had a datasheet for something called a "TC4429 6 amp power mosfet gate driver", that took a logic level high input of 2.4 or higher volts. (TC4429 - inverting; TC4420 - non inverting) Later I found on line at Digikey a "FAN3111E" that looked good, and noted another number 'IR44212' whose datasheets I didn't have time to examine. But there again any of these would be 3 chips for the 3 outputs. Wasn't there a triple or more power mosfet driver somewhere, with logic level inputs, to do all three with one chip? I didn't find one. In the end I went on line again to Digikey, and from a horde of search results I ordered a simple 4 pin chip that I didn't even remember the number of: Ground, Vdd (up to 15+ volts to drive power mos gates), logic input, and drive output. (It was a Micrel MIC4417, with inverting logic. I should have used a MIC4416, non-inverting, but I hadn't seen those when I was ordering. "There are more types of IC chips littering the Earth than are drempt of in your philosophy, Horatio!") I also ordered a selection of resistors, 100 of each, since mine were (sigh) still in the U-Paks in Victoria. (Crap, after all those resistors... I forgot to order the 3.3V regulators! I meant to write up the shopping list before I went. I was pleased enough that I remembered to write down my Digikey user name and password before I left for town, or the trip could have been a waste.)
   Obviously the next version will incorporate all desired changes, and obviously there will need to be one. Success would want migration to a better 'pre-production' prototype. Failure might be because of weak, slow drive switching to the coils owing to low voltages dictated by the original poor circuit choices, so a second, better one should be tried anyway.

Further Programming of the Board

   On August 15th and 16th I cleaned up some of the programming and added new sections. I decided to do everything in the main event loop with interrupts disabled. (I changed this later to handle timerA 'ticks' in an interrupt for precise timing of the coil drives.) I made the effective analog input reedings (output volts, wiring temperature) show the average of the last 4 analog to digital conversions (ADC) to filter noise out. I added user input buttons for Start converting ("ON-strategy 1") (lower left button) and "OFF-Stop converting" (lower right). I added automatic "OFF" if the unit got too hot or if the output voltage got too high. Restart for now is manual only. But I had yet to determine just what numbers to expect for what temperature and what voltage.

   On the 17th I broke the power trace to the MPU and put in a diode (a surface mount 1N4148). As it ran I checked the voltage and it was 4.6 volts instead of the expected 4.3 to 4.4. That turned out to be because the "5 V" power adapter was 5.2 volts. So 4.6 was still .6 volts less than it had survived previously, and if it should continue to work at 4.6, the 5.2 to the IRF7307s is just that .2 bits extra to drive the power mosfets with - still 1.8 volts or more below optimum.
   I also started to tackle the problem of just how and when to turn each coil on and off in software. I decided for the first strategy that only one coil would be on at a time, in a rotating sequence like a motor. If it's a linear relationship, then the faster the pulses, the more output there should be. Or perhaps at some frequency each one should be starting just at the right time to catch the 'DC kick' from the previous one, and in the rotation, Steven Mark's "kick of the kick of the kick" should kick in to produce as much electric current as desired from small changes in frequency. In that case, the adjustments will be fine and rather critical compared to the first case.
   How long should each coil be turned on? I decided to try 1 TimerA 'click', then 2, 3, 4 ... As with the motor controllers, the longer the pulse was on, the higher the current would be. Once the pulse was turned off again, the number of timerA 'clicks' that went by before turning on the next one would determine the overall frequency of the 'rotation'. That would be the thing to adjust 'on the fly' during operation. But it might be that the pulse widths would also need to be set, perhaps depending on the load. 10 watts output load was almost sure to need substantially different settings than 1000. For testing I decided to use two of the buttons to increase or decrease the pulse width manually, 1 to 7 'clicks' of timerA. I made it so changing the setting would also turn it off, requiring manual restart. We don't want the voltage to suddenly multiply by 2 - or perhaps by much more - if the ON time was doubled from one tick to two. On changing it, the display says the new pulse width, eg, "PW=2" ("W" isn't a letter a seven-segment display does well. I did it like a "u" above a "u" - which might be viewed as an "8" with the top segment missing. Best I could do!)
   On the 18th I got some of the "real time" coil switching sequence done. Each coil is to be ON for time "PulseWidth" TimerA clicks, then OFF for "PulsePeriod" clicks, which will presumably be substantially longer than the ON time. When the OFF time ends, the next coil in the sequence is turned on and the ON and OFF are repeated for each coil, in rotation.
   PulseWidth is set by the user per above. PulsePeriod, to say it simply, is to be determined by monitoring the output voltage, shortening it if it is too low and lengthening it if it is too high. That's the initial "Strategy 1" for trying to coax the lambda rays to release their energy into the wires in the "moebius strip" double loop collector.
   After a few days off, on the 24th I did some more, and decided that the actual switching of coils ON and OFF would be done in the TimerA 'tick' interrupt, as would checking the readings and updating the automated aspects of the output control, once as each pulse was turned on. I decided I should try for 120 volts output, and run common electric heaters and incandescent light bulbs for loads. (I'm not worried about burning out light bulbs: I have a whole shelf full of them since changing the house to LED lighting!)

   As the days went on I got more into the details, and became very exacting over every small point. This is good because glossing over seemingly trivial program details has crashed space launches and various other side effects. I wrote not only code but extensive comments in the code, explaining every variable and every little instruction sequence in great detail. Most of the text of the program is explanation, with a few actual machine instructions inserted here and there. If somebody comes along to make their own version converter with some other microcontroller and other hardware, they should have no trouble understanding what each little bit of code is supposed to do and how it all fits together. Poorly commented code is the bane of the next person to read it - not excluding the original writer of the code, who won't remember it when sufficient time has passed. (I once wrote a "paint" or "fill" routine for the MC6809 with almost no comments, back in the days when every byte was precious, and some years later I was unable to figure out what was going on in order to write another one for a different CPU.)

   Aug 25th: An analog input of 0 V gives a reading of binary 00,0000,0000 ; an input at or above Vref=2.5v gives 11,1111,1111, or Hexadecimal 3FF or Decimal 1023. Assume Vref=2.56 volts and you get 2.56/1024=.0025 volts/step. If the temperature resistor is 2.5K ohms, with the AD590 sensor putting out 1 µAmp/°K, then 0°C=273°K=2.5 mV/°, so a reading of 273 would be 273°K=0°C and (eg) 328=328°K=55°C. The output voltage is divided by 200 before being fed to the ADC. [5 KΩ/1 MΩ=5 mV/volt] so a 100 volt output would be .5 volts. .5 V/.0025 volts/step=200. So the reading is double the actual voltage, or each step is .5 volts. If I wanted to make the reading number correspond to the voltage in volts, I would have divided the output voltage by 400. These are rough values, since the actual ADC reference voltage is stated as being 2.35 to 2.65 volts - plus or minus 6%. Close enough, I guess.


Steven Mark Review

   Because Mark had originally used vacuum tubes, I had guessed his work probably started in the 1970s or so, not seeing any dates on or in the document. (OTOH, there was no such thing as e-mails back then!) But on this reading I noticed a compilation date, 2007, and a couple of dated e-mails from 2006. It mentioned Mark had worked for 15 years on the units. So perhaps he started a little before 1990. (Where could one still buy vacuum tubes in that day and age?) I'm not sure this was the same copy and version (it said "version 2") as there were a couple of e-mails I didn't remember and I didn't remember it having any illustrations, of which there were several. But where could I have picked up another copy? So probably it was just my memory... which was a good reason to review it again after so long. There were still the same e-mails about the units and how they worked, and the ones saying the US government had been threatening to arrest him for trying to explain to people how the energy conversion process worked, with the meeting at his attorney's office with DOJ, FBI and (!)AEC. (Of course, AEC's funding would be curtailed if everyone had a non-nuclear source of plentiful energy. In fact, it threatens every organization with a vested interest in "pay per kilowatt-hour" and "pay per gallon" energy rationing as it exists on our planet today.) One suspects Mark has probably been in jail since 2007. It doesn't look like there's any "due process" of justice if any official in the government decides you're working against their biased idea of "American national interest", especially since due process and fair trials have been officially removed in legislation, for example in the so-called "Patriot Act". (What Orwellian doublespeak!) And it's probably even worse in Brazil where the two inventors who were apparently successful in 2015 and put up a video on youtube about it, hoping to sell their devices, were quickly arrested and (having had no time to defraud anyone of anything) charged with "fraud". (If only "the law" was 1% that fast with frauds in the billions by the big banks!)
   I read through this 63 page PDF file (Aug.14th?) and came out with at least one new factoid that I had missed in the reading before: Mark always used 3 or more horizontal collector toroids, each with its own set of 3 or more vertical control coils. These formed 3 rings on top of each other. Then that had another set of vertical control wires around the whole thing. He said that gave "more options" and was easier to keep under control. Newly understood by me was that when he spoke of "striking a chord" with three related harmonics that caused the great amount of power conversion, it seemed one frequency was put into each collector ring. This may mean that my one "moebius strip" collector ring by itself won't work. And yet I had seen others that were simpler and claimed to work. The one "moebius strip" ring "Otto Ronette" experiment whose coils had I copied had struck a disaster when all the connected test equipment was damaged or destroyed. That was certainly indicative of a sudden power conversion "success".
   Mark also noted that one had to have a manual "kill" switch, a temperature sensor "kill" switch, and an overvoltage "kill" switch. Especially the overvoltage can happen much too quickly for human reaction and destroy everything, with melted wires and burned out parts. With microcontroller control, one pushbutton can manually tell the unit to stop, and I put in a temperature sensor input in case it gets too hot. The voltage sensor is intended to tell the processor the voltage, and have it adjust the pulses to regulate the output. It shouldn't need an overvoltage "kill", just to skip a beat or two or adjust the frequency, cycle by cycle, as the voltage goes up or down. It's the critical control that Mark's units evidently lacked. He never spoke anywhere about computer or microcontroller control of the units, only of tuning or detuning oscillators.
   Mark noted that the output is mainly high voltage DC but with smaller high frequency AC components - 5 KHz switching frequency was mentioned several times. I was going to put in a diode to make the output DC. Apparently that shouldn't be necessary. It just needs some filter capacitors at the output to damp out the smaller AC component. I have no idea what voltage and current to expect. One sees Mark's RF 'flames' in one image, and mentioned in a couple of the e-mails, and thinks of high voltage. But it may not be so.
   Ideally from a practical standpoint it would output 12 VDC, since there are so many mass-produced things that run on that voltage, including inverters to turn it into 60 CPS, 120 VAC. But it may be that it has to put out 100 volts or more to get "avalanche" conversion of the VHE lambda rays.
   Another thing Mark noted was that he had to put the control circuitry in the center of the toroid owing to all the electromagnetic noise getting into the transistors, turning them on and off at the wrong times. (I've had enough of that sort of thing in my motor controllers!) I guess it all canceled at the center of the ring. I plan to put the whole toroidal assembly in the breadpan to shield it (really, to shield everything else from it) with the control board outside, but I'll run the wiring through the top cover at the center to hopefully achieve the same effect.

   Moving along to later experimenters, I may want to beef up the physical coils and their supports. Mark had mentioned some strange effects, such as the device acting like a gyroscope when in operation even tho nothing was visibly moving. (Could that relate somehow to "anti-gravity"?) The writer of "2magclashtpu-V1_3_2_1.pdf" (or "2 FREQ-MAGCLASH TPU Ver. 1.3.2.1  - 05-23-2007 by ronotte") notes:
The TPU is embedded in cork to stop it from vibrating to pieces. Each time the magnetic field moves between the torsion state and the Electric state it creates a small jerk and makes a physical motion of the wires. The TPU exhibits an inertial momentum, or gyro effect because its Proton layer is in a spinning motion of its magnetic poles, this creates a "forced precession".
   Whether most of that describes the physics well or is mumbo jumbo, there's nothing unusual about wires getting vibrated around by magnetic fields. That's why motor and transformer coil wires are solidly varnished or epoxied into place - otherwise they don't last. Apparently, and not surprisingly once it's mentioned, such techniques may also be appropriate for the coils of a VHE ray converter.

Notes for Version 2

* change mosfet gate drivers per writeup.
* Board Voltage: 9 volts to 15 volts to the board (gate drivers & coil drive voltage), 3.3 V regulator for the MSP chip.
   (Apparently there's no use for usual standard 5 volt supply.)
* Add ground pin to unit sensors header connector (5 pins instead of 4, or drop the 'extra' AIn pin).
* Connect coils thru header connectors instead of soldering the wires to the board. Then the whole board can plug in and unplug.
* Since there are only a couple of "display-controllers" left and no more are to be made, find an alternative SPI display & pushbutton system. (Of course that will mean reprogramming for the display and user inputs, but these can doubtless be simplified once control procedures for different loads, etc, have been established and stabilized.)


How it Works: A bit of Speculation

   On rereading some material and thinking of the RF burns suffered by some, I started thinking of the VHE ray energies as being released not as current flowing in the collector coil directly (well, duh!) or even in the control coils, but as electromagnetic energy appearing in space in the vicinity of the coils, and thence inducing current into the collector, in the same manner as a radio transmitter can induce an incandescent light or a fluorescent tube to light up simply by it having an unconnected wire or even a finger (a receiving antenna) connection and being in proximity to the transmitting antenna. (A big thanks to electronic technician Maynard Atkinson for demonstrating these things to me at Comox airport middle and outer ILS (Instrument Landing System) marker beacon transmitters in 1975!) The "collector coil", then, is essentially a "receiving antenna" for the liberated electromagnetic energies. It follows that the shape, size, orientation and position of the collector coil wires determines their effectiveness at picking up the radiated energies - and that other metal things nearby might get hot or might pick up electrical power as well, as happened when "Ronette's" unconnected aluminum heatsinks got hot.
   I suppose much of this isn't so different from a transformer, where the voltage across the primary windings is turned into magnetic flux in the core, and the magnetic flux is turned back into voltage and current flow in the secondary. But here the "primary" is the control coils whose sudden high voltage spikes cause the release of VHE radiant energy into the "core" area, the "core" is an airspace (in most designs), and the secondary, at 90 degrees to the expected magnetic alignment for a transformer, is the "collector coil" (perhaps more appropriately the "collector loop", since it is usually just a turn or two).
   Per some writers, there may be some use for a permanent magnet to 'directionize' some of the energies, as there is in a magnetron microwave/radar tube, but I have no clear idea about that.

   One of the items seemingly missing from the earlier experimenters' circuits was a flyback diode on each coil. Usually this would be "Aha, another mistake!", an omission from the circuit design, but in this case it seems likely that the high voltage flyback spike generated as each pulse is turned OFF, is the actual suddenly changing high voltage, the "DC kick", that induces the lambda rays to release their energy. Otherwise, the 500 volt rated mosfet transistor driving each coil seems out of place in a 12 volt circuit. That this high positive voltage spikes up suddenly with the same polarity every time, and then decays more slowly toward ground on the return, might explain why the units produce fluctuating DC current rather than AC.
   In the "Ronette" design the flyback voltage is further amplified by the coils having a still higher voltage "secondary" as in a transformer (or perhaps a step up as an autotransformer, depending how I wire it - the schematic is vague on this point). Here we get up to the sort of voltage ranges of those working with vacuum tubes - hundreds of volts. Most of the successful converters used vacuum tubes. It took Mark a long time to succeed with solid state circuits.

Reflections

   If I dropped this project twice, lastly for another year and a half, it's partly because other recent experimenters apparently haven't been very successful, leading to me thinking that it must be pretty difficult to get it to work... and why would I succeed where others have failed? That, plus competing attractions - other more interesting (if seemingly less valuable) projects. With no moving parts, the converter somehow just seems unexciting.
   And also I have been somewhat fearful of it. By all accounts the VHE ray/lambda ray/short space ray energy is there and is strong, perpetually raining down upon or perhaps passing right through our planet unnoticed. It sounds like Tesla accidentally harnessed it around 1900. Now headed for a century ago, T H Moray (who also invented the semiconductor, making the first germanium diodes, as well as the multi-stage radio receiver & audio amplifier designs in use ever since) harnessed it and recognized that it had to be radiation from a band beyond the gamma ray band, a band which astronomy and physics have only recently begun to discover and explore. A number of credible people including government officials signed statements that they had seen Moray's devices driving multi-kilowatt electrical appliances, even when set up on a back road of their choice nowhere near a power line or other potential electricity source. Steven Mark, in spite of not knowing the source of the "DC Kick" energy, has demonstrated a basically replicable means to harvest it. But few have really succeeded, some have lost their electronic test equipment, and some (including Moray) have gotten serious RF or radiation burns from stray energies around the wiring. It can make glows and arcs I didn't understand at all until the day I wrote of them in the above "speculation" passage. But the long delays, with the taking it up again afterward, seem to have helped with understanding various aspects of the operation. There's less total mystery and more glimmerings of realization of how it must work.

   As I begin to see that Tesla, seemingly, made power with primitive electronics and without realizing just what he was doing, and that not-very-successful experimenters inspired by Mark a decade ago weren't using a properly designed transistor circuit to drive their coils, I gain confidence that if done right, and with computer monitoring of the output for pulse by pulse control adjustment, it may not be so hard as it seems to get it working. The metal box should be a good shield for the stray energies.
   I would also remark that the previous experimenters have done an extremely valuable service in fully writing up their work and putting it on line. Without it I wouldn't have investigated the "TPU" design or its inventor Steven Mark, and I've gained a number of valuable insights from reading their work. That gives me pause to hope that Turquoise Energy News may be doing the same for some future researcher(s) in some of the various areas that I've covered, with project successes or not.



http://www.TurquoiseEnergy.com
Haida Gwaii, BC Canada