Turquoise Energy Ltd. News #90
  covering July 2015 (posted  August 2nd)
Victoria BC
by Craig Carmichael

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

Highlights: BDLC 4:3 Motor and Controller: A superior BLDC system, prototypes finished and working

Month In Brief (Project Summaries)
- BLDC4:3 Motor & Unipolar Controller - an order for CAT 12VDC plugs, sockets, wall plates! - fickle ceramic capacitors - Variable transmission/Chevy Sprint progress (with Kelly BLDC motor controller) - DES battery electrolyte (I'll make it, I hope) - Electric Caik Outboard/fishing/anemometer for boat speed - Motor failures - Kia Soul production EV.

In Passing (Miscellaneous topics, editorial comments & opinionated rants)
- Resistance to Change - Greek Example
- PETE/PET is burnable plastic - Life in the Solar System... On a Comet? - Pluto - More on (Ugh!) Chem Spraying - Turquoise Energy taxes/SR & ED 2014.

- In Depth Project Reports -

Electric Transport - Electric Hubcap Motor Systems

* "BLDC4:3" Electric Caik Motor and Unipolar Controller
* Variable planetary gear transmission with Centrifugal clutch in Chevy Sprint
  - Installation in car
  - Kelly Controller over-revs motor with old magnet attachment
  - Will replace rotor... with BLDC4:3 configuration and good for much higher RPM.s
* Electric Caik Outboard: fishing trip & problems - planning next improvements
* Copenhagen Wheel Electric Bicycle Wheel - link et al
* Original GMC EV1 Brochure (attached)

Other "Green" Electric Equipment Projects (no reports)

Electricity Generation (no reports)

Electricity Storage - Turquoise Battery Project (NiMn, NiNi), etc. (No reports)

No Project Reports on: Magnet motor, Lambda ray collector, evacuated tube heat radiators, CNC gardening/farming machine, Electric Weel, battery making, aquaponics.

July in Brief

   July was almost solely about electric transport systems for a change. Except that I wrote quite a lot in In Passing. There were some interesting space mission news items that I give my take on, in particular the Rosetta mission to comet 67P/Churyumov-Gerasimenko with its lander Philae. Readings on the comet's surface appeared to indicate there was life on it, according to astrobiologists Chandra Wickramasinghe and Mike Wallis. And of course, the New Horizons probe flew past Pluto after a 9-1/2 year flight, revealing some interesting stuff on a tiny world. And I wanted to cover peoples' powerful resistance to change as related to world politics and social order or the growing lack thereof. And the almost global northern boreal forest fires of early July, probably caused mainly by drought owed to chem spraying, led me to reluctantly comment on that subject again.

BDLC 4:3 Motor and Controller

    I started the month with an experiment, a development and some testing of the new BLDC motor and unipolar motor controller configuration. More inductance seemed to make minor difference to the energy recirculation characteristics. Later I got the "CRM", "current ramp" variable frequency modulation, working. That completes the essential functionality of the BLDC4:3/SRM motor controller.
   I have now to make something more along the lines of a "production prototype", which I thought would need a bigger circuit board owing to the addition of several large capacitors for better filtering. (And that might not be doable with the "cheap" Eagle PCB license. The full license as I recall isn't cheap.) Later I decided to mount and wire the mosfets and capacitors off-board (the mosfets screw to copper bars anyway), which actually shrinks the board instead.

   Another aspect is the 'active generator' system both for regenerative braking and for the Weel hydro generator. I've been assuming that one simply drives the motor in the opposite direction to the one it's turning in, and the back EMF adds to the voltage instead of subtracting, so the energy in the coils will be higher than what was put into them, so more will come out than went in. But I went to the July EV club meeting, and someone from Australia who has been converting cars since the 1970.s (!) said the Mitsubishi iMiEV had better regenerative braking than all the others, and that people were trying to reverse engineer it and figure out how it works. And I remember being disappointed by the current being returned during braking in a converted Sprint I tried out at Canadian Electric Vehicles a couple of years ago, compared to what went in to speed it up. I assumed at the time that the reason was mechanical losses - the lossy automotive transmission. But maybe that's not the whole story! I hope my setup will produce good results. There's some chance it'll work better because the return energy circuit drains the coil energy to zero volts (plus one diode forward drop) instead of just down to the power supply voltage (plus two diode drops).
   Then I decided "active rectification" would help in the energy recovery circuit. The diode forward drop was too high and causing them to heat up. Heat isn't just a nuisance to dissipate, it's lost energy. In "active rectification" the mosfets are turned on just in the instant the motor coils return their stored energy to the power supply, making it a much lower resistance path hence with low voltage drop. On the night of the 17th I figured out a "self switching" circuit (several discrete components) to have then turn on at the exact time of the turn-off pulse and only then, and over the next two days I layed out a circuit board to solder to the feet of the diode-mosfets, which are bolted to a copper bar. On the 24th, in looking for additional information about synchronous rectification as applied to regular BLDC motor controllers (trying to find some theoretical info to help determine whether my controller would actually be better or not), I found a chip (IR1167) that drove mosfets specifically as synchronous rectifiers, doubtless better than my circuit of discrete components... but they cost over 5$ each - ouch! And they hardly reduce the component count.

CAT 12 VDC Wiring System

   On the 8th I got another order (Kansas), this time for a good number of CAT plugs, sockets and wall plates for an off-grid dwelling. They said they were pleased to see there was something better than the car "cigarette lighter" types. Alright - that's why I created them! I fired up the 3D printer on the 11th and started printing them all. It became obvious that I needed something to just punch out the blade pins for the plugs, as the whole order and especially those took much to long to put together. (Even aside from that I'll have to raise the prices a little.) And I'd really like to find something better than the easily-bent, poor grip, Pico zinc coated (unsolderable!) .205" socket pins. I did find some later in the month - gold plated in fact - but they came in little packages of just a few with rubber insulating "boots" that I have no use for, for a relatively high cost, instead of in bulk. But perhaps if I search I can order some in bulk, or find some others?
   I have no idea whether people I don't know are adopting the CAT system - of how many CAT wiring items may have been made overall since I posted the 3D printing files for them at "thingiverse.com". Anywhere between zero and a few thousand, I suppose.

Capacitor Variances

   Someone wrote to the OSMC list on the 10th about how variable and amazingly far off the printed value larger ceramic capacitors can be, depending on temperature and applied voltage. He gave a link to a web site on the subject: http://www.maximintegrated.com/en/app-notes/index.mvp/id/5527  I had never seen this in my electronics courses or elsewhere, and I had paid no attention to the mysterious three symbol sequences often seen, like "X7R" and "Y5V", that indicate tolerance curves.

Variable Torque Converter Transmission - Chevy Sprint

   Also on the 10th I decided I really had to get on with the long delayed variable planetary gear transmission with centrifugal clutch in the Chevy Sprint. I do very much want the improved EV! I wasn't particularly confident it would work, but it had been mostly ready to try out since last fall. I mounted it in the car and then started in on figuring out the wiring and installing of the Kelly KBL36301X motor controller that I bought for its 'regular' 2:3 BLDC motor last September. On the 13th I finally removed the gas tank, just because it was now in the way, and I ran a heavy 'cab tire' battery cable so I could have the batteries in the back, where they'd be convenient at least for testing.
   When I started in on the controller, I realized the 12V relay I had should be 36V, and that I'd want an ampmeter and a few other bits and pieces. So I ordered them from KellyController.com . They came to around 300$ where I had hoped 100 - 200. Everything adds up.
   While I waited for the parts, I contented myself with figuring out what batteries to use and where to mount them, and changing the car's light bulbs to LED "bulbs". I ran out of 1156 and 1157 bulb socket LED.s and had to order more of those too, from DX.com.

For now the batteries are just three size 27 RV
batteries in the back on a sheet of plywood.

   The Kelly parts arrived on the 21st.
   I got curious about the shunt material. Apparently an alloy called manganin (Cu:Mn:Ni ) has a very stable resistance over temperature and is the substance usually used for electronic resistors including shunts. It replaced an earlier alloy called constantin (Cu:Ni 60:40) that was good but not as constant. Both were created by the same person, Weston, in the late 1800.s. I note that constantin is close to the same formula as the cupro-nickel 70:30 that I bought for battery electrodes, so perhaps I'll try making my own shunt resistors from that.
   After a couple of weeks wiring and making up a "panel" for the Kelly controller, I was ready to test on the 31st. After sorting out a small problem with the Kelly handheld controls box, I ran the motor. Everything worked right, except that the slightest turn of the throttle potentiometer would send the motor racing at high RPM. After running it forward and backwards a just a few spins (verifying that everything worked right), the over-revved motor suddenly seized. The magnets, as a group with their epoxied polypropylene strapping, had all come detached from the rotor and flung out against the outer rim. So I didn't get to do any further testing, much less to hook up the control pulley and try out the transmission. This motor was put together in 2011 or early 2012, and it had the zinc coating that I later learned would delaminate easily, and of course the original strapping arrangement. The later techniques are much better.

The magnets complete with their PP-epoxy strapping flew off all the way around the rim.
Only magnetism holds now them to the rotor.
The new magnet mounting techniques should put an end to this sort of problem,
even at much higher RPM.s.

The Transmission (with planetary gear variable torque converter [without slip pulley cable] and centrifugal clutch),
plus Kelly BLDC controller and various parts mounted on an aluminum panel.
(Kelly handheld controls box lying to left. Oops, I didn't get a picture while the motor was in.)

   But now I might as well rebuild the rotor in the new 'BLDC43' configuration. And then it'll need the new controller, and I'll be discarding the Kelly. And all this means, of course... the car electrifying project needs to await the pre-production unipolar controller and is on hold again, for now.

Deep Eutectic Solvent (DES) for Battery Electrolyte

   The DES that Leonard Elionix said he sent from Italy, which I wanted to try out in my batteries instead of salt water, didn't arrive. On the 20th I ordered from acpchem.com the ethylene glycol and choline chloride that it's made from, to try and make some for myself. Unexpectedly it was a 350$ order - more money flying out! - as the choline chloride was about 260$ for a litre. (Hmm... no word from ACP as of August 1st!)

Electric Caik Outboard - Fishing?

   About mid-month I uncovered the boat and put in the Electric Caik outboard, my (old type) BLDC motor controller, some PbPb batteries, and the combo volt-amp meter I got from China, intending to go fishing. The amp meter part had disappointed me, the readings jumping around wildly. It was useless with a modulated motor drive. Then it occurred to me to filter and average out the current to the meter with a resistor and capacitor across the shunt, 330Ω and 2.2µF ceramic. This worked quite well, except that the current readings seemed to me a little low. I changed the 330Ω to 100. It didn't seem to make much difference. But now I don't have to hook up, in addition to everything else, the DC current clamp and a voltmeter to have an idea of the power being used. (I should however hook one up to find if the readings are right or if I have to apply a scaling factor.)

Filtering the shunt voltage gave steady current readings.

   I also bought a hand-held "weather station" especially for its anemometer. Not only would this be good for checking wind plant sites, it said it could be immersed in water, which meant it could measure boat speed - if you didn't fall overboard trying to read it. (I thought it was in the fishing tackle box. Apparently I've misplaced it - arg!)
   After I finally got a trailer license (50$) and a fishing license (20$), the days were pretty windy so I didn't go out. 70$ for a fish or two is bad enough. 70$ for not even fishing is a crime. I finally went out on the 23rd. I forgot a few things (bait, anemometer...) and after some time, my line tangled in the propeller. I couldn't lean over the back so far to untangle it without inviting the increasingly large waves into the boat, so I rowed back to the boat launch (I didn't go out too far!), where I managed to untangle it and retrieve my tackle at the dock.

   Then I tried to motor over to the launch ramp to trailer the boat, and found the motor was seized! And it was hot. The metal shaft and bearings were quite hot, and the coil bolts were hot. (The new model has nylon coil bolts, not subject to magnetic induction.) The composite body itself felt very warm.
   I found the "joy" coupling connecting the motor shaft to the outboard drive shaft had forced its way down the shaft, pressing the motor shaft up and the drive shaft down. The latter made gear noise in the outboard, which I could hear from the start of the trip. Perhaps I should have paid some attention to it! As it pushed the motor shaft up, it finally - or eventually - deformed the polypropylene-epoxy end cover of the motor until the rotor rubbed on the stator wall and finally seized against it. (I should have paid attention to the slight rubbing sound I could hear before the trip, too!) Reinforcing the actual rubbing, a thin gap setting makes the motors run rough and generate excess heat. The end plate showed little sign of wanting to flatten out again later. Probably it was only especially pliable when it was overheated.

   The culprit? The set screw in the joy coupling pressed against the shaft key rather than the shaft itself. That may be good for holding the shaft key in place, but it seems like - and has now been proven to be - a poor arrangement for holding the unit itself in place on the shaft. I'll drill another hole and add another set screw, this one to grip against the shaft.

   Sometime it feels like I'm just treading water... I thought that the outboard, at least, was finished and working pretty reliably! And my one working Electric Hubcap motor's rotor - albeit from early 2012 and older methods of fastening the magnets - flew apart. Other things now need work too, like there's a "burning rubber" smell in the RX7-EV after 3 or 4 miles, and the Tercel seems to need a new brake cylinder, as well as jobs always getting behind around the house. At least I got Turquoise Energy's 2014 income tax/SR & ED claim done.

   Anyway I'll look on these component failures as creating the opportune time to change all my motors over to the new (and I believe superior) BLDC4:3 magnetic configuration, with the new (and superior) unipolar controller, and with the new improved magnet strapping technique, safer for higher RPM.s. 4500 RPM for the Caik might just get the boat up on a plane, and 3000 or 3500 for the Hubcap might just make the difference between starting the car to move reliably as currently configured, and not! And I can't help but feel efficiency as well as reliability will rise with the unipolar controller - synchronous rectification or not.
   And when I design the next Electric Hubcap and Electric Caik motor molds, I'll be looking for every way to reinforce (thicken) the center areas where the bearings attach, so they'll be less prone to deforming.

   As a final extra note, on the 31st Tom Sawyer said he wanted to go down and test drive the new Kia Soul electric car. We drove to the Kia dealer in the RX7-EV and he got his test drive. I rode along. The regenerative braking seemed good. Of the production electric cars I've ridden in or driven so far I like this one the best. It was comfortable, powerful, took bumps well, and had nearly a 200Km range. It incorporated novel organic composites like (IIRC) corn-epoxy in the interior plastic and upholstery. Of course the price, in the upper 30,000.s of dollars ($C currently =.77$US), puts it way out of my budget.

   And, I still think I can make a vehicle that goes the farthest on a given battery energy storage. And (if I ever get there) I still want to do an Electric Hubcap system - or maybe even a couple of types - that can be simply added to a gas car's wheel. (To Whit: A BLDC4:3 motor with planetary gear torque converter and centrifugal clutch (or flywheel), or a high RPM switched reluctance motor with belt drive to the wheel, per TE News #86.) With my motors becoming safe for higher RPM.s and the mechanical component development creeping along albeit at a snail's pace, these things are starting to look more doable.

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

Correction: PETE (or PET) Burns Fine

   In talking about burning of polyethylene and polypropylene last month, I mentioned polyethylene terepthalate, recycling symbol "1" and abbreviated PETE or PET, as being one of the substances that gives off noxious fumes if burned. But it turns out it's nothing but carbon, oxygen and hydrogen - a hydrocarbon. According to Wikipedia, it's an excellent candidate either for recycling (it's made into carpet fibers) or for incineration.
   As long as it doesn't join the gigantic islands of plastic floating in the oceans, I'm happy. I've been noticing tiny bits of plastic all around the yard lately, which have been accumulating semi-unnoticed over the years, and other bits like stickers on fruit, things you peel off wrappers, packages and jars to open them, and bread bag closers. Other than burning them, what happens to all those bits?

Resistance to Change

   I was disappointed that nobody seemed interested enough to join an e-mail list that would allow governance of our municipality by general consensus of involved citizens and probably involving occasional votes by the populace to be presented to our elected officials. It seemed so simple and straightforward to me, and I was sure that once there were a few subscribers, it would grow of its own accord. I feel we desperately need to reverse the trend towards autocracy in governments, where there is an ever-growing disconnect between the agendas and motives of the elected and the desires of the populace.
   But while a few said "That's just what's needed!", no one at all subscribed. Apparently most people think the proposals are wild or even crazy. It seems most people haven't given political/governing systems the sort of thought I have over the years, and new ideas, however sensible they seem to me, are alien to them.

"The remarkable thing about that form of democracy is that democracy is the last form of the governance that makes rational sense to large numbers of people.  What has not been appreciated is that democracies must evolve. Even if it were in existence today, the skeptics among you would be reluctant to participate, as they think it would be a fraud." - Machiventa Melchizidek

( My related concept web sites: www.CampaynCentral.com - www.DirectVotes.ca - www.saers.com/recorder/craig/FundamentalDemocracy.html )

   People are born into a system that works a certain way, and it seems regardless of the systemic problems that ensue, most of them feel that system is "right", and that it's just the people involved today that are the problems. 'If we just elect ___ instead of the present heathen(s), it'll all be fixed.' They don't seem to see that the reason for an unsatisfactory list of candidates, bad election results, and why those elected become rulers of the people instead of their servants, is because the procedures, institutions and systems of governance were set up hundreds of years ago in more primitive times and are badly in need of updating. They are obsolete and they are gamed by the unscrupulous and power hungry. Generally, real leaders and social contributors have been shut out of the whole process.
   Yet people seem to consider their current system - whatever it is - to be some sort of "divine order", "set in stone" by some sort of omniscient men of long ago. No one today should tamper with them. In fact these systems themselves were of course new inspirations once, revolutionary ideas that somehow overcame all resistance, overcame all the people who thought that they were "crazy" and those who expected they would surely cause society to crumble, and got adopted.
   The American constitution in particular was revolutionary at the time, and a truly brilliant work, but its authors couldn't pre-test the plans or foresee all the outworkings of the untested ideas they were proposing should govern the land, nor could they account for the scientific, technical and social progress that would take place over the next two centuries - which were in large measure spawned by the freedoms and opportunities created by the new governing system itself. Governing systems need to be dynamic and changing to meet human social needs as they develop and change. The big problem is that while quality of life and equality were at least partially taken into account, none of the governing systems in use today had or have provision for growth and change of themselves as needed. This one of the three cosmic core values of sustainability is missing.
   Here is where a Department of Progress should exist within government, to sift and coordinate not only ideas for technological inventions but also those of a social or political nature, to be tested perhaps in local areas, and when a superior way of doing something is found, presented to the people and to legislatures for enactment - and modification or repeal if it isn't working out well. The BC Citizens' Committee on Electoral Reform (~2002-2004) almost got two electoral reforms passed in one shot. But the committee was formed at the whim of one premier [~= state governor] for a single shot deal. The real reform would have been to make it a permanent committee with a larger mandate to seek out improved ways and means. If they had continued to exist they could focus on one smaller change at a time, and some of the better changes would eventually start getting adopted.
   And concerned citizens need to be able to initiate referendums on any subject, that if passed, legislatures will then be expected to act on. If we had had that, the citizens' committee or some portion of it could have chosen to continue to meet, and independently of anything else, present their proposals to the public for enactment. The right to submit a referendum has brought about some progressive legislation at the state level in the USA. So far no one has proposed any further systemic changes, such as adopting the choice ranking vote to make election results fair and reduce the influence of money and corruption on the outcomes, and the initiatives aren't so far available at the federal level where they would most count, nor at local levels to help resolve local issues.

   But the crux is as usual education and knowledge. Alternatives need to be invented, presented, and discussed by everyone. There has to be a general social ferment: it needs to be generally understood that the changes so badly needed are needed. As long as most people think the system is fine as it is, and don't hear or think about alternatives it's hardly possible to make real changes... even as everything progresses towards a social collapse "of Biblical proportions" (as Gregory Mannarino would put it on Youtube). Apparently real and serious suffering will be needed to start that ferment and discussion.

   Change comes, whether it's planned or not, wanted or not. The US constitution provided that there be one congressperson for each 3000 citizens. Citizens had pretty good control over their congressional reps. With population growth, by 1911 that made for an impractically huge congress. Change was essential, but it hadn't been planned out. Congress simply voted to limit its own size to somewhere under 500 people. Today, each congressperson represents some 700,000 people. Thus the reps have lost touch with the electorate, and into the void have stepped lobbying corporations and vested interests, each putting their own agenda ahead of the public interest.

Greek Example

   We see the results of the Greek situation. Political things are less "overcontrolled" than in the USA, and in January a rather single platform government promising to combat the austerity that has been causing so much suffering and perhaps hundreds of demonstrations in Athens over five years was elected. This was unexpected by the "power brokers" as the political partisan group (party) that was elected only had a few percent of the vote in the previous election. (If the corrupt who rule behind the scenes had had any inkling this Syriza party would be elected, you can be sure they would have been in there interfering with the process to get people to change their minds, and to corrupt the leadership and members of this party from within. As it was, an uncorrupted group with real leadership was elected by surprise.) The first thing they did was tell the truth and admit it wasn't just a 'liquidity crisis', that "Greece is bankrupt.", which had been obvious for five years.
   Those who made private unsecured loans to Greek interests in dubious circumstances, those banks who offered "credit default swaps" to guarantee those doubtful loans, and those who pushed the Greek government to take responsibility for those loans in order to protect the investors from losses and their banks from insolvency, must share responsibility with those who borrowed more money than they could pay back -- all expecting the unsuspecting people of Greece to "backstop" and pay for their reckless gambling. But it is the same everywhere, not just in Greece!

   As the money ran out and the next payment couldn't be made, the Troika's [ECB, IMF, European Commission] only proposal was for the Greek government to steal perhaps the last remaining unpilfered monetary asset available in Greece (other than private bank accounts): Greek pension fund money, and make the next few payments on the unpayable debt, which would cause still more suffering and solve nothing, only "kick the can" about five more months down the road. The government decided they had no "mandate" to accept this proposal, having been elected to do the opposite, but nor did they feel they could reject it, having received only 40% support in the election. So they decided to let the Greek people decide in a snap referendum, called for only about a week away on July 5th. That wasn't long enough for Europe to browbeat and threaten the Greek people into submission, and the result was a pretty solid "oxi" ("no", 61%) to accepting the Troika's proposal. This was acclaimed as "a victory for democracy" by many, and as "a circus" and "irrelevant" by the head of the European commission. (Would he have said that if he had liked the result? He's an acknowledged critic of the present government, and almost in the same breath he said "The Greek government should respect the people of Greece." !?!)

   But was it a "victory for democracy"? Why did it take five years of increasing misery and and a "freak" election result before the people "were granted" any say in the matter? Why don't they have the right to initiate a vote themselves, instead of having them only on the rare occasion when it's convenient to those elected and at their behest? Does a referendum cost money? Gosh! How much did keeping insolvent banks open cost? How much did five years of unproductive protests in the streets, with the people battling their own police and ignored by their government cost, instead of having this vote before or when the government of the day had agreed to backstop private debts with public money, and long before so much of the Greek people's wealth had been pilfered away by the central banks? If the banks are going to make bad loans, let it be them that go bankrupt, not the population of the nation they're housed in. If the people had the right to initiate a referendum themselves, surely this vote would have come years ago.
   And why have the American public had no say since 2008 about grossly inflating their money supply to continually bail out their biggest banks and their unsustainable central banking institutions? Or Canadians, and other lands' citizens? Surely if "the silent majority" everywhere had the right to speak and be heard, things would never have come to this pass in the first place.

   Later in the month, Tsipras negotiated a new bailout deal with the troika, the worst one yet. He completely caved in. Why did the people of Greece not get another referendum to ratify or reject it? Is democracy only at the convenience of the elected really democracy?

Life in the Solar System... On a Comet?

   First the "fluffy" character of the surfaces of Ganymede and Callisto were discovered by Earth-based diurnal temperature observations in the 1980.s. Then polycyclic aromatic hydrocarbons were found on these same two worlds by the Galileo spacecraft in the late 1990.s. They were shrugged off as being "tholins, left over from the formation of the solar system." (Seemingly ignoring the older "fluffy" findings?) Then more were found by the Cassini spacecraft on worlds orbiting Saturn, most notably on the small outer moon Iapetus, where the leading hemisphere is "very fluffy", dark and organic, and the trailing face is largely bright ice. (...The term "dark side of the moon" applies legitimately to Iapetus!) It was noted that this organic material seemed to have "migrated throughout the Saturn system", and that Iapetus must resurface itself "at the very most in 100000 years and probably much faster" to have so few bright ice craters on its dark side. Apparently (in spite of what was said at the time about it being "highly interesting") the fluffy organics were of little interest, because there are no plans for the Cassini to have a close look at Iapetus in the entire remainder of its mission.
   Then this March the Dawn spacecraft reached the largest asteroid, Ceres, and found it too was mysteriously dark for an icy body. No one seems to have reached any sort of conclusion.
   Finally, the Rossetta spacecraft has been orbiting a comet named 67P/Churyumov-Gerasimenko since 2014, and it landed a probe, Philae, on that body. After several months of recharging its batteries, the probe has come to life and found (guess what)... polycyclic aromatic hydrocarbons, surely of the same variety as the earlier missions. Finally, on a little comet, two astrobiologists, Chandra Wickramasinghe and Mike Wallis, are saying in effect 'There must be life there!'

Link to article at RT: http://rt.com/news/271843-philae-probe-comet-life/  (More at Wikipedia, etc, too)  From the article:

. . . . .

Wickramasinghe interpreted the data received from the comet as proving that micro-organisms are “involved in the formation of the icy structures, the preponderance of aromatic hydrocarbons, and the very dark surface”.

“These are not easily explained in terms of prebiotic chemistry. The dark material is being constantly replenished as it is boiled off by heat from the sun. Something must be doing that at a fairly prolific rate,” Press Association quoted Wickramasinghe as saying. [IMHO the dark material surely is the plant life, on the surface.]
The European space craft, Rosetta, which is orbiting the comet, has also picked up unknown “clusters” of organic material that could be viral particles. [IMHO these are probably the same seeds as the one picked up by the high altitude balloon above Earth, or seeds or spores of other species of the same sort of plant life.]
 However, both Philae and Rosetta lack the right equipment needed to search for evidence of life. Moreover, the two astronomers revealed that an earlier proposal to equip the mission with the right technology to detect life was reportedly dismissed as a joke.

“I wanted to include a very inexpensive life-detection experiment. At the time it was thought this was a bizarre proposition,” Wickramasinghe, who was involved in the planning of the mission 15 years ago, said.
“Five hundred years ago it was a struggle to have people accept that the Earth was not the center of the universe. After that revolution our thinking has remained Earth-centered in relation to life and biology. It’s deeply ingrained in our scientific culture and it will take a lot of evidence to kick it over,” Wickramasinghe said.

. . . . .

   This comet orbits between Earth and Jupiter with an aphelion of 5.7 AU (850 Gigameters, slightly beyond Jupiter's orbit), perihelion of 1.2 AU (186 Gm), with a year of 6.44 Earth years. Probably at 1.2 AU (1.2 times farther from the sun than Earth) the comet's water ice will eventually vaporize off and it will then be unsuitable for this form of airless life, but it spends more of its time out towards Jupiter, farther from the sun. If the life originated on Ganymede, that's the right zone.
   What's the difference between a comet and an asteroid? At least in this case, it would appear to be simply that the comet gives off a "tail" of volatiles as it approaches its perihelion, while the asteroid doesn't - perhaps all its volatiles have already vaporized off, or its perihelion is a little too far out for major vaporization.

   At first I didn't think there was much to indicate plant life visually, then I saw the image below (in a youtube video), which is full of features that seem highly unusual for dead geography. Without knowing the exact pixel scale, I'm certain this must be by far the closest view ever of such a surface. It's certainly convoluted, seemingly "fractally" at scales down to the limits of resolution, and too dark to be ice extrusions. My take on it is we're seeing some bare areas, like glare ice where nothing could grow, and in other places projections from the surface, in various directions in the very low gravity, which would appear to be some form of strange, perhaps "crystalline" forms that look quite botanical in nature, even something like a forest canopy in a few areas. If the scale were much, much smaller, one might think it looked like moss; or maybe fuzzy mold. (What cost, decent COLOR images of all these many worlds?)

The surface of Comet 67P
Are there areas that might broadly be described as 'fluffy'?
Could all those strange protrusions (reminiscent of similar forms on Ganymede, etc.)
be some sort of strange plant life, as the spectral findings suggest?


  Most readers will have noted that after a nine and a half year journey, the New Horizons probe flew past Pluto on July 15th (or 14th GMT) (a world about 1/3 the size [volume] of Earth's moon), its small moon Charon, and several other orbiting "oids". It was once thought to have been the last significant unphotographed body in the solar system, but while the probe was en route, Mike Brown's sky survey uncovered several more small spheres, including Xena... I mean Eris, which is about the same size as Pluto and which orbits the sun far from the plane of the rest of the solar system in a huge polar orbit 2 or 3 billion miles out. Additionally, Brown had hoped to find a world "the size of Mars", and recently other astronomers now figure there must be a couple more worlds way out there to cause gravitational effects seen. If these are dark in albedo the survey could have missed them.

   But back to Pluto. New Horizons had to do a one pass, high speed flyby because it was naturally impossible to speed the probe up enough to make this 3 billion mile journey in such a short time and then to suddenly to be able to match Pluto's stately orbital speed (other than by crashing into it and making a large crater). Probes to Jupiter swing by Io, using it as a gravitational brake to help slow them into Jovian orbit, but no significant body is available at Pluto, and the relative speed of the probe is much higher.

   There is again some dark, perhaps organic matter - small lines of it in crevasses between polygon blocks of terrain. (Seems similar to the "cantaloupe" terrain seen on Triton?) Pluto must be so cold, and dark, it's almost inconceivable anything could live there... unless gravitational tidal interactions with its moons cause it to heat up internally. That might explain how dark vegetative material could form, and why it would be in the crevasses rather than on the main surface: opposite to the 'almost too warm' comet 67P, they're the only places with any heat to speak of. Again spectrographic information hasn't been released so far -- nothing saying it's definitely organic, much less polycyclic aromatic hydrocarbons -- and nothing has said it's "fluffy", so it's all in the realm of "very speculative". (But all the dark stuff everywhere else on airless bodies seems to have those characteristics.)

   Of special interest to me is some odd mountainous terrain. In my write-up on the probable formation of Pluto, its moons, and Triton by tidal disruption of a larger world that I've called "Pluton", via passing too close to Neptune (like Shoemaker-Levy 9 at Jupiter), it might make sense that these odd mountains were formed where two bodies separated, where they were pulled apart by tidal forces like two clumps of bread dough. If that is the case, we might also find there are such mountains on the opposite side of Pluto to the first set -- and nowhere else. (unless Pluto was the outer end of the original body, in which case there would only be one area of such terrain.)

   The terrain is said to be 'young' owing to the paucity of craters, less than 100,000,000 years old. It's not likely in this case that this is a result of biological activity. It now seems more likely to me that the breakup of "Pluton" was a relatively "recent" event in solar system history, occurring well after most of the original meteoric (crater making) material was used up, perhaps sometime in Earth's Cenozoic era. Or maybe one chunk of "Pluton's" material became the asteroid that apparently crashed into Earth about at the end of the Cretaceous period and reputedly killed all the dinosaurs, pterosaurs and ichthysaurs worldwide. (Let me not get into that specious theory!)

I also cover this subject on my web page, which I've updated pretty much with what I've said here.


Geoengineering, Chemtrails: Forest Fires

   I wasn't going to mention this subject again, but here we are. On July 5th I woke up to find everything strangely dark, the sun hidden by "clouds" and giving off a weak, orange light. It looked like the whole landscape was being lit with 40 watt incandescent light bulbs. The sun wasn't visible but later in the day could be discerned as a pale orange disk, easy to look at with the naked eye. Then it was obscured again. It certainly wasn't clouds, nor did it look like any previous chem spray. LED lights in the house, and the white computer screen, looked startlingly blueish by contrast after outdoor light. I couldn't figure it out until someone told me it was smoke from a forest fire in Washington state's Olympic Peninsula, just across the strait. The next day it started to thin out and by nightfall the brightest stars could be dimly seen.

   BTW, a correction: It turns out the rumor I heard last month was distorted. It wasn't the ocean but the fresh water fishery that was closed, owing to low water levels in lake and streams, and to allow fish stocks to recover from overfishing. Perhaps nothing to do with chem spraying unless it caused blooms of algae related to the decision... except that it's probably the cause of the low water and high temperatures.

   In the tinder dry conditions with low snow pack and melt, seemingly created largely by the chem spraying since last winter and its disruption of the northern jet stream, there are hundreds if not thousands of forest fires, very early in the season, blazing away and burning up forests almost engulfing the Pacific Northwest, Alaska, Northern Alberta and Northern Saskatchewan. Eastern Siberia was literally freezing - in July - while the Western half was hot and on fire. On the 6th I heard there were over 200 forest fires burning in BC alone. It appears the whole northern boreal forest ecosystem, almost around the globe, is being devastated. Weather everywhere is exceptional - wildly out of wack.
   And still the distant rumble of the heavy jets continues - I heard three distant takeoffs as I initially wrote this article, and I saw countless chemtrails over the month.
   One day I saw one jet leaving an ordinary jet contrail. It seemed almost miraculous that the trail faded out and disappeared after a while instead of spreading and spreading over the sky.
   And it seems that they aren't just spraying coal ash from present burning, but bringing it from the landfills where it's been stored for years and decades and spraying that too. A silver lining is that some of the substance may be having a beneficial fertilization effect. After all, that's what causes algal blooms, and I myself sprinkle ash from the wood stove on my lawn and garden.
   But at least my thought it would be hot with no rain all summer proved false. I ran out of stored rainwater. Then it sprinkled on the 24th and 25th, and on the 26th there was a thunderstorm with a torrential deluge in the afternoon - another very rare event for Victoria BC in late July. This may be the hardest rain I've ever seen, or at least it ranks with it. The gutters and downspouts could by no means handle it - in a minute water was spewing everywhere. I changed the rainwater collecting hose from a 1/2" hose to a 1", since most of the water was going to waste. Luckily it soon eased off. I guess this is what rain is like now... nothing, nothing, until the whole sky falls at once, making the record-breaking flash floods that have been in the news lately in so many places.

   Still the jets come over day after day where there are no passenger routes, flying back and forth and leaving their trails of nefarious crap. As I proofread this about noon on August 1st, the spreading linear "cloud" trails are interfering with the solar collectors in recharging the RX7-EV after a drive, causing it to go into alarm again and again as they cross the sun.
   This all these things seem to be an incredible demonstration of unforeseen (but in this instance not unseen) side effects of burning petroleum and coal as fuels, and of just how badly we need to cease doing so.

Turquoise Energy Tax Return; SR & ED

   Around mid-month I finally got on with and finished my tax return and SR & ED tax credit application for 2014. But although the amount I could reasonably apply for (~10000$... without claiming for aquaponics, battery research or solar PV & low voltage wiring project expenses that might get shot down) was just half what I had been getting in 2011 and 2012, and might well get chopped again by zealous auditors (who themselves get a secure salary from the taxpayer that they don't have to spend weeks justifying every dollar of), it should at least be money.
   Since the auditors last year had claimed my documentation (for 2013) was insufficient, I figured they hadn't looked at TE News in spite of my referring to it at every turn, and earlier I compiled the relevant sections from each newsletter for each project. It came to about 140 pages and I knew the printer would run out of toner, black and all three colors, part way through printing them. (That was one reason I hadn't finished the tax return - I kept putting off or forgetting to buy toner. But also it's much more interesting to invent new things than to fill in paperwork, and the work this spring and summer has been exciting and successful.) After two days of printing this material and the various CRA forms (150$ for toner), and two more days of calculations, filling out the forms, checking it over and photocopying it, I finally delivered it to CRA's Victoria drop box on the 22nd in a letter size paper box nearly an inch thick.
   I also did a very late "payroll deductions" submission for February 27th, which tied in with the main return. (Do B before completing A. Do C before B. Do A before C...) No doubt I'll get a 100$ late fine for that. But at least after much complaint on my part they refunded me a bunch of fines and interest they had levied on me related to salary issues for 2010, 2011 and 2012 when they themselves hadn't responded to my written notes, questions about what I needed to do, until 2013. It seems CRA is all very compartmentalized and one hand knows nothing about what any of the other hands are doing. I was very glad to be finished with tax stuff for a while. If I'm lucky I'll get the tax credit from them before 2016 to pay off one of my credit cards with.

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

Construction Manuals and information:

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

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

(Will accept BITCOIN digital currency)

...all at:  http://www.TurquoiseEnergy.com/
(orders: e-mail craig@saers.com)

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

July 1-4: Finished May TE News (#89)
6: BLDC4:3 - Tried doubling energy return decoupler coil from 400µH to 800. (Result seemed to be slight improvement.)
7: Purchased "Weather Station" with wind - or water - speed indicator. (For boat speed, and potentially for windplant siting.)
8: installed Electric Caik Outboard Motor in boat with BLDC motor controller, lead-acid batteries and metering system.
10th: mounted variable transmission unit and motor in Chevy Sprint (after washing it).
11: BLDC/SRM Motor Controller: Changed 10K trimpot to 47K resistor & tested. Sprint/Variable Transmission: Put ends on 3 heavy wires for phases A, B, C. (Took a ridiculous length of time.)
12: Got the variable rate modulation working in the motor controller. THIS EFFECTIVELY COMPLETES THE PROTOTYPE CONTROLLER & ESSENTIAL BLDC4:3/SRM MOTOR CONTROLLER DESIGN.
13: Variable Transmission/Sprint: purchased heavy battery cable, removed gas tank from chevy sprint, and ran the cable from the back (battery place) to the motor controller under the hood.
14: Purchased a few parts for Kelly controller for Sprint. Tried to order some more from KellyController.com web store but it wouldn't do the last step, 'confirm order'. (This was handled by e-mail the next couple of days. Seems to me this is what happened last September, too!)
16, 17th: Printed stuff for SR & ED return. Also investigated what batteries to use in the Sprint and where to mount them if it should become street-worthy. (Hoping for that 100Km range with fewer KWH of energy storage!) Unipolar motor controller: Designed a circuit for activating synchronous rectification.
18: Refined Sprint battery ideas, measurements, figures. Started laying out circuit board for synchronous rectifier operation.
19: Unipolar controller: Finished laying out sync.rect. circuit board.
20: Worked on CRA tax returns.
21: Checked out some of the wiring and electrical in the Sprint, and changed most of the bulbs to LED lights; ordered more LED car lights. (Kelly controller parts arrived.)
22: Worked on CRA tax returns, delivered them to CRA drop box. (Done! -- free at last!)
23: Sprint/torque conv.tranny - started assembling a mounting plate with Kelly motor controller, main circuit breaker, main contactor relay & diode, controller fuse, precharge resistor, ... all wired together per KellyController diagram of course. - Fishing trip w. El. Caik outboard, test - something seized!
24: Sprint car wiring, controller wiring.
25: more controller wiring; created 3D printed plastic box to hold 12V relay from car ignition key switch.
26: Checked out Electric Caik Outboard failure to see what happened and what might be done about it. More controller wiring.
27: More Kelly controller wiring for Sprint.
28: More wiring: 36V/120A APP power plug/socket.
29: Wired Kelly handheld Control Box to match motor sensor plug.
30:  Completed Kelly controller installation; figured out headlight problem: wrong polarity for LED.s!
31: Tested Kelly controller and motor. It worked, but over-revved the motor, and the rotor magnets failed. (I'll be rebuilding the rotor in accordance with new techniques and in the BLDC4:3 configuration, and making a new unipolar controller for it.) Checked out new Kia Soul EV at Kia dealer.

Electric Hubcap Motor Systems - Electric Transport

"BLDC4:3" Electric Caik Motor and Unipolar Controller

   An addendum to last month's schematic: The 200µH power line coil is about 60 turns of #14 wire wrapped donut style around a 2" O.D., 1" tall toroidal iron powder core, and painted with ilmenite in sodium silicate to complete the magnetic circuit. (Ilmenite in plasti-dip sticks better.) It's actually one of the earlier motor coils from before I started doing 21 turns of #11 wire and putting the coils in series instead of in parallel. But it gets pretty warm, and to handle full current it'll need two+ in parallel or some other improved configuration. It might need a bigger core too.

   Something to try out seemed to be to try a different inductance for the power line decoupler. I added a second coil in series, presumably making it 800µH instead of 400. (The motor coil phases are 90µH.) This didn't seem to change much. I had to turn the dial up a little higher to stay at 1000 RPM, but the supply current then seemed to be the same, or perhaps slightly reduced from 4.6A to 4.5*. If anything the energy return pulse widths seemed to go from 300nSec to maybe 370, and the return diode forward voltage drops seemed slightly reduced. It's probably something to fine-tune, but for the moment, the operational difference between 400 and 800µH isn't something I'm going to spend a lot of time on.
   Something I hadn't noticed before is that the energy return coil(s) gets hot, probably hotter than the diodes, in spite of the low supply currents with no load on the motor. It's #14 wire where the motor coils are #11, so two or more should be put in parallel anyway, but if it gets proportionately hotter at 100 amps than at 5, it's going to burn out. I don't think it will. If anything (according to theory), the energy return should be reduced as load increases.
   Also, employing the "CRM" cycle termination and adjusting components to reduce the 'PWM' frequency except at high current start-ups, should drop the number of switchings and hence return energy pulses per second under most steady-state conditions. I tackled that next.

* BTW some may notice a discrepancy from some lower figures last month, that I got with the motor open and running with just one bearing. The top cover and bearing was back on the motor. It takes substantially more amps than with it open and just one bearing to turn. The efficiency losses mainly aren't in the electromagnetic design aspects! (I suppose I should turn that new shaft to help straighten the bearings.)


   On the 11th I started on the variable frequency "CRM" control. I took out the 10KΩ trim pot and put in a 47KΩ resistor. This lowered the base frequency from 10 KHz or higher to 2.2KHz, which frequency could be heard in the motor coils. It also prevented the motor from drawing more than about 3.7A and exceeding 700 RPM. It seemed that unless the control was set very low, every ON cycle ended on "overcurrent", as might be expected. The next thing to do was to shorten the shut OFF end of each cycle so that in that case (as well as the low power case), the frequency would go back up. Since the MC33035 has no signal to indicate "overcurrent shutoff" to facilitate more advanced methods of current control, I put in a 4025 NOR gate (15 volt rated) from all three phase drives to simply indicate "no power is on to the motor", which (via a diode and a lower value osc. charge resistor, R4) speeds up the oscillator to rapidly end the cycle and start another.
   (Another way to do it might have been to invert the FAULT* signal to use as the speed-up. That was my original intent when I chose the 24 pin MC33035 instead of the 20 pin MC33033, where there's no FAULT* pin. But I became concerned it might not turn on and off at exactly the right times, and there are other things besides overcurrent that turn on the fault indicator. In the end I never tried it.)
   I reconnected the diode and used the upended 1KΩ and a new 4.7KΩ resistor in series to make 5.7KΩ. This resulted in an oscillator frequency of 42.6KHz with the motor off (instead of 2.2KHz), because with the motor off, the signal is always high. (Trace 1)

Variable Frequency "CRM" Modulation
Trace1 - motor off (43KHz); Trace2 - motor running (27KHz)
(Osc. probe 10x, so actually ~3V ramp)

   With the motor running, the slope of the oscillation ramp was much shallower in the powered part of the modulation cycle, representing the 2.2KHz rate. In the unpowered part, the 42.6KHz rate applied, resulting in lowering the modulation frequency to under 30KHz, variably depending on the control setting, and on whether the cycle was terminating earlier on overcurrent or regular PWM setting shutoff. (Trace 2)
    This is the desired effect, providing up to about 40KHz when a lot of torque (high current) is needed to get the motor started under high static load, and a slower and slower rate at higher loads and higher speeds (probably well down into the audible range) where high speed switching causes unnecessary switching losses. As usual with CRM "Current Ramp Modulation" it's hard to set a steady RPM on the unloaded motor. It wants to stop or go up to max. But the outboard motor shows that when it's running a load it gets easier to control.

   Probably the maximum 42KHZ should be reduced a little, maybe to 25, 30 or 35 KHz, and the minimum 2200 Hz down to maybe 1000 or 1500. But these are just resistor or capacitor value "fine tuning" adjustments.

Energy Return Losses

   Whatever one may think of the potential for losses owing to the unusual energy return system with the series coil, it appears to work, and it is a fact that the motor coils' return energy only passes through one diode drop instead of two, potentially halving that cause of energy loss.
   Furthermore, in other controllers, the return energy pulse is applied across the power supply voltage. When the voltage in the spike is reduced to the power supply voltage (plus two diode drops), no more energy is returned. The unipolar controller drains the motor coil pulse down to zero volts (plus one diode drop) regardless of the power supply voltage. I don't know the exact calculations or the coil pulse figures to start from, but I figure one will get more energy from draining a pulse down to zero volts than down to 36 volts.
   Running some hypothetical numbers... If the voltage pulse from the coil is (or would be with no load) 180 volts and the power supply is 36 volts, and the energy remaining in the coil is proportional to the square of the voltage, then (180^2 - 36^2) / 180^2 = 96% of the energy would be returned, versus 100% for the unipolar. If the coil pulse voltage would only have been 90 volts, (90^2 - 36^2 / 90^2 = only 84% of the energy would be returned. If the pulse voltage would have been 360 volts no load, the loss is just one percent.
   If the energy remaining was simply proportional to the voltage (not squared), (180-36)/180 = 80% return, or for 90 volts, (90-36)/90 = only 60%, or for 360 volts: (360-36) / 360 = 90%.
   I suspect the energy must be proportional to the square, and I suspect the voltage pulse potential is moderate to higher rather than lower, but in the bipolar system there is some theoretical loss involved however small, where the unipolar controller is theoretically "lossless" in this regard except for the one diode drop voltage.
   It will be exciting to see if there's any actual notable improvement, but I haven't noticed anything so far. The fact that it's not the very same motor as was tested previously (the one in the Electric Caik outboard) makes it hard to determine the source of small differences, whether better or worse. They could quite easily be mechanical.

Synchronous Rectification

   After deciding I was finished, at least a couple of things still bothered me a bit. One was electrical noise. But how much can that be reduced in a unit switching high currents? Keep the runs short! Another was the heat in the powerline coil for energy recirculation. That may drop when I make coils with heavier wire, probably using two of the iron powder cores instead of one.
   Even more was the heat in the energy recirculation diodes. The Kelly controller manual said theirs used synchronous rectification, where the mosfet is turned on when the body diode would be active, substantially reducing the voltage drop - and hence waste energy and heat. I had rejected out of hand the idea of putting mosfets in in place of diodes because of the possibility of shoot-through currents. But I had ended up using mosfets anyway because their body diodes were both better and cheaper diodes than simple power diodes. (Huh?) And they still got hot.

   As I woke up on the morning of the 15th, an idea that was vaguely forming in my subconscious jelled. These would be high-side mosfets. If the floating gates of these "active diodes" were driven, not as usual by transistors from some circuit that used logic to decide when to turn them on and off, but by the coil drive itself, they could hardly activate except at the exact moment the coil turned off - exactly when they were needed. The operational aspects are as follows: When the coil turns on, the coil signal is driven to ground. Since it's going negative WRT the mosfet, even a big spike to the gate pin can't turn the mosfet gate on. When the coil is on and the signal is at ground, the floating gate charge capacitor charges, in a few microseconds. While the coil is on, no other coils are switching since just one coil is driven at a time. So there won't be any spikes that could couple through and spuriously activate the gate/mosfet, except when the coil is off anyway. As the coil is turned off, it spikes high. This is where the body diode of the mosfet conducts to return the coil's energy to the power buss. But the gate signal goes high through its now charged capacitor, so the mosfet will turn on, reducing the diode voltage drop at exactly the right moment, active for a few microseconds. (A 3:1 voltage divider will be used since the power is ~36 volts and the maximum gate voltage of most mosfets is 20 volts.) A resistor is sized to bleed off the charge of the floating gate capacitor in a few microseconds - well before the coil would turn on again. Slow turn-on of the mosfet is of no concern because it occurs as the coil voltage is rising and before it goes positive. In fact, it may contribute to faster turn-off/rise time of the coil, and actually draw current momentarily from the supply for that. This is my only real concern: a short region of linear operation as the coil voltage rises toward the supply voltage. The mosfet gate's linear signal region in the decay voltage doesn't matter because the coil by then has no voltage across it. The same would apply to any "off" spikes that might manage to partly (or fully) activate the gate from the other two phase coils. And since the voltage is backward at the active time, the body diode would still take the load if the mosfet is only partly on, preventing even partial turn-ons from causing high power to be dissipated inside the mosfet.
   Thus, the synchronous rectification is gained and is synced by the very signal being rectified, improving efficiency and reducing heat, but with only the most miniscule possibility that any shoot-through current could occur.
   After I thought I had all this figured out, I realized that the mosfet would turn on the whole time the source pin was pulled low by the coil - much more than a transient shoot-through! It appeared some sort of switch would be needed after all. And the circuit started to look complicated - at least a couple of transistors and associated components. Maybe I wouldn't bother for version one - just heat sink the diodes well and get on with making the controllers.

   At this point it occurred to me to put the 3 mosfets on a copper bar as a fourth bar in the row of bars sticking out, and have the transistors as well as the associated resistors and capacitors on the circuit board. At this point I'd be really sure I'd need a bigger circuit board than Eagle will let me make with the "cheap" license... Except that there's only about 6 connections to the rest of the circuit. What about making it a separate board and run six wires between the "logic" and "power" boards? The power board is mainly heavy current connections that should use heavy wires anyway. And some of the big filter capacitors should go on it too. So now the original board would need few changes. TaDah!
   With only 6 connections between all the "brains" parts and the high power components [Ground, Battery+, A, B & C gate drives, and current sense], they could be plugged together with a 6-pin header strip and socket. Then the power components could be swapped out for repair without removing the control parts and control wiring. Then, if the big filter capacitors were on the bottom of the power board and the mosfets were on the top, the power board could be a lot smaller. Perhaps the electrical noise could be kept from the logic section, and I might even still get it all into those little extruded aluminum boxes. Hmm, hmm!

   In fact, the mosfets could just about as easily be screwed to the copper bars (as they are anyway) and hard wired instead of circuit board mounted, leaving only the smaller logic board, under 3" x 3". The only components not on this board would be the mosfets, the filter capacitors, and the current sense "shunt" resistor... and the synchronous rectification components.
   The rectification might well have its own little circuit board hanging off the feet of the "synchronous diodes"...  which (come to think of it) should be powered by the +12 volt logic supply instead of the variable battery voltage. That would eliminate the resistive voltage dividers and concerns about different battery voltages causing different gate drive voltages. And if "version 1" doesn't have synchronous rectification at first, it can easily be added later. (To supply Vcc and Ground to the synchronous rectifier board, the logic board connector will need 8 pins instead of 6.) But after a couple of late evening doodling sessions, I came up with a fairly simple single transistor circuit before retiring on the 17th, and I designed the board the next day. Soon after that, I found a chip designed for this very job, so it probably does it better so I'll probably order some and change the design. (On the other hand, it's a bit pricey at 5$ and hardly reduces the component count.)

   As August starts, it looks like I'll be converting all my motors immediately to the new 'BLDC4:3' configuration, which I'm sure is better. That means I'll be making some unipolar controllers sooner rather than later.

Variable planetary gear transmission with Centrifugal clutch in Chevy Sprint

   I'll start by discussing my largest concern with the system as built; the biggest reason my confidence level it'll fly as built isn't really very high: I made the variable planetary gear system in 2012 thinking it would run the car. The tests showed the idea worked great -- but only once everything was in motion... how to get the car moving other than by placing it where it would almost start rolling by itself?
   One possibility was a giant flywheel on the motor itself so that one could rev it up and then tighten the control pulley tensioner. The energy in the flywheel would start the car moving as the motor slowed, even with no torque conversion at first. I goofed when I tried that technique. I mounted a flywheel on the control/slip gear instead, because that was much easier. Only later did I realize that there it did nothing useful - the force of slowing that pulley didn't transfer to the wheels, only to the rope tensioner, making heat. The flywheel had to be on the motor itself, and really I had no place or shaft to mount a flywheel on the motor.
   The other way would be to use a clutch. That would let the whole gear train start into motion to activate the torque conversion so the car would start moving under all conditions when the clutch was let out. In 2013 I tried to make one that let a flat belt slip when the clutch peddle was pressed, replacing the chain drive. But it was... well, poorly designed and made. The slipping belt slipped off the pulley sideways and it was probably too narrow for sufficient grip anyway. A wider belt would get more grip from the same tension, but there wasn't really room for a wide belt with its wide pulleys.
   In 2014 I made the centrifugal clutch. I made it big, 9" diameter and 3" wide for lower RPM. That's probably a great way to do it. And I added some 3" "stand-off" blocks to move the motor over and effectively widen the unit by 3" so everything would fit. (Yes, a 3" wide belt would now fit, too.)
   But still I fear the RPM the clutch disk gets from the planetary gear will be too low to start the car moving. The trouble is the RPM of the motor is reduced by something like 1.8 to 1 in the planetary gear, so the clutch will only be getting 1100 RPM. How much centrifugal force can that low an RPM exert on the drum? I'd be more confident if the motor itself would do a higher RPM, like at least 3000 instead of 2000 so the clutch disk was doing 1650 RPM or better. If it looked like it almost worked, I'd probably try to run the motor up towards that RPM anyway. (and stand off to the side!) If magnets flew off the rotor, it'd be time to make the new 4:3 configuration rotor with the slots to run the PP strapping through. Then it'd be the new improved type, and surely good for 3000 RPM too. Of course I'd want to do that eventually anyway, but I was hoping to put it off until I tested the transmission and hopefully either moved the car or found it needed changing. Also if rotor magnets flew off, it'd be a good test of the solidity of the outer wall of the rotor compartment.
   Moving ahead... on the 31st I tried out the Kelly controller. It spun the motor so fast so quickly that after just a few spins, the rotor failed and I never did get to try anything else, so the above rebuild and reconfigure will apply. The outer wall was scuffed up, but held okay. I didn't get as far as attaching meters, so don't know what RPM was reached - it could have been 3000, 3500 or 4000. The motor with the new attachment techniques should easily withstand 3500 RPM and probably 4500 or more.

   If I can't make it work even at 3500 motor RPM, I'll probably have to reverse the whole planetary gear configuration so that the RPM at the clutch is increased by 1.8 to 1 instead of decreased, providing a good 3600 RPM from 2000 RPM at the motor. That would take much rebuilding. On the other hand (now that I think about it), with the 4:1 chain reduction following, it would give a good matched set of RPM.s: motor 0-2500, high RPM.s to 4500 to activate the centrifugal clutch, and 0-1125 RPM to the wheels for around 0-100 Km/Hr. I have a funny feeling I'll have to change it, but I'll certainly try it out as is first. Now it'll have to be with the new motor configuration and new unipolar controller.

Now, on to the month's work.

   On the 10th, after about 9 months of diversion, I finally put the transmission and motor into the Sprint... after a time out to wash nearly four years of accumulated pollen and moss off the car. It's hard to believe it's been so long since I bought it thinking to make a quick electric conversion of some sort with it. I guess the Mazda took its place for that -- and also convinced me that getting rid of the inefficient regular automotive transmission is essential for good results: lower power requirement, less battery storage and longer driving range.
   I got out the Kelly 24-36V, 0-300A BLDC motor controller that I purchased last fall, and screwed it too under the hood - a temporary mounting. Of course now I feel the 'standard' BLDC motor and controller are obsolete. But I wasn't going to change over what's (hopefully) ready to use at this point when it would delay everything still more. Assuming I get the Kelly working and it continued to do so without failing under heavy loads like my controllers. There were lots of wires to connect and to get right so as, hopefully, not to have everything go up in smoke.
   The next day I made up the three power wires to the motor. For some reason this seemed to take hours, perhaps because I was busy doing 3D printing for an order of 12VDC CAT sockets, plugs and wall plates at the same time. (When I went to put the wires in the car, it started to rain - a light sprinkle for the first time in a month. The water on the house roof didn't get past the moss.)

The transmission having the planetary gear variable torque converter with slipping
control pulley (tensioner not connected yet) and large, lower RPM centrifugal clutch.
(Somehow I don't seem to have taken a picture while the motor was on it.)

   On the 12th I went to KellyController.com and, at last downloading the manuals, started looking up how to connect and configure everything. It looked like there would be a fair learning curve. Should I swap everything over to my new controller and make a new 4:3 rotor for the motor? Notwithstanding that I now view the 'standard' BLDC motor and controller as a temporary arrangement I decided to persevere. With luck the default settings will be acceptable... and, I just might learn a few useful things about motor controller setups. But studying the wiring diagram, it didn't look so bad - pretty similar to mine, really.
   The rotor position sensor connections to the motor are the same as mine, except the controller supplies 5V instead of 12V. The Hall sensors will work either way.  (Optical sensors with the 3 LED.s in series would need a higher voltage to turn on.) In fact, in spite of the 5 volt supply, the outputs were at 0 or 10 volts.
   The temperature sensor range is also 0-5V, but I didn't find what kind of sensor is expected, and anyway, 5V isn't enough to run the AD590 temperature sensor as configured. Maybe just a potentiometer to fake "temperature okay"? Finally I found "thermistor" in the J2 pin diagram (the wire just says "temperature"), and on a search for that word in the manual, under "Features" I found, very specifically: "Configurable motor over-temperature detection and protection with the recommended thermistor KTY84-130." (A specific thermistor only? So much for "configurable"!) But the wiring diagram showed the temperature pin unconnected - yay! The wiring diagram also only showed the "J2" plug and not "J1". It didn't need all those wires to run? Yay again.
   However, it started to look like I should make another order with KellyController.com, for various "accessory" parts - especially the ampmeter. (Why didn't I get that with the controller?) If I was in it for 600$ or so already, another 100 or 200$ wouldn't kill me. (To buy a laptop computer just to run the Microsoft Windows based config program would definitely be annoying.) By the time I was through at the web site, I was at 210$US, about 300-350$C when the smoke settled.

   While I waited for the controller parts, I hooked up a 12V battery and started converting the light bulbs to LED.s. I ran out of larger ones and had to order more. It's always a surprise how many 'bulbs' a car uses. I couldn't get the LED headlights to work. Not until the 30th did it occur to me that the circuit in the Sprint was backwards, putting +12V to the common pin and then grounding the low and high beam pins to activate them. Of course reverse voltage doesn't work for LED.s. I could reverse one set easily enough. Getting both low and high beams without taking the dash and switches apart was going to be difficult. Anyway, "high beam" seemed to be just turning all same the LED.s on brighter than "low beam". Hmm... On August first I took the relevant part of the dash apart.

The Kelly Controller Panel in the car
(Again no picture with the motor connected.)
   The parts having arrived, from the 23rd I picked away day by day at mounting and wiring the controller. I made an aluminum back plate to fit where I wanted to mount it under the hood, and then put the components onto it one by one. I realized first that the power relay was turned on by the same switch that turned on the 36 volts to the controller, but that the ignition key had a 12 volt coil. I couldn't use the 12 volt relay, and I couldn't supply 36 volts from the ignition key. So I ended up first buying a 200 amp contactor relay with a 36 volt coil (in the Kelly order), and then adding a small 12 volt relay to turn on the 36 volts when the car key was turned on, for the contactor.
   I got finished the wiring and put 3 batteries into the back of the car on the 30th. Double checking that all the wiring seemed to make sense I tried the system on the 31st. There was a problem with the Kelly hand control. The motor controller LED blinked a pattern indicating the throttle was shorted. Really, it was just a very good potentiomenter that went right to zero. I tried turning it half way up before turning the power on, and then it complained about that with a different blinking pattern. I tried disconnecting the hall sensor cable, and it gave a another complaint pattern, followed by the first one, alternating. I must admit the trouble codes, combined with the manual, saved a lot of trouble tracking down the problem - it's a good reason for computer overcontrol of the controller. I soldered in a 75 ohm resistor in the ground leg of the throttle pot., and finally got a green light. When I turned it up even slightly, the motor spun! In fact, it spun up really fast with the slightest turn of the pot. It didn't always seem to like going backward, but when it did, it spun up just as fast. So the hall sensors and motor phases all matched, with the same connections as my controllers. (Wow!) After just a few spins there was a "clunk" and the motor was seized. I didn't even get to checking out the Kelly controller features, much less to installing the torque converter tensioning rope and trying out the transmission.

   When I got the motor apart, all 12 magnets complete with their polypropylene-epoxy strapping had flown loose from the rotor, spread out, and jammed against the outside wall, all the way around! So that was that for that! Maybe I should have tried 24 volts at first instead of 36. How fast did it go? I hadn't got as far as connecting a meter to check it. I've run that motor at maybe 2200-2400 RPM at least a couple of times with loads, with no problems. So I'll guess it was well over 3000 - probably over 3500 or 4000 RPM. Furthermore, it seems this rotor had the "cold galvanizing compound" zinc anti-rust primer on the rotor, and red (urethane?) spray paint on that. The zinc especially has a weak bond, as I found out with the first Electric Caik motor. This motor was made before that.

The motor, somewhat gouged, and the rotor, looking almost normal but for the paint ending where the magnets used to sit.

Rotor from the top (Top bearing still on shaft)
Only magnetism now holds the 'sprung' magnet assembly to the rotor.

   So now I'll rebuild the rotor with all my latest magnet attachment techniques on bare steel, and it should be good for much higher RPM.s. It reminds me of someone who once talked about an alternator or starter motor for a particular model car, which was notorious for failing - the armature wiring would fly apart. After repairing a couple and then re-doing them again, he changed the "glue" in the armature - I forget what, from motor varnish to epoxy or something. They never failed again, and he soon became known, not just locally, as "the guy to go to" to get them fixed. They were the same motors. Technique was everything.

   As long as I'm rebuilding it, I will of course do it in the new 'BLDC4:3'  configuration: magnets NSNSNSNSNSNS instead of NNSSNNSSNNSS, and with an exact placement magnet placing jig similar to the one I made for the Electric Caik. It'll want the new unipolar controller, so after all that expense and installing, the Kelly (which seemed to work fine except for getting any fine control over the motor speed/power) will be out.

Sprint Batteries and Battery Wiring Considerations

   On the 13th I bought 4m of #6 AWG 4-wire "cab tire" (very tough cable with a thick rubber jacket), removed the gas tank from the Sprint (a long put off job) and ran the cable from the back to the front under the car, intending to put the batteries at the back. I'll double the wires as two pairs, which makes them the equivalent of AWG #3.
   The gas tank weighed 25 pounds, not counting the 5 liters of gas I recovered. (2011 gas - ugh!) That makes 275 pounds of gas stuff that's been removed from the 1710(? 'curb weight' IIRC) pound car, making it around 1435. Then about 85 have been added of electrical transmission and motor, and the fat 10 pound battery cable makes 95. Then there's the batteries...

   Since the voltage is only 36 volts, it seemed to me it should have over 300 amp-hours, to have at least 10 KWH of energy and plenty of current drive for the high amps of the motor. I still only have 7 KWH available in NiMH (by stripping them from the Mazda), despite all the cells I've bought over several years. There are 240 AH golf cart batteries weighing about 65 pounds, 7-1/8" x 10-1/4" profile, for about 150$ plus taxes each. So six of them for around 1000$. It looked like they would just fit in the spare tire hole. I like the size and the price, and they would weigh under 400 pounds. But 240 AH would still be a little light on energy storage and on current drive. They could be doubled, but then it would be twice the price and weight, and they certainly wouldn't fit in easily.

   It seems the next size up is 380 AH golf cart batteries - 60% more energy - for about 350$. Obviously that's more cost per amp-hour, over 2000$ total -- more costly than doubling up the 240 AH ones! I can't say I like that very much. Well, I chose to use just 36 volts, and that obviously meant using high amps and high amp-hour cells. 380 AH at 36 volts is 13.7 KWH and 600+ pounds, perhaps 1/2 of a "typical" lead-acid EV battery storage and weight. It should provide a lot of driving range in this small vehicle with an 'ultra efficient' drive system, and the high currents should impose relatively low stress on these huge batteries.
   It works out to 150 $/KWH instead of 105$ for the 240 AH ones. But the batteries are supposed to be good for around 600 cycles. Compare that to NiMH dry cells or lithium at 1000 $/KWH. That's about 4000 cycles for the same price, albeit with some fuss in changing them 6-1/2 times more often, having substantially more weight in the car, lower voltages under load... and then probably not getting the rated performance and life from them anyway.

    They weigh around 100+ pounds and are considerably larger than the 240 AH ones, 7-1/8" x ~12" (and very tall - 16"?). They wouldn't fit in the spare tire well, but they would still fit in a single row across the back, just inside the tailgate. The loss of luggage space wouldn't be severe (partly because the back seat folds down when more is needed).
   A friend, Tom Sawyer, pointed out that putting them at the very back - which seemed so convenient - would put a lot of weight (probably over 600 pounds) behind the rear axle, and said I should put them at the rear seat position, between front and rear, for balance. We both stood on the rear bumper (together 365 pounds). It wasn't bottomed out, but it certainly went down a long way. Thanks Tom for wrecking a great plan!
   Maybe I should put 350 or 400 AH of NiMH together as one 12V battery and just buy four of the 380 AH golf cart batteries? That would at least save some weight and money, but it wouldn't solve the problem. (8 sets of 50AH, 12V NiMH @20 pounds for 400AH would be 160 pounds instead of 200+. With the other 400+ that's still too much weight at the back.)
   Next plan: Two of these giant batteries would fit under the hood. (Three might be managed, but not at all easily.) That would leave only four to put at the back -- or two plus a big stack of NiMH.s.
   Or... the original plan was for 300 AH. I seem to be upping that in my mind to match the golf cart batteries' 380. I could just make up two 12 V, 300 AH NiMH sections from my existing stock, and only add two of the big golf cart batteries for the third section. I already have 300 AH of soldered batteries in boxes and 240 of 'giant flashlight tube' pipes. Another 6 pipes (60 AH) would get that up to 300. It leaves me a few cells as replacements for any that may be bad. That would also be only 250 pounds (NiMH) in the back instead of 400 (PbPb) - and more cargo space - and 200 pounds (PbPb) under the hood, total 450 pounds. And it's the lowest immediate cost. The extra 80 amp-hours of the front PbPb pair can be the lowest 12 volts to run the lights and all. That seems very agreeable, so it's the plan.

   Again using separate chargers for each 12 volt section allows use of non-identical batteries. As they age the golf cart batteries can be replaced as a pair (or perhaps even singly by using two 6 volt chargers?), reducing the sudden major 'fueling' cost. And since the NiMH.s seem mostly to weaken rather than to fail, a few more cells here and there can simply be added to increase or if necessary maintain the range, probably for quite a long time.

   So I'll leave some room for a few more amp-hours of NiMH in the back for additional range, and of course there'd be room to change the ones at the front to NiMH or lithium if I want to spring 3000$ some time.

   For testing I just put 3 size 27 RV batteries in the back on a sheet of plywood, nothing up front. While it's good to have an arrangement worked out, permanent battery considerations are academic until the car is otherwise roadworthy.

Electric Caik/Honda75 Outboard test - fishing trip

   Well, I finally went on my fishing trip on the 23rd. From the start things went wrong - I forgot the bait, a net or gaff, and a club. I was told it wasn't rough, but the waves increased as I fished. I was jigging for cod or halibut, but they probably weren't interested in the plastic squids. Then I tried trolling with a salmon lure. Evidently I was drifting sideways as much as going forward, because the line tangled in the propeller. I could see that if I leaned far enough over the stern to try and untangle it, the waves would come over the transom. The tackle seemed to be still attached, so I rowed back to the launch ramp. I wasn't venturing too, with the chance of some failure always in mind. Back at the dock I untangled the line and reeled it in. I saved my 1970.s flasher, hook et al!

   Then I tried to motor over to the launch ramp to trailer the boat, and found the motor was seized! And it was hot. That is, the shaft and bearings were hot, and the coil clamping bolts were hot. (The new model has nylon coil bolts, not subject to magnetic induction, so that last won't happen in new ones.)

The top (stator end) of the motor was pressed up by the coupler at the bottom of the shaft until
everything got hot, the case deformed, and the rotor magnets started rubbing on the stator
compartment wall until they jammed against it.

   I found the culprit: The "joy" coupling connecting the motor shaft to the outboard drive shaft had forced its way down the shaft, pressing the motor shaft up and the outboard drive shaft down. The latter made for gear noise in the outboard, which I could hear from the start of the trip. Perhaps I should have paid some attention to it! As it pushed the motor shaft up, it finally - or eventually - deformed the polypropylene-epoxy end cover of the motor until the rotor rubbed on the stator wall and finally seized against it. (I should have paid attention to the slight rubbing sound I could hear before the trip, too!) Reinforcing the actual rubbing, a thin gap setting makes the motors run rough and doubtless generate excess heat.
   The set screw in the joy coupling pressed against the shaft key rather than the shaft itself. That may be good for holding the shaft key in place, but it seems like - and has now been proven to be - a poor arrangement for holding the coupling itself in place. I'll drill another hole and add another set screw to grip against the shaft. (Both top and bottom coupling halves.) The better way to prevent a shaft key from sliding out, if the slot is long enough that it can, is to take a hammer and cold chisel or something and put a big dent at the end (or both ends) to prevent the key from sliding.
   And when I do the new Electric Caik body molds (with the slightly enlarged rotor & coil circle diameter) I'll look for ways to thicken and strengthen the center area where the bearings attach.

Noting that the needle bearings seem to have a lot of friction when I install them new, I've mentioned that I thought they would loosen up over time and as the excess of grease came out. Here is the lower bearing from the outboard, which has now had a few hours on it, oozing green grease. But it's still somewhat stiff. The top bearing shows no grease, but it now turns pretty freely.

Copenhagen Wheel bicycle motor system - & Battery Issue Comments

   Someone sent me a link to an interesting and well hyped-up bicycle "pedal assist" motor system: https://superpedestrian.com/
(Copenhagen, Denmark, has more bicycles per capita in daily use than anywhere else in the world.) Evidently everything comes in one kit. There was a video, from 2013. The next video was a 'production update' from a year later, November 2014. It reminded me of the trend in my projects... the prototype product being proved, then improved and improved, with some really exciting sounding stuff, but nothing getting produced. Here it was claimed they'd be available 2nd quarter of 2015. But I didn't see a "Purchase" option on the web site. I hope they get there.
   A down side in the specs was that it was only 350 watts. E-bike affictionados around here use more powerful motors as any visit to the monthly VEVA meeting will disclose. [usually 1st Wednesday evening of each month. See www.VEVA.bc.ca - VEVA Islands] I've ridden a couple. A couple were more like clandestine motorcycles.

   Of particular interest to me was that the specs said the lithium batteries were only good for 1000 cycles - much less than the several thousands of cycles claimed for many. Nickel-metal hydride dry cells were derated without comment from 1000 to only 500 cycles on the all-battery.com website. Although heavy and not well suited to bicycles, lead-acid golf cart batteries are rated for over 500 or 600 cycles, and are far cheaper per kilowatt hour.

GM EV1 Brochure

   Tom Sawyer said he had seen a 1998 brochure for the original GM EV1 for sale on e-bay. I bought it. I scanned it and I reprint it here fairly large because the writing isn't legible if it's further reduced. It contains some fine humbuggery such as why you could only lease one and not buy it: in one place, 'because they'd use better electric motors in two years, so the car would then be worthless' and then (not to be too consistent, bottom center of side 2) 'because they were expensive'. (and who set the high price?) When did any car company refuse to sell a gas car for either of those reasons? And when did a new model ever make the old model suddenly depreciate to zero? This was one of several insidious things presented in the brochure designed to turn off prospective customers or clients - and to excuse the fact that you couldn't buy one. In fact, among all the features touted to appease the California authorities, the majority seem to be presented in the worst light instead of the best to deter interest. For example, any advertiser worth his salt would have said at the top left of side 1, something more like "Futuristic styling and aerodynamic design maximize the battery powered EV1's driving range." (In fact, even today only Tesla model EV.s have surpassed the EV1's driving range of 160 miles (255Km).)
   Aside from a few later estimates from those at the suppressed company, the lowest being "200 $/KWH in mass production", we never did find out what the price would be for large flooded nickel-metal hydride batteries in mass production, and as long as they're suppressed, people must settle for lead-acid (which should have been out of production for years by now) or pay - at least - double or triple that figure for lithium types.
   Aside from being electric, the EV1 first introduced some novel features now common including larger wheels (15" IIRC) holding tires with narrow sidewalls. If you haven't seen it yet, do watch the documentary Who Killed the Electric Car? about the manufacture, leasing only, and suppression of the EV1 and most other electric cars and their NiMH batteries by every car company, in the late 1990.s in California, and their final destruction about 2002. (It was on youtube last time I looked.)

Side 1 of the brochure

Side 2

Victoria BC Canada