Turquoise Energy Ltd. News #38
Victoria BC
Copyright 2010 Craig Carmichael - April 2nd 2011

http://www.TurquoiseEnergy.com = http://www.ElectricHubcap.com = http://www.ElectricWeel.com

Feature: Electric Hubcap Motor Making Kits: $499.99
(no where near $500) - Taking orders now.

Month In Brief
  * Plus opinion... Natural resources: public property?
  * Plus news item... The End of nuclear power? A study says that in an ongoing trend, the cost per KWH of electricity from solar panels has dropped below that for nuclear plants. (It crossed over in 2010 for new construction... and that's before Japan!)

Electric Hubcap System & Motor Building Workshops
  * PP-Epoxy Stator Ring Mold - for stator production
  * PP-Epoxy Stator/'End Bell' Rings: Same stator for any configuration of motor.
  * Bearing holders, 1" rod axles
  * INSOLUBLE Sodium Silicate coil coating
  * "Regular Motor" configuration motor made, CNC setups...
  * Ilmenite seems to work even better than ceramic rutile - even lower motor currents!
  * Flat steel rotor disks - cut by abrasive waterjet (it's much cheaper than I expected). Motor only 4" long.

Electric Weel Motor Project (Electric Wheel Motor... Rim Motor...)
  * Flat plate rotor cut by abrasive waterjet
  * more pieces to be cut, taper lock bushing

Torque Converter Project
  Sigh, no time!
  * Moving escapement pivots from inside the drum to outside: 6" torque radius instead of 5".

NiMH Dry Cell Car Battery Project - Get the lead out - of your car! Reduce vehicle weight! Reduce Fuel Consumption! Green batteries! from $300. (e-mail me!)
  * AA cells arrive, assembled new car battery - sure enough, 9 pounds instead of 12

Turquoise Battery Project
  * Tests: runs for days with 61 ohm load, but at pathetic ever dropping voltages - mostly under 1 volt.
  * Repeated tests show similar performance: it's rechargeable.
  * Improved case: thicker walls, better(?) '+' terminal connections.

Newsletters Index/Highlights:

Construction Manuals and information:
Electric Hubcap Motor
- Turquoise Motor Controller
- 36 Volt Electric Fan-Heater
- Nanocrystalline reflective rear electrodes to enhance DSSC Solar Cells
- Simple Spot Welder for battery tabs, connections

- Electric Hubcap Motor Kits, Parts - Build your own ultra-efficient 5 KW motor!
- Sodium Sulfate 4x longevity additive & "worn out" battery renewal.
- NiMH Dry Cell Car Batteries
(please e-mail me for batteries)
- NiMH Custom Batteries
(EVs, E-Bikes, Scooters, etc. - no extra charge)
- NiMH individual Dry Cells (D, AA)

are all at:  http://www.TurquoiseEnergy.com/

Electric Hubcap Motor Making Kits

499.99 $Canadian - Electric Hubcap Motor Kit

   Although I don't have complete car installation systems yet, I think some people have been waiting for the motors alone as they can have many uses. The last details are finally falling into place as described in the Electric Hubcap topic below, so I think it's finally time to start taking orders for them as kits, with the caveats that it may be a few weeks before they are actually shipped, that specs are only roughly determined, and that the kits may be just a little "rough around the edges". (I will of course try to minimize unwanted 'features'.) A $100 deposit will put you first/next on the delivery list.

Electric Hubcap kit configuration, Open edge view and closed.
(The disk brake rotor shown is being replaced with a flat plate rotor,
with the main body of the motor just 4" across instead of 5".)

   The kits include all the parts to make one Electric Hubcap Motor except they EXCLUDE:

* the 12 supermagnets (about 2" x 1" x .5" or 50mm x 25mm x 12.5mm) Purchasers must buy their supermagnets separately - I can't supply them for the sort of prices they're available at off web sources since that's where I buy them.
* Epoxy resin. The magnets are glued to the rotor with a small amount of epoxy resin, and then pieces of polypropylene strapping (supplied) are glued over them to ensure they don't come loose. Although a cup of resin is plenty, it's locally available everywhere and I think liquid would complicate requirements for shipping.
* Brushless Motor Controller. That's a separate item. Unfortunately mine aren't ready to sell yet.
* Mechanical Torque Converter to hook them directly to car wheels. I don't have this working yet.

What the kit DOES INCLUDE:

6" x 1" 4110 HT SR machine shaft axle
SDS 1" Bushing (holds rotor to axle)
2 - 1" trailer bearings
2 - 1" trailer bearing races
10" diameter x 5/16" mild steel flat plate magnet rotor (without magnets)
1 meter of 2" wide polypropylene strapping
Rotor compartment PP-epoxy cover bell (or end plate and ring for walls - a last pesky detail)
2 - sets of bearing holder parts
9 - pre-made stator coils for 36 volt operation, with:
 * low loss iron powder toroid cores
 * ilmenite coating to improve magnetics and reduce losses (I don't think any other motor has this!)
Inner stator plate/rotor-stator compartment separator
Outer stator plate
Magnet sensor PCB with 3 hall sensors with 5 pin flat rubber trailer lights plug for motor controller connections.
APP connectors (70 amp size) for coil connections to motor controller
Angle iron 'feet'/mountings.
Stator vent cover screen
Miscellaneous nuts, bolts, washers.

Some Electric Hubcap Motor rough specs
approx. 12" diameter x 4" body length, rotor 10" diameter, center of force: 4" radius.
Brushless, 3 phase, axial flux, 120º Hall effect magnetic polarity switches, air cooled.
5 KW, 36-42 volts (nominal), 0-127 amps. (12-14 volts, 0-380 amps with parallel wired coils.)
Stator coils (9 of) to rotor supermagnets (12 magnets 2" x 1" x .5", 6 poles) flux gap: 14-18mm / .55"-.7".
Torque and speed vary with flux gap: "adjustable specs" motor. (Nominal 0-2000 RPM; .17 N-m/amp torque constant @ .65" flux gap)
Efficiency: around 95% peak, around 75% at max rated load (these are both excellent figures)

March in Brief

   I put off filling out Revenue Canada SR & ED project forms, corporate income tax and my Personal income tax until March 20th, then spent many hours on them. (The last item remains to be finished.) Until the 20th, work on the motors took first place.

   At about the start of the year, a friend took it upon himself to start rebuilding my rather primitively formatted website and give it some pizazz, which he had been threatening to do for some time. He's a great database and web design programmer. This project started bearing fruit in March, and TurquoiseEnergy.com is getting a new and more professional look. At the same time, the work has led to me revise some obsolete info. But so far I haven't been able to pay the web site revisions the attention they deserve.

   Paperwork and web site aside, the next thing to do was to get set up for making all the special parts of the Electric Hubcap motors, so I can offer them as kits.
   I made a polypropylene-epoxy composite 'stator ring' using the plywood mold. It was gooey and frustrating work. But if you're trying to do the best motors, they might as well be made of the best materials. I made a better, deeper mold with 'butcher block' HDPE rings that would allow a simpler technique: pile in shredded fabric, pour in epoxy, put the top on and press it down with C-clamps. The molds themselves can also be produced and sold. In spite of that, since the coil making has been much simplified, these 'rings' are now the most labour intensive part of making a motor.
   The new rings are 'universal': a stator made with them can be used for any configuration of motor. They can bolt onto a trailer wheel hub or an axle flange, or hold a bearing race plate or a hub. The rings can be on both sides of the stator coils, and they can form both 'end bells' of the motor. They can be made thick or thin by varying the amount of PP material and epoxy used.
   I decided to offer a 'regular motor' configuration kit, with the rotor and stator enclosed in a shell with a bearing at each end and the axle sticking out one end (or both ends). That meant creating some small metal plates to hold bearing races at the center of the rings. Very large washers (4" OD) with CNC drilled bolt holes, one turned to the bearing race size, turned out to be ideal to clamp onto the plates.
   With those parts done, only revised tiny printed circuit boards for the magnet sensors and miscellaneous bits remained. Perhaps I should get a small CNC machine suitable for PCBs. I haven't made any myself since the cave days with film, photoresist and ferric chloride etching fluid, but APC charges just as much for these simple , almost trivial, boards as for complex ones. On the last night of the month, I roughed out the board layout.

   The new motor had ilmenite coated coils instead of rutile, and the no-load power drain seemed even lower. I mentioned the motor on a list. Among several, I got a response from the owner of SuperMagnetMan.com, who is involved with the magnet end and has helped design a fine motor himself.
   Among other things, he said: "I am excited to hear about your advances in this area. [snip] I have worked with lots of people through this group and other places that are looking for something similar to what you seem to have operating." And: "I have been wanting this kind of motor for vehicles for the last 10 years [snip]."
   I think customers for motor kits are out there if I can (a) find them, (b) deliver what's needed (c) with decent quality and service, and (d) at an affordable price. And doubtless there's some DIYers who'll make their own, maybe buying some of the parts here - hopefully after I get around to updating the (now pretty obsolete) instruction manual, and maybe coming for a workshop to ensure they have the finer details right (and maybe install the motor in its intended application).

   When the AA cells arrived I did a lighter car battery - 9 pounds, still 30 amp-hours. It started the car, but it wasn't as strong as the D cell battery (12 pounds) and I switched back after a day or two. But with the high energy storage per weight, NiMH AA cells should make excellent electric car/PHEV batteries.

   For a week or more, most of the work was nice in that it required little creative thought - just plug away at it. But there seemed little excuse or reason to stop between morning and midnight, and the desire to finish things - especially the motor in its new form - was strong. Those hours can't be good for anybody, and everything else falls behind.

   For my own MnMn batteries, I made a new battery cell case with thicker sides that I hoped wouldn't leak. But I didn't get around to filling it. I ran some long tests on the present cell which showed that although self discharge is awful and the discharge voltages and maximum current are pathetic, it will put out small levels of current for days, and do so again and again upon recharging, with even slight improvements at times: it has the essence of a rechargeable battery. A light doping of osmium that probably didn't contain much osmium on the separator sheet made a minor improvement. I'll try to ensure more of it gets employed in the next cell, which will also use salvaged dry cell MnO2 instead of cheap (impure?) pottery supply MnO2. That might reduce self-discharge.

Opinions: Dropping Solar Power Prices - Natural Resources are a Public Trust

   It seems two people once entered a nuclear plant in Ontario to test security, and found that they could easily have walked out with all the 'spent' nuclear material they could carry to make dirty bombs or whatever. Ontario Hydro denied that there had ever been a security breach and said everything was perfectly secure. The nuclear industry puts much pressure on politicians and governments to adopt nuclear power plants, with the same bland assurances that everything is perfectly safe and nothing could happen, and that the country can't do without nuclear power.
   The USA has hollowed out a whole mountain to store nuclear waste in. It will need to be guarded for longer than recorded history at taxpayers' expense. A large area around Chernobyl including the town is uninhabitable and will remain that way for a long, long time.
   The worst potential radiation effects - probably alarmist, but who knows? - of the nuclear plant disasters resulting from the Japanese earthquake and tsunami appear to be (a) the loss of mountainous Japan's main agricultural land area, (b) the end of many Japanese cities including Tokyo, and (c) the end of fishing in affected areas of the Pacific. And at times, the Pacific coast of North America is directly downwind of the dying Japanese reactors via the jet stream. The Canadian government - before it could possibly have had time to check out the facts - quickly assured us there is nothing to worry about and that we should take no prudent advance measures, however simple, in case radiation hits us here.
   In case these lessons aren't abundantly clear on their own, there's now a new reason to phase out nuclear power generation, and never to build another new station:
   The cost of photovoltaic solar panels has gradually but continually dropped ever since they were first invented,  while the cost of nuclear power continues to rise. In 2010 a milestone was reached: a report says it is now less costly to build installations that generate electricity from solar cells than from nuclear energy. The report may not be entirely unbiased (any more than pro-nuclear reports), but the ongoing trend seems clear. Perhaps it is solar electricity that will one day be "too cheap to meter" - a one-time boast of nuclear power exponents.


   (...and all this before Japan, and without my nanocrystalline borosilicate reflective glaze for DSSC solar cells!)


   Someone said they had heard on TV that among the 1% top income 'earners' who now rake in 1/4 or 1/3 of America's wealth each year are those who feel more at home with Saudi princes than with citizens of their own country. It can well be imagined that some of these people would feel uncomfortable around the citizens of the civilization they're pillaging, and even be ashamed look them in the eye. The words of the song One Tin Soldier come to mind:

 Go ahead and hate your neighbor!
 Go ahead and cheat a friend!

   May they wake up and see how shallow and unfulfilling is the obsession with wealth for its own sake, and how destructive it is to their own personality and sense of self worth when it's knowingly and willfully gained by manipulating and exploiting others. "The love of riches all too often obscures and even destroys the spiritual vision. Fail not to recognize the danger of wealth's becoming, not your servant, but your master." - Jesus


   Other than the need for patent reform and for the missing government Department of Progress, one outstanding societal structural flaw promoting gross inequity would seem to be the permitting of private ownership of natural resources, such as oil, before they are extracted and processed. I believe the principle needs to be established and made sacred that natural resources are public property, to be under the control of a public trust or perhaps (in Canada) a crown corporation.
   Private contracting corporations may extract resources from the ground and process them as called for by the trust, but they will not own them. The trust will sell the end products without discrimination to whoever needs them, and of course some products like oil and gold can help the government fund services with less taxation. This rather obvious step should eliminate some of the unaccountable vested interests and their obscene wealth, and their oppressive economic power to prevent progress and keep civilization static and stagnant.

Electric Hubcap Motor System

Polypropylene-Epoxy Stator Rings

   I made another stator with the plywood stator mold, this one as a flat ring to which would attach to some sort of metal center. It was frustrating and gooey work.
   One change was that instead of cutting 30+ circles of polypropylene, I took some of the scrap pieces, and I took some bigger pieces, of both the nonwoven mat and the cloth, and cut and ripped them into smaller pieces. This was intended to eliminate the tedious cutting, and to mingle the layers so it couldn't delaminate, as I had to a small extent experienced earlier. Only a top and a bottom circle were cut. But in a sense this was like starting all over again. I had no idea how much cloth or how much epoxy would be needed, and there was no way to get a good estimate except to see what came out of the mold.
   I used 150 grams of epoxy on the first try, and I neglected to weigh the cloth. The proportions seemed good, but the piece was only 1/8" thick. Next there was just 150g of that epoxy left, so I did it again. This time I used 72g of fabric. This ended up still under 1/4" thick. Proportion was okay - it might have used a little more cloth. Then I used around 235 grams of new epoxy and "some" more fabric. It came out the desired 3/8" or so.
   I had a problem weighing the epoxy as the digital scale turned itself off twice while I was pouring it slowly in. (I also went months before figuring out how to turn this same scale off manually: you have to hold the "On/Off" button down for about 3 whole seconds. Another poorly designed product: the hardware is fine but the software sucks!)
   Anyway! It seems it took about 400-450 grams of epoxy and 200-250 grams of shredded fabric.

Old ring mold (not a fungus)

  The next day (6th), I made a second ring. For this one I used 360 grams of epoxy and 120 grams of PP fabric, and it came out just under 1/4" thick, and seemed to have good proportions.
   It was still gooey and frustrating work. One problem was that the fabric wouldn't lie down and kept coming out of the shallow mold while I was trying to put it in. Then, in spite of waxing them, the epoxy stuck to the plywood and to the wooden center I made. I decided to make a new mold out of 1/2" high density polyethylene 'butcher block', that the epoxy wouldn't stick to. I spent the afternoon of the 5th at my brother's near Industrial Plastics, waiting for them to cut my piece of HDPE.
   There was no way I was going to try cutting these somewhat pricey pieces by hand. I had just learned the programming and technique for routing curves and circles with the CNC drill/router, so that was the obvious way to go. I cut a 6" length of 300 mm PVC pipe (same as for the motor bodies) for the outside edge of the mold, to solve the fabric flopping out problem. Instead of brushing on the resin to wet down the fabric and get it flat enough to work with, I'd simply throw in fabric and pour in resin, then cram the top ring on it and clamp it down with C-clamps.
   Then I realized I'd need gigantic C-clamps to fit over the 6" sides. I trimmed the pipe down to 4-1/2", and bought 3 more 4" C-clamps that would fit over that. The mold had now cost $60 and several hours work, exclusive of an unexpected $40 and more time to fix my air compressor, which the CNC machine needs to raise and lower the tool.

CNC machine with router, cutting HDPE stator mold piece

   The final piece was a 2.75" OD center pipe. A 4" piece of aluminum pipe 2.85" OD was found and given to me at Smith Bros. That at least didn't cost a lot, but I spent a couple of hours on the lathe shrinking it to 2.77". (partly because my cutting tool was dull - it really helped when I finally thought to try another one.)

New ring mold
Drop in bottom mat (left), then some shredded pieces of fabric, some epoxy and repeat, then place top mat,
top mold ring over that, and compact it down with C-clamps.

   The first ring I made with this new mold had 150 g of fabric, 450(?) g of epoxy, and came out 1/4" thick and was a little dry in spots. The next one used 150 g of fabric, 510 g of epoxy, and took 45 minutes from start to putting it in the oven - a bit laborious, but much less so than previously.
   That seemed fine, but when I took it out, it had large dry sections at the outer edges and it was .33" thick on one side to .42" opposite. Since I used no more cloth and more epoxy yet had bigger dry areas, it would seem one must pay very close attention to distribution when piling in the fabric, distribution when pouring in the epoxy, and the tightness of the C-clamps, which I had done up less tightly this time, fearing I might be bending the plastic mold pieces.
   I'm getting better at this, but I think the real answer is to find still more improvements to the technique.

   I calculated co-ordinates for the coil mounting bolts and some others - lots of sines and cosines on a calculator - entered them into a CNC drilling program, and made two PP-epoxy rings into a stator and one into an end bell for the other end of the motor. Some other holes I drilled by hand and will add to the CNC program later, and differentiate between the inner ring, outer ring, and end bell pieces.

   When this is all ready and programmed into the CNC machine I can sell not only finished rings, I can sell the CNC-made molds - and manual drill guide templates - to help others do more of their own making of motors as well. This is part of the counter-intuitive business strategy: If this catches on, there are 250 million vehicles in North America almost desperately needing electric propulsion. The more molds, etc, that I can produce, the faster it'll spread, and it will take many years to saturate the market no matter how many people start making them. To paraphrase a famous saying, sell someone a motor and you have one electric car - sell them means to make motors, and they'll grow a whole crop of them.
   Or perhaps some sort of franchise system could be worked out.

Metal Ring Centers - bearing race holders

   Next, what goes in the hole in the middle of the ring? For motors with a bearing at each end, I had been trying to come up with some sort of plate. I went out and bought some huge washers, and ended up using two that were 4" outside diameter, bolted together and sandwiching the plastic ring between them. The center of one was turned on the lathe to fit the bearing race, while the other prevented the race from pushing right through.

   Next I set up the CNC to drill 5 holes in the washers for 5 bolts to hold them together. Using the machine is the only way the holes will all line up properly, and they can be cranked out easily when and as required.

   For the thicker rings, center washer spacers that fit are necessary to prevent the race from falling in between the big outer washers. Unfortunately the best spacers I've found still require work - grinding 5 indents where the bolts go through to accommodate them, as the outer diameter is just a bit large on the size where the inner diameter is good.

The Form of the Motors - "ordinary motor" type

   I got a trailer hub - the new stator ring would bolt right onto it. Then I started thinking about shapes and forms. My axial flux motors have both bearings at the stator, whereas most motors have one at each end of the body - one on each side of the rotor. Mine can exist and run without having a body over the working parts, which, as I stop to think about it, is rather unusual.
   But whatever form is best for a motor with a torque converter, for an ordinary motor that spins a shaft, the shaft will be stiffest with bearings at opposite ends. I'm going to use the new 'stator rings' also as 'end bells', with the turquoise PVC tube for the bodies as originally planned.

Electric Hubcap in an enclosed "ordinary motor" guise.
This was a first "mock-up". Then I asked myself why it needed two layers of outer case over the rotor, and eliminated the outer one.

   This will be one more of several unique EH rotor-stator-axle-bearing configurations that may be adopted for various purposes - one of general utility and familiar to most motor users. By making the stators as flat rings, one standard design of rotor and one of stator can now be employed to form any of the configurations.

Axles: 1" Steel Rod Shaft

   I couldn't fit the rotor and stator properly onto a trailer axle. But after trying the encased motor layout, I realized that an axle end nut wasn't needed, and that the optimum axle for it would be a simple 1" hard steel shaft such as "drill rod", which could be cut to any desired length, with the "SDS" taper lock bushing to hold the magnet rotor and provide a base for the bearings. At Metal Supermarkets I got a hard steel rod called "heat treated, stress relieved, 4140 machine shaft", which sounded like the best and was also (unlike drill rod) economical. It only needed to be 6" long, with the bearings at 0" and 4" leaving a 2" projection. Such a short axle is inherently very hard to bend, so I'm confident the strength is more than adequate.
   Once I had this piece along with the others I'd made or gathered, the steps for putting together a motor fell into place.

The new motor without the rotor cover
The stator and the rotor form two separate compartments
Cooling air flows by the coils from the rim to the interior, through holes to the center of the rotor,
is flung outwards by the rotor magnets ("fan blades"), and exits through holes in the left hand cover.

All assembled

March 28th: I had a steel flat plate rotor made by abrasive waterjet cutting.
I re-assembled the motor with it to check the fit. The motor body is just 4" long (or is it 4" thick?),
The rotor cover will be just over two inches across.
The thickness of the SDS bushing that clamps the rotor to the shaft prevents a further 3/4" reduction.

   The main thing left to do before I can sell these motors as kits is a revised circuit board for the magnet sensors, and I layed it out on the night of March 31st. I would also like to change the shaft end bell to a one-piece bowl shape with integral sides. This will better ensure proper alignment, and also that the motor can't be run without the protective cover over the rotor the way it's seen in a couple of pictures here.
   I think I'll also sell the kit sans supermagnets. I don't do anything with them here, and I would just have to import them, and then in many cases re-export them, all at extra expense in taxes, fees and shipping. People will can just as well order their own. It lets me knock off around 125$ from the kit price for around 75$ in magnets. The price of the complete motor kit (without magnets) will be about 500$.

INSOLUBLE Rutile (Titanium Dioxide)/Sodium Silicate Coil Coating

   The rutile in sodium silicate coating on the coils definitely improves the efficiency of the motors. But the water soluble sodium silicate medium seemed to portend poor durability of the coating.

   In trying to figure out ways to make the rutile/sodium silicate coil coating insoluble, I had been forgetting one of sodium silicate's most interesting properties - when heated to 99-105ºC, it loses its water of hydration... and thereupon becomes insoluble, all by itself. Problem solved! The coating doesn't look any different, but it no longer will dissolve in water.
   The wire insulation and the epoxy are rated for about 200ºC, but it does mean taking the coil cores above their 70ºC rated temperature for a few minutes to do it. The manufacturer says that's no problem - the coils degrade, but slowly over time, if they're too hot. A once-only overtemperature is thus different from having it as a continuing or repeating condition while running the motor.

   The other problem, the coating flaking off, is a bit more tricky. Painting on a stiffer mix (less water) seems to help, but it really doesn't bond to the cores. Perhaps I should rough them up with sandpaper?

   The other experiment, painting on 2 or 3 coats to see if a thicker layer helps even more than thinner, also remains to be tried.


   As a further experiment, on the coils for the next motor (seen above) I tried using ilmenite, a mineral blend of titanium and iron. (FeTiO3). The TiO2 alone (rutile) is much better than air, but perhaps the mix of nano-size paramagnetic and ferromagnetic attributes would create a still better magnetic path from the outside of the wires into the core. Ilmenite responds weakly to a magnet, suggesting (along with the oxygen content) that the iron is in more of a ferrous (FeO) valence state than ferric (Fe2O3).

The ilmenited coils on the new (painted) stator ring. The center hole holds a bearing race.
The rotor magnets, acting as a fan, will be right behind this ring. Another ring will enclose the coils,
and the vent holes will pull cooling air in from the stator outside edge, across and through the coils.

That ilmenite might well work even better than rutile was suspected... just look at that rich,
creamy chocolate color! (doubtless supplied by the iron content.)

   When the new motor was finished on the 19th, I measured the no-load currents. The gap was about .6", and the coils are 20 turns of old cotton insulated #11 wire (3 phases with 3 coils in series) whereas the previous motor is 61 turns of #14 (3 of 3 in parallel), and the flux gaps were different. In addition, the bearing setup is a bit different with one bearing now on each side of the rotor. So some difference in figures may have other causes than ilmenite versus rutile. I'm not making enough motors (so far) to try out just one small change per motor as they've so rapidly evolved.
   Interpolating last month's rutiled coil results for a .6" gap estimates 38 watts at 500 RPM and 105 watts at 1000. The new motor, with the ilmenited coils (and the comparison-disruptive factors mentioned), produced the table below. Power was two NiMH dry cell batteries providing about 26 volts. Results compare favourably with the previous, at 31 and 88 watts, a further 15-20% idle power reduction.

Current (Amps)
Watts (A * 26 volts)

   The coils felt cold after the tests, but the 1/4" x 2" coil mounting bolts (at least, their heads) were noticeably warm. Zounds! Perhaps I should use thinner or non-metal bolts, too! I think I'll try #10's next time. With the double ring stator, they should be stiff enough.

Iron Powder Cores Info

   I called Micrometals and found out that (a) the iron powder size is on the order of 100 microns, and (b) that it is pre-coated before making it into cores so that the particles don't short into one big iron conductor, with all the nasty eddy currents that would result in.

   I should think that having nano-size iron particles would be theoretically optimum, though improvement on near perfection may be a bit academic. Problems with producing iron nano particle cores would be: (a) getting iron particles so small. It's easier to get 'nano' with oxides, but iron metal is best. (b) coating such tiny particles to insulate them. And even if they were coated, it would have to be a nano-thin coating as well or it would be mostly inert coating with little actual iron.
   The glassy ceramic substrate I was trying to do with ferrous (FeO) and paramagnetic TiO2 particles might work (if successfully made - in a reducing atmosphere kiln), but I'm not confident it would be better - it would be better than air, and almost 100% lossless, but one might need substantially more core to do the same job, meaning a bigger motor for the same power.

The Electric Weel Project:
The Car Motor of the Future?

   I decided this giant diameter motor should have a solid steel plate for the rotor after all - if only a thin base. I'd hate so much to have a weld give way and see a huge spinning rotor fly to pieces! I thought I could really use that pulsejet steel plate cutter that I didn't finish making last fall - cutting out a 26" round steel disk with the angle grinder has very little appeal.
   Instead, I called Waterforce abrasive waterjet cutting in Sydney, and found the price for cutting such disks is much less than I feared. So I asked for a 26" disk of 1/8" mild steel.

26" diameter rotor disk of 1/8" steel (20 pounds)
with 10" EH disk and chair for comparison.

   Next (latest plan), I'm having a 26" steel ring cut 2.125" wide from 3/16" stock. This will go around the rim as a stiffener, a thicker base to mount the magnets on, and to carry the magnetic flux between magnets. The inside piece left over from that ring is to be the center of the stator, which needs to be non-metallic in the outer area under the magnets. I hate to think of the mold it'll need and all the epoxy it'll take to do the giant 'stator rings'!
   And it occurs to me that with the SDS taper-lock bushing attaching the 10" rotor to the 1" axle seeming to work so well, all that's needed for the bigger motor is bigger SDS tackle. SDS bushings for shaft sizes up to 2" are in fact available so I bought one.
   When I have the steel pieces - and some time - I can continue the project.

   I still think the pulsejet steel cutter would be a cool tool to have!

Mechanical Torque Converter (MTC) Project

   I decided to change the torque elements, and to have three curved 'strike' teeth 120º apart on the motor rotor, that would be 'hit' first at a very shallow angle, that gradually steepens so as to operate quietly and produce a smooth pulse of torque. Then three 'escapement' arms to strike them, on the output drum.
   Variants of two, three, five and six teeth and arms can be tried pretty easily. Near the end of the month, I realized I could put the pivots on the outside of the drum and just have the points sticking in - this would give the largest radius of force, about 6" instead of 5" or 20% better torque from the same pulses.
   I may also put springs on the escapements, causing one hook to try to stay down. This isn't in accord with the oscillating masses operating principle, but it could help transfer more force from the motor to the output during each pulse.

   I also looked at a truck this month at Canadian Electric Vehicles, with an electric motor driving a set of planetary gears whose output turned the drive shaft to the rear differential. That motor must really rev up on the highway. Speeds of single gear ratio electrics are evidently often limited to about 80 Km/Hr. The value of a torque converter is again brought home: to provide the required torque at low speeds yet keep the motor revs down on the highway - and to do it all with greater efficiency than the gears.

   I hope I can find some time in April to work on it!

Ni-MH Dry Cell Car Battery Project

 NiMH dry cell battery #2 in a better fitting enclosure under the hood: 30 AH, 30 D cells.
Starts great even at -6ºC! Under 12 pounds - the car is 30 pounds lighter!
(Strap is aluminum. ...though d
uct tape might work fine if you can wrap it on somehow!)

Best Charging Voltage?: car alternator's 13.8 volts not only works: it's Ideal!

   For a typical battery charger, charging up NiMH cells from low to full, voltages up to perhaps 1.41 or 1.42 volts may be used. Then the charger shuts off. In the car, the battery usually doesn't get discharged very far, and driving continues until a destination is reached, not just until the battery is charged. But NiMH cells don't stay up at 1.4+ volts, they discharge to between around 1.37 and 1.33 in a few hours or in a day or two - even lower after some days or weeks unless they're the 'low self discharge' type. Let's call them 'overcharged' if they're above that. If a cell keeps getting strongly charged for extended periods when it's already full, it may shorten the life, and anyway it's doing no good. My first in-car test showed that the car alternator's 13.8 volts is acceptable, but what really is the best constant voltage to charge at if a battery is going to be left on charge for long periods such as while driving?

   I ran a test to find out. I charged one of my NiMH car batteries (sets of 10 cells in series, so voltages are *10) with the lab power supply at 14.10 volts. The current gradually settled to about 4 amps and stopped going lower.
   I reduced it to 14.00 volts. Since the battery was charged nearly to 14.10, the current stopped. But when checked a couple of hours later, as the 'overcharge' reduced, the battery was again drawing abut two amps to hold it at 14.0 volts.
   I reduced it again to 13.95 volts. Again the battery stopped drawing current for a time, but 3-4 hours later, it was up to 400 milliamps - much better, though still a higher trickle charge than desirable (and a needless 5 watt power drain).
   Then down to 13.90 volts. This time as the voltage dropped the current came up to only about 80 milliamps. Divided amongst 3 banks of D cells (or 12 banks of AA cells), this is a very acceptable continuous trickle charge current.
   Next test was at 13.85 volts. It took longer for the cells to drop from 13.90 to 13.85 since they were much closer to staying up there by themselves. The current to keep them at 13.85 was trifling - the current meter needle on the power supply barely moved.
   Then down to 13.80 volts. It took longer for the cells to drop down to the new voltage with each drop, and this one was took some hours. I would have had to connect a milliamps meter to read the trickle current.
   It took overnight to drop from 13.80 to 13.75 volts. This might mean it was a few percent below 'full charge', eg 97%. Not much in the overall scheme of things!
   It took about a day to drop from 13.75 to 13.70 (temperature around 12 to 15ºC), so considering typical self-discharge rates, one suspects that could mean at 13.70 it's not quite fully charged, eg, it might sit at around 90-95%.

   Conclusion: for a constant voltage charge of indefinite duration, a good voltage to charge at is about 13.75 to 13.90. This will keep the battery 'topped up' to 100% without overcharging it. By some fluke of chance, car alternators put out 13.8 volts as a standard, making NiMH batteries ideal car batteries. And they are more ideal than lead-acid, because until they're considerably discharged, they sit up over 13 volts, so headlights will hardly dim if the engine is off. This will be especially useful for a car running on electricity instead of gas.


13.65 - too low, won't be well charged
13.70 - okay, might stay slightly undercharged, eg 90-95%
13.75 - good: virtually no trickle charge. If undercharged, it's only by a very few percent
13.80 - good: trivial trickle charge (10mA?) <- standard car alternator voltage!
13.85 - good: trivial trickle charge (20mA?)
13.90 - okay: draws an allowable trickle charge (eg, 80mA)
13.95 - okay but a bit high: a considerable trickle charge (eg, 400mA) stays on when battery is fully charged
14.00 - too high: keeps drawing a couple of amps when fully charged

   On measuring, I've discovered that MY car's alternator is putting out 13.95 volts. It may well be that on older cars, the voltage is really regulated by the current, so it probably rises to give the NiMH battery the same current as it would give a lead-acid battery. It's still workable. I expect that newer cars made in the solid state era have better regulation.

Speaking of Chargers...

   Okay, so how do you charge NiMH 12 volt batteries that aren't connected to a car alternator? I found a good 6 or 2 amp charger for $40 in the automotive department at Canadian Tire, model 11-1526-8. It's said to be for lead-acid batteries, but inside are two trim pots that can be adjusted. The one towards the rear changes the voltage at which the green LED comes on saying it's charged and the current to the battery shuts off. With the nearly charged battery connected, turn it clockwise to reduce the shutoff voltage to about 14.1 to 14.15 or maybe 14.2. It takes about 20 seconds to react - not to mention you have to wait until the battery charges to the right voltage - so patience is required. (The other trim pot more towards the front seems to adjust the 2 amp rate current setting.)

NiMH dry cells technical advances: Improving EV energy densities?

   The first NiMH AA dry cells I bought in 1996 were an impressive 1.6 amp-hours - over triple the last NiCd's I'd had. Today 2.3 to 2.6 is the norm. With 2.5 AH AA cells, GM's EV-1 could have had about a 200 mile range instead of 100 miles, with the same total weight of batteries.
   I recently saw a NiMH AA cell with some odd brand name claiming to hold 2800mAH of charge. It weighed 27 grams, 10% lighter than most. I couldn't take it and run tests on it, but taking the specs at face value:

   2.8 AH * 1.2 V / .027 Kg = 125 watt-hours per kilogram. That's over 20% higher than the previous best, and well up into the range of the high density lithium ion types. It's also over double the energy of NiMH flooded cells I've seen specs for such as those used in the GM EV-1.
   These sound like great EV batteries. If 2.8 AH, 125 WH/Kg, is the coming standard, in the choice between lithium ion and NiMH, it may boil down to price, cycle life and performance: which is really better value? Here, eg, the 7 pound, 10 AH electric bicycle L-ion pack at $1000 would have to last a very long time or come down a long way in price to outdo a 7-1/2 pound, 11.2 AH $300 or even $400 NiMH pack. And the NiMHs are inherently good performance, and environmentally nontoxic at the end of their long life.

   The chemistry? One surmises the positrode must have manganese in it, charging to permanganate, or that much of the nickel charges to NiO2 [valence IV] - or both. Otherwise it's pushing the limits of the nickel to an amazing degree. The traditional nickel charge from Ni(OH)2 [II] to NiOOH [III] is theoretically only 289 AH/Kg, and it really only attains around 217. So the weight of the Ni(OH)2 powder alone would have to be 13 grams - half the total weight and probably 3/4 of the bulk, which doesn't seem to leave enough for everything else.
   The alloy of the hydride must be very good, too, one of the best sort of 1000+ AH/Kg types. This is about the same in theory as my Mn negatrode, but at a lower voltage. The overall chemistry is obviously getting very good.
   If the positive is charging to MnO4 and probably NiO2 (both at almost the same voltage), it's the same 'mixed valence' Ni-Mn system I'm using, except mine is in neutral (KCl salt) solution where the reaction voltages are double what they are in alkali (KOH), and I've added gas recombining catalysts to reduce pressure and chelating agents. Except for the lower voltages (of both electrodes) and hence lower energy density, higher operating pressure preventing large cells, and the higher materials costs, it would seem the latest NiMH dry cells accomplish - in miniature - much of what I've been hoping for from my own 2 volt cells.

120 AA Cell Car Battery: Tenergy D cells prove better than Tenergy AAs for starting cars

   I got 200 AA cells, ordered in February. I used 120 to make a 30 amp-hour car battery. This would theoretically operate the same as the 30 D cells battery. But it weighed 9 pounds instead of 12, each AA cell being less than less than 1/5th the weight of a D but holding 1/4 as much energy.

      AA: 100 WH/Kg (energy density by weight)
      D:   75 WH/Kg.

120 AA cells soldering jig
The skew at the center was to make room for terminal bolt heads in the smallest possible case.

The cells kept getting out of place, falling over, etc, so I put the jig on a slant.

Having now soldered together many hundreds of cells in my long career without ever having any  resulting problem or battery trouble, I suggest that no one who solders should hesitate to solder together batteries from dry cells. I think reluctance to solder dry cells has been a major unrecognized factor holding back utilization of the best small cells such as these for greater purposes.

   As assembled, the 360 WH battery weighed 4.00 Kg and was thus down to 90 WH/Kg. Using lighter wire, skimping on the solder, and having no enclosure are probably not the best options. Smaller, lighter terminal bolts could shave off 40 or 50 grams - I've been using big 3/8" ones whose big washers can clamp down car battery cable clamps, but that's unnecessary for EV use.

   In operation however, the AA battery delivered lower voltage supplying the 200 amps to the starter motor. Voltages down to about 8.2 volts were caught on the DVM, versus 9.3 volts and up with a D cell battery. And sometimes the car hesitated to start although it was cranking over fine. Although it's hard to get better than a rough approximation of a very short duration minimum voltage with a simple meter that samples the voltage at a 'random' time, it would seem a D cell can put out the same high current as 4 AA cells with 10-15% less voltage drop, even though the amp-hours ratings add up to the same. Perhaps that extra weight in the D cells goes into heavier internal conductors. (It probably also needs a thicker case to hold the same pressures in a bigger cylinder.) To get the same performance as the 30 D cell battery, the AA would need around 150 cells instead of 120, bringing the weight up to 11 pounds (and the amp-hours to 37.5) and costing more. Of course, it's possible this may apply only to "Tenergy" brand cells, which is the only brand I've tried, and maybe even to these two specific models, so a hard conclusion of "a D cell provides more current than four AAs" may be premature.
   With what I have though, I've decided that the D cell battery is more practical for a regular car battery, and switched my own car back. Three 'extra' pounds means little -- 12 pounds is still 20 pounds lighter than lead-acid.

   The situation is different for EVs and PHEVs, however. In that case, enough cells are needed to provide a desired electric driving range - say from around 2000 AA cells up. With so many cells - and an efficient drive system - the current demanded of each cell isn't so high. Instead, it is probably more important that the 2000 AA cells weigh 60 kilograms (132 pounds), versus 500 D cells at 81.5 Kg (180 p). Increasing from that minimum to 24 KWH to push the driving range up to perhaps 150-200 miles shows more significance: 530 pounds for 8000 AAs is almost 200 pounds lighter than 720 for 2000 Ds. Those new(?) 2.8 AH, 27 g AA cells would be only 193 Kg/425 pounds - another 100 pounds off.

Versus Single-use Alkaline

   On a list someone said he had discharge tested 4-packs of single use alkaline batteries versus nickel-metal hydride. His graphs show the NiMH to be superior in every way - and they can still be discharged 999 more times, while the MnZn alkalines are spent.
   The only value of the MnZn type is that the self discharge is very low. It can be kept around for a long time without attention until needed.

4 AA cell packs of same brand tested with 2 ohm load (note that the voltage and time scales are unequal):
At this heavy load, one charge of the 2450 mAH NiMH battery ("5 volts") outperforms the single use ("6 volts") alkaline battery's only charge,
not only in total run time but also in voltage and current output over virtually the whole discharge period.

The Car Batteries for Sale ad:

30 AH (210 amps cranking, small car) - $300
40 AH (280 amps cranking, mid size car) - $400
50 AH (350 amps cranking, V6?) - $500
60 AH (420 amps cranking, V8?) - $600

Custom EV, PHEV, cycle battery configurations (any voltage, current) - $proportional, nothing extra for being custom.

Individual NiMH cells:
10 amp-hour D cells: $10 each
2.5 amp-hour AA cells: $2.50 each

Turquoise Battery Project

Wasting Time Plugging Leaks

Salt encrusts leaky test battery

   I decided it might be nice to try using 2" ABS pipe endcaps for small test battery cells since my square glued together ones nearly always seem to leak, and on the 4th I went to make an electrode compactor for it. I couldn't get pipe and steel rod that matched in diameter, so I had to make my own, turning plate steel into round bits on the lathe.
   I spent some hours making it, getting everything just right - only to find when I fetched the ABS pipe cap that I'd made it all 1/4" too small - the pipe size instead of the cap size. Back to the leaky glued rectangles!

Wasting time making a compactor that doesn't fit!

   It then occurred to me that I had found some 3/8" thick ABS sheet a few months ago. Before that, I thought the thickest it came was 1/4". If I used 3/8" thick for side walls, there was a much better chance of getting cells that don't leak: more glue, less flex, wider gaskets.
   Along the same lines, I had earlier tried making a transparent acrylic plastic cover for one cell, but immediately inside that goes the graphite sheet or sponge rubber, blocking any view of the internals. This time I made two of the sides of 3/8" acrylic, so that two cross sections can be seen. This may not show much of visual interest either, but there's more chance than with either end face!

Thick walled battery cell case
white and black ABS, and clear acrylic sides.

With graphite sheet and carbon rod with large contact area -- trying to get lower resistance connections.

Osmium: a better dielectric?

   I bought a gram of osmium powder well over a year ago now as a dopant for the separator sheet. It's over 100 $/gram and forms osmium tetroxide, which evaporates as a poisonous vapour. I hadn't dared break open the container. Instead I've used ferric oxide, but I think osmium will work better - with luck a half an amp might become ten. Osmium is quite volatile, reacting with air and water, and it is one of only three elements attaining a valence of plus eight, usually as osmium tetroxide. (The other two are ruthenium and xenon - which is evidently less noble than the other noble gasses.) Osmium will absorb protons and could be a good metal hydride if it wasn't for the fact that it is so rare, and that it reacts with alkali, the usual metal hydride battery electrolyte. Of course, I'm using salt instead of alkali.
   By its chart, it appears it could perhaps also make a great positive electrode in spite of its high atomic weight, potentially moving a whopping 8 electrons per reaction (at almost a volt in acid... reacts with alkali... ??? for salt):

   Of course, I just want it as a dopant on a thin film in the separator. Near the end of March, I finally opened the tiny bottle and sprinkled a bit of the powder into a small test tube with a few cc's of acetaldehyde. I hoped it was enough to have an effect. A bit of this mix I painted onto a piece of cellophane with a q-tip. (The rest went in the fridge with a rubber stopper on it.) I took the battery I'd been testing (leaky, above) apart and placed the new film next to the positrode layer.
   By using the same battery, changes of performance owing to the film should be more readily apparent than by starting fresh. It appeared to be a 15 or 20% improvement, with slightly higher (but still pathetic) voltages and currents - not the sort of "night and day" change I was hoping for. On the other hand, it was a tiny fraction of a reservoir of acetaldehyde with a fraction of a gram of osmium in it. Perhaps it wasn't enough. Also, I shook it but didn't really stir it, and as I think about it, the osmium (the densest element of all, at s.g. 22.6) may have mostly settled to the bottom. The Q-tip only reached down to the top of the liquid. Next test cell I'll sprinkle in some more osmium, and dip a small paintbrush down to the bottom and stir.
   I'll also use the MnO2 salvaged from the dry cells, as that from the pottery supply may not be very pure - that could be part of the high self-discharge problem.

Victoria BC