Turquoise Energy News #8

Craig Carmichael   October 6th 2008


The Month in Brief
** Classes/Workshops: Battery & Electric Hubcap Motor Making
Electric Hubcap Project Detailed Report
rewiring; tests; new motor controller; tests
Turquoise Battery Project Detailed Report
a bit of progress; Lead-acid: 1/4 cost of gasoline

The Month in Brief

   Inventing and developing things would be so much easier if you knew all about them from the beginning! Several assumptions I made when I started proved, one at a time, to not apply to PMSM motors, and I decided to strike out for a better battery chemistry instead of simply copying known Ni-MH workings. The schedule has been set back and back. I'm not driving on electricity yet and I haven't had a chance to look a wave power for months, though a successful test seems so close.

   But if I have a talent, it's to stubbornly continue to explore ways and means when I think I have a good idea, and to recognize and admit to myself mistakes and do an about face. Some important redesign and reworking followed the realization that an ungeared direct drive motor on the car wheel or axle needs hefty torque (hence high magnetism hence heavy current) to start the car rolling, rather than power per se. The Electric Hubcap direct drive concept thus needs a quite unusual motor not only with low RPMs, but utilizing high currents and low voltages.

   I reconnected the motor coils for the high current at low voltage, then designed a circuit and circuit board and made a new 240 amp motor controller. Jacked up, the wheel spins with seemingly great force, but on the ground, at 24 volts the car barely moved. (Conclusion: cars are heavy!) Next I'll try some "tweaking" for more performance before increasing the voltage to 30 or 36 volts.

   Looking ahead a bit, once the essential design and construction details are worked out and the car is running on electricity, the motor and controller will be essentially ready for "beta testing" and refinement - and then perhaps limited production. The question is, what comes once it's running and how do I finance it, or better, start recouping some of my expenses and pay off debt I've incurred? That's the topic of the next section.

   I've also done a bit with the Turquoise (nickel-lanthanum) Battery design. It doesn't charge up yet, but I've finally figured out how to make an acetal ester "glue" for the lanthanum, and I've gelled the negative electrode together with agar agar. A complex job, done so easily for the positive electrode simply with Sunlight dishsoap.
   I also worked out that even using today's limited life lead-acid batteries, it would appear that a car with a direct drive "Electric Hubcap" type motor should cost about 1/4 as much to drive per kilometer as with gasoline. And seemingly with much less environmental impact - batteries are recycled and remade, but gas exhaust goes straight into the air.

Classes - Workshops

   I've decided that the immediate goal once the projects are up and running will be to train others, with (tax deductible) classes and workshops. I've begun writing up instructional manuals to explain the workings and construction details of both the batteries and the motors. (Doing manuals may seem a bit premature, but it also helps me organize and document my own thoughts.) For the participants, the objective of the course will be to build their own working motor project (hybrid or electric car, marine use...), or high energy Ni-La batteries.

   The Electric Hubcap Motor Making Workshop will likely cost around $3000, which will include the parts to make one motor. (Two motors for an extra cost. A deposit will be required to cover parts.)
   Successful participants are to end up with a working high efficiency plug-in hybrid electric/gas vehicle, or other completed electric motor project of their choice... and will know how to make them and the theory behind them. They'll not only save on gas, they'll be at the cutting edge of electric transportation technology. Perhaps alternate configurations, mountings and design improvements will emerge with multiple people engaged. Maximum 5 participants.

   For the Turquoise Battery Making Workshop, again I will order the requisite materials and chemicals so they are on hand for the workshops, and will require a deposit to cover the cost. The participant decides what size battery he wishes to make. (We may start with a small 12v car battery for practice and then do bigger ones.) Again, theory will accompany the actual making, and the workshop will be limited to a small number of participants.

   Both workshops will include some class time to detail theory so the participant gains in depth knowledge of the subject.

   Please let me know what you think, or if you're interested in attending a workshop! (This newsletter gives you advance notice... I plan to put out ads if I don't find participants readily.)

The Electric HubcapTM Vehicle Drive Motor
September Detailed Report

Voltage, Current, Power and Torque

   When 60 volts, 60 amps, eating 4.8 horsepower in electricity, didn't budge the car at the end of August, I did some rethinking. A starter motor isn't very powerful, but it can move a car in low gear because it's geared down about 20 to 1 to the engine, and in low gear the engine is geared down again to the wheels. The force it exerts is magnified (just for example) maybe 50 times. Even the Tesla Roadster's motor is geared down, though much less. (4x?)

   "El Hubcap" turns the wheel directly, one to one, so it needs 50 times the torque of the starter motor. Its static torque is proportional to coil magnetism interacting with the supermagnet magnetism. Supermagnet magnetism depends on the number and strength of the magnets and their physical layout. For a given coil layout, coil magnetism depends precisely on the current flowing in the coils, not on power or voltage.

   The big question for September, since 60 amps didn't do it, was: how much current was needed? I estimated 120 to 180 amps, as a design guide for the next version of the motor controller. It was made to withstand up to 240 amps.

Safety Feature

For safety I've always wanted the car to start braking somewhat as soon as the driver's foot is off the gas, even before it reaches the brake pedal. I've now realized that with the microcontroller in the motor controller, a neutral pedal point can be chosen. At this point the car will be freewheeling, and pressed farther, power will be applied. However, the farther the gas pedal is above "neutral", the stronger the regenerative braking action, braking most strongly with the pedal fully up. The proportions of these effects can be adjusted in software.

September Motor Works - an Acid Test

   After frying a MOSFET or two in the old motor controller, on the 8th I took a fresh approach to testing the motor. After all, essentially the motor controller switches the battery voltage across two of the three motor power leads at a time. Why not forget the controller for a moment and just take out some batteries and jumper cables, and touch the leads in sequence? Hitting the right pair for the current rotor position should provide motive force. If it's sufficient the car should move.
   So I took three batteries out and placed the car where the ground was about level. I got the following results:

1) A lot of big sparks, burned connections, hot wires, and smoking coils. Amazing that the power MOSFET transistors take this sort of abuse multiple times per second! This sort of motor would likely not be practical with purely mechanical switching such as commutator rings.

2) 12 volts didn't have enough force to move the car. 24 volts turned the wheel, rather sluggishly - performance with one motor might be marginal. (Verified later - the car only drove down a very shallow grade, not up it.) 36 volts (probably over 200 amps) looked like it would surely move the car, at least on level ground or up a shallow slope, but without a lot of power. (30 or 36 volt driving test has been delayed by rain. Maybe today.) Certainly no spinning rubber.

3) 36 volts also looked like it would fry the coils in short order. They got smoking hot quickly, even for the high temperature epoxy. I'd rather not use 36 volts without limiting the current based on the microcontroller reading the motor temperature.

4) The 10 gauge power wires into the motor got hot. 8 gauge or 6 or even 4, would be better. Luckily the power wires between the controller and the motor are quite short. (I've changed the battery cables to #4.)

5) One of the old batteries started sputtering out some acid. This made it truly an "acid test"!

   If the car starts rolling well, there is probably no good reason it wouldn't accelerate to at least a decent city speed at least on level ground. Torque generally drops with RPM. At 21 volts in a free spinning test, the wheel started with a strong kick, but the wheel stopped accelerating at only 680 RPM, 68 Km/Hr for my car wheels. That doesn't bode well for the car attaining highway speeds as now configured.

Ideas for Increasing the Torque

   More torque is gained by increasing the magnetic interaction between the rotor and the stator. Presumably, any increase in magnetism, either the supermagnets on the rotor or the coils (electromagnets) will increase the torque. Any or all of the following could be helpful:

1) Two motors. I plan to put on left and right side motors anyway for balance. But aside from this obvious measure...

2) Cram more poles into the existing size motor, meaning 12 coils and 8 magnets instead of 9 and 6. If there's enough room, that would be 33% extra magnetism (and 33% more current). If on the other hand the components have to be made much smaller to fit, it might be self-defeating.

3) By going to larger rotors, eg 11 inch diameter instead of 9.5, one gains space to add more coils and magnets, eg 12 and 8 (+33%). Also, it increases the distance of the pushing force from the center of the wheel, giving more torque from the same magnetism - effectively "gearing down" a bit. So maybe 50% more torque overall. That size or even larger will certainly be the way to go for a larger vehicle, if not this one.

4) Deepen the magnetic fields. The coils are an inch long. They might be made 1.5 inches. That would probably turn 60 turns of wire into 90 and 24 volts into 36 to keep the same current, and add maybe 25% or more to the torque.

5) Various magnet arrangements can (and will) be tried to maximize the rotor's non-electrical torque contribution, deepening the field by stacking two magnets, and-or broadening it by doubling or tripling them around the rotor. Intuitively one would think of increasing the size of the magnets, but 1" x 2" x .5" seems to have become a standard and they cost much less. And it's a hard enough size to handle safely - bigger would be worse.

6) The magnets I used are strength 35 or 37. I've seen up to 50 plus recently, and 42 or 43 are pretty common. 42/35 = 1.2 or 20% stronger, adding perhaps 10% more torque.

   Well... I think I've just talked myself into trying more with the magnets before worrying further about the electrical end of things!

   I happened to open a fishing tackle box (I haven't been fishing in about 20 years!) and found a 25 pound spring scale. (Once my dad's... how did I get it?) I had been thinking of getting one, to gauge the static torque to ascertain which arrangements are actually the more effective. However, I hooked it to the wheel (at about 5" radius) and the first zap of 12 volts to a coil "instantly" rammed the spring to the far end beyond 25 pounds and broke off the ring that the scale hangs from! The ring flew off somewhere - I can't find it. (That still doesn't mean the car moves even with 24 volts.)

   The voltage needed to get the current for the torque is just waste power that heats the copper. To be able to supply the same current at 1/5 the voltage is a definite power advantage, as far as getting the car to start moving is concerned. With the coils now in parallel instead of in series, only around 24 to 36 volts will be needed instead of 120 to 180 volts, "wasting" 1/5 the kilowatts.
   With superconductors, one could supply whatever current was needed at almost no volts and hence almost no wasted power. The thought is of course academic until and unless someone comes up with a "room-temperature" superconductor. Copper, though element number two in electrical conductivity (slightly behind silver, which is pricey), has resistance to overcome.

   Direct drive of the wheels still seems to me to be the way to propel a car, but I'm starting to see why people who started with more knowledge than me intuitively rejected it and opted for more familiar higher speed motors, geared down to the wheels, with lighter torque and lower starting current. Once again, I throw my ideas out the window: my thoughts of "higher voltages at tens of amps" like typical grid powered motors gives way to "hundreds of amps at as low a voltage as will push those amps through the coils". The good news here is that I'm unlikely to electrocute myself, and the car should run with just a few batteries.

New Motor Controller

   In order to do the beefed up motor controller, I decided to make a proper PC board and eliminate my cluttered MOSFET wiring, which was starting to look something like this:

   My last PC board (an entire computer CPU board) having been done in about 1987 with mylar peel-offs, donut pads, tape and photographics, the first week was consumed figuring out up from down in the freeware Eagle PBC Layout Light program.
   Using a PIC microcontroller would save some board layout work, space and soldering, so I designed it with one, and the next step is to humbly learn how to program PICs in "C". (I've already written more software than anyone should in one lifetime, always working in assembly language.)
   Many thanks to Ian Soutar for pointing me to the Eagle Printed Circuit Board Layout program, and especially for holding the PIC microcontroller workshops and inviting me along.

The new motor controller with "fan" heatsink, 12 - 120 amp MOSFETs, PIC 16F690 microcontroller (socket) and IR2130 MOS gate driver chip, mounted in the wiring box with 200 amp breaker, motor filter "run" capacitors, and... wires! A bare motor controller PCB is to the left.

   Of course, now that the PCB is done (and once the software is done), the motor controllers can be duplicated pretty easily.

Electric Hubcap Motor Factoids:

* One small but powerful hubcap motor supplied with 30(?) volts has the power to drive a motor vehicle to city driving speeds (up to 60-70 Km/H or so on level ground) instead of using the gasoline or diesel engine.
* Most installations are expected to use two, for left-right wheel balance and better, balanced, regenerative braking.
* Only the car's wheel turns. The only moving part in the motor is a thrust plate to keep the stator centered on the wheel.
* The virtually frictionless magnetic link to the wheel magnifies useful power by transmitting it all directly to the wheel. There's no losses from a transmission or gears, it requires no gear shifting or other attention by the driver, and it's quiet.
* Permanent magnet synchronous motors also have the highest intrinsic efficiency of all electric motor families, further leveraging the efficient power transfer. As a guess, one might perhaps expect up to 50% greater range than other (geared) electric motor systems from the same energy, and correspondingly better performance for the same kilowatts of electricity used by the motor.
* Installation requires no connections with or changes to the car's existing mechanical components and systems.
* When not in use, the motor has no more effect on the car than any other 35 pounds of luggage.
* The motor sticks out just 4" from the wheel or a couple of inches past the fender, less protrusion than the outside rear view mirror.
* The RPM with 13 inch wheels is about 10 per one kilometer per hour of speed, that is, 450 RPM at 45 Km/Hour. Most electric motors prefer much higher speeds, but the "Hubcap" has good low RPM torque and power. 120 Km/hour is just 1200 RPM, a stately pace for most electric motors but a good upper range for the "Hubcap".
* The rotor is a 10 inch steel disk brake disk mounted on the wheel lug bolts, 6 poles using 6, 12 or 18, .5" x 1" x 2" NIB supermagnets, glued and-or bolted on.
* The stator is a similar 10 inch brake disk (but with cooling vanes), with 9 epoxy cast coils bolted to it, each of 60 turns of #14 wire, in 3 phase "Y" configuration. Magnetic flux is axial.
* A unique design breakthrough is that the stator coil iron is strips of regular nail gun finishing nails in the coil cores instead of custom die cut iron laminate sheets. With this and no axle or other moving parts, the motor is simple enough to make at home, or the coils could be wound by machine and cast, for super economical mass production. Individual coils can be easily replaced.
* The motors dissipate their waste heat via air cooling, avoiding the complexity of liquid cooling systems. There's maximal coil air exposure and heat sinking with the magnets blowing air in front of them, an air scoop on the front of the fairing and air guide vanes, plus a temperature actuated electric fan in case all else is insufficient at low motor RPMs that don't move much air and high power (eg, climbing hills and mountains).

Motor Controller Factoids:

* The controller switches the DC power from the battery onto three power wires that go to the motor's stationary magnet coils, in a six state drive sequence timed to continually push/pull the supermagnets on the rotor around in one direction.
* Three optical sensors looking through slots on the rotor tell the controller the rotor magnet positions, to time the switching.
* The amount of torque is controlled by pulse width modulation of the power, proportional to depression of the accelerator pedal beyond "neutral". Reverse torque to slow the motor (regenerative braking) is provided by differently timed pulses proportional to the release of the accelerator pedal above its "neutral point".
* A reverse switch switches the signal polarities to reverse the push on the magnets.
* In accelerating, the motor uses energy from the battery. In decelerating, the motor generates energy, which goes back into the batteries.
* The microcontroller chip in the motor controller is the "brains" of the switching system, reading also motor temperature, car speed and direction, and battery voltage.

Turquoise (Ni-La) Battery Project
September Detailed Report

   Once again I have had limited time to spend on the batteries. There's progress but nothing definitive to report.

   The most brilliant (and seemingly obvious) thing I did do to simplify testing of new ideas was to make a small sealed battery container of ABS pipe fittings, with a screw-on lid and the terminal wires poking out the sides through tight fitting holes. Theoretically I can now just dump out electrode formulations and try others instead of making a whole new sealed battery every time. Currently it's bubbling out at one of the wires, but the concept is good.

A resealable test battery case,
wherein new formulations can be tried and then dumped out.

   And, it appears I've managed to change some acetaldehyde (made last month) into an acetal ester. Lanthanum chloride has good "lewis acid" properties that enable this to happen, so I added some HCl to the negative electrode powder to turn some of the La(OH)3 into La(Cl)3, then added the acetaldehyde.
   Then I added agar agar, stirred, cooked and then refrigerated it to gel it. The intent is to "glue" the negative electrode materials into an inert ion and electron conducting "cheese" that one may hope won't deteriorate over many charge-discharge cycles.

   A study of "standard reduction potentials" at http://www.webelements.com/compounds/ reveals that lanthanum seems to have the most desirable electrochemical characteristics and price. Furthermore, the lanthanum [as La chloride] can change the acetaldehyde into acetal ester. I bought lanthanum originally with the idea of simply copying Ni-MH batteries, but it seems to be the right stuff for something better!

Battery KWH, Vehicle Range and Cost per Km

   (main stats in brief?... just read the bold print.)

   When attempting to anticipate vehicle range with any given batteries, I've been using a GM EV-1 figure I saw somewhere that showed the EV-1 as using 230-250 watt-hours per mile. I was assuming that the EV-1 would have been somewhat lighter than a typical small car, since it had no gas engine stuff and an aluminum frame. That would somewhat counteract the higher efficiency of the Elactric Hubcap motor, so I just used 250 WH/mile as a rough round figure.
   However, I've looked up the actual EV-1 GVW and I find it was originally 2900-3100 pounds with the lead-acid batteries, which weighed 1200-1300 pounds. (Different web sites seem to have varying figures.) The later Ni-MH gave it a longer range, but it was only 200-300 pounds lighter, making the car around 2700 pounds.
   This figure is around 500 pounds heavier than, say... my Toyota Tercel Wagon. With 150 pounds of batteries (assuming 100 AH, 36 volts) and 50 pounds of motor and wiring in my car, call it 300 pounds lighter. If mileage is approximately inversely proportional to weight, and if the low RPM Electric Hubcap PMSM is about 1.5 times as efficient as a high RPM induction motor operating through a gear, my mileage should be about 1.7 times that of the EV-1. Instead of 240 watt-hours per mile, it should use 145.

   Thus, just three 100 amp-hour, 12v, deep cycle lead acid batteries (=3.6 KWH, =150 pounds) should provide around 25 miles, or 40 Km, maximum electric range (though it's best not to completely discharge lead-acid batteries). This is much better than the 15 miles the EV-1 would get. Double the range with two sets if desired. (300 pounds is still just two passengers worth of weight.) With a very small car, even better figures could apply. Of course, all this remains to be put to the test, but it looks promising.

   One can compare lead-acid battery operating costs to gasoline. If the batteries got 400 cycles of life with about 25 Km driving per cycle, that's 10000 Km on a set of three batteries. Add 2/3 of a cent per kilometer for the actual electricity (@ 6 c/KWH): $67. If those batteries cost $100 each, that's $357 for 10000 Km on Pb-A batteries or 3 cents per kilometer. If the car on gas gets 10 Km/L and the gasoline is 1.40 $/L, gas for 10000 Km would cost $1400 - over four times the overall cost of running on electricity (excluding two oil changes and other gasoline engine periodic maintenance, etc).
   Environmentally it should be remembered that although lead is toxic, recycling car batteries into new batteries is standard practice - the lead doesn't end up in the environment. Gasoline exhaust goes straight into the air. Thus, the idea that lead-acid batteries are worse for the environment than gasoline seems untrue. Perhaps cost per kilometer is a good indicator of environmental impact as well as economic sensibility.

   Similar or slightly better mileage figures should apply to 100 amp-hour (or maybe better - up to 150 AH) Turquoise batteries (6" x 6.5" x 12.5", 24 volts), except that they'll cost more initially, weigh 1/4(?) as much, and running them right down isn't expected to shorten their "indefinite" life expectancy... in short, they should have no worries. Over "indefinite" periods, cost with Ni-La or other "ideal" batteries approaches $67 for 10000 Km.

   I may be disillusioned if I actually try to run the car with lead-acid batteries, but poor though they may be, as used in a plug-in hybrid with limited electric range, and gasoline "backup" for longer trips, the figures above indicate we'd all save a lot of money by driving electrically.

   A Stanford University report* shows over 75% of trips as being under 20Km and 50% of all total daily driving as being under 50Km. If the vehicle can be plugged in at the common usual destinations (home, work...) then at least the bulk of one's driving can be electric.

   Nickel-lanthanum batteries could fill the bill for an all-electric vehicle, almost tripling the range over nickel-metal hydride -- with efficient direct drive wheel motors extending that by a further 50%!

* Document: ev_battery_assessment.pdf , downloaded from somewhere or other.

Victoria BC