Turquoise Energy Ltd. News #7

August was busy, but when I could I picked away bit by bit at the motor and controller with the new "optical commutator" design. It worked great! Here is a video of the spinning car wheel: http://www.saers.com/~craig/ElHubcapRunupTest.AVI.

I'm becoming chary of such battery voltages, so I've rewired the motor coil connections from series to parallel to use around 40 volts in place of 120, at what looks like it might be a couple of hundred amps max current. The motor controller will need beefing up.

Hopefully I'll be able to give the new Ni-La (3.25v cells) Turquoise batteriestm some attention in September, having done but little on them since May. I don't think a lot more is needed to get them going. (But then the motor has seemed tanalizingly close since June, so you never know for sure.) Just twelve cells would give the 40 volts, and they should - theoretically - be good for the very high amperages that are going to be needed.

The question of whether we want battery and-or motor manufacturing here in the west, or just to give away the new IP/technology and then import everything from China, is now looking for decision - a Chinese battery maker is already potentially interested.

In the obviously coming switch to plug-in cars, I should think making things here would improve our balance of trade, and getting into the market early would place Canada as a market and technology leader in the field. It should be a fairly easy thing to do, but in my experience it's paddling upstream to move anyone around here who has power to make things happen, so dealing with the Chinese, who want to play ball, might well be the best - or only - way to get them into mass production.

Apparently I've not explained the whole purpose of the Electric Hubcap in recent issues for newer readers, so I'll recap: it's an electric motor add-on system to change gas (or diesel, propane...) vehicles into highly efficient plug-in electric hybrids that normally run on home electricity, turning on the gas engine and putting it in gear only on longer trips when the batteries have run down.

For all its absurdly small size and light weight, this powerful electric motor has the torque to turn the car wheels directly, with no gears or transmission. Someday not so far off, all vehicles will be propelled this way.

First, the old ICE drive system and mechanics is unaffected. The vehicle can run on either electricity or gasoline to fulfill all driving needs, with much less pollution, fossil fuel usage, and cost.

Second, because it is the most efficient type of electric motor and because it turns the wheels magnetically with no gears or transmission, it uses perhaps 50% less energy. It will also have regenerative braking, which has been estimated to further extend range 20-30% in city driving. Thus, the electricity goes farther, minimizing the battery power required and the impact on the electrical grid should people switch en masse.

The frugally used batteries need only power a typical day's driving, eg 30-50 Km, beyond which the "regular" gas engine is used. (Three deep cycle lead-acid batteries might total perhaps the weight of an extra passenger, not half a ton. and will repay their cost with only about six avoided gas fillups!) If the Ni-La Turquoise Batteries complete as expected, even the weight and long-term cost of lead-acid batteries will be cut by four-fifths. The "regular" car battery under the hood will be about the size of a couple of coffee mugs.

I plan to put in optional "battery charge while driving" on gasoline, so on an extended highway trip one can alternate between periods of gasoline and electric driving, which I hope will reduce overall gasoline consumption somewhat even beyond plug-in range.

Wiring changes: The Electric Hubcap system needs to be enabled by the ignition key on "Acc" or "Run" to prevent theft and accidental vehicle movement, and either the turn signals must operate with the ignition key on "Acc", or an additional switch needs to shut the gas engine system off even with the key on "Run". These changes will not affect normal gasoline operation.

For city dwellers who rarely do extended highway travel, a gasoline additive to prevent the gasoline in the tank, fuel line and carburetor or injection system from going "stale" after six or more months may be advisable.

I managed to clarify (never hurts) the new motor control design over the first week or more, and then picked away at making it bit by bit, and finally "went at it" to try out the new system, ignoring most other things, about the 14th to 17th. It worked great! By the 26th I had wired up a single box housing all the parts for the back end of the car: a main breaker switch, spike filter capacitors, motor controller, plug socket for the motor, and all the connections between them, and I mounted it in the back of the car and tested on the 28th.

But I found the car didn't roll even with 60 volts of batteries, and going to 72 volts surely won't make the difference. (Two motors *might* have, but I still expect one to move the car.) Evidently my original estimates of using 120 volts were more realistic, and I was misled by the stopped motor currents being higher than I expected: the currents drop rapidly with RPM to the ranges I'd originally anticipated. However, I've reconfigured the motor to use around 40 volts (to be determined) at triple the current. It's much harder to electrocute yourself with 36, 40 or 48 volts, and the minimum number of batteries for operation is lower. The heavy wires are very short, with cables less than two meters in all. Things seem to be headed straight to the specs ranges of the commercial car motor controllers: low voltages and hundreds of amps of current capacity.

Regenerative Braking: Braking by generating electricity that becomes available again, by recharging the car batteries, or by feeding it back into a power grid (eg trolley busses, electric railways).

The new "open air" motor design (made in July) seems to cool well. The coils haven't been more than a little warm in tests so far. This bodes well for running it at substantially higher powers. Temperatures will be monitored with a sensor on the motor over more extensive operational trials.

The "Lazy Susan" thrust plate (July) did the trick for stabilizing the stator mounting, but it didn't last long. It was quite "fried" after only 15 or 20 minutes of 1000+ RPM motor testing. I made a better one from a car wheel hub and bearing. Since the optics used a slotted disk mounted on the lazy susan, I had to change all that too, grinding slots in the hub and remounting the LEDs and phototransistors. I just finished making it today.

Then I tried out the motor, again with the car wheel jacked up. It ran great on 12 and 24 volts. The currents were lower than I expected, but a 40 amp phase drive fuse blew at 24 volts, indicating over 60 amps from the batteries. Starting gently prevented a recurrance.

The first test of the new design (details below) was on the bench on August 13th. This was evolved from the occasional hours of study and R & D I could squeeze in here and there: figuring out the timing and the design, remaking part of the motor controller, assembly and testing of the optical system, and some resistor ohms changes based on those tests.

In defiance of Murphy's law, it spun up great on the first try! With 12 volts, despite a one inch air gap, in less than one turn in 1/2 a second it would get moving so fast that the heavy but unaffixed rotor slid askew by centrifugal force.

For the first time, there was the sense of the POW of HorsePOWer! What could it not do with a 1/4" gap, a doubled up magnets rotor, and 60 volts!?! That would have to be settled with the motor properly mounted on the car and with a proper pulse width modulator driving it, not jury rigged on a table with a simple on-off switch.

After making and testing a PW modulator the next day, August 14th, I put it on the car on the 15th, and it ran well but needed adjustments. On the 17th I tried again. Timing is everything: With the photo couplers at the right angle, the RPM rises and the current drops. At 24 volts, the current dropped from about 25 amps with the wheel stopped, or 20 amps labouring along with poor optics alignment, to under 10 amps (240 watts, 1/3 HP) spinning away. I finally ran it up to several hundred RPM with 36 volts supply, but chickened out of going to 48 volts.

After some time away, on the 26th I completed an 8"x8"x4" "main box" containing a main breaker switch (100 amps), physically large motor filter ("run") capacitors, a "dryer" socket to plug the motor drive wires into, the complete motor controller unit (relocated from its original separate box), and all the heavy wires connecting everything up.

I removed some gate diodes recommended by International Rectifier from the controller, that I was sure in this design were just going to cause trouble in the form of exacerbated switching voltage spikes. Any spike that exceeds maximum voltage ratings will probably fry the motor controller, and I'm getting tired of fixing it.

So I felt more confident about going up to 48 volts and beyond, and I installed the box and tried operation at 48, then 60, volts on the 27th. The voltage spikes seemed acceptable. The jacked up wheel spun up to over 1000 RPM, but on the ground again the car, though it shifted a bit, wouldn't roll on the gravel. Doubling the voltage again to 120 volts (to provide four times the power) looked like it might tip the scale. A second motor would ensure it.

But I've decided against high battery voltages for electrical safety reasons and have reconnected the coils for 36-40 volts in place of 120, with maximum current of up of to perhaps as much as 300 amps instead of 120 amps.

In typical "dumb" AC power installations (as best I understand it), synchronous motors have to be nursed up to speed, eg with another motor. They run synchronously with the AC power frequency, but if the load is too great, they suddenly stall and stop. I experienced this in the tests with the electric hupcap. It didn't seem like a very good way to run a car!

Synchronizing the drive signals to the actual positions of the magnets instead of assuming the magnets will follow the drive signals, changes everything. The motor is "synchronous" only with itself. Power and hence speed is easily regulated by pulse width modulation (PWM), essentially rapidly turning the power on and off. For more power, it's on more (wider pulse width), and for less, it's off more of the time, with the ends of the range at always on or off. A pulse width modulator to work with the driver chips can be made with a few cheap components, including a potentiometer under the gas pedal.

One could synchronize a drive signal to the shaft position with a traditional commutator. In this case, the simplest design could be three rings and five brushes, subject to wear and arcing on a rotor that doesn't itself use electricity at all, but making it operate like a "dumb" DC motor.

However, to provide this with pulse width modulation, reversing, and regenerative braking still requires a solid state power controller that would still need four power MOSFET transistors - and heavier ones - in place of the six that are used in the brushless solid state control.

Enter optical sensors and a slotted ring. With three LED/phototransistor pairs spaced apart by the angle between coils, and a simple on-off slotted disk turning with the rotor and running between the LEDs and the phototransistors, the signal timing is generated by the magnet positions. The new thrust plate fortuitously provided a ready means of mounting the rotating disk and these optical magnet position sensors.

To turn them into drive signals they're simply fed into the IR2130 MOS gate driver which which turns the appropriate outputs off and on, feeding the high power MOSFETs which drive the actual motor coils.

This makes the motor controller, together with the optical components in the motor, essentially into a subcomponent of the motor, an "opto-electronic commutator" that can easily be made to also handle PWM, forward-reverse, and regenerative braking. Now it works very specifically with the Electric Hubcap motor that has the matching optical sensors, and the motor will work with this controller only. Perhaps it's as well I made my own controller instead of buying one!

The controller is changed, having now three control inputs from the front of the car: PWM for the amount of power, forward/reverse, and brake (derived from the brake lights). For driving we recognize four conditions: go ahead, stop, go back, stop going back. The motor knows only two: accelerate clockwise and accelerate counterclockwise. Regenerative braking is just trying to go the other way from the way it's going. However, the "brake" signal feeds the controller logic to keep the car from zooming off backwards after you brake to a stop.

Now I need to make the driver's dash control box with two PWM's (for gas & brake pedals), change the car's signal lights to work on "Acc", tie in the brake lights... and actually make the Brake handler inside the motor controller.

The "photo" end of the "commutator". Optical sensor pairs (LED<-->Phototransistor) protruding from thrust plate shaft. These are alternately clear and blocked by the slotted disk that turns with the thrust plate, magnet rotor and wheel - a most convenient second function for the thrust plate!

After the testing, the lazy susan was worn right out. So just a week after trying it out, I've had to make a new "version 2" thrust plate with a car wheel hub and bearing. Naturally, the optics mounting and board also had to be redone to match.

"Version 2" thrust plate with hacked car wheel hub and bearings (optics weren't mounted yet in this picture), in motor rewired in parallel for 1/3 voltage, aiming at 40v batteries (at 300 amps) instead of 120v (at 100 amps).

August 30th: Speaking of turfing cherished designs, it appears a microcontroller might be a better choice than "simple" standard logic chips after all. My latest design with regenerative braking isn't bad, but a controller could reduce the 5 logic chips to 1, and 14 passive components to 5, at the small cost of adding a five volt power supply. (The original logic ran at 12v battery voltage.) That's quite a bit less PC board design and less soldering. It would also provide more options such as operator displays for temperatures, battery voltages, and so on.

The controller converts the DC power from the battery to a variable frequency approximation of three phase AC power, on three power wires that go to the motor to create a variable speed rotating magnetic field in the stator, the motionless part of the motor. That field pulls the permanent magnet rotor on the wheel around with it and hence rotates the wheel.

Optical sensors tell the controller the position of the magnets to time the drive pulses. The amount of torque is controlled by pulse width modulation proportional to depression of the accelerator pedal. Torque to slow the motor is provided by differently timed pulses proportional to the depression of the brake pedal. A reverse switch switches the signals to reverse the rotation of the magnetic field.

Funny, last month I quoted a blurb from a custom motor manufacturer extolling the virtues of the permanent magnet synchronous motor (PMSM), and one of their applications was large submarine motors.

In small subs, counter-rotating props are probably the best way to keep the entire vessel from spinning around opposite to the propeller rotation, but to have them both on one axis without gears means the aft motor has to have a hole right down the middle for the other shaft. As it happens, the hubcap can do that easily, and the low RPM PMSM is especially suitable for subs. And it happens that two "Hubcap" motors can also match the designer's target motor input power.

"It's possible that Cobasys (Chevron) is squelching all access to large NiMH batteries through its control of patent licenses in order to remove a competitor to gasoline. Or it's possible that Cobasys simply wants the market for itself and is waiting for a major automaker to start producing plug-in hybrids or electric vehicles.*"

This is one of the main reasons I decided to try to make batteries: Chevron et al seem to me to be abusing the whole purpose of patents, buying it specifically to block access to the technology, while manipulating the marketplace by buying up Cobasys, the company that was all set up and raring to go to provide us en masse with great Ni-MH car batteries. Instead of simply shutting down the company, which could encourage a new company to start up, it would appear they have cleverly limited access to Ni-MH batteries larger than "D" cells and raised prices far above economic levels, to delay as long as possible the advent of electric cars, and then having done the world this service, to be ready to take as much financial advantage of the batteries as possible.

"Ni-MH technology is good, but it's so expensive!" they say. It would appear they've deliberately made it so, and that they've managed thereby to steer everybody away from Ni-MH, by far the most effective proven car battery technology so far, even though there are other hydrides they don't have a patent for, including at least one better one. Anyone else who now wants to start a new Ni-MH plant must recon with the likelihood that Cobasys/Chevron/Texaco will suddenly drop their prices to bankrupt them just when they're most heavily committed for expansion. (and then raise the price again)

I was originally going to make Ni-MH batteries on a small scale, and I hoped to find a way to make them easily enough to publish instructions on the web for any handy people to make them at home.

Then I happened across the lanthanum-lanthanum hydroxide reaction. A new and better chemistry, Ni-La, could change the whole picture. (If it can be made at home, so much the better, but it's looking like there are enough tricky things to do that not so many people would care to tackle it.)

The well known nickel [oxy]hydroxide positive electrode [+0.55 volts] that needs hydroxide ions to charge, may be used with various negative electrode chemistries that all give off hydroxide ions as the battery is charged:

The biggest practical difference between these chemistries is the higher voltage with lanthanum. There are a few other higher voltage elements. A factor in this type of chemistry is that the material can only be recharged if its hydroxide conducts electricity. Thus, aluminum [-2.3v] can be used in non-rechargable batteries, but the Al(OH)3 is evidently an insulator, and so the process can't be electrically reversed.

Another difference is construction. The big nickel-iron cells are traditionally made quite differently to the cadmium and metal hydride "AA" cells. Ni-Fe might be smaller and lighter if packaged similarly to Ni-MH or Ni-Cd.

On the other hand, the bigger Ni-Fe cells with more electrolyte are renowned for their robustness and decades of longevity. Similarly, the big Cobasys Ni-MH batteries used in the GM EV-1 looked like they might well outlast the cars. But the EV-1's weren't on the road long enough for a definitive test.