Turquoise Energy Ltd. News #34
Victoria BC
Copyright 2010 Craig Carmichael - December 1st 2010
= http://www.ElectricHubcap.com


November in Brief (summary)
   * A month of further development and refinement of the EH motor

Electric Hubcap System & Motor Building Workshops
  * Electric Hubcap Motors get Better and Better!
  * EH Motor Testing: proving the efficiency and power; the motor with adjustable performance specs!
  * Magnet Rotors: Epoxy-polypropylene composite skin strengthens magnet bond to rotor
  * Welding bearing race hubs: preheating parts with torch improves hub-to-rotor welds
  * Jigs, CNC drilling improve motors & speed making them
  * Motor controller: Designing for serviceability is paying off!
  * New Motor Controller printed circuit board designed - smaller PCB uses improved A3938 chip.
  * First motor workshop student completes his motor!
  * Small business opportunity? for Victoria BC residents: make & sell Electric Hubcap motor coils, magnet rotors (sales opportunities per market demands).
  * Making & marketing motor components online to simplify making more marvelous motors
  * Lower price parts sources found: estimated parts cost to make an EH motor down from almost $400 to under $300. (That's still buying parts at retail prices.)
  * Lower cost for workshops: $700 or less.

Better Blades for Windplants (just some cool ideas... not a project)
  * Residents object to nearby windplant noise.
  * Here are ideas for quieter, also durable and reproducible, windplant blades.

Torque Converter Project
  * Plan for modifying existing escapement converter -- 5-at-once torque hits instead of one at a time should substantially boost performance.

Pulsejet Steel Plate Cutter Project
  * One-way air intake made, main parts cut out, welded... a couple of parts left to make.

Electric Outboard Motor Project
  * Steeper Pitch Propeller installed.
  * On the water: Runs great, quiet!
  * Movie Clips are on YouTube (Search there for "Electric Hubcap Outboard" or "Turquoise Energy".)
  * not 'speedboat' as prop is 'stuck in first gear' with motor using only 1/4 power at its max RPM.
  * but it should move a considerably larger boat just about as fast as the small one, drawing more on its available power.

Turquoise Battery Project
  * Latest carbon electrode backing experiment: hexadecane & graphite.
  * Some Behind the Scenes Battery News... and of course an accompanying editorial.
     - new developments in lithium and nickel-iron
     - what the slippery big oil people try to get us to believe, and to not realize, about bateries.

Newsletters Index/Highlights:

Construction Manuals for making your own:

* Electric Hubcap Motor
(latest rev. 2010/09/xx)
   - the only 5+ HP motor that can easily be made at home?
* Turquoise Motor Controller
(latest rev. 2010/05/31)
   - for the Electric Hubcap. (Probably there are commercial controllers that would work, too.)
* 36 Volt Electric Fan-Heater
   - if you're running your car on electricity, you'll want a way to defog the windshield and keep warm.
* Lead-acid batteries: Sodium Sulfate 4x longevity additive - "worn out" battery renewal.
* Simple Spot Welder for battery tabs, connections (in TE News #30)

all at: 

November in Brief

   This month was devoted in large part to further development and testing of various things related to working Electric Hubcap motors and motor system components, rather than to experiments with partly working or new inventions that aren't ready to use.

   Tristan Money finished his motor, the first "motor making workshop" made Electric Hubcap motor and (AFAIK) the first one not made by me, on the 18th. With designs and techniques being continually improved and refined over the last 2 or 3 months, it's also the best EH motor to date and likely to be pretty much the final design. He plans to use it for an electric motorbike and the project is underway. I made a shop stand to hold the motors and I'm running some tests on his motor.
   On the last Sunday, a second workshop student came to help make the next motor for the experience.

   I got a steeper pitch propeller on the electric hubcap outboard, dragged out the boat, and got things ready to sail. The steeper propeller would push somewhat harder at lower RPM: the gas engine was probably around 5000 RPM, whereas the EH was 1750. On the first outing (Saturday the 6th) with Tristan, the boat had decent speed, but it wasn't going to make a big wake or get up on a plane. The motor loafed along near its maximum RPM drawing only 1150 watts of power (33 amps at 35 volts), perhaps 1.4 HP. We did a couple of blurry (tricky camcorder) video clips - total 4 now on www.youtube.com; just search there for "Electric Hubcap Outboard" or "Turquoise Energy".
   Still, the boat made several short trips with no hitches and the motor performed reliably. It didn't even get warm, and the controller stayed cold. If the propeller wasn't geared down 2.75 to 1 in the leg I think it would have given a nice fast ride. "Stuck in first gear" as it is, presumably it would work as hard as necessary to push a considerably heavier boat almost as fast as it ran the light 14' aluminum one.
   The test against the induction motor outboard in a heavier boat was delayed and finally scrubbed for the month by preoccupations of the owner, by his boat having heavy marine growth all over it, which I scrubbed off in 3 or 4 sessions, and by cold weather.

   I used Bill Metcalfe's homemade CNC drill/router machine to drill the 29 holes in three EH motor stators and the 9 in each of a couple of heatsink bars. Then I purchased it from him. He had been getting little use out of it in recent years and it was rusting and dusting in his unheated shed. I could see various possible uses for it. It will be valuable for making the new motors and controllers, and for prototyping new designs with more precision and accuracy.
   We disassembled it, loaded all the parts into his van and my station wagon, and brought it here. Making floor space for, reassembling and setting up the CNC machine was a small project in itself. (oh boy... another project!) It occupies 1/4 of my whole machine shop and is larger than I need, and the investment was all I could afford. But it works! Making my own CNC machine would have been "problematic", as they say.

   Having the CNC machine now makes it practical to offer pre-drilled stators as "parts for making Electric Hubcap motors", now listed on the web site. I hope these will be the first of a growing list of DIY EH motor making parts. It will take a lot of sales to recoup the cost... classes of electrical students making motors, perhaps?

   Of course, all the activity to do with the outboard and with setting up a CNC machine ate seriously into my R & D time. Running the motor making workshop didn't take a whole lot of time per se, but with the attention and thought to them, the motor design and construction techniques became more refined as things progressed. I designed some small circuit boards to hold the magnet sensors and a temperature sensor on the motors.
   I didn't think I'd get much done on the batteries or torque converter, but I did a small experiment in carbon electrode making and came up with a good design for the next torque converter prototype (only slightly modified from the current one but should work much better!), and I snuck in an hour here and there to work on the pulsejet steel plate cutter and got it pretty much together except the propane connection.
   By the end of the month, the need for more motor controllers, and for circuit boards for the magnet sensors in the motor, were getting pressing, and I spent some late nights laying out boards - the new motor controller board using the A3938 motor controller and three little boards for the hall effect (magnet) sensors, one also with a temperature sensor - not that any of the coils have gotten more than slightly warm so far.

   And somewhere along the way, I had an inspiration for how to make improved windplant blades, written up below - ways to make them quieter, strong, replicable - which I don't have time to pursue. If anyone is interested though, I'd be glad to go over it, and photocopy some blade dimensions and shapes from a book.

Electric Hubcap System & Motor Building Workshops

   I read in a newspaper that a Canadian sponsored "Around the World 80-Day Zero-Emission Race" of electric cars is in progress. It crossed Asia from the Geneva starting point on August 16th, and hit Vancouver November 12th en route to Cancun Dec. 5th to hit the World Climate Change Conference, and back to Geneva in January.
   It's a pity there's no "World's Most Efficient Electric Car Motor" contest, because I think the Electric Hubcap would win it. I've said it before, but perhaps it bears restatement:

1. The brushless permanent supermagnet motor is inherently the most efficient family of electric motors.
2. Axial flux layouts with wide magnetic flux gaps are the most efficient of those, and the Electric Hubcap has a 1/2 inch gap.
3. The lower RPM range of the EH (0-2000) results in lower running losses, and having no brushes also reduces friction losses.
4. The donut shaped coils filled with core material to form fat "hockey pucks" magnetize the maximum ferromagnetic core interface, nearest to the rotor magnets, with the minimum amount of copper wire, for the lowest possible copper resistive losses.
5. As well as being insulated between strips, the soft-magnetic iron alloy nail gun finishing nail strips typically have poor conduction along their length, further minimizing iron stray conduction losses. (Even lower iron losses are possible, and feasible with capitalized production methods... The ideal core "traditional alloy" form is even thinner, individually coated iron alloy wires packed tightly together, all aligned straight up-down in the coils. If they were nanocrystalline to much reduce hysteresis losses as well as stray conduction losses, so much the better. I think the ultimate core is nanocrystalline ceramics, which would have virtually no losses at all.)
6. For electric transport, the motor itself needs to be carried around. A lighter motor is effectively more efficient. The latest EH design, bare bones, weighs only 32 pounds. In contrast, my 7.5 HP sawmill motor is about double that weight for just 50% more power - and that is in fact much the smallest 7.5 HP induction motor I've seen; typically they are considerably larger. Many radial flux permanent magnet motors are smaller for their power, but their high RPM ranges would lower their efficiency.

Performance Testing EH: the motor with adjustable specs!

   Better than just saying how good it is in theory will be actual tests to prove it. I thought I'd need a pretty fancy setup to do so, but someone on a motor controller chat list (!) says it can be done with some fairly simple techniques and measurements, and then calculate some of the specs like efficiency from there. I've done the first of these: the coil resistance measures to be about .064 ohms between phases - almost exactly as calculated for the length of wire in the coils and the resistance of #14 AWG copper wire. Since most ohmmeters including mine only go down to .1 ohms, this was measured by applying power to the coils from a power adapter, and measuring the voltage across and the current through the coils between any two phases. I = E/R, and also R = E/I. The voltmeter is much more precise than the ohmmeter: .0001 volts.

   For the next step, I had to turn the motor with another motor and measure the generated voltage at several different RPMs. That entailed building a stand to hold the motor securely and making some way to connect it to another motor. I made a stand from angle iron (to be  C-clamped to a bench) that holds a EH motor vertically, and drilled holes in a 9" V-belt pulley made of sheet metal, so it can be mounted on an EH rotor disk. Then I used my radial arm saw with 2 or 3 smaller pulleys to spin it, and recorded the voltages. (farther down) Then there's some calculations and some results, including efficiency.

Testing EH motor on the new motor holding stand as a generator,
using radial arm saw and various V-belt pulleys to turn it at different speeds.
A 6 diode bridge for the voltage, and a power supply for the magnet sensors
to get the RPM, were required along with the meters.

   To get the maximum current, I need to be able to put a steady load on the motor and measure the current how hot the coils get after a while, but someone good at math can derive the horsepower (roughly) from the acceleration of the motorbike now being made, so I won't need to know or measure just how strong that steady load is.

   Some interesting specs are different currents and RPMs measured on the same motor with the flux gap set to about .5, .65 and .8 inches. The farther the rotor is from the stator, the lower the current at a given RPM, and the higher the maximum free-spinning RPM is. Naturally, the torque, and the horsepower at a given RPM are reduced as the gap increases. Within limits, the motor has "adjustable specs" with respect to torque and speed.

The "Adjustable Specs Motor" table:
No-Load Current versus RPM with the flux gap set to three different widths.
(@ 24 volt operation - I really must treat another supposedly "worn out" old car battery
with sodium sulfate for use in the shop so there's 36 volts handy!)

Amps, Gap .5"
Amps, Gap.65"
Amps, Gap .8"
(not measured, max RPM not measured)

13.9 @ 1150 (max)

11.9 (max 1440)

   One interesting thing I might try related to this is to increase the gap in the EH outboard motor. This would up the RPM somewhat, and so the boat would go faster - up to the limit where the torque stops being adequate to get the motor up to the higher speed under load. I doubt if going to an inch or beyond would leave a whole lot of torque and power.
   The present gap is roughly .57" and the maximum RPM is about 1770 (at 36 volts). If I upped it to .75", it might well go up to around 2500 RPM. On the other hand -- Ooh that's fast!

Note: More on performance testing is farther down... I seem to have accidentally split the text into two areas, with considerable duplication, and I don't have time to correct it.


   But low losses and high performance are only one part of the equation. Motors need to be affordable and practical. With the strengthened magnet rotors - improved twice in two months - and generally improving parts and construction techniques in all areas, the Electric Hubcap motor is getting simpler and faster to make, and I have confidence that it is also getting to be rugged and dependable to a degree probably unmatched by other axial flux supermagnet motors.

   The improvements lead to the tentative offering of parts to help make the motors. Some of them (eg, the stator) can be CNC drilled. I'll be offering stators complete with turned, welded bearing hubs on a limited basis. Other labour intensive parts (coils, magnet rotors) can perhaps be made by people who might like to start their own little business. (Any volunteers living around Victoria?)

Hubs - Welding

   My biggest remaining qualm was that the welded hubs hadn't turned out to be as strong as I expected welds to be, owing to the weakness of cast metal and especially after the heat of welding. If pounded a bit, the welds would break where they met the rotor metal. The welds themselves also tended to crack during cooling, which appeared to be a separate heat stress problem - probably related to the cast hubs shrinking when heated by welding.
   I was thinking that perhaps there might be some better part than the 1-1/2" pipe couplings to use for hubs. But on further reflection, that's not where the welds had been breaking - changing the hub wouldn't solve the problem. Changing the rotor disks isn't practical, unless perhaps one brand proves to be notably better than others.
   On consideration, I decided to try pre-heating the cast parts, as suggested at one welding store with agreement by another that it should be helpful. Perhaps that might pre-shrink them, or at least be an advantage. I used a "swirljet" propane torch to preheat, and though I couldn't get all that metal glowing red, it would have been much more than warm. It was the best I could do at that moment. Then I welded, using lots of stainless steel weld, all the way around the circle.

   That seemed to do the trick. I did two rotors, and considerable hard pounding with a hammer didn't break the hub away from the rotor or otherwise seem to have any effect on either of them. I stopped short of probably breaking the rotor or hub themselves, or seriously mushrooming the edge of the tenon, with the hammer. (One of the rotors already had a minor crack, that probably developed during the welding. It seemed unaffected by the hammering.) Unless there are future problems, I'm calling the problem solved, providing the preheating is used and the hub is welded all around with the stainless (or special cast metal) welding rod.

Hubs - Turning - CNC Lathe?

   Another small concern is getting the hubs lined up true with the rotors so the rotors sit absolutely straight. With a clamp and careful welding they're 'good enough', but ideally they would be welded on and then the races would be turned, but I can't fit the finished rotors onto the lathe. The answer to this must be either a special plate that would allow finished rotors to fit onto the current lathe, or a bigger lathe. (...and just where would I put that?)
   Later I figured out a way to mount the rotor so the outer face (only) could be turned after assembly. This has proved to be vital as the hub, at the outer tenon end, so carefully turned to exact size, seems to distort and shrink a bit during welding, and the bearing races won't fit in. It would seem that the outer tenon can and should be turned after the rotor is assembled. That should improve alignment. (Hub shrinkage also explains why the welds are prone to cracking as they cool, and shifting alignment during welding despite the hub being securely clamped to the rotor.)
   My latest thought for a solution to getting the rotors onto the lathe facing the other way, since I'll probably be doing a lot of rotors, is to cut 1/4" off the lathe bed, thus making the 5" radius gap zone just that vital little bit wider. Then nothing else is needed.

   Now I'm thinking... A CNC lathe to turn the hubs automatically would be ideal. But isn't it just a matter of the tool moving against the spinning part? That would mean a minimal CNC lathe could just be a CNC lathe tool turret mounted on any old lathe... even mine. In fact, it would seem it may be just a matter of replacing two hand cranks with stepper motors. The hard parts would be (a) getting the cutting point set to some "home" so it's moving from a known reference point relative to the hub and (b) in my lathe, there's some play in the threads: the position doesn't start going backwards to the previous direction until this play is taken up. That would make for uncertain positioning by the computer, which can't account for that.

Rotor: Magnet Attachment Improvements

   After having another magnet fly off a rotor in October I realized first that the epoxy glue I'd been using loses its grip after a couple of years, and then further that something more than just gluing would be best regardless. The problem was slight when the motors originally were supposed to turn with the car wheel, where they would never go more than about 1200 RPM (at 120 Km/H on the highway with 13" wheels), but it increases with the square of the speed. It would seem they want to go up to somewhere around 2000 RPM when spinning freely. This time I used a liquid epoxy resin as glue. Then I coated the whole rotor and the magnets with a thin layer of epoxy, which gave 'glue' around and going up the sides of the magnets, rather than just a layer underneath.
   Still, I wrote last month that I wouldn't want to be around if my 2000 RPM motors were turning at over 3000 RPM. At 3200 RPM the centrifugal force would be 10 times stronger than at 1000; at 5000 RPM, 25 times stronger. But these are being billed as "DIY" motors, and what people might do with them in experimentation is unpredictable. Someone with no tachometer, or unaware of the danger or not even knowing the motor specs, might, eg, boost the voltage to 48 volts or more and actually get it turning at 3000 RPM or higher.
   Owing to this, and on general principles of strength and durability, I decided the magnets needed to be attached even better. I decided to do polypropylene-epoxy composite, and tried it out on the outboard motor's rotor. "PP" cloth is lighter and stronger than fibreglass cloth, and non-itchy. (Its great tensile strength is why polypropylene is used for that rather coarse common yellow rope.) Drawback as a composite cloth is that it likes to hold its form and any creases - it won't lie flat like fibreglass - it "floats up in the resin and so is harder to work with". (It would also be degraded by UV light, but what I found only comes in opaque black anyway.) Although this information is on the web, nobody in town seems to have ever heard of polypropylene cloth, much less has it in stock. Unexpectedly, I seem to find myself at... well, near... the leading edge yet again, even in such a 'mundane' area as 'fibreglassing'!
   But at Capital Iron I found heavy PP strapping ("web") up to 2" wide, such as might be used in backpack straps, cargo straps, etc. Can't get more solid than that! They had it in both the marine department and the fabric department - same stuff at somewhat different prices. (Aha!, it's true... boat owners pay extra!)

   I first planned to wrap a long strip around the outside of the rotor and cut slits in it, and fold the pieces over the magnets from the outside edge. Then I would paste some around on the inside, slit it, and fold it over the magnets from the inside. That would cover each entire magnet and hold across much of the rotor surface.
   But the PP strapping was too stiff to work with so easily. I ended up cutting a separate trapezoid piece for each magnet, and then making an aluminum form (x 12, with non-glueable polyethylene sheet liners) to press the cloth flat against the magnets while the epoxy set. I put small pieces of steel on top of these, and the magnetism pushed the aluminum pieces down firmly.

Covering the rotor with epoxy/polypropylene (re-enactment):
12 aluminum clamps (with polyethylene under to keep them from sticking)
 held the stiff cloth pressed against the magnets

Finished Magnet Rotor (with another coat of epoxy).
Lots of epoxy and the polypropylene strap/cloth pieces overlap --
I don't think there's any chance now that any magnets will come loose at any sane RPM.

   On December first I inadvertently got an EH motor turning slightly overspeed (2300 RPM) myself, during testing of its specs as a generator. I was glad for the solid assurance no magnets would fly off.

   I think next time I'll try to leave more of the strapping's rougher surface rather than have so much of it 'flooded' with smooth epoxy - after all, golf balls have little pits all over because it actually cuts air friction. (See the 'better windplant blades' article for my idea of the 'ideal' texture.)

Magnet Sensors

   The only other thing that's been bothering me is finding a better holder for the magnet sensors, one that can be 'mass produced'. ...But never mind, as I write I've just thought of a nice circuit board arrangement to bolt to the angle bracket mountings!
   Later in the month I designed these boards in Eagle PCB layout program, ready to send off the Alberta Printed Circuits along with both types of motor controller boards when the A3938 board design was ready to try, which it was on November 29th.

Coil Clamp Bar Drilling Jig

   When Tristan got to the nylon coil clamp bars, I cut a heap of them from a nylon sheet on the radial arm saw in a few minutes, and I had him make a jig for drilling the holes in the clamps exactly 1.5" apart. Two thumbscrews clamp the work in place, and they just need to be loosened slightly to slide it out by pushing the next one in. It worked perfectly, and made it a snap to drill 9 clamp bars for the 9 coils of a motor.

Drilling the nylon coil clamp bars with 1.5" apart holes jig

   I then put the 1/4" - #20 threading tap sticking straight up from a bench vise, and turned the nylon bars around it instead of turning the tap. That was much easier, though to call it "a snap" would be exaggerating. Still, I had 9 bars drilled and threaded in under 1/2 hour.
   Later I found it was much faster to use a variable speed (cordless) drill to turn the tap to thread holes, and I did it also with the six #6-32 MOSFET mounting holes in some heatsink bars. The cordless drill has adjustable "slip" settings so it won't snap the threading tap if it's accidentally turned too hard. This is a great production speed-up!

   I used 3/16" nylon for the bars, but with a 1/2" flux gap leaving plenty of room, I may increase that on the next purchase to 1/4" for a longer threaded thickness.

   The reasons for using nylon bars to clamp the coils down are:

1) Bars instead of just nuts prevent laminate pieces of the coil under the nuts from breaking off, which allows the rest of the coil to slide forward and hit the spinning supermagnets. (Experience! But the coils that did that also had a weakness around the bolt holes, now eliminated.) The bars hold across the whole coil including the wire, so it would have to break into little pieces before it would come unclamped - and even then the coil of wire would be held back.
2) Metal bars would interfere with the motor's electromagnetism and decrease efficiency.
3) The nylon makes for a friction fit, providing "nyloc" lock nut function so the bolts won't work their way out.

   The bolts are placed left and right on the coils because that's the direction of magnetic thrust, so it holds the coils stiffest. Otherwise a vertical bolts and a vertical bar (that is, pointing inside to outside of the rotor) would be preferred for cooling air flow. (...a single mounting bolt might let the coil twist and work loose.) Also to aid cooling it's just a bar, rather than a big round piece covering the whole coil.

   With the CNC drilled stator holes and the jig drilled clamp holes both exactly 1.50" apart, everything lines up perfectly and tightly with no custom fitting, hole edge filing, or need to drill oversize holes, which multiply the hours of assembly time, as often happens with hand-drilled holes.

First EH Motor Workshop Student Completes His Motor

   Tristan Money got his motor running on Thursday the 18th, after about 38 hours of actual working time during October and November. (Yes I was keeping track.) He did everything including the welding and the lathe work (which jobs I had pretty much expected to do myself), except for a couple of painting jobs that I did between sessions just so the paint would be dry and not hold things up. The magnet rotor has the polypropylene/epoxy composite shown above.

   But now there's the CNC drilled stators, preheating and defined better procedures for welding and turning the hubs (...or complete, ready to use stator plates for those who don't want to weld), the strip lengths chart and simplified assembly techniques for the coil cores, and the jig for drilling the coil clamp bars. Sometime soon there will also be little printed circuit boards for the magnet sensors to simplify their wiring and assembly.
   I anticipate all the improvements together will likely cut around 14 hours off the building time for future motors, making it somewhere around 24.
   If ready-made coils could be made available, they would cut off another 7 or 8 hours, altogether cutting building time to less than half. In lieu of some success at making nanocrystalline ceramic coil cores (so far elusive) that are also economically practical, this would be dependent on someone wanting to make coils to sell.

Tristan's motor on bench. 32 pounds; It ran great!

Turquoise Motor Controller: Designing for serviceability pays off!

   I ran Tristan's motor the next morning to take a few sample readings, and suddenly it quit working. The problem appeared to be with the motor controller. I opened the case and wiggled the control wiring sockets, and the sound changed. Having made it integrated into a wiring box chassis and easy to dismount the motor controller side piece, I did so. There are just a few chassis screws, 5 heavy wires and two small plugs to disconnect, and out comes the whole controller to troubleshoot and repair (or send for repair), without disturbing the 'junction box' wiring to the outside.
   I unscrewed the small circuit board, and found two pins of the controls socket had been left unsoldered when I made it. Of course I thought that was the problem and looked no further. On reassembly and reconnection, it still didn't work right, but wiggling the plug now had no effect. The unsoldered socket pins on the circuit board could have caused intermittent problems, but in fact none had previously been evident.

   Next I tried measuring voltages and found that a pin in the motor power connector wasn't pushed home in its socket properly and had come disconnected. When I fixed that the motor did violent things.

   Then I measured some resistances and found there was a short-circuited power mosfet. I took the controller apart again and replaced it. Since it was mounted to the heatsink and was connected with wire and not to a circuit board, replacement was pretty simple. Everything worked fine after that. The original problem was doubtless the improperly inserted power plug pin - it probably made a spark that 'took out' the mosfet, as it came disconnected while the motor was running.

   Problems including blown components are more likely in high power circuits than in other electronics, and the units are usually more costly. At Canadian Electric Vehicles I heard blowing controllers was common in new installations owing to incorrect connections. And in my experience, with poor connections in any of the high power wiring. Once things are properly installed, trouble is rare.
   With most motor controllers I suspect opening, troubleshooting, disassembly and repair would have been a much bigger job, and I had to do it 2 or 3 times, so it saved me some grief. With a construction that's hard to disassemble, test components, and access for repair, it gets to the point where you order a whole new $500 (or whatever) unit - and possibly more than once before the problem is debugged - instead of taking a few readings and replacing a $4 transistor.

Motor Controller

   Tristan did up an initial schematic for the A3938 motor controller PCB. I fixed up or changed a couple of things and layed out the circuit board for it, "Rev 1", finished on the night of the 29th.
   In looking a bit further into regenerative braking, I realized it's not done the way I thought, and it probably would be complicated with a permanent magnet motor. With an electrical field creating magnetism, the field voltage is increased to raise the generated voltage ("back EMF") above the battery voltage, hence charging them with the rotating energy of the motor.

   It occurs to me that with the "adjustable specs" EH motor, one could have in the brake mechanism a means to reduce the flux gap. This would raise the generated voltage and induce regenerative braking at higher RPMs (which is where the bulk of the energy is anyway).
   But that would certainly complicate construction of the motors!

The CNC Machine!

   The CNC drill/router machine maker/owner asked when I was coming to do the rest of the stators on the evening of the 3rd, and we arranged for the next morning, so I got ready to go over there and do the two I had on hand, and since I was going, I wrote a CNC drill sequence for the heatsink bars of the motor controllers and a jig to hold them in place for drilling.
   I did the rotors and bars, then we had lunch, and he said (again) that he didn't want to go to too much trouble adjusting it for smoother operation since he had to move it somewhere - it was collecting dust and rust, little used in the unheated garage. And, again not for the first time, he suggested an option was to move it to my place. This time, I had uses for it and readily agreed. We took it apart and brought it over in my station wagon and his van.
   A couple of days later I decided where to put it - a corner of the machine shop - and removed the prior furniture and drill presses, and started setting it up. It takes up 1/4 of the whole floor space of my shop (ugh!). I did a considerable amount of work on it and spent some money on air line, plugs and switches.
   Then I thought that having someone else's machine set up in my shop was eventually almost bound to prove unfair or troublesome to one or both of us -- either he would gradually start to feel he was intruding to come over and use his machine and I would essentially 'acquire it by osmosis' as time went on, or he would find a major use for it himself and the frequent use actually would have become intrusive, or he would take it back after I'd invested my time and energy and was using it regularly and was counting on it. So I bought the complete machine off him. The investment was more than I've put out at once for any single item besides property taxes in many years, and it pretty much wiped out my meager bank accounts, yet it was doubtless a great deal for an "industrial strength" CNC unit that I already knew does what I need it to do.

   I can now do a better job making prototypes where precision is important, make motors faster and more accurately for testing and evaluation, and also "mass produce" certain parts for the motors and controllers - initially pre-drilled stators and heatsink chassis parts - and put them up for sale on the web site, to make building EH motors and controllers easier for others.

Torque & Magnets (how many is optimum?)

   Last month I wrote that with 6 supermagnets (2" x 1" x .5") on the EH outboard's rotor I could prevent it from starting to turn even at full power with my hand. I tried it again with all 12 magnets and I couldn't hold it back. (These tries were just with my hand on top of the smooth rotor with nothing to grab.) I still think 18 magnets would be overkill, though, and have no plans to try so many again.

Unexpected Motor Current Measurements

   Before I had the DC clamp-on ampmeter, I had no way to measure higher DC currents. Measuring up to 30 amps earlier had apparently "done in" my good multimeter's test leeds, causing measurement problems later on, especially with some battery resistance tests. I was taking the AC current flowing in one phase, as measured with my AC clamp-on ampmeter, and multiplying by the square root of two, per my BCIT 1975 Electric Power course notes.
   This month, I tried measuring both the DC and the AC currents, and I found the situation was wholly different than I had imagined. I ran a motor spinning with 3 amps DC coming from the battery per the DC ampmeter clamp attachment and checked the three phase currents. Much to my surprise, the AC current readings averaged about 4.3 amps! According to formulas, there should either have been 2.1 amps AC (with 3.0 amps DC), or 6.1 amps DC (with 4.3 amps AC in each phase).
   What is the answer? Either the AC meter doesn't like pulse width modulated square waves and its readings are out to lunch, or else the AC current isn't in phase with the voltage, and so the volts * amps, volt-amps or "VA", doesn't translate directly into watts, where the DC current times the battery voltage does. The latter idea would seem reasonable on a motor with inductive coils, but the former seems indicated by the fact that by changing the range on the AC meter, the reading changes: on the 0-20 amp scale, it says (eg - they fluctuate) 5.67 amps; on the 0-200 scale, 4.6 amps, and on the 0-600 amp scale, it just reads 1 amp!
   Assuming then that the DC current draw figures are to be trusted over the AC, I have some recalculating to do! Further specs testing is regardless coming up.

Motor Testing and Performance Categorization

    So far, I've built the Electric Hubcap motors and measured a few things, but they haven't been properly measured as to efficiency, total power, etc, except in vague terms or in certain aspects. Now someone on a chat list has said I can calculate some of the specs such as efficiency with a few readings and without much special equipment. And Tristan pointed out that horsepower can be derived from knowing the weight of his motorbike plus load, and timing the acceleration from 0 to whatever. Only maximum steady-state currents and power levels really need some sort of mechanical load in the shop, and a temperature probe to check the maximum temperature of the coils over time.

   But minimal equipment isn't no equipment! Last month I made the second motor controller, for testing motors in the shop.
   On Nov. 26th I made a motor holding stand from pieces of angle iron so that the motors would sit still and securely while running. It can be attached to a bench with my favourite bolt-down devices: C-clamps.
   Finally I drilled holes in a large V-belt pulley (a 'solid' sheet metal body pulley) so it could be attached to a trailer (motor) axle flange or to a rotor, and on December 1st I fitted it out and got it attached. This is to allow an EH motor to be turned by another motor in order to measure the generated voltage, which when a motor is running is called the "Back EMF".
   Later I plan to make another stand to allow two EH motors to connect, and then one can be used as a test load, eg, powering car headlight bulbs, to test the other's motor characteristics under load.

   To digress a bit into motor theory, the generated voltage increases linearly with RPM, and is subtracted from the supplied voltage. This is why motors accelerate quickly to a certain speed and then run steadily at that speed. If the motor runs at 36 volts and really only needs 2 volts to keep it running at top speed, it will draw lots of current and accelerate rapidly until the back EMF approaches 34 volts. The motor won't go any faster unless it's pushed mechanically (as below), in which case, if the voltage rises above the 36 volts of the batteries, the push starts generating power that goes backwards, into the batteries, recharging them.

   Below is a table of the generated voltages for Tristan's motor, which should be typical, with about a .65" flux gap, at different RPMs obtained with the V-belt pulleys I happened to have on hand that had the right bore size to mount the radial arm saw. The two low RPMs were turned by hand to a ticking clock and readings are approximate. (One could wish for a couple more in-between figures, eg 400 & 1500 RPM.) The voltage measured was the 3-phase rectified DC voltage. Since the diode drop across the rectifiers read .765 volts on the meter, this figure was added to the measured readings.
   The 2300 RPM reading is interesting because it's the fastest any one of these motors has ever spun, and because the voltage is over 36 volts, so this speed wouldn't be attained with a 36 volt supply - it would put charge into the batteries instead of drawing it out. With a wider flux gap, the generated voltage would be lower and the 2300 RPM might be attained. A 42 volt supply would probably do it, too.
   At 2300 RPM with the motor mounted on a stand and clamped to the saw table, everything seemed quite solid and there was no serious vibration or other problem. In addition to voltage, it generated a feeling of confidence that all was as it should be.
   Following these tests, I ran the motor with no load at the same RPMs (except 2300) to find the idle currents. Perhaps only the 840 and 1040 RPM readings are reliably precise.

Generated VDC (diode drop corrected: +.765V)
Running, No Load: DC Amps @ same RPM
roughly 1.5
roughly 2.8

Mechanical Torque Converter (MTC) Project

   I had little time to work on this in November, but some time to consider it. In the first rendition of the 'escapement' torque converter, the original light escapements had provided little thrust, but when just 100g was added to each one, their force was strengthened by a surprising amount. I had made the 25 'fan fold' drum rim, and spots for six escapements, but tried out just three. However, these ratios, 25 to 3 or 6, meant each escapement provided its pulse of force at a different time, and (obviously) non-symmetrically around the wheel. But part of the idea is to provide enough force at one moment to overcome any slack, inertia and static friction and start the wheel turning and the car moving. It is then assumed that the next force pulse will come before it has a chance to stop moving again, so it will accelerate. So having one escapement at a time provide a push probably wasn't a very good idea.
   My next idea was to change the ring of teeth to a different shape that would increase the inner diameter and allow 27 teeth that would fit the same escapements. That way, three teeth at 120º intervals would 'strike' at the same time. Although the average force would be the same, the force at that moment would be triple. I didn't get around to making this new ring. Then I realized that it would put the teeth farther from the center, which would mean drilling yet another six escapement mounting holes, probably almost overlapping the previous set and maybe at or beyond the outer rim of the rotor. And theoretically the escapements would be just a slightly different shape for the new radius, though they were probably close enough.
   While considering these complications, it dawned on me that another way achieve the same thing - even better - would be to change from six escapements to five, evenly spaced (72º) around the rim, and the spacing of those would match the current tooth spacing and all five would act in unison for a quintuple pulse of torque. For a while I didn't want to drill the escapement mounting holes, as some of them would have to be behind magnets, and the drill and threading tap would surely poke through and damage the magnets. But on further consideration... too bad for the magnets - it's a prototype! I'll spray some paint in the holes to seal the magnet alloy from the air again. As it happens, I even made just five of the previously intended six escapements.
   Hopefully I can complete this and try it out in December. There would seem to be reasonable hopes that this version will move the car well beyond the "just barely" of the last couple of tests (I can push it harder by hand)... which are now at least a couple of months ago.

Quieter, Reproducible Blades for Windplants

   Denmark gets 20% of its electricity from wind and exports power surplus to its needs to Europe on windy days. Even here in BC several windplant sites have been or are being set up.
    I've heard that a prime objection of residents to nearby windplants is noise, and a small one deployed near here by the city certainly makes enough air/propeller noise. The shape of windplant blades is 'optimized' in cross section for lift, and smooth, which one would think would provide the least noise and the most efficient operation... but does it?

   There are indications of at least two ways to improve the seemingly "no brainer" optimum designs. First, research on the notched flukes of humpback whales has pointed the way to quieter, more efficient cooling fan blades incorporating notches and ripples. (This was on TV a while back.) Second, textured surfaces (eg, the holes in golf balls) can have less air resistance than smooth ones and may be quieter.
   These ideas could both be tried and likely used to advantage. A windplant blade akin to designs of fan blades with notches or ripples might be quieter, and I think a blade surface texture, especially such as "pebbly" (I visualize sand size and somewhat larger rounded surface bumps - varying random sizes) would also help air slip past and quiet them. Any such measures that made a blade quieter would probably also decrease wind resistance and make it more efficient as well.

   As a production idea, tough polypropylene cloth or strap + polyester or epoxy (resin) could make lighter, stronger blades than fibreglass + polyester or epoxy, would be easy to reproduce in a mold or jig (even with odd notched shapes), and would be more durable than wood. An opaque "sand pebbles" surface layer might be applied afterwards (to also provide UV protection), or the surface texture might be incorporated into the mold.
   Perhaps a single blade could be hand made, and a mold cast from it for duplication.

First drawing of the design ideas.
(Date of conception is approximate - I added the text to the pictures later, on the 20th.)

   In spite of the intriguing possibilities I don't propose to add this to my already extensive project list any time soon. However, if anyone in Victoria would like to pursue it I would be pleased to help give direction and support to the effort, including  typical blade profiles and overall windplant designs, web links and perhaps a lawnmower motor. There currently seem to be very few sources for windplant blades on the web (I only found one), and none for smaller, eg lawn mower motor, sizes. (Much less doing what are probably improved designs!) I'm not sure if anyone has thought of polypropylene-polyester composite construction either - perhaps not a surprise when the main fibreglass/plastic store in town hadn't even heard of it!

Pulsejet Steel Plate Cutter Project

   I drilled out the telescoping holes, finished the floating ball bearing one-way valve, and cut out most of the parts.

Pulsejet 'exploded' view.
L to R: inverse conical exhaust tube, main combustion chamber, nut & sparkplug.
Lower, inset, from bottom, the one-way air intake: ball chamber & outer (air sealing) stop,
'floating' ball, notched (air passing) inner stop.

Assembled. Threaded exhaust (left end) allows for trying different size nozzles.

   Then I welded them. I burned a hole through the stainless tube by the spark plug nut, and pretty much burned the "spark plug seat" surface off the nut itself, so I cut the end off, got a new, thicker, nut, and re-did that. That almost finished the main body.

   Then I got busy with other things.

   Remaining items were:

* a means to make the spark plug spark. There are a couple of ways that this might be done. Tristan's motorbike engine yielded a spark plug transformer, which itself now needs a means to activate it... but a low voltage means.

* the propane connection. I picked up a .28mm nozzle with 6mm threads for attaching it... to something. (I lost it somewhere and had to get another one.) I also got two 20 pound propane tanks someone had tossed in the bushes on the boulevard, and a hose to connect the tank to the nozzle. The outer end was a fitting with an outside thread that was fortuitously the right size inside to thread for the 6mm nozzle.

* Some handle or way to hold it - pulsejets get red hot and more in operation. One idea is to make a long handle that was also the propane intake duct. Another is to put a long tube over the air intake - but as that's threaded on, it might turn loose.

Electric Outboard Motor Project

   The original guzzleine engine had worked with the prop speed reduced about 2.75 to 1, because it ran at a high RPM (eg, 5000?) to get its 7.5 HP. The Electric Hubcap is a low RPM motor and at its maximum RPM (1750) the prop was spinning too slowly, because of the now counterproductive gear reduction. It's like having your car stuck in first gear - there's power to spare for a truck, yet you can't get going very fast.

   So the first task was to fit a new propeller with a steeper pitch. The steeper pitch prop moves more water at a lower RPM. I found an "11 pitch" Johnson propeller of the same diameter that looked like I could make it fit, propellers for the Honda evidently being out of production. I was told the standard Honda one was "10 pitch", but the new one appears considerably steeper - so I suspect my original was actually shallower, perhaps 8 or 9. It took a couple of days, but with some drilling, turning down and sawing slots in a shaft collar, turning a nylon "diameter matching fairing" (the prop's diameter was smaller than the outboard's) and other cludjing, it fit on.
   Next was another water barrel test on Saturday the 6th. With the first propeller, the outboard had drawn 29 amps at 36 volts and churned up the water "about so much" (for which the movie must be the record). With the new one, the current rose only to 30 amps - but it did seem to move noticeably more water. A little even slopped out of the bucket.

   After the water barrel my motor making workshop student Tristan Money and I took it down to Esquimalt Anglers boat launch and took it out in the sheltered harbor. On the 14' aluminum boat with two passengers and 140 pounds of batteries (3 "size 27"), it drew 33 amps at about 34-1/2 volts (gradually dropping to 32 amps at 33 volts over the course of several short outings) at the motor's maximum RPM, 1740. That's 1100 watts, under 1.5 HP, and naturally with the motor loafing along like that the boat didn't get up on a plane or even make much wake. The prop would have been turning only about 630 RPM. But it did run well and certainly moved well above trolling speed. The heat sink bars on the motor controller stayed completely cold to the touch. Once I stopped on the water, opened the hood and felt the motor coils. They were perhaps body temperature.
   Tristan shot some videos with my camera, but not being familiar with its idiosyncrasies, he didn't realize it was set to "close-ups" and all the footage was quite blurry. (A couple of short clips are nevertheless on www.YouTube.com - search for "Electric Hubcap Outboard" there.)

Images from the movie clips.
The motor controller (I'm turning the speed control) was clamped to the seat.
Another movie details the insides of the motor.

   The owner of the induction motor outboard said they also had had problems getting enough RPM out of the propeller. It seems they doubled the speed of the motor in the induction motor frequency drive controller - from 0 to 60 Hz (normal) to 0 to 120 Hz (well above the motor's rated RPM specs, I'm sure). They also placed the motor behind the drive shaft, turning the shaft with a chain drive. The chain sprocket gear sizes also pretty much doubled the RPM again, so they had 4 x RPM increase to counteract the built-in decrease (Yamaha 9.9 HP). Neither of these is an option with the PM EH motor, driving the shaft directly.
   I looked at the gears of the outboard and saw nothing that looked like it could increase the prop speed short of making different gears to fit. Later I thought: perhaps I could, with more or less difficulty (probably more - the two shafts were quite dissimilar), reverse the gears. Then the ratio would be 1 to 2.75 instead of 2.75 to 1. At only 1000 RPM on the motor, the prop would be turning 2750 RPM... assuming the motor could get it going that fast! That would be like being stuck in fourth gear instead of first gear... or maybe fifth, sixth or seventh? If it isn't geared too far in the other direction, perhaps the speedboat  is in there yet. But it would be a difficult job.

   I guess I should just look at the silver lining: it's solid, quiet, and will get you out to... well, I'm going to say it (hey, it's my newsletter... you can skim over it!) ...to where I've been hearing the RCMP now lie in wait to levy big fines against those who haven't kept abreast of the latest regulations, bought the latest required gear, acquired a "certificate of competency" to run anything over 0.0 HP, and paid admittance fees to trespass on private BC public waters. This should chase most casual boaters off the water, which may gradually kill pleasure boating entirely the same way provincial park use has dropped off with pricey "day use parking fees". I think our governments want everybody to just sit at home and watch TV.
   I probably don't have the details entirely right, but just the thought of probably being pulled over by the RCMP almost as soon as I get out on the water would have dissuaded me from doing an outboard if I hadn't already invested a lot of time in the project. That I am not alone in disliking such "big brother" overregulation is attested to by the fact that, when only boats with outboards over 10 HP had to be registered - just the boat registered - by far the most popular outboard size for use on lakes was 9.9 HP. The government could have inculcated safety consciousness and safer boating practices as well, and more easily and cheaply, with educational pamphlets placed in marine and water sport related stores, without creating a whole new odious system of regulation and enforcement. It seems our governments try to protect people from themselves by attempting to control them, passing restrictive laws that impact on everyone in unforeseen - and sometimes unseen - negative ways, to control some minor problem caused by a few, instead of simply pointing out better ways.
   Our "tolerant society" of the latter half of the 20th century has become intolerant to the point where using one's "freedom of speech" plus honestly holding "politically incorrect" beliefs can land you in jail, "zero tolerance" is even a slogan, and official advice "Just say no." even shuns common courtesy - what happened to "No thank you."? It seems to me we are gradually losing rights and freedoms in every direction and edging towards dictatorship.

   Naturally I wanted to test the EH outboard against the induction motor outboard, on the same boat drawing the same amount of power from the batteries. The tests were delayed by preoccupations of the owner of said boat and outboard, the need to scrub heavy marine growth off the unused boat, not to mention by very cold (for the coast: -7ºc) November weather.

   A single set of three 'deep cycle' lead-acid batteries should last a couple of hours at the low 30 amps current draw. Two sets (280 pounds) might last five hours.
   Again I reflect on how lack of - and indeed sabotage of - commercialization of economical batteries has been sabotaging economical electric transport. For more on that, read the next section, Battery News Behind the Scenes.

   If I get batteries using the NiMn-Mn chemistry working, you'd lift a cheap maybe 30-40 pound battery into the boat for the trip instead of a 30-40 pound gas tank - and on the ocean another one for a spare for an extra 5 hours reserve.

Turquoise Battery Project

Carbon Sheets

   The battery project seems to hinge now on creating impervious, conductive carbon sheets that won't deteriorate, to connect the entire outer surface of the positive electrode briquette to the positive terminal.

   The traditional carbon electrode uses "pitch", evidently a mish-mash of higher numbered hydrocarbons. In a store one day, I saw "Diesel Kleen Cetane Boost". This is mainly hexadecane, C16H34. (or more precisely, an alkane - a rather heavy, acyclic, saturated hydrocarbon string: CH3CH2CH2CH2CH2CH2CH2CH2CH2CH2CH2CH2CH2CH2CH2CH3.) It came as a black liquid in a plastic bottle. (It may, or may not, be some additives that make it a liquid.) Although pitch is quite thick and the "cetane boost" was a liquid, pitch is also a liquid - simply a very, very viscous one at room temperature. It needs to be heated in hot water on the stove to become workable: hexadecane (evidently) doesn't. But the lighter alkanes such as methane (CH4), ethane (C2H6) and propane (C3H8) are gaseous at Earth's temperatures.

   Perhaps it would do the same job - without the heating? I mixed a small quantity with graphite powder until it was a thick paste with quite a low electrical resistance - x 10s of ohms; upper x 1s with the electrodes close together. Then I rolled it out into a sheet between two pieces of polyethylene with a "mini rolling pin" - a yellow pencil crayon that was handy. (Another color might work okay, too.)

   Oddly, the polyethylene, inert to so many things, started curling up. So I transferred it to an aluminum sheet, and left it to see if it would dry and harden or otherwise change. The smell indicated something was evaporating. It would have to be considerably dryer before I could try to compact it.
   The next morning, the sheet had evidently dried considerably and was cracked in several places. It wasn't hard, but it had become quite crumbly.
   I waited a couple more days, then tried to compact it. It remained very crumbly. Then I heated it to 450ºF in the oven for 75 minutes. No noticable change.
   It was just strong enough to get some ohmmeter readings: as low as 2.1 ohms with the leeds close together, and under 4 ohms anywhere to anywhere. Interestingly, the same readings were attained whether the sheet was on aluminum or plastic, which would tend to indicate that much of the resistance is at the contact points rather than within the sheet. But pushing on the meter leeds harder would only have broken up the sheet.

   I think the hexadecane may be a good ingredient to add to the pitch as part of a mix, but it doesn't have the right characteristics by itself.

Battery News Behind the Scenes & Editorial

   I continue to hear things about the lamentable state of battery commercialization today and to give them thought. In battery news:

* I heard Michael Moore tell Larry King that 1% of Americans "earned" (my quotes) 25% of the nation's total income last year. I'll tie that in with that university researchers found a way to produce nanotubes that magnified the power density of lithium batteries tenfold.
   The potential value of this work to electric transport is zero: It was patented, and Chevron has bought the patent, another corpse to add to their graveyard of murdered battery technologies. No one will be able to manufacture transport size batteries using the advantageous new technology - they'd be sued into the ground. The 1% control the economy through such means.

* In India, Ni-Fe dry cells, made possible by research in 2004 at Bangalore University, are now being produced in research quantities and tested in applications such as telephone exchanges. If they perform well, more than one interest has ideas about producing them to sell. (We may well see attempts to suppress the new aspects of this technology as well.)

* Europe may adopt Ni-Fe batteries as a standard to replace Ni-Cd, owing to the toxicity of cadmium (and lead) and Chevron's suppression of the green Ni-MH technology. The Ni-Fe's will last much longer than Ni-Cd's, too.

* Ni-Fe battery standards are being reviewed and reinstated by IEEE. Evidently Exide lead-acid battery company long ago convinced IEEE to drop the standards. With standards in place, Ni-Fe batteries may be purchased by telephone companies and others who have large needs for reliable storage, which could increase quantities manufactured and reduce prices. Even the original Ni-Fe cells Edison developed were substantially better than lead-acid for electric transportation, and "the 1%" played a substantial role in getting them off the market.

   And here's yours truly's inevitable accompanying editorial:

   Regarding the first item: if Chevron (more correctly, the unscrupulous owners of big oil and transport in general via any of their subject companies) can seize enough lithium battery technology patents they may be able to stop manufacture of EV size lithiums altogether. (Then they'll tell us that lithium is too scarce to squander on batteries.) How do these people sleep at night, knowing full well they're working directly against what the world wants and needs? What does it profit a man to gain fortunes of money and lose his own soul?
   It shows how patenting new transportation technology only guarantees it can't be used. How long will these greasy reprobates continue to bleed us all white and sabotage non-petroleum transport before people rise up and demand the changes to the patent system needed to rein them in - and, oh by the way, perchance to allow inventors to get paid in some way commensurate to their contribution to society rather than to starve - the supposed but non-functional raison d'être of the whole patent system in the first place?

   Big oil's sabotage of superior battery technologies includes not only buying up patents and companies that try to start producing better batteries, and leaving out the sodium sulfate from lead-acid batteries in order that they'll corrode too quickly to consider as being truly viable for electric transport, but also strategically spreading propaganda that blows the weak points of the best technologies out of all proportion while staying silent on the good points and on the fact that lead-acid has many worse features. As there is no major organized group with an interest - or even the knowledge - to counteract these gross distortions of the facts - in effect lies - they get accepted as being true in everyones' mind, steering research and potential manufacture away.
   I'll single out two much-distributed statements about Ni-Fe for scrutiny:
 "Nickel-iron batteries have a very high level of self-discharge."
 "Nickel-iron batteries have a low charging efficiency."

   Without qualifying context these over-emphasized statements effectively give the impression that the speaker must be experienced in the matter and know what he's talking about, and that after a few days, at best, the charged battery must surely have little useful energy left, and that it is very wasteful of electricity to recharge it. Is that why Ford and Edison wanted to use it for their great new mass production electric car in 1914? Is that why somebody went to a lot of trouble and risk to burn all Edison's Ni-Fe battery factory buildings down? Rational qualifications and comparisons -- that Ni-Fe's charge dissipates only somewhat more rapidly over the weeks than lead-acid's does, and that it's only marginally less efficient to recharge -- are never made. Furthermore, the lead-acid battery will start to corrode unless continuously maintained fully charged (it's under charge, using energy, all the time when your petroleum car is running!) and it should only be discharged 50-60%, whereas the nickel-iron recharges good as new whenever it's next wanted (even years later) and it can be discharged down to 'empty' with no worries. Also not mentioned: a battery is most likely to be used within a day or two of charging rather than weeks later anyway - self discharge is a very minor issue.
   The overriding truth is that nickel-iron (Ni-Fe) batteries, notwithstanding certain weaknesses, are excellent batteries that last for decades -- and they could be much better and cheaper if the designs and manufacturing techniques were updated. New Ni-Fe developments virtually ceased when Jungner and Edison stopped developing them a century ago until the above mentioned dry cell work in 2004 in India.

   The industry has similarly been kicking away to undermine nickel-metal hydride, notwithstanding its recent excellent success - nay, because of its success - at running electric cars in California:
 "Lithium is the 'holy grail' of battery research." (We smell money! Forget everything less costly than scarce lithium - we've bought the mines!)
 "Yes, those Ni-MH batteries worked great in the EV-1, but battery technology has moved on." (Lithiums probably won't last 1/4 as long, and we can use the high price as an excuse for not making and selling economical electric cars.)
   Actions speak louder than words: Is Chevron's hoarding (via its puppet Cobasys) of all the patents it can lay its hands on related to this "worthless", "obsolete", "too expensive", battery technology, and their point-blank refusal to allow it (in useful car battery sizes) to be manufactured, or to be imported from countries where they can't control the manufacturers, consistent with the 'casually' publicized statements?

   Ni-MH batteries are probably 20-30% better than Ni-Fe overall and evidently of similar super long life. They probably cost somewhat - but only somewhat - more than 20-30% more to produce. More important, having the technology and being able to manufacture and sell it - economically - with a slimy oil company sitting on a stack of relevant patents, are two different things. Even for small dry cells prices seem to remain fixed at or above 1000 $/KWH even in bulk - too high to hook up hundreds or thousands of them economically for an electric car - instead of dropping below $500 and probably well below.

   I keep seeing lithium AA cells in stores now. The whole idea seems absurd: the dubious advantage of lithium is that its high reaction voltage of three volts (theoretically) allows greater energy density. What is the point to making it into 1.5 volt batteries? The energy density advantage is wasted, and they cost more to make and won't last as long as green Ni-MH cells. I may have a suspicious mind, but this would seem likely to be a strategy to quietly phase out the superior, very long life, green Ni-MH batteries. to gradually get them off the shelves, off the market, while spreading slander about them that will poison Ni-MH in peoples' minds and prevent a revolt. Then before the prohibiting patents all expire everyone will have forgotten there ever could have been an economical alternative available, to (deliberately) crappy lead-acid and to (pricey) lithium.
   Many of them also seem to be single use, non-rechargeable. If lithium is so scarce, why is it being squandered in single use batteries? Something's fishy there. And for all the boasting about "lasting longer" on the packages, they never seem to actually say how many amp-hours they hold. I bet it's less energy than a single charge of a typical Ni-MH cell!

Victoria BC