Turquoise Energy Ltd. News #33
Victoria BC
Copyright 2010 Craig Carmichael - November 2nd 2010
http://www.TurquoiseEnergy.com
= http://www.ElectricHubcap.com

Contents/Highlights:

October in Brief (summary)
  * Feature: The Electric Hubcap Outboard Motor: a practical, working unit has been made to demonstrate that the "EH" motor is not only a simple, robust 5+ HP "DIY" motor for making without a motor factory, but that it performs as well as the best or better.

Electric Hubcap System & Motor Building Workshops
  * Welding hubs to rotors: there's special welding rods for welding cast steel parts... or use stainless steel rods
  * First student is progressing well on motor
  * Hall effect magnet sensors: a mounting system
  * First CNC hole drilling of a stator: to automate the job, improve precision
  * Wider flux gap (.57") greatly improves motor performance
  * Second motor controller made, for testing motors in the shop (& for outboard) - has improvements

Sodium Sulfate Longevity Additive for Lead-Acid Batteries: (a quick correction supplement - no further report)
  * A bad test leed on the multimeter used to determine load currents (recently discovered and replaced) gave skewed readings, which results were propagated throughout the tests. New DC current clamp corrects figures.
  * The "5 amp" (car headlights) load actually draws 6 amps at 12.0 volts; the "10 amp" load is actually 12 amps; the "25 amp" load is actually 35 amps; all headlights on reads 51 amps instead of 40.
  * This means that all the batteries tested throughout the project actually performed somewhat better and had somewhat more amp-hours than indicated in the newsletter reports.

Torque Converter Project
  * a 3rd place that mass might count... on the motor rotor

Wave Power Project - solar green energy systems
  * No wind, no waves, not ready: no tests
  * Direct Solar power installations - Sun, acres of Mirrors, and Steam Turbines
  * Backyard solar power?: mirrors and Stirling engines, or DSSC solar panels (earning over 360 $/month with net metering).

Pulsejet Steel Plate Cutter Project
  * Parts acquisition
  * The air intake: ball bearing check valve
  * shape: inverse conical?

Electric Outboard Motor Project
  * Purposes: demonstrate another application of EH motor, demonstrate its superiority
  * Outboard Disassembly - frustrating process
  * Motor installation
  * New Motor Controller
  * Initial testing - gear systems burn energy (duh!)
  * Magnet glue failure; re-done 'epoxy encapsulated' magnet rotor virtually certain to hold
  * Water pail test: runs great, but light load coupling owing to ~4:1 propeller speed reduction
  * Bought steeper pitch propeller (more push at a given RPM)

Turquoise Battery Project
   * A couple of carbon/graphite experiments

Newsletters Index/Highlights:
http://www.TurquoiseEnergy.com/TENewslettersIndex.html

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:  http://www.TurquoiseEnergy.com/

October in Brief

   I decided to make an electric outboard motor using the idle October 2008 prototype motor, to prove (or disprove - fat chance!) that the EH motor would produce significantly more thrust than a similar power outboard made with an induction motor, and to demonstrate another practical use for the motors besides car drives. Put simply, I think the Electric Hubcap motor will prove itself to be as good as or better than the best motors out there anywhere, as well as being easy to make without a special motor factory.
   My original plan was simply to run the motor at 24 volts, where it would match the 2 HP induction motor outboard I wanted to compare it against. But as the project proceeded, I started to think "Why limit it to 24 volts?" At 36 volts, it would be around 5 HP. Since it's said for marine things that one HP electric equals 2.5 HP gas, and since the extra efficiency might make that more like 3 to 1, could it be I would have the equivalent of a 15 HP gas outboard? That would get my 14 foot aluminum boat moving faster than ever before - not just, "see, it's electric", but real, speedboat performance! Why not?
   I ended up having to strip my own "Honda 75" 7.5 HP outboard. I thought that it might take a week or so to make, but vexingly it took that long than that just to disassemble, strip and un-seize, and remove the gas engine from the diabolical, corroded outboard, without even getting to the electrical part. That part proceeded about as planned, and the motor was running on the 16th. It proved to be an excellent fit, the motor fitting under the hood with little to spare at front and sides; plenty at the top. Most smaller outboards would have been too narrow - and my later motors would have been a bit too fat, too.
   However, various problems (seized steering, axle a bit too short, broken hub welds, later decision to use external motor controller instead of internal, redoing entire magnet rotor after a magnet flew off, repainting... see detailed report for gory details) conspired to keep me working on it until the end of the month, each day thinking it was almost done. But I learned more about getting better performance out of these motors, in particular that the flux gap should be not 1/4", not 3/8", but over 1/2", and I came up with improvements to make the magnets more secure and for the magnet sensor mountings. I can't help but think the most recent prototype torque converter would have moved the car nicely in September if the motor had simply had the higher RPM that the wider flux gap seems to give it.

   When the motor had nice fresh paint and fancy labels on the 24th it inspired confidence that it just had to work great, too - how fickle we humans are! That's when the magnet flew off.


The painted Electric Hubcap outboard

   After removing all the magnets from the rotor, I used a liquid epoxy resin for glue and put the first six magnets on, and tried it with just those at a wider .57" gap. The performance was amazing - idle current at 800 RPM dropped from 20 to 10 amps, a great improvement in efficiency, and full speed RPM went up from 850 to 1460 (33 amps). And that was still with just 24 volts instead of 36. After I put the rest of the magnets on, I gave them a thin, and then a thicker, coat of epoxy, more or less encapsulating them in epoxy plastic. This gives much more confidence they can't rip off than simply gluing them on the bottom.

   Thus, while other things mostly related to motors and controllers proceeded apace, October was largely consumed by this one project. In it I learned things of great value about the operating parameters of the motors I created (especially the flux gap) and better techniques for making them, and will of course pass these on in the motor building manual. I pressed hard daily to finish the construction as I didn't want one more project (supposed to be a couple of weeks) dragging itself out into months.


Finished at last! Churning up water on October 30th.
It works great, but owing to the geared-down propeller (almost 4:1 inside the leg),
the motor is "loafing along" at 1+ HP (34v, 29a) near its maximum RPM.
Options: increase the voltage (from 36v to 42v) to increase the top RPM;
found a propeller with a steeper pitch, to push harder at a lower RPM.

   Sea trials will have to wait until November. (If the winds of November come early, I'll test the wave power instead - no need to emulate the Edmund Fitzgerald.) However, on the 31st I removed the flat deck from the trailer and put on the boat guide rails.


Under the hood: the 5 or so HP Electric Hubcap outboard (less magnet rotor on top).
It just fits within the Honda 75 outboard hood.
By all accounts it should have way more push than when it was a 7.5 HP gas outboard!
Small chassis motor controller *was* to fit in behind motor - na, too much extra work!

   Concurrently, I made a new (and improved) motor controller, also completed and working on the 16th - it was with this that I ran the outboard. I started it quite a few months ago when I got some polyimide ("Kapton") tape for heat sink insulation, but I didn't get around to ordering more MOSFET transistors (had 8; needed 12) and finishing it until this month. This unit is for testing motors in the shop, but I'll also use it onboard for the outboard.


New motor controller for testing motors in the shop.
(...Now if only the chip wasn't being discontinued, forcing redesign of the logic board.)

   I bought parts (except circuit boards so far) for the three remaining MC33033 controller logic boards. After those, with the MC33033 now said to be obsolete, the newer A3938 controllers will need new circuit designs.

   My first workshop student has done several work sessions making the parts for his Electric Hubcap motor. It's progressing well - the bearing hubs and the stator coils are made, and the stator holes are drilled. It is most helpful that he's very talented and has considerable fabrication experience. He had an idea for tracing out the lengths on paper to break off each row of nails for the coil core, and this template seems to speed things up somewhat. Making the cores is the most tedious part of making the motor.



Motor under construction by workshop student:
Hubs turned, coils made (3 shown), stator holes drilled...

   For my part, I've been revising the EH motor making manual (needs much more work) and trying to work out improved systems for doing things. I calculated the exact positions for drilling the holes in the stator and in the rotor relative to the center so they could be drilled by CNC machine. Then I made a solid mount (from a 12" diameter finned rotor - that's just big enough to C-clamp onto the CNC's holding grid.) to hold the stators rigid on the CNC drill machine, and two plastic plugs to stick in the center hole to center the machine on the rotors - one for 'blank' disks and one for those with with center hubs (smaller center holes).
   I spent the day of the 29th with the owner of the CNC drill/router machine getting the "G-Code" drilling sequence program perfected and working through the various challenges that arose. But a rotor was drilled and it's ready to duplicate them! It can be the first custom part available for sale to help people who want to make the motors.


'6129' rotor: straight from a brake shop junk bin to first ever CNC drilled Electric Hubcap motor stator!


A small part of making an EH motor is painting the nail strips to insulate between laminations.
I did this job myself so the shop would be habitable during the workshop sessions.

   I got the wave power unit "essentially ready" to test in September, but calm weather in October up until the 23rd would have precluded testing it even if I had got a nice new planned load/test box completed, the trailer licensed, and the floats and gear loaded onto the trailer to be ready to take advantage of good winds if they struck. By then the electric outboard was about ready to go. I licensed the trailer and hoped for calm water for boating.

   I met a retired guy this month who said he used to work for General Motors in the 1960's-70's. Among other interesting things, he said the engineers there used to cry about all the patents, great ideas for better ways to do things, that were acquired by upper management and tossed into drawers, used only to prevent anyone else from making the product or technology available to the public. There's the scoop, straight from the 'front lines', from an actual employee! This is a symptom of why western civilization has become static and unprogressive in key areas, and our biggest problems go unsolved. Corrupt vested special interests are permitted to man the decision switches on the tracks to the future and send the trains around in circles, when progressive solutions are in fact all around us. (...it also shows how stupid the patent system is, that it can be and is routinely used to kill whole new technologies.)

   I have written tirades about the gangsters running the transportation industry and killing all but petroleum based transport, and also about the dysfunctionality of our still primitive governing institutions that allow such cancers to persist and dominate our society, and these are not without their point. But there is a great principle that shouldn't be overlooked: change begins with each individual. As we become spiritually liberated through faith, as we consider the entire human race to be one family under a spirit father and take a real interest in human and spiritual affairs, we gain broader and more discerning viewpoints. We become world and universe citizens, and we start to evolve a better world that will marginalize the effects of the machinations of these antisocial manipulative elements, by creating accountable social institutions that fill the power vacuums that they now seize and control. When people are working in co-operation instead of against each other, our planet's culture will grow unimaginably.
   The universes are evolving according to the all encompassing, infinite and inclusive plans of the infinite first source and center, the uncaused cause, often conceived by relationship as being our heavenly father or spirit parent. His mandate is that we evolve and become perfect - individuals, worlds, and larger administrative units. In universe terms, though they may work temporary harm and seem to prevail for a brief cosmic time, error and evil are self-correcting and sin and iniquity are mentally destabilizing and self-destructive.

Electric Hubcap System & Motor Building Workshops

   First, I may perhaps point the reader to the Electric Hubcap Outboard Project write-up (below). In putting this unit together I learned some valuable things about characteristics of the motors themselves, which are written up under that heading.
   The most important one is that I found .57" to be a much better flux gap than .35". The amps for a given RPM goes way down, and the maximum RPM goes way up. This last would be because the (more distant) supermagnets are generating less voltage back into the stator coils at any given RPM.
   I realize now I should have experimented with different gaps more, earlier. I imagine if I now increase the gap in the current car motor, efficiency and top RPM will also increase.

   Another important lesson was learned: that the "Epoxy Steel" glue I've been using loses its grip after a year or two. Flying magnets are dangerous. I've thought perhaps getting magnets that screw on as well as glue on might be advisable. And yet most of the magnets are epoxy coated and that coating doesn't come off, and other motors have magnets simply glued on. So I decided to try a liquid epoxy resin as glue on the outboard's motor. Once the magnets were on, I decided to paint another coat over the whole rotor around the magnets. This made them in fact epoxy encapsulated, though only thinly. Then I thought, why not make it a very thick coat? 'plasticize' the whole rotor? It's hard to imagine the epoxy under the magnets, and also going up the sides thickly, giving out in a couple of years!


Epoxy "plasticized" magnet rotor: will prove durable I trust!

   Last month I wrote that I welded the machined hubs onto the rotors with difficulty. Both are cast steel parts. Then someone chanced to say in some unrelated conversation something about "you can't weld to cast metal". This month I thought maybe some advice would be good and I took the welded parts into a welding store.
   It turns out that there's special welding rod for welding cast steel parts. Two places didn't seem to have any. The second place sold stainless steel welding rods by the pound, which I wanted a few of for the pulsejet project. He asked me what kind of stainless - seems there's enough difference between types of stainless steel to warrant different types of stainless steel welding rod. I told him it was the sort of tubing you'd find on sailboat railings. As I pulled some out of a box he held out to me, he said "Actually, these would be good rods for cast metal too, oddly enough." Alright, problem solved!-- the same stainless rod for everything! They only had it in 3/32" diameter size, and I'd never had much luck trying to weld with anything bigger than 5/64" rods before, but these worked great, and on the cast metal. These rods were titled on the receipt: AVE-E316L1725 Electrode, Stainless Steel, Avesta, E316L AC/DC. 2.5mm (3/32") x 3.63KG (8LB), AWS E316L-17, AVE-60610125. (Remember that when you're looking for them.)
   Stainless seems preferable for an outboard, and I was told the rods for cast metals were best run with a DC welder (mine's AC), whereas the stainless ones (per receipt) were AC/DC - good to know. There were many interesting hues in the finished welds - yellow, red, purple. Makes me nervous about the fumes during welding: yellow - cadmium?; red - copper?; purple - manganese?
   Later I was trying to pound a bearing race in and the welds broke. More precisely, the cast steel of the rotor broke where the welds were. There's no question that welding to cast metal isn't as strong as to rolled or extruded metal. For the repair I did what my neighbor suggested in the first place and welded it all the way around (well, more or less), not just some spots.

   Another improvement was a better defined system for mounting the three magnet position sensors. It uses three angle brackets bolted between three pairs of coils on the backside of the rotor, but with the vertical angles sticking up just outside the stator's edge to near the magnet rotor. (See picture of open EH outboard.) By bolting them on from the backside, they can be removed for servicing without disassembling the motor. Some sort of circuit board or PWB is bolted to a couple of #6 or #8 threaded holes drilled in that vertical leg, with the actual Hall effect sensor soldered to it and sticking though the top original hole in the bracket. Two of the motor mounting holes are shared by the brackets. Timing can be adjusted by bending the bracket left or right a bit (crescent wrench) until the sensor is dead center between two coils.

   And I did a technique for placement of the holes in the stator: a cardboard photocopy of the drawing is used as a template, with the center hub cut out of it with scissors. This is placed over the center hub of the actual rotor and taped in position so it can't rotate. Then a center punch is used to position the drill holes.


Rotor with hole center-punch template - a photocopy of the drawing


Workshop student's motor stator, holes per template
(Showing how coils & magnet sensor brackets fit,
with bracket & motor mounting bolts in the small 'corner' gaps between coils)

   Next, though, is CNC hole drilling to automate the process. I worked out the stator hole positions in 1/1000ths of an inch from the center. I couldn't find a drawing program that would rotate things around a center and give me the numbers, so it was oodles of sines, cosines and multiplies with a calculator. I used a 12" finned rotor as a hefty holder to hold the rotors completely rigid while they're being drilled, and made a centering ring that fit over the center hole. I worked with the guy with the CNC drill/router (homemade!) to tweak my "G-code" program, and to get the somewhat neglected machine working smoothly again. We drilled one stator, and more can now be done repeatedly, automatically. And then I expect we can expand it to the rest of the motor and motor controller pieces.
   Marketing "ready to use" parts could perhaps bring some income while making the job easier for those wishing to make their own motors - and without the potential liabilities of selling complete motors having a novel new - and therefore "untried" in the commercial sense - construction.

   I think that if four specific custom parts can be made available, building a motor will be reduced to "bolt it together" and "wire it up". Those parts are:
* Pre-turned bearing hubs
* Ready-made coils
* The three magnet sensors, mounted on angle brackets and wired - ready to install
* The pre-drilled & tapped stator (29 holes)

   Of these it could be said that the ready-drilled stator is the least valuable, because drilling the stator is the easiest of the above. One can always print out the drawing and tape it to the rotor for use as a template to center punch for the holes, then drill them manually accurately enough. Nevertheless the ready part represents a labour saving that isn't trivial.

   I put together a Turquoise Motor Controller for testing motors in the shop, completed on the 15th.  I especially needed it to test the EH Electric Outboard Motor I was making - and maybe to run it on boats. Initially it didn't work. I finally found not one but three main problems:
* One of the 12 MOSFETs was shorted to the heat sink through the "Kapton" polyimide insulating tape. It's tough, but it's very thin. (I *think* I sanded the aluminum bars quite smooth when I made them...) I suspect there may have been some little grain of metal present that punctured it when the transistor was bolted down. I simply added another layer of tape, since better solutions involved a lot more work. That seemed to fix it.
* I had put the two header sockets on the PCB board the wrong way around, so that when I plugged in the plugs, they were backwards. (That kind of negates the value of using polarized connectors!)
* A resistor was missing for the overcurrent protection, and it thought the current was always way too high. (Well, it was way over on an obscure corner of the circuit board, see...)
   These glitches solved, both the car motor and the outboard, newly ready, were run on the evening of the 16th, using two batteries for 24 volts.
   Then, since it was for shop use, I mounted an OFF-FWD-REV switch (FWD-REV direction arbitrary) and a volume control (to make the motors louder or quieter) on the top of the controller box itself, near the breaker switch.
   There's an improvement or two in this unit. Especially, the terminal blocks for the heavy 3-phase wires have been moved from the wiring box back cover onto the motor controller side, much reducing the chances of the vulnerable mosfet leeds getting yanked off during servicing. In company with this, their mounting block now has an angled base in order that screwdrivers are inserted at an angle -- thus the handles don't have to magically occupy the same space as the far wall of the controller. This will avoid much aggravation.


Mechanical Torque Converter (MTC) Project

   On one test the escapements were just light aluminum, and the torque was poor. When mass was added to the escapements, the torque increased surprisingly, verging on enough to move the car. But the light torque converter output drum was vibrating back and forth against the pins to the wheel. Last test, I tried adding a steel plate to the drum, but the results were inconclusive.
   There is however one more place where mass might count, and that's in the motor rotor, the input to the torque converter. It's already a beefy car disk brake rotor, but adding some weights might be a good experiment. If it's being slowed too much as the escapements hit the output drum, that could reduce the torque.

   In addition of course I must finish the experiments I had already planned: having six escapements instead of three (bound to be even noisier?), 27 teeth on the drum instead of 25 so that the three escapements all hit at once with a force that's balanced around the rim, the single 'full circle' escapement (maybe), and a larger diameter for more torque leverage with the heavy cast 15" wok.

Wave Power Project

   Wave power tests had to await strong winds from the southeast or southwest to bring waves to area boat launches. These were not forthcoming until late in the month, plus I was busy with the electric hubcap outboard and didn't get everything ready to go for when wind finally did strike.

Solar Powered Electricity Plant with flat mirror collection

(What does this have to do with wave power? Well, not much!)

   I have been advocating wave power on the west coast in lieu of the Peace River Site C dam, but there are other neglected clean alternatives too besides wind. A great example is power directly from the sun. Sunlight has about 100 watts per square foot or 1000 W/sq.meter, so over a few square meters, there's quite a lot of power.
   A large power site, such as one recently installed in Arizona or California (I'm going from memory here), consists of many acres of large ordinary mirrors mounted on swivel bases, all reflecting the sun onto a collection tower to heat water, to power a typical steam turbine system.
   A computer system aims all the mirrors, and a cleaning machine automatically goes up and down the rows spraying dust off the mirrors.
   The radiant power from the sun on the Earth's surface with no clouds is about 100 watts per square foot. Thus an area about 100 by 100 feet, directly facing the sun on a sunny day - a couple of city house lots - gets about a megawatt. Okay, we won't do that well in Canada in the winter. And the site will have to at least survive snow if not continue operating in it, perhaps feathering mirrors to vertical - not a concern in warmer climes. Say we get 23% on flat land in an area that gets lots of sunshine per year (probably not the west coast). Let's say we'll need an area about 208 by 208 feet (one acre) per megawatt. A 900 acre site, 1.4 square miles, or, more practically, the same overall area made up from a number of smaller facilities, would be required to obtain the same 900 MW power capacity as the proposed Site C river hydro. That's far, far less land than is to be flooded for the dam. And taking advantage of south facing slopes could increase winter effectiveness and reduce land use requirements.

   I can't undertake to estimate what the capital cost would be. But like wave power and unlike the giant river dam, it could be built in stages, each of which would start contributing power as soon as it was built. From the first installations (as well as taking advantage of suppliers, services and lessons available from existing installations elsewhere) would come the practical experience and infrastructure to create cheaper, better additional sites adapted to our conditions.
   Large capacity sites are considerable projects that will be difficult of realization unless the government is on side (although perhaps not as impossible as with wave or river power. Hah! - Try and imagine Site C dam going ahead via BC's "private power producer" program with a few million dollars from the ICE fund and no other government support or "okays"!)

Backyard Solar Electricity?

   At a glance small sites might not seem very practical - you won't find steam turbines or computer controlled 5' x 7' mirror swivel mount systems at Home Depot. But talented hackers could create - or perhaps even produce for sale - smaller sites using Stirling engines instead of steam turbines to drive a generator, if a system can be worked out for tracking the sun with a parabolic dish or relatively smaller flat mirrors. A system of "n" individual mirrors in any flat field, throwing 10 x 10 meters of sunlight at a Stirling engine on a stand, might produce 30 or 40 kilowatts (peak) - a sizable 'reverse metering' installation for selling power to the grid from home and repaying (at only 5 cents/KWH) over $360 per month in a sunny area. This is at least double the power of solar photovoltaic collectors and the components could be cheap in principle.
   Such a project would ideally include at least a mechanical designer to do the Stirling engine and an electronics designer to do the mirror tracking system with (as I see it) microcontrollers on each mirror, a central override to aim the mirrors away (for servicing) and stepper motors. Individual equatorial-mount mirror stands might point due south and also have an arm pointed at the stirling engine to indicate to the microcontroller where to aim the sunlight, making setup adjustments trivial. Seasonal 'tilt' adjustments could possibly be manual. The rest is 'regular engineering'. There is information on Stirling engines in Wikipedia, various plans for making them on the web, and an aluminum wok lid solar powered Stirling engine video - among others - is on YouTube. [http://www.youtube.com/watch?v=7Q4UENGN_Yk]

   But I gather that, theory aside, making a larger, efficient Stirling engine is a considerable project. It could be argued that photovoltaic panels might be more practical. In particular, lower cost DSSC solar panels that might be almost 20% efficient could be made with my nanocrystalline ceramic rear reflectors, which would be heading towards what might realistically be achieved by small Stirling engines, and without the moving parts. Depending how much sunlight the panels could actually absorb without overheating, they could have angled mirrors at top, bottom and sides to reflect more light onto them, and if desired they could follow the sun on equatorial stands in place of the mirrors, to collect full sunlight over much of the day. And since each one would face the same way, a single sun tacking system mechanically linked to many panels would be practical.
   If they can withstand a whole pile of light, they could be at the focus of the mirrors in an arrangement similar to the steam turbine or Stirling engine, further reducing the number of collectors needed.

   (This makes me want to get back to the DSSC solar cells project, and make actual working cells rather than just the nanocrystalline rear reflectors for them. Mass-produced, they may have much more real energy potential than I thought!)

Pulsejet Steel Plate Cutter Project

   I got some pieces of stainless steel tube from from the stainless steel fanatic next door. I got three virtually telescoping sizes, and a piece of a .303 rifle barrel. It seemed to me to be a good size for the output jet. (I could also have had a piece of .22 rifle barrel - perhaps that would take less "kerf" from the steel?) I also got stainless steel welding rod and some steel balls - ball bearings.
   I thought it should work better with one moving part: an air intake one-way valve to keep it from firing flame out the air intake port when the fuel burned. I'd rather have the entire jet concentrated against the steel plate that's being cut. The V1 'doodlebugs' had some sort of flaps, but the hobby pulsejets don't. I was told people had tried making flaps, but the reliability was poor. (It may have been poor in the doodlebugs, too, but they each flew only one flight!)
   My idea is, instead of a flap, to use a ball bearing ball that free-floats in a small chamber. It gets shoved against the air intake pipe by the expanding air when the jet fires, closing the pipe off. The 45º slope of the seat is such that it would be hard for the ball to jam. At the other, inner, end of the chamber are air slots so that when the ball is pushed to that end it can't impede the incoming air. I drilled out this cavity from the rifle barrel piece and threaded the inner end on the last day of October. The inner piece with the slots, outside threaded, will be next. That will be welded to (or threaded into) the body of the pulsejet torch. (intake pipe: ~4/16" I.D.; ball: 5/16"; chamber: 6/16" I.D.; threads: 7/16" NC.)
   Materials acquisition, the concept for the one-way ball and an overall shape, and the making of one side of the ball chamber were the only progress on the project during October.

   The jet wave pulses travel at the speed of sound, and ideally, I think the entire chamber would be a tuned, inverse conical bore like a recorder (the musical flute), perhaps about flautino/sopranino size. This idea was supported by a set of pulsejet plans e-mailed to me in which the main part of the combustion chamber had that inverse cone shape.
   The "Supercorder" alto recorders I developed a few years ago sound fabulous, but I'm not about to make a "flute bore reamer" capable of milling stainless steel. So I decided to just use the two telescoping tube sizes, and then "telescoping" drill bits of progressively smaller size, deeper and deeper into the rifle barrel. This may require finding a couple of long, thin drill bits.

Electric Outboard Motor Project

   Making an Electric Hubcap outboard quickly became October's main preoccupation. The project's main goals are:

1. To compare the performance of an Electric Hubcap axial flux permanent magnet motor to a typical radial flux induction motor (an existing electric outboard motor) consuming identical watts of electricity. Even once the motors are running cars, it will be difficult to make such a direct demonstration, since it would be impractical to make an induction motor version of the car drive.
    I'm expecting a good increase in thrust. In fact, I suspect that virtually no other motor will match the EH's performance except maybe some other axial flux supermagnet motor. This project is the best way to demonstrate that.
   However, since the induction motor outboard is 24 volts, 2 HP, the EH outboard will be run with the same limiting parameters, whereas it should be capable of about 5 HP at 36 volts.

2. To demonstrate a practical use of the EH motor other than as a car drive.

3. I also hope to find that the thrust from the 5 HP Electric Hubcap outboard is noticeably greater than that of the 7.5 HP gasoline engine that previously powered this same outboard. That engine could just get my 14 foot aluminum boat up on a soft plane with one person on board. If the electric hubcap motor can do the same thing with one person and 140 or more pounds of lead batteries on board instead of a 20 pound gas tank, it's better. From what's been said in marine circles about 1 HP electric being equivalent to 2.5 HP gas and the Electric Hubcap being in the most efficient (axial flux) branch of the most efficient family of electric motor, I'm optimistic it will do even better than that and provide "speedboat" performance, equivalent to around 15 HP gas.


At last, almost ready to roll... er, fly... er, swim.
Movie here: Under the Hood: Inside the Electric Hubcap Outboard

   I spent some fruitless hours searching for a scrap outboard to use for this project. One recycler said fall was the wrong time of year - outboards get tossed out in the spring. An outboard shop said the ones they'd been seeing were from the 1960's and 1970's, looked awful, and were going straight into the scrap bin -- and that that had been emptied just two days ago. Another outboard dealer had a whole storage room full of dead outboards for parts, but wanted $200-$250 for any of them. I passed.
   So finally I decided to dismount the gas engine from the top of my 1976 Honda 7.5 HP, trusting that it could be put back on later. It seemed about the right size. I hadn't run it in about 15 years anyway! It turned out to be much more complex than I'd envisioned, the engine extending into the outboard leg in a mechanical nightmare arrangement instead of simply being mounted on top like on a gas lawnmower, and soon I realized I'd disassembled so many things that I'd probably never want to try to put it all back together again. Scratch one perfectly good gas outboard!
   Then several bolts broke off instead of unscrewing, including some critical ones for the water pump. And the drive shaft was pried out of the gear case through the water pump to get the leg apart, instead of separating at either of two points where it should have slipped apart. These seized "slip-on" joints had to be heated red hot and pounded loose later. The drive shaft got much bent up in the process, so it had to be carefully pounded more or less straight again.

   Overall, it took most of a week to disassemble and clean up the diabolical and corroded unit - my vision of the total time it might take to have it initially running - without even starting on the electric motor part! On the bright side, it's a great fit, in an outboard leg of about the right size and power capacity. Most smaller outboard hoods would have been too narrow for the EH motor.

   The electric part went easier. I mounted the motor stator in a day, making an "origami" single piece main mounting bracket from strap steel, that used existing bolt holes on both the outboard and the stator. And I turned a hub from my supply of "hub blanks" -- 1-1/2" cast threaded pipe couplings -- and welded it to the (cast) stator disk with stainless steel welding rod, a recommended type for welding cast parts. I didn't worry about how I might seal the bearings -- none of the trailer type bearing seals were going to fit. This motor won't be collecting any road dust, but salt water may be another matter.

   The next day I picked a scrap (car part) 1"D x 9"L splined shaft  for the axle, the requirements being a little different than for the car drive. It proved to be just long enough. The bottom end I carefully ground down to a square to fit into the splined socket (alas, a smaller spline than the axle's) on the outboard's drive shaft. Square fit in, driving four of the sixteen splines. I also found an "SD" type compression coupling for 1" shaft to attach the magnet rotor to the axle, which I had no idea how I was going to do until I saw one on a shelf... AHA! (BTW This magnet rotor was the extra-scary one with 18 magnets on it. I thought it would be asking for trouble to try to get them off and redo it with just twelve.)


The (car transmission) axle, with the end ground down to a square that fits
in the outboard's splined drive shaft socket, here sitting on the stator.
The 3 hall effect sensor brackets on the stator are seen at far left.

   On the 14th I installed three angle brackets to mount the magnet (hall effect) sensors on, put on the magnet rotor, and then started on the motor controller. I finished soldering that together and also installed a circuit breaker and did some of the chassis wiring.
   Later I put the top cover on the outboard. There was lots of room from the magnet rotor to the top of the cover. I reached in through the starter rope hole and turned the motor by hand. One magnet stuck out about 1/8" farther than the rest. It, and it alone, rubbed on both sides and the front of the cover as it turned!
   I tried to cut off the protruding end with a zip disk on the angle grinder, but suddenly the whole magnet broke loose. So I removed it. Then I was worried about balance, so I tried to pry off the antipodal one. It too came loose, without trouble. Great! But I started worrying about magnets breaking loose, and I became more convinced than ever that they should be screwed on as well as glued. But maybe I should try some other type of epoxy before giving up - they just glue lawnmower motor magnets on, and those... except for on my lawnmower... it seems rarely fail.
   It's lucky my original (well, first *successful*) prototype motor had a slightly smaller diameter than the later ones and that that was the motor I used for the outbaord project. (And I almost remounted the coils on a larger rotor - nah, too much work.)

   On the 15th I finished putting together a second Turquoise Motor Controller to have one in the shop - it was needed generally and I wasn't about to drag the outboard out to the car to plug it into the one there. It didn't seem to work (with the car's EH motor either). On the 16th I mounted the hall sensors on the outboard, and got the controller working (Details on that under "EH Motor Making").  I didn't yet have the proper power cable plug for the outboard's motor, so some brand new skinny little alligator clip leads (#24 wire) were given a real workout. Though quite hot, somehow they survived some short bursts of several tens of amps.
   I had bought an AC/DC amp probe that hooks up to a volt meter, so at last I'm able to measure the total DC current coming from the batteries, which I did with the speed control turned right up. Some of the figures (steady state, no load, 24 volts) were interesting:

CAR MOTOR: 10 amps at 800 RPM. (240 watts) The three motor phases didn't seem very balanced, reading 7.2 A, 5.5 A and 7.5 A.

OUTBOARD: 22 amps at 850 RPM. (550 watts). I didn't measure the AC phases.

   For a brief time I was thinking that seemed like a lot of power - over twice as much - to keep the slightly smaller motor on the outboard turning not much faster than the large one. The differences in the coils didn't seem like enough to explain it. Then I realized the main difference was that the car motor had no load, while the outboard's motor, though it had no "real" load, was turning the drive shaft with its 90º gears and the propeller. The gears are in an oil bath. The gearing seemed to be about 11:3, so the propeller would have been turning about 230 RPM.
   This seems like a good illustration of the power consumed by a gear transmission drive system. It definitely takes more force to turn the outboard's motor by hand than the free one.
   But later I discovered that a wider magnet flux gap (.57") reduces current and increases maximum RPM by surprising amounts, and the car motor's gaps, while under 1/2", were wider than the outboard's, doubtless accounting for part of the difference.

   Next question was whether to make a motor controller especially for the outboard, in a compact box that fits under the hood, or just to use the external shop one. It would be cool to have everything enclosed inside the outboard, with the controls where the starter pull was and just the two battery cables coming out. That would be the obvious choice if it was to be used more than rarely. I finally decided on the external unit, since another controller was another small project in itself, and doing a mini-box controller might well have unforeseen headaches, too. The outboard motor "side project" was already stretching into some weeks.

   But the outboard wasn't through tossing in nasty surprises. The steering wouldn't move, and I had assumed there was some sort of silly little interlock tied in with the tilt system that was misbehaving. Wrong! Inspection disclosed that there was no such interlock, and that in fact the stainless steel pivoting tube was somehow totally seized in place by the plastic sleeves. Even after I got it to move, only powerful brute force could swivel it left or right, and a week of moving it back and forth and pouring in WD-40, oil, solvent, toluene, methylene chloride and methyl-ethyl keytone wouldn't loosen it up.
   After hours of brutal manipulation in two fruitless sessions with two people, in which the stainless tube was worked out just an inch and actually got bent, and I hurt a rib and my back, I finally got the idea to blast a propane torch flame right through the tube, which was open at both ends, to soften or melt the surrounding plastic. The plastic caught fire a few times, but after 3 or 4 torchings, the tube finally came out. Once it was *finally* apart, the top and bottom plastic sleeves virtually fell out, leaving the long middle one in place, and it was easy to ream and file out the three sleeves until they fit easily over the tube. On pounding the tube straight again and slipping it back in, the steering was free and easy. The plastic presumably must have gradually swelled or something to put such friction on the tube. It seems strange that either stainless steel or "inert" plastic should have swelled or somehow changed over the years to produce this problem. By some small miracle, none of the vital cast 'pot metal' body parts got broken in all this. (Earlier I had broken a piece off the anti-cavitation plate when trying to get the leg apart - not good, but not vital.)

   I had started by thinking the 26cm axle shaft was too long and might have to be cut shorter with the angle grinder, but once everything was in place, the rotor holder was clamped to less than an inch of the axle at the top, and the rotor wasn't sitting quite straight. While I had the motor apart again for the steering, I decided to lower it 1/2" - all the clearance available - to get a bit more shaft at the top to clamp the rotor to. That meant re-bending the metal mounting bracket and getting everything lined up nicely all over again.
   In this, I tried to pound a bearing race in farther than it wanted to go, and the hub welds broke. One of the four welds had spray paint in it, which meant it wasn't welded to the rotor in the first place, only to the hub. Next time, I'll put down even more weld, and more carefully melt it to the rotor before getting it to the hub, but on this particular motor, with the axle held centered at the bottom by the drive shaft, it appeared (for the moment) that fortunately the hub welding wasn't really necessary anyway, and that adjustments were easier without it.

   By the 23rd the adjustments, painting et al were ready. I bought narrow headed bolts for the shaft collars, cut a keyway into the axle to match the one on the shaft collar, and put the magnet rotor on. Still not very straight - oh well! That 1/2 inch gap between rotor magnets and stator coils has its advantages! I ran the motor again, with about the same results electrically and similar amounts of vibration.

   Then I decided not to make a special motor controller in a miniature chassis to fit under the hood since it would be another small project in itself. Instead, I would just use the new shop controller in the back of the boat. But I'd made the motor cables too short for that. With the weekend on and ECS Cable closed, I painted the rubber hood base dark green for effect (I'd sprayed it white with the hood), then found some transparent adhesive labels [Staples Office Supply] and made up a logo saying "ELECTRIC HUBCAP .com" with a frame of lightning bolts. With nice fresh paint and fancy labels it inspired confidence that it just had to work great, too - how fickle we humans are!
   I got a rubber "cab tire" cable for the #8, 3 wire, motor power on Monday the 25th and did the rewiring.


The painted electric outboard

   Presumably the motor was at last finished, but in the next test (with the cover off) one of the supermagnets flew off like a bullet. It scored a bullseye on some empty plastic sodium sulfate jars on a shelf, sending some flying. The one it hit had some deep scratches and was coloured with a rectangle of the magnet's paint. I decided that the "epoxy steel" glue I've been using must not really be very good - it seems to loose its grip over a year or two - and I decided to try taking more magnets off - hopefully all of them at that point - and re-gluing with a different kind of epoxy. Sure enough, they all came loose with a sharp tap of the hammer -- with much of the striking force provided by the magnet itself attracting the hammer. One seemed better stuck than the rest and I chipped it (small chip), but finally it too came loose. As a silver lining (besides having not been hit by the flying magnet), I could reduce the "overdone" 18 magnet rotor to the "optimum" 12 magnets and have 6 left over for another rotor.

   On the 26th I put six magnets on the rotor with a liquid epoxy resin, and on the 27th decided to try it with just those before adding the rest. I also increased the flux gap from 1/3" to .57", which gap (.55") I'd seen mentioned for another axial flux motor of similar power. The motor had less torque - fewer magnets at a greater distance. In fact I could hold it stopped with my hand even at full power. (...drawing 40 amps - gosh, I'm really glad I finally got a DC clamp-on ampmeter, even if it was $90.) But once it was moving, it was smoother, and it was unstoppable! At only part power it was doing 800 RPM and drawing only 10 amps, a great improvement in efficiency. At full power it was at 1460 RPM and drawing 33 amps. The unwelded hub was rattling fiercely. Even with the hood on the motor, this was scary! (Obviously the unwelded hub wouldn't do, and I re-welded it a few days later.) And all this was still with just 24 volts battery supply instead of 36. I was almost scared to try it at 36 volts -- in air it might over-rev!

   This experiment led to a couple of valuable conclusions about the Electric Hubcap motors that I hadn't caught before:
1) The motor works much better and much more efficiently with a flux gap of at least 1/2" than with the smaller gaps I've been using like 1/3", and still better than with even smaller gaps like 1/4" or less. However, even with supermagnet magnetic fields the flux drops off rapidly with distance, and I can't see going much wider. (Of course, I couldn't see going to 1/2" before, either!)
2) Smoothest operation appears to be with single magnets in each pole position - double magnets introduce more vibration and probably torque ripple. The motor made strong vibrations with some of the magnets missing as I was taking them off. (Yes, I tried running it with various opposite-side pairs of magnets missing as I took them off, to see how it might work. Towards the end, sometimes it had to be started by hand, when magnets were too far away or too unevenly spaced from the Hall effect sensors to get them to work properly.) However, with the wider gap, evenly spaced double magnets seem to introduce trivial vibration.
3) 12 magnets are still required for good torque and maximum power - especially with wider gaps. But it may be that some uneven spacing of magnets, eg with pairs having like poles closer together, will give smoother operation or better torque. This could still use some experimentation, but I'm happy enough with even spacings.

   On the 28th I put the other 6 magnets on the rotor, and on the 29th I put a thin coat of epoxy over the whole of the magnets and their area of the rotor, virtually encapsulating it (if not very thickly) in epoxy plastic. On the 30th, I decided that seemed like a great idea and put on a second, thicker, coat. I have much more confidence that the magnets should never come loose.

Water Pail Test

   Then when that had set, I put the rotor on and tried out the motor. It seemed to run fine. With my workshop student's assistance we dragged the finished outboard and three batteries outdoors and ran it in a garbage pail of water. It ran great, the meters showing 34 volts, 29 amps, and about 1740 RPM. Lots of water was being churned up strongly, but I was disappointed that it wasn't spewing out of the bucket. (Would make better video!) The indicated power, 1 KW, was only 1-1/4 HP, the low current indicating that the motor wasn't doing a lot more work than just loafing: the load was unexpectedly light. The motor controller heat sink remained cold, and the motor coils were still almost cold after a few minutes of running.


Electric Hubcap Outboard: first in-water test
Movie Here

   The problem is of course the 11:3 gear reduction from the motor to the propeller. At full RPM, the motor isn't turning the propeller as fast as it ought to be going. One to one or even two to one gear ratio would load the motor down much better and move a lot more water, but the ratio is built into the outboard leg. Two things that can be tried are (a) to up the battery voltage, say to 42 volts with an 'extra' 6 V battery, which will up the motor's maximum RPM some more, and (b) to try and find a steeper pitch propeller, which will push more water at a lower RPM. Above 42 volts is endangering the 60 volt rated mosfets in the motor controller (considering voltage spikes from the coils), as well as starting to get dicey for humans, especially in a wet boat, and heading towards speeds well over my recommended 2000 RPM upper limit, that might be hazardous with the fat EH motor - epoxy encapsulated magnets and all. (The centrifugal force is proportional to the square of the speed. If one is ever running over about 3000 RPM, I don't want to be in the same room with it!)
   Still, even at 36 volts, in a boat and turning still water instead of water that's already circulating in a pail might make for a notably greater load, and the boat's performance might be fairly impressive. And if not, the silver lining: the batteries should last around two hours at that low current draw.

   On November first I found a steeper pitch propeller. It's not the right one or an exact fit, but the diameter is right and the shaft size is close - I'll get it to fit with a few custom measures.

Plug-in Hybrid Boat

   Someone suggested that the electric outboard could be run from a generator. A portable generator on board makes a great "plug-in hybrid boat" idea, especially for longer trips: it would eliminate the danger of being stranded on the water if the batteries got low.

Sacrifice... or Upgrade?

   Though I initially regretted sacrificing my Honda outboard for the project, the EH outboard should outperform it and run more quietly too, so assuming the battery project gets successfully completed, I'll probably never want a gasoline outboard again. (...or any decent battery project... where are those cheap 70-80 WH/Kg nickel-iron dry cells from Bangalore, India? Perhaps Tata Motors in India will have them produced to use in their $1,999 electric cars [importing these is banned in USA - unfair competition for all those American electric cars you see advertised on TV] and for the electrical items in their new 300 Km range, fast refill or home refill, petroleum free compressed air engine vans!)

Turquoise Battery Project

Carbon Experiments

   I've been trying to learn how to make highly conductive carbon sheets for the positive electrodes, but instructional information is vague and even the few tidbits of information about the raw materials are confusing and sometimes in conflict with each other.

   In the search for carbonaceous materials, I found that "pitch", "road tar" and old "creosote" from a wharf seem to be pretty much the same thing.
   Then, "lamp black" and "carbon black" are evidently the same thing (but not to be confused with "black carbon"), and may perhaps be describable as "surface oxidized graphite". carbon-black.org says: "Carbon black is chemically and physically distinct from soot and black carbon, with most types containing greater than 97% elemental carbon arranged as aciniform (grape-like cluster) particulate." It also says that it's used to make non-conducting products conductive, whereas the Wikipedia article says it's a non-conductor owing to chemisorbed oxygen complexes on the surface.

   What was that compared to the stuff that I'd cleaned out of the chimney? Though usually called "creosote" or "soot", It seems more like light, airy charcoal to me, and I thought the Wikipedia article on "charcoal" seemed to hint it probably had that sort of composition. "Charcoal" it seems is 50-90% carbon. But the article was unenlightening on the non-carbon composition, electrical or crystalline properties, etc.
   It ground into a coarse powder with mortar and pestle. It was hard to get a resistance reading, though if the probes were virtually touching, a low to medium reading might be had across, perhaps, single grains. This is quite different from graphite powder where simply dipping the probes into a jar of loose, uncompressed powder gives a reading of only hundreds of ohms. I guess it's not good stuff to use.

   Then again, what about that compressed, graphite impregnated plastic? GraphiteStore.com had some intended for fuel cells - doubtless closely related - in fact, probably intended for the very same purpose. Hmm, a compressed composite... And the dry cell electrode rods were a dense carbon substance.
   So a good experiment, since the more conductive carbon things all seem to be higher density, might be to try to compress one of those light sheets of graphite to see if it could indeed be compressed in the electrode compactor, and if so, if the resistance dropped by much. That would be easily tried!
   The initial thickness was 1.6mm, and the resistance readings were typically about an ohm anywhere across the thickness and 1/2 an ohm if both leeds were on the same side of the sheet. (This is considerably more conductive than I thought, which is probably related to having just replaced one of my multimeter leeds, which had been going bad over the last few months.)
   After compaction, the thickness was uneven, 1.0 to 1.5mm, with one corner seemingly little compacted. Had the compactor "bottomed out", since I hadn't planned on compacting that thin? The resistances on one face were about the same, but the resistances between faces were now also down to about 1/2 an ohm.
   I decided to try again, with a plastic insert to push the die down farther. Sure enough, it compacted down everywhere to between about .8 and 1.0mm. Resistance readings between any two points were about 1/2 an ohm. Well, interesting and informative. Compacting is part of the answer, but simple graphite sheets will doubtless still degrade and swell in  the battery.

   If pitch mixed with powdered graphite can be made with similar low resistances, perhaps the pitch would prevent the graphite from swelling. If only it wasn't so sticky, and didn't have to be so hot when working with it, it might be great stuff! As it was, I stared at the tarry cookie sheet for a good portion of the month before I went at it with paint thinner to clean it so it could be used again for the next experiment. More rewarding to work on the electric outboard! Of course, once that's working, all the more will I want superb batteries!

http://www.TurquoiseEnergy.com
Victoria BC