Setback...

Back online and with heat, light and cable TV after 6 days without. We live at the epicenter of the worst storm damage in ConEd history. Winds up to 70 mph blew all day and into the night, ripping shingles off the roof, and ultimately pulling the entire garage shingle layer off just before dark.

At about 7pm on March 13th, our 75 year old, 95' tall oak tree uprooted cleanly from the ground, which had been saturated by heavy snow pack and two weeks of rain. The tree came down, snapped off a telephone pole, pulled the power and telecom lines down onto our driveway, entrapping our cars, and landed in the middle of the street- all the way across it, actually.

Five minutes later, 3 of our 5 60' tall Honey Locust trees uprooted, and they came down on the house, hitting the chimney, putting two holes in the roof, snapping 3 rafters, and scaring children young and old. We had water coming into the kitchen and threatening the basement, no heat or light for many days, and no way to get our cars out of the driveway until today.

Within 1000 feet of our house, throughout Orchard Hill, there are about a half dozen more scenes just like this- huge trees across the road and a house, lines and poles down, etc. I counted over two dozen utility poles down or broken on a 1/2 hour walk. There are hundreds of trees down, and many, many big ones. Its incredible.

This has nothing to do with the Moonray, other than it is a pretty big impact on our family, requiring about all the time, money and energy we have to spare for a while. Bummer.

The good news is that no one was injured in the neighborhood, although a motorist was killed by a falling tree and power line a mile down the road. That's really all the good news there is at this point...

Notice the chimney sparing the roof corner, and the north wall of the house, from full impact...

Its hard to visualize how big this tree is. It was as high, though not so big around, as the tree still standing on the right- over 90 feet tall. The root mass in front of the house is sticking 10 feet out of the ground.

 Just look at how much of the formerly level ground was disturbed when this monster pushed over. On the right of this picture, my neighbor is about to get his car stuck in the spongy ground trying to drive from his driveway to the other neighbor's driveway, around the shattered pole and downed lines in the street. The pole still standing at the corner, with the street light on it, is pulled over and cracked at the base.

Getting the right parts...

Well, its been a while, and the model is now officially a little behind schedule. That's OK, and the subject of another post...

Until then, let's talk about parts.

With a curb weight of just 400lbs, (160 of which are batteries,) I'm obsessed with weight of every component. I'm also very cost-conscious. Mark Twain wrote: “The difference between the right word and the almost right word is the difference between lightning and a lightning bug.” For the Moonray, almost the right parts won't do. Getting the right parts has become an ongoing exercise in lightening. Three cases in point: the steering wheel, the front suspension arms, and the front wheel/tire package.
 
Initially, I thought the steering wheel (handlebar?) from the very interesting Honda Odyssey FL350 Off-Road  Go-Cart of the mid-1980s would be ideal for the 'ray. It has brake levers, a throttle and cable, headlight and other switches, all built into a cool aluminum aircraft-style yoke. It says Honda on it, too. I bought one for $95 on ebay, complete with cables, switchgear and wiring harness. It really is cool, but when I got it in my hands, I realized its very rugged and heavy. Sure enough, it tips the scales at almost 7 pounds, about 10 with the $35 steel steering shaft that I bought with it. Since I would still have to rig up some motorcycle cable-actuated brake master cylinders, (I bought one of those, appropriately sized for my Wilwood PS-1 front calipers, and had trouble figuring out how that was going to work...) the whole "fully integrated hand controls" thing started to become more of a project than the actual drivetrain, steering and brakes combined, and it was already too heavy for the weight budget. The lightning bug of steering wheels. So I reluctantly put the super cool Honda Odyssey yoke on ebay, and it fetched $100 from a disabled guy in New Zealand who wants it for a hand-controlled dune buggy. The shaft is still listed, and the brake master cylinder sold for what I paid for it, about $25. Consider it time well wasted...

Meanwhile, I found an even cooler steering yoke, from a 1959 Beechcraft Bonanza. Its a smaller, super-trick aluminum casting with an emblem surround that would support a little domed moon in place of the Beech "B" emblem there. Fantastic, and only 1 pound, 5 ounces! I snapped it up on ebay for $85, and was thrilled. No controls on it, but having switches, the throttle and brakes on the steering wheel was adding complexity and weight to the vehicle, so this is better. It is very hard to beat aircraft parts for lightness. Here's a comparison between the Beech and the Honda yokes:

I was so happy, at least until a week later when I saw an early Ercoupe steering yoke on ebay- its more steering wheelish and looked super light. I didn't really want to give up the Bonanza yoke, but I won the stylish Ercoupe yoke for only $22. Then I discovered its much simpler arrangement for bolting it to any old steering shaft, and, oh yeah, it weighs just 9 ounces! Lightning has finally struck the steering wheel of the Moonray. The Bonanza yoke is currently back on ebay. Here's a comparison of these two yokes:

 

From 7 pounds to about a 1/2 pound in two easy steps...

I've had a similar mission of refinement with my front suspension (unequal length) arms. I'm using the front suspension from a Suzuki LT250R Quadracer. Its a good choice, because the Quadracer is an old-school design from the mid-1980s, with less suspension travel, and arms that operate closer to horizontal than newer quads that tend to have arms at a 45 degree angle to the ground. There are also improved (lighter, stronger) replacement arms designed for racing and available in the aftermarket. My plan was to use the stock used arms for the initial build, and move up to the lightweight racing arms as my weight and money budgets dictated.

I bought a used set of 1986 front arms, complete with spindles, steering arms, hubs and brakes, for $100. They were cheap, but upon further study I discovered Suzuki improved the arm design and mounting, while using the same spindles and hubs, in 1987. So my '85-'86  arms are less common than the later '87-92 arms. More important, all the available aftermarket ChroMoly racing arms, (that are lighter and stronger than the stock arms and offer adjustable alignment settings) are also the later design, and incompatible with my early arms. Worse than that, I suspect my arms have worn ball joints at the spindles and bushings on the frame side. These parts are over $175 new, so I've been in the market for stock or aftermarket '87-'92 arms for several months. Stock ones in good used shape could be $75-$100 a set, and aftermarket racing arms are three to five times that, and harder to come by used and good.

Then lightning struck again. For whatever reason, a set of fully restored 1987 arms, freshly stripped, sandblasted, powdercoated yellow, and fitted with all new ball joints and bushings, came up on ebay and somehow attracted no attention. I was the only bidder at $45, and I was giddy unwrapping them. They are perfect, and were a plain steal even with the $29 Shipping. If I make weight with them, I'll probably forget about upgrading to aftermarket arms unless the handling requires tweaking. Aren't they just beautiful? Take a look:

 

Nice, but racing arms are camber/caster adjustable- and lighter.

I struggled with the front wheel/tire problem for a long time. I bought stock Suzuki 10" x 5 1/2" wheels in a couple of styles to evaluate using them with auto tires. Even 10 inch car tires are really overkill for the Moonray, and at 11-12 lbs. each, they are relatively heavy. Add the stock wheels to them at 3 1/2 lbs. each, and the total package is over 30lbs for the front wheels and tires. More like a June bug than a lightning bug. 

While I was worrying about all that a drag racer at the QuadracerHQ put up for sale his custom-made Douglas 10" x 4" (standard width is 6") aluminum front wheels, complete with mounted, DOT-Approved street legal 3.5" x 10" scooter tires. These are ideal for my purposes, and under 7 1/2 pounds each! That's the lightest possible setup, already tested to over 100mph, and road ready. Incredible, and fairly priced at $150 delivered. Its enough to make the air crackle with electricity.

Way cooler than it looks...

So, for now, its all about the parts. When building a vehicle from scratch, there are many more choices for how things can be done. With constrained vehicle mass and cost budgets, each of those choices needs to be made carefully. I'm not unhappy about revisiting the choices several times, so long as the outcome improves and the expense is modest. Its been like this for almost every component, just gradually finding exactly the right thing. Of course, this is fun and easy only so long as nothing is built that these components attach to.

Once chassis fabrication begins, swapping parts in and out will be much harder and more expensive...

1:3 Scale Mini-Me

To help with the dimensions and packaging, I created a 1:3 scale model of me for the project. First, I found this awesome anthropometric pilot model that the US Air Force used to validate cockpit dimensions in their aircraft designs. Its known as a Two-Dimensional Drawing Board Manikin, and it was actually patented in 1977:

The model is of a male representing the 95th Percentile size of USAF personnel, which equates to 6'2" tall. Since I'm 6'1.5" tall, its about perfect, and it came with several arm and leg variants to fine tune the size. Even better, it already has a helmet on. Incredible...

I measured myself carefully, selected the arm and leg variants that fit me best, and enlarged the manikin image until it was exactly 6'1.5" tall in 1:3 scale (24 1/2 inches.) Then I cut out the pieces, glued them to 1/8" masonite, and cut them out on the jigsaw. I drilled out the mounting holes to articulate the parts accurately. Since I wanted the manikin in 3 dimensions and not two, I made two sets of arms and legs, as well as two sets of hips and ribcage parts. I then assembled them with screws, threaded rod, washers and nuts to create a three dimensional, perfectly articulated pilot model exactly my size. Cool. I haven't decided if I need to add foam "flesh" to the masonite to round out the body and limbs, but its sure a great start for building the interior around the pilot:

I also found the the Cockpit Geometry Design Guidelines published with the manikin in 1982 in the Human Factors tome Anthropometry and Biomechanics: Theory and Application, by Easterby, R.K., K.H.E. Kroemer, and D.B. Chaffin (Eds.) Then, my brother-in-law sent me this way cool link to an audio MP3 of Dennis Wolter (an approx. 57 MB mp3 download), founder of Air Mod, discussing human factors, aesthetics, and safety considerations regarding the interiors of homebuilt aircraft. Its in the Experimental Aircraft Association Historical AirVenture Forums here:

Dennis Wolter of Air Mod Talks About Aircraft Interior Design in July, 2008

These resources will really help me get the driver's seat and controls right, and make the Moonray more comfortable and safer to operate than I would ever have been able to figure out on my own. 

Getting Started on the 1:3 Scale Model

After a long holiday season, I'm back to putting time into the Moonray. The first phase of construction is actually building the vehicle in 1:3 scale. That will allow me to work out the dimensions and design problems relatively easily. There aren't any ready-made parts I can use in this scale, like there would be if I were modeling much smaller in a scale where plastic cars and motorcycle kits were available. Almost everything has to be made from scratch. My 1:3 model will be about 4 feet long, though, so it isn't exactly fine detail work...

I would very much like to find wheels for the model that don't require much work. In my collection of such things I have a wheel I got from American Science and Surplus years ago (I know, don't ask...) that's just about right for the rear wheel. I also had a pair of fronts salvaged from my kids' stroller that were just right for the wheel/tire package I thought I was going to use, but then I scored a uniquely cool pair of front wheels from a Suzuki quad drag racer. They are ultra-rare custom made Douglas 10" x 4" wheels (two inches narrower than stock) with DOT-approved street legal scooter tires already mounted on them. That set is 16 1/2" tall, so I now need to find 5 1/2" wheels to model them accurately on my 1:3 scale Moonray.

I'm currently modeling replicas of the suspension parts and shocks that I have in 1:3 scale. When that's done, I'll work on creating the custom chassis to mount all those parts. Then I'll build the body shell, probably out of cardboard or coroplast, but maybe out of foam. I haven't decided how well to finish the model, but for now just getting the dimensions and packaging right is the goal.

Initial Specifications

I've been thinking about building an electric three wheeler for many years, but I've only been seriously looking at the figures, components and performance for a few months. I was following the development of a new electric vehicle motor from a start-up company (really a one-man operation) called EnerTrac on Long Island. The motor is the largest hub motor currently available, and designed for converting motorcycles up to 400lbs to electric power. The EnerTrac website is here:

EnerTrac Corporation Home Page

The public discussion during the motor development went on here:

Markcycle's Endless Sphere Hub Motor Thread

As this motor came into existence, I realized that it would be a breakthrough powerplant for a high-performance electric reverse trike, that is, a three wheeler with two wheels at the front, provided my vehicle was under the motor's maximum 400lb vehicle weight rating .
 
That's now the big challenge- keeping the Moonray under 400lbs. It would be rather straightforward to build it out of mild steel square tubing, and then put a wood and fiberglass body over it, but it will weigh 600 or more pounds so constructed. Components, materials and construction techniques all need to be carefully selected and executed to keep the weight down.

Batteries are the single heaviest component. Traditional lead-acid batteries with sufficient power to deliver the 300 amps the motor can take at 120 volts or so would weigh over 500 lbs. The lightest of the new Lithium-based batteries (from A123 Systems and used in cordless power tools) would be about 100lbs for the same amount of power, but they are still incredibly expensive- probably $10k for a pack of them, once I figured out how to get all of them put together. They are hard to deal with, requiring precision assembly of hundreds of about C-Size batteries into a large traction pack, and given the cost and complexity of that, it isn't the way I'll go.

In the last year or so, large format (about 1/4 the size of a car battery) Lithium Iron Phosphate (LiFePO4) cells have become reliable and widely available. They are half again as heavy as the best A123 batteries, but they are less than half the cost, and will be just fine for the Moonray. Moreover, all the skittishness, horror stories and dire warnings the EV community lavished upon early users of these batteries seem now to be quieting down as they are being installed and used to great satisfaction. Jack Rickard, in particular, has greatly contributed to the available information about these batteries by installing, testing and driving them in his converted Porsche kit car- and publishing his data for all to see. The batteries are fine, and they won't need an expensive or complex Battery Management System in the Moonray. Initially, I wanted a 100 mile range, which would have required about 240lbs of batteries. No matter how I work it, that much battery weight puts the vehicle way over 400lbs. I even lost 70 pounds myself, and it isn't enough to make up the difference. I just can't carry more than about 165lbs of batteries, which will cut the range to around 50 miles, but it has to be that way. At least the battery pack will be cheaper...

The design is simple to describe, but the devil is in the details. I will be using the front suspension from a Suzuki LT250R Quadracer ATV, and the rear suspension (swing arm) from a 1985 Kawasaki KZ125 motorcycle. I'll be adapting air-adjustable spring/shocks units from Honda motorcycles- the pair from the rear of a CB900F in front, and just one of two rear shocks from a GoldWing 1100 on the Kawasaki swing arm, which uses only one spring/shock unit. The air-adjustable units will allow me to balance the trike and trim the ride height. I have both the LT250R and KZ125 frames, and I could conceivably just weld them together to make a chassis, but that isn't why I got them. I needed the dimensions of the suspension attachment points to duplicate for my own custom chassis, which will be fabricated from Chromoly 4130 steel. I actually considered welding the motorcycle and quad frames together to build a "mule" chassis to test the motor and battery systems, and it would be a good idea in the abstract, but I think those systems are reliable enough that it would be a waste of time.

Since I'm not a computer-assisted design (CAD) guy, I'll be working with paper and pencil making the Moonray. If I had a CAD-versed collaborator, that would certainly improve the process, but its OK as is and I can't afford the time or expense of the CAD learning curve for this project. To compensate, I'm building the entire vehicle in 1:3 scale, which I am good at, and it does solve problems and refine the design to go through the exercise. The exterior shape will be determined mostly by the wind, but also by the components, configuration, and the available parts I choose, like exterior lights, a canopy, etc.  Still, I have a vision of what I want, and some references for styling and aerodynamics that I'm going to follow. First my drawings:

 Moonray, 2009- Tom Alvary

Now the three-wheel designs I think are the best to emulate:

 

California Commuter, 1980- Doug Malewicki

The Polliwog, 1963- Bob and Bill Summers

There are many others, but these are my favorites, and the direction I'll be going. Note the similarities between these two vehicles, which could not have been more different under the skin. Clearly, they had the same (and in both cases, record-setting) approach to the air flowing around them. Interestingly enough, the Polliwog is actually a four wheeled vehicle, with the rears in tandem (one behind the other) on a bogie, and front wheel drive. Yet another breakthrough Summers Brothers design. More about them later and elsewhere...

I'll flesh out my reasons for each choice as we go along, but the initial specification is:

Length: 11-13 feet
Wheelbase: 85"-95"
Width:  49"
Front Track: 43"
Height:  45"
Weight: 400lbs Empty
Capacity: 240lbs.
Speed: 75MPH
Range: 50 Miles

Components and construction materials and details are being collected on a spreadsheet to make it easier to see and juggle the total weight of the vehicle. That's going to be where the all the action is, and on the model, until the basic problem of coming in under 400lbs can be solved...

Process & Project Management

Building something as complex and expensive as an electric vehicle requires planning and good work flow management to stay on track. Its a hobby, yes, but it won't be any fun if it never gets done or costs far more that I anticipated, or both, and in any case I will surely hear about it from the family, which won't be too much fun, either...

I will need to be very practical in getting tasks handled on time and within budget. I also need to be thorough to get both the details and the major issues addressed without undue delays. For now, while I have the luxury of putting some things off, and I can still experiment and learn to a fair degree, but progress must be made consistently to stay on track. I'm less concerned with the time I'm spending now than with the money, so I'm basically finding the more common components cheaply and solving problems I can handle myself. As the Moonray becomes more complete, that will have to change, but for now I am accumulating parts, studying other people's designs, and evaluating materials. That's fun and relatively cheap, but at this point I need to start building something.

I've decided to build at least one 1:3 scale model first, so that I can validate the design and refine it at minimal cost before I start spending full-scale time and money on the Moonray. I'm not a CAD guy, and I don't think it is practical to become one (another steep learning curve?) within the next few months, so I'll be making the Moonray the way I have made models, furniture and structures all my life- on paper, and then with rulers and calipers. I'm a pretty fair modeler, with 40 years of experience in everything from assembling plastic and paper kits to scratchbuilding flying planes, rockets, model railroads and architectural models. Its fun work for me, and it will be satisfying to have something come together quickly that I can see and test. I will also likely save some expensive and frustrating trial-and-error work on the full-size Moonray by developing the model until at least the obvious problems are fixed. I could also really use a well-finished model to raise money, sponsorship and support for the project as the design nears completion and the really expensive bits need buying.

Part of the value of writing this blog is that it focuses my thinking, and creates the archive of managing the project through its various phases. While its far more time-consuming to document and post the process online than it would be to just do it, there's good karma in blogging it and I hope it is fun and informative to see. It also puts my work out for review and suggestions, which can only improve the final product. This has already been going on at the Endless Sphere EV website where I participate in an online discussion group for electric trike builders. Ultimately, maintaining the blog also helps me stay on task with the project. Its not a job and I don't want to make it a chore, but its a pretty big project and anything I can use to help manage and move it along is a good thing.

Finally, there are three pieces of advice I've heard about car projects that make great sense, and I'm determined to stick with all of them on the Moonray project:

1. Do some work on the car every day, even if its just 5 minutes. It will probably never be 5 minutes, but its a great practice to get into, and doing it ensures that project won't stop, which is the most common way projects never get finished. 10 minutes a day is over 60 hours a year, and even a minute a day will keep the Moonray alive.

2. Write down every penny spent on the project, from paper towels to transportation. Without doing that, budgeting and costing is pretty much meaningless, and having detailed expense records will be a great benefit in many situations, from proving up an insurance valuation to selling the vehicle. 

3. Try to map out the time line from now until completion, and keep it current and realistic. Well, that's good advice for any project, but too many hobbyists, (and businesses for that matter,) don't follow it enough.

My current road map looks like:

12/2009-1/2010:
Complete the 1:3 scale Mannikin and suspension component models.
1/2010-2/2010:
Complete the model chassis structures and finish the model.
3/2010-4/2010:
Revise and finalize full size design; create workspace and production schedule
5/2010-8/2010:
Build subframes, suspension, steering and hard points
8/2010-11/2010:
Layup the Monocoque body, and complete the chassis
11/2010-2/2011:
Build and test the drive, brake and electric systems.
2/2011-4/2011:
Finish systems and final assembly- lighting, paint, etc.
4/2011-6/2011:
Begin testing and operation.

As always seems to be the case with long projects, the closer-in tasks seem rushed and the later ones seem to be easier than the amount of time allotted. I also note that EVERY commercial EV project misses its deadlines and winds up with a delayed product, but those folks are working on many other things I don't need to worry about. At any rate, we shall see...

Project Goals

Even building a very small car is a pretty big undertaking for an individual, especially since I have no experience in several of the subject areas where expertise is required- hub motor and LiFePo4 battery systems, composite design and fabrication, etc. There is plenty to learn, a fair amount of work that will probably need contracting out, and hundreds of details to handle. Inevitably, there will be some trial and error getting things right, which can get really expensive and time-consuming. The trick is to do as little of that as necessary, but more than that, good planning and effective project management are critical to the Moonray hitting its targets.

First, a very clear set of project goals needs to be formalized, which will then largely dictate a set of specifications for the vehicle. These specifications aren't etched in stone; in fact they have already moved around quite a bit and I expect that to continue. Being practical is important, and when I learn the limits or costs of materials and components, being flexible enough to modify the specifications around what is easy, hard or likely impossible is essential. The goal is simple:

"Moonray will go faster and farther than any commercially available electric vehicle in its weight class, and will be practical to operate as a street-driven daily commuter/runabout EV."

Notice that it is "faster and farther" because coupling those two things, and producing a licensed, insured road vehicle, is critical to defining the project goal. There is a class of EV single-seaters, somewhat lighter than Moonray, that compete for distance. They are called Electrathon racers, and they go very far, like 50 miles, on a single car battery, but they are slower than the 'ray, and not practical or even licensed for street use. There are also a few even faster EVs, which are designed specifically for bagging speed records, but they aren't licensed, either, and I don't want a race-only vehicle. The Moonray aims to be the fastest street-licensed trike in its weight class, leaving the ultimate speed and distance marks to no-compromises off-road-use-only vehicles that you really wouldn't want to commute or run errands in, anyway. Moonray isn't a contest vehicle; its for daily use.

A secondary goal for the project is to showcase recent EV technology, and to show particularly how much can be done by the hobbyist with essentially off-the-shelf components. The batteries I'll be using have only really been available for about a year, and they are improving rapidly. The motor has been available for a couple of months. Without those leading edge components, Moonray simply wouldn't be able to achieve its goals, but they are neither terribly expensive nor hard to come by. What I'm doing isn't rocket science, either. Anyone with the time and skills to do meticulous work, or the modest money required to buy other peoples' time and skills, could easily build something very similar. I'm not making plans or kits, though, because its my hobby and not a business. I'm not an engineer and I'm not willing to have my work certified by one so that other people can confidently rely on it. The Moonray will be a demonstration of what is now possible, not necessarily a program for how to do it.

My other goals for the Moonray are logistical: I want it in service by the spring of 2011, with a total cost at or under my budget projections. I have a budget in mind, but it still has to be validated. I don't have unlimited time or money to spend on it, and frankly I'm not aware of any project with neither time nor economic constraint that turned out satisfactorily. Limits on resources seems to be a necessary ingredient to the development of anything good. 16 months from now feels like a reasonable amount of time, but let's see where we are in a few months. I'll talk about my budget once I understand it better, and have cleared it with my family.

So that's it: farther and faster than any commercially available lightweight street-legal EV, on the road by Spring 2011, and within the yet-to-be-determined budget. Piece of Cake. Let's get started...