This project has been cancelled and I have decided to pursue converting a bicycle to electric drive instead.  The range to speed to investment ratio is much more attractive when converting a bicycle given the much lighter weight. I will start a new blog describing my electric bicycle conversion project.

72V Curves, Tysonic, Charger, Motor, Deployed PV 3-D models

September 17, 2009

I have several updates today:

• The Factory Service Manual is on backorder and I would still rather wait for it before beginning to disassemble the gasoline power train.


• I discovered another sealed lead acid AGM battery option from Tysonic that is significantly cheaper than B&B and fits perfectly into the available space. It would provide similar range as the Lithium Ion pack I am considering, at one third the cost, but its life may be up to a tenth shorter than Lithium Ion and the range will suffer degradation throughout the batteries life. LiFePO4 is still the better value. I also created models of the new battery option, battery charger, motor, and deployed PV panels. See the video for more detail.


• Todd Kollin at http://www.electricmotorsport.com/ and John Fiorenza at http://www.marselectricllc.com/ were very responsive in emailing me the 72V performance curves for the Etek-RT (ME0709) motor I will be using. I haven't yet studied the curves in detail to begin my performance estimates and to select the proper gearing and top speed targets.

Testing 45Deg PV Panel Concentrator

September 13, 2009







I tested the first version of a concentrator to increase the power output of the photovoltaic panels. The concentrator consisted of 4 white pieces of foam core board set at 45 degrees to the panel surface. I set the panel outside in my backyard facing due South and at 32 degrees with respect to the ground. Thirty two degrees is the optimal angle for my location in September. I took measurements of panel voltage and current throughout the day to calculate power (Power[watts]=Volts X Amps). Measurements were taken with and without the concentrator in place and I compared power levels to see the effectiveness of the concentrator. I found that the concentrator produced shading early in the day and late in the day. As a result of the shading, power output turned out to be lower with the concentrator in place even though it did help to increase power near noon. I confirmed the shading issue with a 3-D sketchup model and will use sketchup to help redesign the concentrator with a better angle to eliminate shading. The redesign of the concentrator will include:

• Concentrator set at a lower angle based on Sketchup shading simulation (likely between 30 and 40 degs)
• Surface of concentrator painted with high gloss white paint to increase reflectivity
• Improve mounting system of foam core board to PV panel
  

Front Tire and PV Panels Arrived

September 12, 2009



The new front Bridgestone Battlax BT45 tire arrived yesterday and was installed this morning. I can finally stop fearing the front tire will blow every time I get on the bike. The two 30watt photovoltaic panels also arrived, appear to be functional, and the dimensions were as listed. I will be performing some measurements on the panels tomorrow to start estimating the added power that may be attainable by using some white reflective panels to concentrate sunlight onto the panels a bit. The service manual is still not in and obtaining Texas plates and safety inspection sticker is still pending. Batteries seem to have a long lead time of 6 to 10 weeks from the lowest cost supplier so I will likely hold off on beginning the dismantling of the bike until a few weeks before the batteries and motor are expected to come in. The Google Sketchup model I showed in my previous post gave me confidence that there will be enough room to mount a 69.3Volt battery pack in the available space so I feel I can order batteries without more precise measurements with the engine removed.

3-D Model of Battery Pack with Google Sketchup

September 10, 2009

Google Sketchup is a wonderful, easy to use, free tool for doing 3-D modeling of everything from small mechanical components to entire buildings. I used it to make a 3-D model of the battery space available and various sizes of battery packs. I took measurements on the stock bike with the engine in place so I will have to refine the model when I take the engine out and can get more accurate measurements with nothing in the way. It looks like I will have enough space to even mount the battery pack in a stylized V-shape in honor of the V-4 soon to be removed. Mounting the packs in straight flat rows would be quicker and easier but I'll study the feasibility of the V-shaped mounting once I have the batteries in hand and an empty chassis to work with. I compare various pack sizes in the following video and show how they would look mounted in the bike. You can download sketchup and tutorials on how to use it at: sketchup.google.com

Photovoltaic Panel Cardboard Mock-up

September 7, 2009

I made a cardboard mock-up to help me start to plan and visualize how I will mount the photovoltaic panels. There seems to be enough room and solid mounting points to attach the panels sturdily and reliably. I am not convinced that exisiting solar charge controllers will work well for my application which requires using a low voltage solar array to charge a high voltage battery pack. I also want the ability to use a battery pack that is not necessarily in increments of 12 volts as current charge controllers are designed for. Having to design and build a custom charge controller will take a bit of time and research on my part, which means the photovoltaic portion of the project will take a bit longer than I had hoped. Freeing myself from the constraints of charge controllers on the market allows me to start with a higher voltage battery pack, however, and I am now thinking of going ahead with a 66V or 72V pack. Friday will be a big day as the new front tire, photovoltaic panels, and factory service manual should come in. As soon as I get the new front tire installed I will start removing the internal combustion powertrain components.

Battery Technology Selection

September 6, 2009

I've been researching different battery types after finishing reading the "Secrets of El Ninja" motorcycle conversion guide. The guide was published in 2006 and it compares different battery technologies as they stood in 2005. The three technologies I was considering were 1)sealed lead acid (Absorbed Glass Mat), 2)flooded lead acid, and 3)lithium iron phosphate (LiFePO4).

I'm discarding the thought of using flooded lead acid. I don't like the idea of a tipped bike spilling Sulfuric acid all over itself and potentially the rider. The constant need to monitor water levels is also not my idea of maintenance free EV bliss. Sealed lead acid or flooded lead acid were the only economical options for conversions until recently. In 2005 the cost of a 4.1kWh lithium based pack would have been over $8000! Pricing in 2009 is now a little over $1300 for a 4.1kWh lithium pack and is very close in price to a comparable sealed lead acid pack.



It now seems that the logical choice is to go with a LiFePO4 pack. I've decided to start with a relatively low voltage pack, 48Volts, using Thundersky LiFePO4 batteries, http://www.thunder-sky.com/home_en.asp. EV Components is the company that currently has the best pricing on LiFePO4 batteries, http://www.evcomponents.com/SearchResults.asp?Cat=34. Chinese battery makers seem to be the only ones currently selling Lithium Ion based batteries in low volumes to U.S. distributors for the conversion market.

A total of fourteen, 3.3Volt Thundersky batteries are required for approximately 46.2Volts, but what amp-hour capacity to choose? Once the system voltage has been decided upon (46.2V), the amp-hour capacity has to be chosen based on the range needed. I need a minimum of 36 mile roundtrip range in case I cannot plug in at work. The two formulas in "Secrets of El Ninja" for estimating range are 1)best case at 30mph on flat ground with no wind, 2)stop and go range:

1. BestCaseRange = kWhrs x 9.8 miles/kWhr Where kWhrs is (pack voltage) x (battery Ah)


2. StopAndGoRange = kWhrs x 7.0 miles/kWhr

I analyzed the range for various Thundersky batteries from 40Ah to 150Ah and found that to meet my range target with 48Volts I need to use the 90Ah batteries or greater.

An issue with battery sizing is that I will likely increase the voltage in the future as funds allow. Increasing voltage improves range much as increasing amp-hour capacity can. At higher future voltages I could get away with a greater number of smaller, lower amp-hour batteries for the same range. I will be making some 3-D models to determine what the best battery size is with future expansion in mind.

Using hypermiling techniques, http://www.cleanmpg.com/, and based on my commute route, I estimated my range as the average of the two formulas above. A unique feature I will be adding in my conversion is two 30Watt photovoltaic panels as rear fairings that will be able to tilt up into position at the correct angle while parked. These two panels I estimate will add about 3.5 miles of range based on the following formula:

• AvgSolarMiles=Watts x Derating x AvgSunHrs/1000 x AzimuthOptimizationFactor x miles/kWh


• 60 x 0.85 x 6.53/1000 x 1.064 x 9.8 = 3.5 solar miles

The Azimuth optimization factor adjusts for the fact that being able to adjust the panel angle will help increase power production throughout the year as the position of the sun changes with the seasons. I would also like to add a concentration feature to increase output further and will experiment with concentration once the panels come in.