Engineer Your Own Electric Motorcycle!

Converting a motorcycle to an electric vehicle is not as difficult as you might think. An EV conversion can be done with 100% off the shelf parts with the aid of various build instructions available for free on the internet. Lots of people have built these things before.

But that's not to suggest that an EV conversion won't benefit from research & engineering design. College research labs and professional racing teams know all kinds of fascinating information, but they choose to share it in long boring doctoral thesis papers, if they share it at all.

What you never see is the awesome engineering work presented in a conversational tone, let alone an instruction manual. My intention here is to summarize all the good sciency stuff in down to earth terms and share it online for other people like me!

This instructable details the process of physically converting the motorcycle AND the decision making process of optimizing the design.

These are 8 main steps to converting a motorcycle that will guide the structure of this instructable:

1. Initial Planning. (Get a Donor Bike)

2. Gut & Clean Chassis. (The Dirty Part)

3. Design & Planning. (The Nerdy Math Part)

4. Mount Motor, Batteries, Controller, Charger. (The Frustrating Part)

5. Mount Peripherals. (Main Contactor, Throttle, Fuses, Etc)

6. Wire Everything Together. (The Be-Very-Careful Part)

7. Build Body & Paint. (The Other Dirty Part)

8. Drive off into the Sunset. (The Put-On-Your-Cool-Guy-Shades Part)

Before you do anything at all you should read about other people's EV conversions while thinking about your goals for the project. Ben Nelson's Instructable, his DVD, and Carl Vogel's Book are all great places to start. Keep the lessons learned from their work in mind and immediately begin looking around at potential donor motorcycles, because it will take a while to find one.

The demand for cheap/used motorcycles is huge pretty much everywhere. It took me months of looking to find a suitable motorcycle though craigslist. I ended up with a partially wrecked 1989 Honda CBR600 with an invalidated title and an oil leak from a huge gash in the engine block. In retrospect, a fool and his money were parted that day.

My desire was to own a simple environmentally friendly hobby vehicle that I could use for my daily work commute, a 60 mile round trip. Did I mention I had never ridden a motorcycle before? (Yes, the arrogance was strong with this one.) So I needed a bike with a small enough frame that I could handle, but a big enough frame to carry a lot of batteries. An EV conversion can be done on any type of motorcycle so let your personal preferences guide this decision.

This initial research step isn't finished until you spend an adequate amount of time researching the topic of motorcycle conversions. Reading every single word in this instructable should suffice. Then go buy the bike.

Once you have the motorcycle you can immediately get to work taking everything off that is internal combustion engine related. Go ahead and cut out the underside of the gas tank and clean off the rest of the bike.

You can recoup a little money by scrapping the waste motorcycle parts but first be sure to take dimensions off the engine drive shaft and the mounting points. It's really important that your electric motor output shaft go exactly where the original engine output shaft was. See last photo above. When the rear suspension travels up and down the effective distance from the driven wheel to the drive shaft can change, which has the potential to mess up your drive chain. The original designers picked that specific drive shaft position as the best place to avoid this problem.

Unfortunately, the motor output shaft & the mounting points are out of plane and very difficult to measure the center to center distance on. I used what I call the 'ghetto CMM' technique (coordinate measuring machine) by taking photos of the engine alone side a ruler for scale and then later viewing that photo on my computer screen and taking measurements from the screen. Just beware of parallax and depth variations.

This step is also the best time to do any mechanical repairs to the chassis. I had to repair the brakes.

This step is complete when the chassis is completely clean and naked, like mine in the photos above.

This is the fun part where you get measure, calculate, and remeasure to figure out which hardware to use and where you will mount it to the bike. The biggest and most expensive things are the motor, controller, batteries, & battery charger.

At first it's difficult to select and size these parts because it seems like you don't have enough information to make any decisions. It turns out that your options are far more constrained than you might think. In theory there are many options but in practice something is only an option if it is attainable. The reasons behind the specific choices made will be discussed in later steps. To make this instructable flow better I have moved these discussions to the end of this instructable and have added '(SCIENCE)' in the step title.

For now know that you want to size the batteries, motor, and controller so that no single component becomes a significant bottleneck in the system. Each component will have its own maximum capacity for power delivery that has to be considered and that all components have to operate at the same voltage.

A huge portion of my time on this project was spent developing an excel document for estimating the finished performance of the bike. Once finished I realized that I could play with individual variables and quantify their effect on the bike's performance. The result is a bunch of really cool graphs in the science section.

I've attached the excel document to this step for your reference.

One big topic I've left out is vehicle dynamics. You can calculate how your bike would perform in various tests of handling, suspension, braking, etc. But your ability to change these things on a conversion bike is very limited unless you are willing to do a major overhaul. The conversion will affect the center of gravity of the bike, but that's about it. If you have a more custom bike project & want to learn about motorcycle dynamics, then I recommend this book: Motorcycle Design & Technology: How and Why

One funny thing I learned is that after having done all that work to figure out the power consumption of the finished bike I could have just used a rule of thumb to estimate it. The back of the envelope calc is that EV motorcycles consume 100 Whr/mi of energy. My actual power consumption ended up being about 110 Whr/mi.

The last thing to note on this step is that making cardboard mock-ups was immensely useful for checking fits. Even though I love 3D modelling, you just can't compare the two methods in this context.

This step is complete when you've done the math, know what you want and where you will put it, and you order all the electronic parts.

This is the 'sweat & be frustrated' part. Physical volume was a huge constraint on this project. Everything is heavy and mounting it inside tight spaces is a good way to smash fingers.

Mounting all these things required the fabrication of custom brackets and some welding.

The motor mount was CNC milled from a 1/4" aluminum plate. Perforated steel angle struts made for excellent battery racks. Long threaded rods & hose clamps were used to provide a degree of adjustability. Plumbers tape was used to secure the batteries to the rack in case I dropped the bike.

The motorcycle engine was actually a structural member of the frame so I had to mig weld a couple new beams in its place.

This step is complete when the frame components are fabricated and all the large electrical components are mounted.

These parts are also known as the 'balance of system' and they included a manual disconnect switch, a main contactor, a cycle analyst (voltmeter/ammeter/odometer), a DC-DC converter, and all the little things like diodes & resistors.

The DC-DC converter is for powering the low voltage system from your main battery pack. In retrospect I could have nixed the converter altogether. I had found enough space for a mini 12v 10ahr battery to power the headlight and remove that load from my main pack. This trick gained me an estimated 0.6 miles of range.

This step is complete when you have found a way to mount everything securely so you don't worry about things falling off your bike on the road!

This is the most dangerous step. Be careful not to accidentally ground things to the chassis or drop a wrench across the battery terminals!

I made the schematic shown in TinyCAD before I started wiring so I wouldn't get confused. For all the low voltage nodes I used wire nuts because it was cheap, but that turned out to be really messy and unprofessional. I recommend using some sort of terminal blocks mounted inside of a project box.

The sizing of the high voltage wire is important. The bigger you go the less resistance you will have, but thick copper wire can get really expensive.

Everywhere I look the standard operating procedure for safely sizing electrical wires seems to be ‘pick a really big one’. So part of the main excel document has a wire sizing calculator in it that I used to make the graph detailing power loss vs wire gauge above. I ended up using 10' of size (1/0) welding wire from evdrives.com

When thinking about wire size, things like current draw and length are import to consider here but interestingly voltage has no direct effect on the ampacity of electrical wire! Wire ampacity is a function of wire diameter and length. Voltage drop is only a function of wire length!

While voltage has no effect on the ampacity of electrical wire, its still important to consider here. Choosing a higher voltage system (say 72v instead of 48v) allows you to transfer more wattage (power) through a given wire (Watts = Volts * Amps). This allows you to operate with lower current which would reduce the power loss within the wire. This is why you find surprisingly small gauge wires inside large industrial AC electric motors; because high voltage motors draw a relatively small current.

On the low voltage system (12v) you have greater limitations on how long that wire can be run. The resistance of a given length of wire is constant, so the voltage drop through that same wire will be constant for any voltage. The resulting percentage voltage drop is significant for low voltage systems. If you’re not careful and run a low voltage wire too long you can give yourself a brown out. You can read more about this topic and find a more in depth wire sizing calculator here: EngineerDog's Free Electric Wire Size Calculator.

This step is complete when your bike looks like a pile of spaghetti and everything works.

The body that comes with the bike is made to fit around the engine. That body will not longer fit so you have to make a new one with fiberglass & cardboard. To all you professional car body people out there, look away, this was my first experience using fiberglass.

I made the body 3 layers thick then covered it in bondo body filler & sanded it to fill the imperfections. One trick I learned was to use a really wide bondo spreader to leave it as smooth as possible to minimize the sanding time. I also used 'filler primer' to help fill in the tiny scratches. Sanding takes a significant amount of time to do right. Everything has to perfectly smooth because imperfections will stick out after you paint it.

After the body was finished I attached it to the chassis using industrial velcro.

I was relieved when I finally finished this thing and it worked...wonderfully! I was actually able to get the bike legally licensed & insured! Of course I failed in my initial project goal of driving to work because it was simply too far, but the bike was great for cruising around town. A lot of people did a double take when I drove by because it was so quiet.

In the end the calculations, design, & build time took 6 months of intermittent work, which was much longer than I originally planned. This isn't due to incompetence, but rather the planning fallacy, which makes projects take longer than expected.

The bike performed almost as well as I'd predicted with the excel calculator. That calculator enabled me to confidently say that I could not have possibly gotten better all around performance than I did for the price that I paid.

The last step of this instructables contains a little video about the bike so you can see it in action!

Final Bike Specs:

(4) 80 Amp-hr AGM Lead Acid Batteries (Werker Brand)

PMDC Brushed Electric Motor ME1004 (10.75 hp continuous)

Alltrax Controller AXE4855 (250 amp continuous.)

Final Bike Curb Weight: 453 lbs

Donor bike: 1989 Honda cbr600

Typical Driving Range: ~25 miles (varies with speed)

Top Speed: ~60 Mph

MPG equivalent: 219

Total Operating Cost: <8 cents/mile

Project Cost: $3,200

Where its at now: I actually built this bike in 2013 and have been sitting on all this documentation ever since. I've since moved to an apartment without a garage or place to charge it so I had to sell off the bike. I wanted to avoid the liability of selling it whole so I took it entirely apart and sold it piecemeal at a huge financial loss.

That's OK though, I hadn't plannied on this project as being a direct financial gain. The money was more of an investment in myself. To give myself the chance to take on a project that had intimidated me and to learn as much as I could. As an engineer I can also say that this project has helped my career prospects as well.

I want to point out that you cannot justify a DIY electric motorcycle conversion for financial reasons alone. If you want to build one then do it for another reason. Do it to learn something new or for fun. Do it to show you care about the environment. Do it for science. Just don't do it with the intention of saving money, because you won't.

Most people do EV conversions out of environmental concern. They understand that the monetary price of goods doesn't always reflect their true cost to society. The end user's monetary operating cost is not a good measure of the environmental and human health cost of operating that vehicle because it does not account for externalities, subsidies, and the like.

Electric vehicles are better than gas from an efficiency standpoint at the vehicle level and arguably also at the well-to-wheel energy efficiency level. (To be fair, calculating the well-to-wheel efficiency of any fuel is really difficult, so the photos attached have Tesla's calculations for natural gas fueled power plants and the EPA's for everything else.)

The EPA has stated that, in terms of well-to-wheel CO2 emissions, EVs are the best option due to their high system efficiency. You can read it yourself here and while you're at it use this website to see where your electricity comes from.

Because it really matters where you get your electricity from. If your part of the US gets most of its electricity from coal, then your EV could be much less environmentally friendly that you might think it is. Ignoring CO2 emissions, there are many other considerations to be made including the impacts of mining, transportation, other chemicals released, and on and on.

**So which is better, gas or electric?

Clearly this comparison is not as straightforward as it may at first seem, and I'm not qualified to make the final call on the subject. But I have shared some insights and can be a guide for open discussions on the topic.

The only thing I can say for certain to the environmentally concerned among you is to use renewable energy, when possible, to power your EV. Also that you are always better off if you can use a smaller vehicle if doing so allows you to replace the use of a larger vehicle.

For this reason, micro-electric vehicles such as you might make with a FlexPV kit, are within the Goldilocks range of practical DIY vehicle design. Things like EV bicycles are cheap, simple, and efficient enough to make sense environmentally, practically, and financially. If you want to DIY something, stick with vehicles small enough to not need a license or registration to legally drive.

Having said all that, I cannot recommend doing a DIY EV motorcycle project for practical reasons. My bike had such a limited range and had to be plugged in to charge for long periods of time. Also, when you DIY build a vehicle you end up with something that only you know how to work on. You have no warranty, and you can't even sell the thing whole without risking a lawsuit down the road. For less money than I spent on the bike I could have bought a new moped that got 60+ mpg. The moped could then drive all day and is only a 5 min gas fill up away from being able to drive home.

You ask: How much does it cost?... I answer: How much you got?

There is one thing that all motorcycle conversions have in common: These projects take a LOT of time & labor. As I said, mine cost $3,200. The breakdown of where that money went is shown in the plot above.

Your conversion cost is completely up to you. Some people have completed similar conversions to mine for less, but they had to sacrifice quality (reduced range, speed, looks) or features (data logging, safety switches).

Your budget is the biggest determining factor of achievable performance at the DIY level. For $10k, I could have put better components onto the same chassis and made a bike that would beat the pants off my current one. But having done the analysis with the excel document made me confident that I spent my money as effectively as possible.

One thing that saved me a couple hundred was to make friends with the guy at the battery store. He sold me the batteries at the prices reserved for b2b transactions because he liked my project. Getting them in person saved me even more on having them shipped from an internet supplier.

The first thing everyone asks me is how far can this thing go? To that I say, it varies with how fast you drive. All vehicles consumer more power with higher speeds due to wind resistance, so the range of your electric vehicle varies with your cruising speed.

I was able to plot the estimated vehicle range vs driving speed. Then I picked a single cruising speed (45mph) for estimating typical performance. To my delight my graph matched the shape of Tesla's own range vs speed graph to the dot. My graph also had the little slump at the beginning like Tesla's, until I added the 12v auxiliary battery devoted to headlights. The source of the slump is that the headlights are a fixed energy cost that is only a variable of time, rather than speed. If you drive too slow then your air resistance is low but it takes too long to get places so your headlights kill your mileage!

One of the single most important variables that affected the vehicle range was the selected gear ratio. If your selected gear ratio is too low then your motor will hit its maximum torque output capacity before you achieve its maximum possible power output. Similarly, if your gear ratio is too high then your motor will hit its max speed at a lower overall power output than it is capable of. Imagine riding a bicycle in a fixed gear either really high or really low.

Fortunately, there is a finite number of possible gear ratios on a motorcycle to help us choose the best one. I picked a driving sprocket of 15 teeth because sprockets with too few teeth cause a racking effect that wears chains out prematurely. From there I varied the number of driven teeth until I found the peak output. 50 was perfect.

The second question people usually ask me is if the bike has regenerative braking, and they always look disappointed when I say no. The truth is that regen is not always economically viable. Regen's overall value depends on your specific circumstances.

Adding regen to MY bike would have gained me an estimated 1.7 miles of range (a 6% increase), but doing so would have cost me an additional $700! (a 30% in total project cost). I calculated the miles gained per dollar spent of adding regen vs buying larger batteries to see where the money was best spent. I discovered that the money was 25 times better spent on buying bigger batteries than by adding regen!

This is not to suggest that adding regen is always a bad deal. This has to be analyzed on a case by case basis because the amount of energy you can recover with regen varies with many factors. (Including how you drive)

For instance, you can recover more power when you have a larger capacity battery bank, such as a commercial car has. Batteries have a maximum recharge rate (~25% of total 'C/20' capacity, that's the normal rated capacity in Ah). So while there is a lot of potentially recoverable energy from the vehicle's kinetic energy, you can only save what you can cram into the batteries safely.This vastly reduces the total energy you can recover.

On the other hand batteries are consumables and spending money on a regen controller (which is not a consumable) would permanently add to your bike's range. Installing a regen capable motor controller won't add additional weight, whereas bigger batteries will. Plus you can only get bigger batteries up to a limit.

Other Regen Considerations:

1. Your motor controller must be specifically say it is capable of regen. Otherwise it cannot be made to do so.

2. Controllers that can do regen are somewhat more complicated to wire up.

3. Regen causes a reversal in the torque direction on the motor, which could potentially slowly loosen the motor mounting bolts.

4. Regen won't work at very low speeds. The voltage produced by a generator is a function of the load and the RPM. (The current produced is a function only of the load.) The vehicle has to be traveling fast enough to generate a voltage larger than the battery bank voltage to enable charging.

Despite all of this, regenerative braking still makes a lot of sense in certain applications. Electric cars are a great choice. They have huge battery banks which can withstand high charging currents. And the cost of adding regen is small compared to the overall cost of a large vehicle.

Regen makes less sense for a DIYer with a small vehicle like myself. The added cost and complexity combined with hardly noticeable returns make for a bad deal.

**Note that regenerative braking is not the same as plug braking. Plug braking is required for vehicles that reverse the motor to change vehicle direction and/or are designed for the electrical braking to supplement the mechanical braking (some golf carts). For applications where the motor only spins one direction (such as a motorcycle), plug braking is unnecessary.

Different Types of Motors: (TLDR: As I said earlier, in theory there are many choices for these parts, in practice not so much due to price and availability of parts. All things considered, for a DIY motorcycle project you will get the most bang for your buck with a PMDC motor. If you have a pile of cash to burn and you want to show off then splurge for the brushless PMDC.)

DC motors consist of one set of coils that moves called an armature located inside another fixed set of coils or (permanent magnets), called the stator. The armature rotates with the drive shaft and the stator 'stays' put. These are the relevant DC motor types:

Series Wound: The field coils are connected in series with the armature coil. Powerful and efficient at high speed, series wound motors generate the most torque for a given current. Speed varies wildly with load, and can run away under no-load conditions. Tougher to do regen with, but it is possible. A controller capable of going this with a series motor will be expensive)

Shunt Wound: The field coils are connected in parallel with the armature coil. Shunt wound motors generate the least torque for a given current, but speed varies very little with load.

Compound Wound: a combination of series and shunt wound. This is an attempt to make a motor that will not run away under no load or if the field fails, yet is as efficient and powerful as a series wound motor.

Permanent Magnet (PM): Magnets replace the field coils found in other motor types. Permanent magnet motors are either brushed or brushless.

*Brushed Motors: Use mechanical brushes to make electrical contact with the rotor at the right time.

PROS: two wire control, replaceable brushes for extended life, low cost of construction, simple and inexpensive control

CONS: periodic maintenance is required, speed/torque is moderately flat, brush friction increases at higher speeds reducing useful torque, poor heat dissipation due to internal rotor construction, higher rotor inertia which limits the dynamic characteristics, lower speed range due to mechanical limitations on the brushes.

*Brushless Motors: These are essentially stepper motors without the cogs that allow steppers to stop the rotor.

PROS: Less required maintenance due to absence of brushes, Speed/Torque- flat enabling operation at all speeds with rated load, high efficiency, no voltage drop across brushes, high output power/frame size., reduced size due to superior thermal characteristics, improved heat dissipation, higher speed range - no mechanical limitation imposed by brushes/commutator, low electric noise generation.

CONS: Higher cost of construction, Control is complex and expensive.

AC Induction Motors: Rotate at a fixed speed and so require the use of a transmission. While they are super efficient and have a high power output, these are typically used in EV Cars rather than motorcycles.

Permanent Magnet AC (PMAC): Expensive and hard to find, but is capable of variable-speed operation and ultra-high motor efficiency.This type of motor is used on professional Brammo electric motorcycles.

Analyzing Motor Specs: The continuous ratings are what is important. Similarly, the 'loaded speed' is going to be your max output RPM. The Rated HP is typically the torque level at which the motor can be run continuously without exceeding the temperature at which the winding insulation beaks down.

Voltage: Higher voltage systems are better, but they cost more. Running a motor at higher voltage generates less heat for a given power rating because the heat generated is proportional to the current draw and a higher voltage will give a lower current draw for the same power output. However as voltage increases so does the need for higher performance insulation, controllers, switches, etc. It is useful to note that input voltage is proportional to motor speed, and the input current is proportional to the motor torque.

48v & 72v are the most common configurations but just for a frame of reference the 'Kilacycle' racing electric motorcycle runs at 374 Volts.

Other Motor Notes:
The physical dimensions will be a significant limitation for a motorcycle project. Making the cardboard mock up was an absolute necessity.

Electric motors have a huge operating speed range, so you don't have to have a transmission. You don't even have to disengage the motor when you are coasting either. The motor won't have any resistance unless it is generating electricity, and it won't generate electricity if it isn't connected to a load. (The controller electrically disconnects the motor for coasting. That is, unless you activate the regen).

There is a devoted section to motors and controllers in the excel document. I used it by plugging in different motors into the calculations to see how they affected the overall performance and required gear ration.

My favorite suppliers for these parts were evdrives.com & electricmotorsport.com. I did NOT like Cloudelectric, who had bad prices and no original documentation.

Hub motor vs Traditional motor

In theory a hub motor does make more sense for energy efficiency, weight distribution, and space saving. Eliminating a roller chain drive could increase efficiency by 2-3%. On the other hand, your hub motor can't take advantage of an optimized transmission system. Oh, and good luck finding one that fits your bike!

I shopped around for a hub motor but there are hardly any options to choose from. Most of the products out there are made for mopeds and would be under powered on a motorcycle. Plus, a wheel is a pretty special part on any vehicle. It is designed to fit within a specific gap in the rear suspension forks and is made for one size of tire. What are the chances of finding a perfect drop in replacement hub motor that matches your specific wheel dimensions?

Controller Considerations:

*There are a lot of EV motor controllers to choose from. Alltrax, Curtis, Kelly are the biggest name brands. Alltrax is easy to use but none of their products have regen.

*All controllers will need programming, but a few controllers will require an expensive special programming device. The Alltrax controllers require a $15 cable that connects to your computer. Programming an Alltrax is easy and involves choosing your throttle input type and how the controller should react to this input.

*My throttle was a variable resistor (0-5000 ohms). You can set the controller so that either the 0 ohm reading or the 5000 ohm reading corresponds to standing still (no output). It is important that you choose 0 ohms as the no output setting! If you choose 5000 ohms as no output then your bike will accelerate if you disconnect the cable!

*Make sure the motor controller is compatible with your electric motor before buying anything. You'll have to find a motor compatibility chart or list provided with the controller. Controllers use pulse width modulation to efficiently control motor speed. Motor and controller compatibility is determined in part by the acceptable range of pulse frequencies.

*Make sure the operating voltage is the same as the motor.
*Make sure the continuous current rating is at least as big as the motor. Beware getting a significantly oversize controller which can deliver more power to a motor than it is capable of handling. (An undersized controller isn't dangerous but it wont fully utilize the potential of your motor.)

Heat Control:

Both the motor & controller will generate a significant amount of heat in use. I made a scoop on the bottom of the bike from a paint drip tray to duct air onto the motor.

Unless your controller has an integrated heatsink or fan then it will need a heatsink of some sort. You can find one online somewhere or you can make your own and use this one simple equation to estimate its overall effectiveness.

Using the excel document I was able to determine the effect of drag coefficient on driving range and power consumption. This gives you an idea of how important streamlining your vehicle is.

My motorcyle didn't have a windshield so I had to mount my own. I measured the position of my head in the upright and crouched positions to determine what angle to mount the windshield at.

You can estimate the frontal area of your motorcycle by photographing yourself on the bike next to something of a know size, I used a piece of paper. Then you open the photo in paint and divide your body into square approximations. Then open the picture in Microsoft Office photo editor and highlight the square sections to get an estimate of pixel count of that area. By comparing this to the area of the piece of paper you can get a good approximation of your frontal area.

Choosing a drag coefficient involves a bit more guess work, but there are charts you can use to estimate based on your sitting position and bike style.

I was also able to determine the effect of weight on driving range and power consumption. This gives you an idea of how important it is to lighten up your vehicle.

I also made note of the tire's rated weight capacity just to be sure that I wouldn't over load it if I added a passenger.

When I was finished with the bike I put each tire on a scale. Interestingly the final curb weight was only 43 pounds more than it was pre-conversion. However, the center of gravity of the bike was definitely higher off the ground than before, resulting in a harder to handle vehicle.

Without question the single biggest limitation on performance of the bike is battery storage. Honestly, what we really need to make a DIY motorcycle conversion more practical is a revolution in battery technology. Here's the low down on making the best use of today's tech.

Lithium vs Lead Acid

Once again, there is much less choice here than there first seems to be. There isn't a good intermediate choice between the really expensive, higher performing, more finicky, lithium based batteries and the tried & true lead acid batteries. (Try finding a Ni-CD or Ni-MH battery large enough for a motorcycle. What are you gonna do, wire up hundreds of AA batteries?)

Here's a rough scale to compare the use of lithium instead of lead acid: Quadruple the battery cost, charger cost, complexity, and performance. You can get a name brand Brammo model for $16,000 so is it worth it to build your own for $10,000?

If you are converting a motorcycle for simple hobby use then lithium batteries are almost certainly going to break your budget. I could not afford the lithium batteries and they scared the heck out of me, so I focused my attention on optimizing the use of lead acid batteries.

Optimizing Lead Acid:

After some research I concluded that Absorbed Glass Matt (AGM) lead acid batteries would provide the best possible performance for this battery chemistry. They are also completely sealed and can be mounted in any position. Unfortunately they are also more expensive than comparable flooded lead acid.

I started by collecting data on every AGM battery I could get my hands on. The result is the chart shown in the photos. This enabled me to make a direct comparison between batteries using the ratio of capacity/weight (Ahr/lbs).

Other considerations:

*Batteries are expensive to ship, so get them in town if possible.

*Lead acid batteries do have a good value in scrap. You can expect to around 15% of their purchase price if you scrap them when your done.

*Pukert Effect- The true capacity of a battery varies with the rate at which you consume power. The faster you discharge a battery, the less capacity it has. This is known as the Puekert effect and is illustrated in the graph above. (I created this graph for a rated 80 amp-hr lead acid battery) The listed capacity of 80 amp-hr is selected because that how much energy the battery has when it is completely discharged in 20 hours, a standardized time interval.

When it comes to selecting a battery size, bigger is almost always better (bigger meaning physical volume and Amp-hr capacity). Here’s why:

1) The larger the capacity of the battery bank, the longer it will take to discharge the batteries, and due to the Peukert Effect as battery discharge rate is increased, the relative usable capacity decreases. This means that an 80 amp-hr battery pack will give you more than twice the range of a 40 amp-hr battery pack.

2) If you were to consume the same amount of power from a large and a small battery, the smaller one will have a lower depth of discharge (DOD) percentage. Using 20 amp-hrs from a 40 amp-hr pack will give you a DOD of 50%, but it would only be a 25% DOD from an 80 amp-hr pack. Battery life is shortened the more deeply it is discharged in each cycle, so increasing the battery bank capacity will increase the usable life of the bank. This is more clearly illustrated in the graph below.

3) As explained in the regenerative braking section, you can only recover as much energy as you can safely cram into the batteries in a short period of time. The larger the battery bank capacity, the higher the safe recharge rate, and the more energy you can recover from regenerative braking.

Interestingly, bigger batteries weren't always more energy dense per unit weight or volume than small batteries. It really varied unpredictably per battery so I guess the variants come from subtle differences in battery design.

Charging: Batteries like to be charged in a certain way, especially when they have been deeply discharged. This type of charging is referred to as multi-step regulated charging, or just 'smart charging'. Make sure you get a smart charger and not a dumb 1 step cram-it-in-there charger. The typical charging profile for a smart charger is as follows:

Step 1. Bulk Charge: up to 80% of the battery energy capacity is replaced by the charger at the max voltage and current rating of the charger. When the battery voltage reaches a 14.6-14.8 volts it jumps to step 2.

Step 2. Absorption Charge: The voltage is held at a constant 14.6-14.8 volts and the current declines until the battery is almost completely charged.

Step 3. Float Charge: This is a regulated voltage and is usually less than 1 amp of current. This will bring the battery to 100% charged or close to it. The float charge will not boil or heat batteries but will maintain the batteries at 100% readiness and prevent cycling and internal discharge issues during extended inactivity.

Selecting a Charger Size: Most battery manufacturers recommend sizing the charger to be at most 25% of the battery capacity (in Ah) or 0.25 C-rate. Thus, a 100 Ahr 48 volt battery pack would take at most about a 25 amp 48 volt charger (or less). Larger chargers may be used to decrease charge time, but risk damaging the battery.

Batteries can be charged across the pack (add up the voltages of each battery connected in series and charge at that voltage) OR you can use 4 chargers or a multiple output charger to connect and charge each battery individually. Battery life can potentially increased by charging them individually vs with a series charger due to the individual attention the batteries get.

Originally I charged all four of my batteries in series with an on-board charger. It may have been too hot or too bumpy though because my charger eventually broke. I ended up replacing it with 4 individual battery chargers from harbor freight. Charging each battery individually gives them more attention and is better for them anyway, but the new charger stack was so big that I had to mount it in my garage instead of on my bike.

Note that charging each of 4 batteries at 12v-10amp simultaneously would take the same amount of time as using a 48v-10amp charger to charge the bank.

Other things to consider:

*Maximum Safe Battery Temperature: 125°F (51.7°C) seems to be the industry standard although some specify 100°F

*To prolong battery life, you should return the energy you use asap after using it. If you don't the battery may sulfate more quickly than it otherwise would have (a gradual and permanent chemical reaction which reduces performance and longevity). Also, when you do start charging the battery, fill it all the way. Undercharging of a battery will still allow sulfation of the battery of the portion not reactivated by the incomplete charging cycle. Note that this is not a 'memory effect'. Lead acid batteries DO NOT have a memory effect. There is no benefit to fully discharging a battery before recharging! In fact, deeper discharges will reduce the number of battery life cycles!

*Batteries are happiest when they are left unused with a trickle charger on them.

*If you can't decide between two great charger options then some cool extra things for a charger to have include: thermal overload protection, thermal compensation (charge differently at different temps), spark proof, waterproof, polarity protection (hooking it up backwards won't break it), and LED indicator lights. I particularly like the 'Battery Tender' line of chargers.

Electric vehicles are the most environmentally friendly when they are powered by renewable energy. I wanted to find out if it would be economically viable to buy a set of solar panels and mount them to a trailer or something.

I calculated that I would need 16,000 watts of panels to fully recharge the bike daily in my area of Ohio. So 100% solar power was not even close to viable here. Maybe if I lived in Florida and had a smaller daily commute.

Even if I just used a small 80 solar power system to recharge a portion of my battery bank the payback period would be 26 years, with a rate of return of 3.8%. There are better ways to save money and help the environment.

Most of the connections on your bike frame should be welded, but the few bolted connections that there are should be guaranteed secure. All critical members relying on fastened connections should have some sort of anti-vibration locking mechanism. Of the many locking options you can choose from, split-washers should not be used here!

Split washers have been experimentally proven to be ineffective locking devices and can even aid self loosening over time. And yet I see these things in use everywhere, so what gives?

In theory split washers (aka lock washers or helical spring washers) are supposed to work by squishing flat between the nut and the mounting surface when you tighten them. At this point the sharp edges of the washer are supposed to dig into the nut and mounting surface to prevent counter-clockwise rotation. In practice a lock washer is unable to gain any purchase and does not actually prevent rotation. The only time a split washer might prove useful would be for fastening onto soft easily deformed surfaces such as wood.

Not to worry, there are better locking options available. Chemical lockers like Loctite, deformed thread lock nuts, and Nyloc nuts should be your everyday go-to locking devices. If you have some money to burn then wedge lock (Nord-lock) washers are probably the best way to go. If you want to guarantee a permanent &strong connection, use a ‘positive locking device’ such as a castle nut or a slotted nut. No amount of vibration will break this kind of connection.

I also recommend reading about these other fastener design tricks here. (Links to support split washer claims also at that link.)

Here is a short video made from the few old clips that I had:

Final note: I've put a lot of time into this project and my future work would greatly benefit from a new camera (clearly) so if you have enjoyed this Instructable then please take a second to vote for me in the Flex PV "MOVE IT CONTEST"! Questions and feedback are always welcome, thank you!