My Electric Bicycle Project

1. Current Version 2. Design Details 3. Early Versions Eric Peltzer Sculpture My BookGem book holder

our sponsor

DESIGN & COMPONENT DETAILS
v1.4 v1.4

BATTERIES

There is no question that the battery is the single thing, the one and only thing, that is holding back the development and mass proliferation of electric vehicles. At this point in time most of us are still stuck with using lead acid batteries. I am now using B&B 16 Ah batteries, made in China. Originally I used Hawker Odyssey batteries model PC545, 12 lbs. each, which are basically the same as the Hawker Genesis G13EP. All of the batteries I have used are deep-cycling, AGM (Absorbed Glass Matt) sealed lead acid batteries (SLA) and are about the best quality lead acid battery available. They can be mounted in any position except upside down and they are sealed and maintenance free. They are much like the Optima batteries used in many advanced electric vehicles like the EV-1 and the Sparrow but are available in a wider variety of smaller sizes. I got them at Thunderstruck-EV.com. They now run about $50 each. It is nice that there is some competition now in this type of battery and the prices are coming down by about half from what I paid for my first set.

A battery is rated in voltage and amp-hours. Voltage times amps times hours equals watt hours and this is the total power available from the battery. So my battery pack has 2x 12V x 16Ah = 384 watt hours. In other words it will give 384 watts for an hour. Of course this is a nominal rating to be taken with a very large grain of salt. Batteries give a lesser amount of total power when discharged at a high rate. They are often rated nominally at a 10-hour discharge rate, in which case they will give more power and thus seem like a better battery. Bikes actually can discharge a battery pack fully in under half an hour, so batteries will not give their rated output under these conditions.

Another choice for cheap batteries might be Panasonic or Powersonic or Sonneschein gel-cell batteries. They are not AGM. I have been told the longevity is not as good, though the stated juice per weight ratio seems better, and they are available in a variety of smaller sizes if weight were more of an issue than range. Hawker makes slightly larger batteries such as the G16EP like my B&B which have a little more juice but are also heavier. The Hawkers are very well made. Another great thing about AGM type batteries is that they can be charged quickly at much higher amps than other kinds of lead acid batteries. Likewise they have the ability to give high current with no problems. AGM batteries are also unique among almost all battery types in that they retain a full charge for years with no 'charge leakage.' One of the drawbacks of of NiMH batteries is that they discharge about 1% to 2% per day. AGM batteries are also marketed to people with antique cars and other vehicles that aren't used very often so that their owners don't have to worry about having a dead battery when it's time to start them up.

All lead acid batteries have two big drawbacks. They are quite heavy for the total amount of power they deliver. And even these so called "deep cycle" batteries really should not be deep-cycled too often, as repeated deep discharging dramatically shortens their useful lifespan. This is where NiCads have lead acid beat, as they are more power dense and can be run down pretty much all the way without much damage. Of course large NiCads are much more expensive and must be recycled due to the toxic cadmium. So until better batteries like NiMH or Lithium Ion are more widely available in vehicle sizes and less expensive, Hawkers and other AGM type lead acid are about the best we can do. More ruminations on the Great Battery Quest below.

Batteries and 120 amp speed controller mounted in the main frame space.

SPEED CONTROLLER

A speed controller is essential in controlling the amount of voltage and power that is delivered to the motor from the batteries. On a bike with a very small motor it might be possible to just have an on-off arrangement, and I suppose there are other less expensive and much cruder ways to control speed but I feel that a decent controller is absolutely necessary with a powerful motor like this. The controller is a rather expensive electronic black box with no moving parts. A modern 'pulse width modulation'' controller is very reliable as long as it's current limit is not exceeded. However a high ampere capable controller is not cheap and will likely cost about as much as the motor. Curtis controllers are well known in golf carts and other industrial vehicles, however I am using a Scoota 180 amp speed controller from 4QD. - www.4QD.co.uk . This is a small company in England, that specializes in smaller and cheaper controllers for light electric vehicles. The controller with a matching thumb-lever throttle was $248. It is capable of regenerative braking, thought you can easily disable the regen. Previously I tried a Scoota 120 amp model that was nearly identical. It failed after a few months. I was just working it too hard and the output connections overheated and melted. The 4QD factory has a nice policy of repairing their controllers and exchanging them for different ones if you want with a very fair exchange allowance. The new 180 amp model gets a little warm but has been quite dependable for a number of months now.

I originally used an Eagle 80 amp controller from the same company. It was $120. It failed due to overheating but that was completely my fault for using a much too-small controller. If I lived on flat terrain I think it would been fine and would have lasted a long time, especially since I added a finned aluminum heat sink from an audio power amplifier - try a decent electronics store, maybe even Radio Shack. You need to size your controller at about 3 times the rated power of your motor. My motor is rated at 41 amps continuous, HOWEVER it will draw a lot more than that at full throttle. Manufacturers tend to state their controllers at intermittent current limits. In other words the Scoota 120 will put out 120 amps for a short time but should not be run that high continuously.

This controller has regenerative braking. Regenerative braking means that as you coast downhill the controller will use the motor in reverse as a generator to re-charge the batteries. On my route to the market I am on flat terrain for a while and then go downhill for about a mile to reach the store. Conceivably this could mean that I arrive at the store with nearly full batteries, which would make the ride back less likely to drain the batteries completely.

Regenerative braking - nice idea but . . . Notice there is no freewheel on the electric side. Of course the pedal-drive freewheel still works, which means that you can motor without the pedals being forced to turn. However, it you pedal the motor will necessarily be turning, which is a drawback if the batteries die. The regenerative braking function requires a solid connection to the wheel. However, after some time with this bike, using the regen function of the special speed controller, I came to a surprising conclusion against using regen. As I say I bought a controller with regenerative braking. With this set-up, I could basically use the thumb-throttle carefully as a fairly powerful rear brake. The amount of braking could be adjusted to make it less grabby, but I never could get it to the point where I liked it at all. It was obtrusive and difficult to control. I suppose if you could hook it up to a pressure sensitive brake lever, it would work better, however, that is a bit beyond the level of sophistication I think is really necessary on a little electric bike. Luckily the regen function can be defeated which is what I finally decided to do.

Realize, all vehicles are "regenerative" going down a hill. Turning the engine off down a long hill in your car and coasting - this is reclaiming the energy spent to climb the hill. Unless the vehicle is very heavy and/or has great aerodynamics, most of the energy going down a hill will just be used to overcome air resistance, with little left over to recharge the batteries. Normally on a bicycle, very little braking is ever necessary, and top speed down most hills is quickly limited by poor aerodynamics. The net result, I felt, is that the regen on my ebike was just slowing me down and creating heavy backloading on the drive train needlessly. It would be better to put the motor on a freewheel and adopt a method of getting up to speed and coasting as often as possible with the motor shut off. This riding technique is apparently very effective at extending your range, probably more so than having regen.

Many people who deal even with larger EVs and hybrids have also formed the opinion that a pure freewheeling function would be better in many instances than regenerative braking. The regenerative braking function of hybrid cars, for instance, is vastly overstated. Hybrids work by generating their electric power onboard using the gas engine; they have a power generation function built-in, it is central to the entire concept. So why not use it for braking as well? Sure, but in reality this produces very little in the way of reclaimed power. The batteries in hybrid cars are charged at least 95% by the using the gas engine as a generator; and the reason they are so efficient in city driving is simply because they can run at slow speeds very efficiently on predominantly electric power. It is only while driving at higher speeds down long steep grades that any decent power is truly regenerated to the batteries. How much of this describes typical driving conditions?

Coasting with minimal drag from the motor is actually much more efficient. While coasting the vehicle uses no juice whatsoever. It is common practice during mileage or efficiency runs with a hybrid like the Honda Insight, to accelerate to speed, then go into neutral and coast for a ways. This actually produces the greatest mileage and efficiency. On an electric bike it is undoubtedly most efficient to pulse the motor on till you get up to speed, and coast on downhills, or pedal on the flat to maintain speed. Having the motor freewheel would be a big advantage. Unfortunately, a decent ready made right-hand drive doesn't seem to be available, at least for a reasonable price. A minor drawback.

THROTTLE

At first I simply got a $2.00 10K potentiometer from Radio Shack, wired it up to the 4QD wiring and mounted it on a plastic spacer beneath the right-hand hand grip. There was no spring return, you had to turn it down to slow down, but on the other hand it was cheap and simple and the brakes can easily overcome the motor in an emergency stopping situation. Also, it worked perfectly as a cruise control.

Then with the new larger motor controller I got a box-mounted thumb lever throttle. This is just a short-throw 10k pot with a return spring and lever mounted on the end. There are now motorcycle-style twist grip throttles with built in pots widely available at scooter shops for about $50, such as the Magura.

The Scoota controller requires an on-off switch - I got a small rocker thumb switch from an electronics store, and mounted it in the throttle box.

thumb-lever throttle and on-off switch

BATTERY CHARGER

I found a great 24-volt 5 amp automatic charger made for the now-defunct e-bike EV Warrior from All Electronics surplus in LA. It was 20 bucks, it's compact and works wonderfully. What a bargain - I got two. These may not still be available but there are a few sites that deal with scooters and electric bikes now that have similar chargers for $50 to $80. In the new world of electric vehicle hobbyists, a good non-12V charger (or any decent component for that matter) is hard to come by cheaply. For instance, I scoured the surplus electric suppliers looking for a decent, powerful, cheap 12V or 24V motor and basically came up empty.

Before this I used a Sears 12V auto battery charger, charging each battery separately. This worked fine but remembering to go out and switch the charger between batteries was very inconvenient. The EV Warrior charger shuts off automatically, the bike is always charged and ready to ride.

COST AND CONSTRUCTION DETAILS

I used high quality parts and the total cost of all the added components including charger came to about $950. Not cheap, but my new mountain bike cost more than that - and you have to pedal it. You can get a turn-key electric motor kit for around $400-700, or a ready made e-bike for $1000-2000, but of course the motors and batteries are all only 1/3 as powerful as this.

Weight - At 85 pounds, this is pretty heavy for a "bicycle," but I think of it more as an ultralight moped. And it is a true "mo-ped" in that you can still meaningfully pedal it. Plus I can still lift it into my pickup or my car trunk if needed. My old Peugeot moped weighed twice as much at 150 pounds with similar speed capability, and pedaling it was a joke - the pedals were geared really low, purely for starting the motor.

Brakes - The old brakes on this bike were okay for 12 mph but hair-raising at 30 mph. A new side pull "vee brake" as used on decent modern mountain bikes luckily bolted right onto the old brake mount posts. I only replaced the front brake, as the rear posts are in the wrong place. Regular bicycles are too light and top heavy to use the front brake as the main squeeze, but as any motorcyclist can tell you, on any heavier two wheeler the front brake does 90% of the work. The vee-brake kit was about 25 bucks on sale at a bike dealer, and came with a new lever for the handle bar.

Frame - I would encourage people doing this kind of conversion to use a heavy duty steel mountain bike frame. The components can take the added force and weight of the motor system. Actually mountain bike components now are amazingly strong and durable - they are made to take off-road abuse. The wider wheels especially are much better for the additional weight and power. Also, a steel frame is easier to weld or braze motor and battery mounts onto. Stay away from aluminum unless you can figure a way to bolt or clamp everything on without welding. My old GT was too heavy for a good mountain bike these days but just the ticket for an electric motor project. The peculiar geometry of the GT rear frame design actually makes a perfect place for the motor.

Construction - I had to weld a motor mount onto the rear frame behind the seat. The chain just clears the wheel and frame and it was critical to get the motor solidly mounted. With this this powerful a motor, the motor mount must be strong, and it's really pretty important to have a ball-bearing motor. The motor mount is removable from the frame with three allen head screws. I have a TIG welder and lots of metal working tools so it was not a big deal for me but I don't see why all this couldn't be done on a steel bike frame with silver solder or bronze rod and a gas torch.

The batteries also have a very solid welded-on mount with a screw-down top retaining rail. The batteries are heavy and need a good solid mount with shock padding. These batteries are narrow and don't interfere at all with pedaling. In this arrangement they are also low and keep the CG down.

LEGAL ISSUES OF ELECTRIC BICYCLES

At 20 mph this bike is basically legal as an unregistered motorized bicycle in Calif. Although technically the motor is over the limit as far as horsepower, I don't see how anyone is going to challenge it if it only goes 20.

Geared for 30 mph this would technically have to be registered and insured as a motorized cycle or possibly as a full motorcycle. Not sure exactly what the difference is but I believe there is a technical distinction. However, used judiciously, most people are still going to assume that it's just a particularly fast bicycle, especially if it's being pedaled.

Actually I have no problem with a legal limit of 20 mph, although I think motor power should not be specifically limited to make hill climbing more reasonable. Slow is not bad. The wind noise is less and you can actually hear the world. A bike limited to 20 with no registration and no insurance and no driver's license (and no noise and no vibration and no gasoline and no mechanics and no smog checks and no smog and no parking hassles and no car payments . . . ) Not a bad tradeoff, to put it mildly.

I would not recommend riding on bike paths or foot trails. Most of these bar motorized vehicles of any kind, even if it's not posted. It is tempting since the bike is silent and by pedaling, most people will not even notice that you've got a motor. I think speed is the key. If I must ride on a bike path, I just don't go any faster than a gently pedaled bicycle. It is hard to argue with that.

I think eventually there will be a lot more electric bicycles and other small powered vehicles and it will probably become a problem competing with pedestrians, roller bladers, etc. There's already enough friction between these groups on crowded paths and trail. This is another reason why keeping maximum speeds under 20 mph is probably a good idea.

FUTURE IMPROVEMENTS, VARIATIONS AND POSSIBILITIES

It's the batteries, stupid. This bike is great fun right now but I can see a lot of room for improvement. Ah, batteries, ever the Achilles' heel of the electric vehicle. Oh how I hate lead acid batteries but - still stuck with them. NiMH, Lithium Ion, or Nickel Zinc batteries would improve the juice-to-weight ratio enormously, and would probably double the range, which is the single least satisfactory thing about this and all electric bikes. The batteries now weigh over 25 pounds and are the heaviest single component. There are some reasons for hope. The increasing numbers of hybrid cars are generally using NiMH batteries which should trickle down eventually and be very suitable for electric bikes. Other battery chemistries as well have made it to market, thought success always seems elusive. For instance, a company called Evercel (www.evercel.com) actually produced large nickel-zinc batteries for a few years, targeting the electric scooter and bicycle market in particular, though mostly they were only boat trolling motors. However, they went out of business in 2004.

The promise of lithium ion. I believe the lithium battery has the greatest potential of all the up and coming chemistries. It has the most energy per pound and is well proven in smaller applications like cell phones and laptops. Progress with this battery seems to be steady. Soon they will be used in power tools, which use bigger cell sizes than electronic devices. They don't use any heavy metals. There are a number of new types of lithium cells that don't burst into flames when overheated or pierced. Toshiba has even demonstrated cells that can be recharged almost fully in just a few minutes. A practical battery with this capability would almost instantly put battery vehicles on an even footing with gasoline powered ones. Current lithium ions have a serious weakness - they die completely after about 2 to 3 years, no matter what. The research patterns on lithium batteries seem to be the most promising.

Well, what about fuel cells? Of course the concept and demonstrations of this still-experimental technology is very compelling, but there has been so much hype on the subject that it is still hard to say if this is all just incredibly successful marketing spin by the FC developers, as well as clever subterfuge by the auto industry to take the heat off the fact that average fuel economy in this country is decreasing due to ever more popular and larger SUVs. It is so much easier and cheaper for an auto company like GM to run a relatively small early research project on fuel cells, than it is to do serious engineering on real live production-ready hybrid development. Even in 2006 there are still basically no economically practical fuel cells in actual use and even their promoters say that any commercially viable model is at least 5 years off and maybe 10. It is the running joke that the day of widespread fuel cells and the hydrogen economy always seems to be 20 years off - no matter when the question is asked. And then, where do we get all the hydrogen? Seems to me hydrogen is and will remain way more expensive than any petro fuel and I have never seen any ideas to overcome this problem. And we store it in 10,000 psi cylinders in the trunk? Uh, OK . . . Believe me I would love to be proved wrong and initially was excited about this technology like everyone else, but increasingly I am thinking the emperor has no clothes. Maybe a loincloth.

Motors are already pretty much perfected, right? Well, yes and no. My Scott motor is starting to look like a dinosaur, it could stand to lose some weight. The power is actually about right for a bicycle but just look at it, it's too big and heavy. I would love to get a Lemco pancake motor, www.lemcoltd.com - they are light, powerful, and efficient but expensive. This company is now making even smaller motors that would be perfect for high-powered bikes but $800 for a motor is a bit much.

And of course my whole motorization scheme, with the belt and chain, is not terribly clean or elegant. I personally think the "hub motor" is the obvious solution for electric bikes, as used on Lee Iacocca's E-bike and the Wavecrest (no longer made) and others. Heinzmann of Germany seems to still be making a variety of hub motors, as are other companies. Try Heinzmann's site www.estelle.de for some pretty interesting ideas - I guess they are now selling bikes, complete kits, motors, batteries and controllers. However, at this point they are still only about 300 watts = 1/3 HP which is about 1/3 of what they ought to be. Also, Heinzmann motors tend to be a little noisy and expensive.A hub wheel motor replaces the normal wheel hub and obviously needs no other transmission, chain, or belt which is a huge simplification - although of course this can also be a slight drawback in that the gear ratio can't be changed.

A very popular and reasonably priced hub motor made in China seems to be the Crystalyte, though I haven't used one myself. It is is sold in kits where it is basically spoked to a bike wheel rim of your choice and you just replace your old wheel with the motorized one. Apparently they can go on the front or rear wheel. The kits include a pretty slick speed controller, throttle, and brake-switches to cut the power. Often batteries are a separate purchase. The hubs themselves are made in various voltages, powers, and rotational speeds to match various wheel sizes and top speed requirements. This seems to be a decent site with links to dealers in the US and Canada - www.evsolutions.net.

The gearing, for my use, could certainly benefit from a nice simple two-speed transmission. One speed for the hill and the other for the flat. Electric motors have such great torque throughout the RPM range that more than two speeds just isn't necessary unless the motor is severely underpowered. The obvious thing to me would be a hub motor with two or three speeds built in. I have never seen such a beast.

I think with better batteries and motor, this bike could easily weigh 25 pounds less or have twice the range, maybe both. The frame and wheels are no lightweights either, at 36 pounds. That's a good 10 pounds more than my new mountain bike, and it's got a suspension fork. So in short with modern components this type of bike could easily weigh less than 50 lbs. Performance would improve, it would be easier to pedal and you could even carry it on a car roof rack.

How about a solar cell battery charger? They sell these to RV owners. Actually there's no reason why you couldn't throw a solar cell on your roof and charge the batteries all day. Imagine - this is real-world fully solar powered transportation, doable today.

As further work on the existing bike, I would love to put a suspension fork on the front and maybe a suspension seat post. Lights would be nice. At this point this bike would be an amazing transportation unit by any standards, not just electric vehicles. Also, this starts to become a pretty intriguing possibility for a trail bike. I used to have dirt bikes but they are dirty and noisy and environmentally a big problem. A dirt bike is a blast but you're not exactly communing with nature. A completely silent electric trail bike would be much more like hiking aesthetically, and would open up long-range trails - can't wait to get some better batteries and go up to the mountains.

I can also see going to a higher voltage system. 36 or even 48 volts would be a lot better and keep the current draw down. I started with a 12 V system for simplicity's sake but the power was low and the current draw was high. I won't go into the basics of electric power except to say that lower voltage is bad because it leads to higher current which creates more heat and requires a much bigger motor controller. Controllers are basically sold by amperage or current capacity, and high amperage controllers are much more expensive than lower amperage ones. The same controller can put out twice as much power at the same amperage level if the system voltage is doubled. I have even considered simply adding another battery on this bike and just over-volting the motor to 36 volts. The motor is probably able to take it (though I'm sure the manufacturer wouldn't recommend it) and my controller can easily be switched to 36 V. The bike would be heavier, faster, and have better range. It's just a trade off. Intriguing. (Note that most controllers have voltage limitations and must be altered or jumpered to make system voltage changes.)

That's what's neat about electrics for the tinkerer - you can change motors, change batteries, change controllers, change gearing - it's all so easy and interchangeable compared to an internal combustion vehicle. Imagine a having an old Vespa scooter, and you think one day "Gee, wouldn't it be interesting to put a nice new Honda 200cc engine in that thing" - you'd have to be half nuts to even consider it.

Turnkey solutions. So you don't have a TIG welder and a metal lathe? You just want to bolt something onto your bicycle? Since I made this bike there have been some developments in the electric bike world. The popular Curry USProDrive bolt-on kit that I eschewed as too undepowered was always a decent kit that cleverly fits onto almost any bike, and includes a charger, speed control, throttle, battery and complete mounting hardware. This kit has a built-in freewheel on the motor which is nice. The batteries are not AGM quality, the stock controller is jerky and mounted inside the motor where it's hard to keep cool, and the drive train is not terribly robust, but all in all a good kit. Also, it is recently possible to make it more powerful. There are two ways of getting 600 watts out of this kit. One is a hop up from Thunderstruck-EV which takes the stock motor and rewinds it for 36 volts and adds an external controller which is less prone to heat shut-down. I hear it works well, thought you do need to get an additional battery to run the 36 volts. The second way - Scott at EVDeals sells these kits and will exchange the stock motor for a different 600 watt motor. This is still 24 volts like stock and the controller is built into the motor. I believe this is available in two different gearing setups depending on whether you want more top speed or more powerful hill-climbing. I may try one of these solutions at some point, provided nothing better crops up. I have not tried either one of these kits so cannot absolutely recommend them, but they seem like the only good alternatives to building everything from scratch like I did.

MISCELLANEOUS THOUGHTS

I love the idea of electric vehicles but the practical electric car is still not quite a reality. The electric bike to me is now a real-world, affordable and practical solution. The limited range is not a big deal for errands around town and short commutes. Given that really nice batteries are going to be expensive, it makes so more sense to use 2 small ones on a bike than try to pay for 30 big ones for a car. Most people won't justify a limited range full-size electric car just for groceries and errands, but many people could throw an e-bike or two in their garage without thinking twice.

Grown adults giggle when they ride my bike. It is very quiet - it's like some genie is pedaling for you. The silence is very important - it just completely changes the whole picture. You can actually hear things and stop and pull up onto the sidewalk and talk to your neighbors when you ride down the street. I have pulled up directly into my ATM booth and people hardly give me a second glance - not even thinkable on a motorcycle. Electrics are not intimidating like a motorcycle or even a scooter - to most people, they're just a bicycle you don't have to pedal. Theres no clutch, no gears to shift, no kickstarter. They aren't just environmentally friendly - they're neighborhood and people friendly.

Another amazing thing to me: this is not rocket science technology. This is lead acid batteries powering a DC motor on a steel bike frame - this bike could pretty much have been made 80 or 100 years ago! Why has it taken so long for e-bikes to take off?

I offer these design and component ideas to anyone who can make use of them and encourage anyone with a garage and some tools to make an electric bike or buy a kit or a ready-made. At my house we have a very fast motorcycle, a car and a pickup, but the e-bike is the most fun - I use it almost every day.

And the most amazing thing that I have learned after all of this - you really don't need 3000 pounds of gasoline powered steel-rubber-plastic-and-glass, $30,000-120-mph marvel-of-modern-transportation-engineering to go a mile and a half for a quart of milk, a loaf of bread and a newspaper.

- Eric Peltzer

> next: early drivetrain versions - simpler and easier to build

- SPONSORED LINK - THIS PAGE BROUGHT TO YOU BY -

Please visit our sponsor - The World's Best Little Book Holder

1. Current Version 2. Design Details 3. Early Versions Eric Peltzer Sculpture My BookGem book holder