The Future of E-Bikes

2021 November 9 Twitter Substack See all posts

E-Bike technology is amazing. How will it improve, and how will we use it? Thanks to Steve Anderson for his engineering tweets and for answering a tire question. Any mistakes are my own!

Why People Love E-Bikes

The best feeling on a bike is cruising down a hill, barely pedaling, and taking in the surroundings. E-bikes are like that all the time. "How do you know if someone has an e-bike? They'll tell you!" goes the joke.

E-Bike Design for the Rest of Us

E-bikes range in price from $1000 to $10,000. The market is still lacking a Toyota Corolla equivalent.

Radical Design Simplification

Hybrid-electric cars are an engineering abomination. They have both gas and electric powertrains, ensuring they are more expensive and complicated than their competitors. Most e-bikes are the same, carrying both a mechanical and an electric powertrain.

What can we do to get a people's e-bike that has 30 miles of range, requires no maintenance, travels 15-20 mph, and retails around $500?


If we have an electric drive, there is no need for a mechanical one. A German manufacturer named Schaeffler recently released its "Free Drive". Pedaling turns a motor instead of a chain and powers the bike or charges the batteries (electric motors run in reverse generate electricity). The system is about 5% less efficient than a chain drive, but e-bikes use battery power, so it's no sweat for the rider. The motors powering the bike are simple in-wheel hub motors that are cheap and efficient.

Many current e-bikes have fancy carbon belts, pricey mid-drive motors attached to the belt/chain, and expensive torque sensors that sense how fast you are pedaling to modulate the speed. Instead, you could have a Free Drive system with an inexpensive motor that generates electricity proportional to how hard you pedal, providing a signal to the bike.

Motor Technology

DC motors have fallen out of favor for "brushless DC motors." BLDC motors have a longer life, better low-end torque, and are easy to control electronically. These are a good solution, but improvements in motor technology driven by automotive R+D should trickle down.

The trend will be lighter motors that use more common materials. They will hand more control off to software, allowing physical simplification.

Motor Configuration

Hub motors come in two varieties - geared and direct drive. Direct drive motors are usually twice the weight of a geared motor because they need more magnetic material to operate at low RPMs. Their performance on hills is inefficient.

Geared hub motors spin at much higher RPMs and use planetary gears to turn the wheel at rider-friendly RPMs. These motors are cheaper and perform better on hills, but several common design flaws remain.

Basic models might have one motor, while faster models have motors in both wheels.


Who needs brakes, am-I-right? E-bikes might not need brakes. The motor can do light regenerative braking. Because bikes are lightweight with terrible aerodynamics, regenerative braking does not add much extra range. But the 5%-10% gained still matters, and it provides low-temperature braking that doesn't wear out.

For emergency braking AC motors, there is a method called DC injection braking. Direct current from the batteries can be injected into the AC motors causing them to lock up. DC motors can do something called plugging or reverse current braking. These methods might end up being safer since they can be done electronically, like anti-lock brakes in cars.

The concept of deleting brakes is pretty raw. Most DC injection braking and plugging applications are on large, fixed motors. Replacing mechanical brakes with motor braking would require a long history of field use before it would be prudent.


Bicycles are already absurdly efficient. E-bike battery packs are small as a result. Their small size means a less energy-dense battery might be better if it obviates the need for a more complex battery management system (BMS).

E-bikes should use whatever chemistry is available, cheap, and durable. Right now, that is lithium iron phosphate (LFP). LFP has more cycles and handles 100% charge better than lithium-cobalt-nickel formulations. It is also less prone to fires. The battery cells in a bike would weigh 2-4 pounds. The controller, pack, and BMS would probably have more mass.

Whether the battery should be removable for charging is a debate. The heavier the bike is, the more removable batteries matter. I lean towards making it removable even if it adds some complexity and cost.

In the future, designs might use cheaper, less energy-dense battery chemistries like sodium-ion. Others might maintain weight to increase range as battery energy density improves.

Rider Control

The faster the rider pedals, the more electricity they create. The motor controller translates the power the rider generates into a set speed. The rider can pedal backward to signal the bike to brake, again with the bike controllers getting signals from how hard the pedaling is. A Bluetooth-linked phone can display any bike metrics desired, eliminating a screen and extra wiring.


The bicycle design we all know today was known as a "safety bicycle." The center of gravity is low and between the wheels, improving stability and braking efficiency. Larger wheels help reduce bumps. Bicycle frames on bikes without chain drives have more design flexibility, but physics will keep them from straying far.

New frame designs need to accommodate batteries and power electronics within the frame and allow easy installation of those parts. Aluminum is a good compromise between cost and weight. Carbon fiber and titanium are too expensive, while steel is too heavy. The bike needs to be reasonable to lift.


Maybe the worst part about hub motors is changing tires. Bike tires get less than 1/10 the miles that car tires do. It makes sense to spend a little more money on more durable tires. Again, we can trade some rolling resistance for durability if needed. We won't get car tire runs, but they can be a lot better than 2000-3000 miles.

Falling Component Prices

Motors, batteries, and power electronics make up a large portion of e-bike costs. The blitz in automative scale and R+D is dropping the price of these components rapidly. Chains, derailleurs, and belts are not decreasing in cost as fast.

Deleting the mechanical drive train saves weight. The weight of the components should fall with improving technology, as well. Current e-bikes are very heavy unless you are paying $5000+ for a fancy, lightweight design. The weight penalty can shrink.

Design in Quality

Most bikes are sold as toys or as overcomplicated hobbyist contraptions. A Toyota Corolla analog needs fewer parts of better quality.

The current bicycle supply chain optimizes for producing poor-quality bikes in small lots (<500). Our bikes need mass-produced automotive-grade parts that will last a lifetime instead.

Van Moof is a premium Dutch e-bike maker attempting vertical integration. Less pricey manufacturers are likely to emerge from Asia. A bike should last for decades with a few tire changes and one battery replacement at most.

How Cheap E-Bikes Change the Transportation Landscape

Faster Speeds in Pre-Car Cities

Cars need a lot of space. Cities that developed before cars have trouble handling one car per person. Average traffic speeds in cities like New York City are comically slow. A 15 mph e-bike can halve the travel time for trips of a few miles. Because they take up less room, speed and throughput increase.

Cheaper Travel in Car Cities

E-bikes could also have a role in suburbia. Cities designed around cars have much less traffic. But needing a car to get anywhere is a hefty expense. Density is too low for traditional mass transit. E-bikes and other light electric vehicles allow fast, cheap travel to Bus Rapid Transit (BRT) stops. Well-designed BRT has stations like subways and dedicated road lanes, allowing high-speed travel. Lacking a car is no longer a killer.

The combination of bikes and buses is like adding together two red-headed stepchildren. It doesn't sell like trains, but it works.

Avoiding Tyranny

Bikes and cars don't mix well. Protected bike lanes and other accompanying infrastructure is a must for riders to feel comfortable. Rebuilding infrastructure and changing land use requires government action.

Rather than using market mechanisms to reallocate space, governments (encouraged by activists) tend to jump from egregious amounts of free street parking to ripping up all street parking and closing roads to cars. There is a middle road of charging properly for parking and road usage.

Paris, France is undergoing a conversion like this. The mayor has a vision for a "15-minute city" where you rarely leave your neighborhood. The concept offends my American sensibilities of wanting to roam free over wide ranges. When I hear these ideas, I imagine it as a scheme for French elites to keep less desirable population segments in their banlieues.

We need cars and bikes in varying ratios depending on city density.

Lower Oil Demand?

A favorite line of e-bike proponents goes something like, "You can manufacture 300 e-bikes using the batteries from one Tesla Model 3." The potential exists to turn over the vehicle fleet much faster using e-bikes than electric cars. Even if we all own gasoline cars, maybe they sit in the garage.

Practically, e-bikes work best in places cars work the worst. The car ownership rate in Mumbai will not match that of Des Moines if only surface roads are available. Rapid e-mobility growth tends to displace small internal combustion engine vehicles instead of hulking SUVs.

Adoption can't rise until cities build the infrastructure for riders to feel safe. The cities with the best opportunities tend to be the worst at building such infrastructure.

We can build more battery factories and more mines. Des Moines will get mostly electric cars, and Copenhagen will get mostly e-bikes. E-bikes won't rapidly change the current trajectory of oil demand until cities build the infrastructure. That means roughly on the same timeline as electric cars.

Travel Abundance

E-bikes can help foster the opposite of being locked in your banlieue. The current mood affiliation around e-bikes is that of scarcity. We can build more road space, like tunnels, to end scarcity. While e-bikes facilitate short journeys, we will use cars, electric planes, and other emerging technologies for longer trips.

If cars mostly travel in high-speed tunnels and highways, pedestrians and bicyclists can ride the surface streets. The faster travel speed across the spectrum means total travel miles can explode. Take the e-bike to the diner, the Loop tunnel to that esoteric used bookstore across town, and the regional electric plane to see Grandma for the day. Freedom is sweet. There is no reason for vehicle miles to stay stagnant.