Bafang E-Bike Projects

---

Back in 2015, I bought a relatively cheap (290€) Prophete 2-S Trekkingrad. This was an emergency replacement for my trusty Peugeot P6 Iseran that was temporarily out of commission. This was a pretty basic bike with one really neat feature: a SRAM Automatix 2-speed hub. As the name suggests, this is an automatic hub, meaning that you don't shift gears manually; it automatically shifts from 1st gear to 2nd. Sheldon Brown's website has an explanation of how it all works, so click here if you are curious. This bike was fine, although the shift point was a bit too early. I later fixed that with a spring from a ballpoint pen by following the instructions on Dave McCraw's website, but before I got around to doing that, the hub developed issues, so I converted it into a single speed. At that point, I had too many bikes, so I sold it to a colleague for 100€.

Fast forward a year or two and the same colleague gave me the bike back for free due to a move overseas. The bike had been left outside during his ownership and needed a bit of an overhaul, so I decided to go a step further and install a new wheel set with a Shimano Nexus 7 internal hub. I now had a much more usable bike, but still didn't really know what to do with it as I had upwards of a dozen bikes at that time. That's when I started to think about an e-bike conversion . . . 

When it comes to building an e-bike, there are three main types: front hub, rear hub, and mid drive. Each type has its advantages and disadvantages. The front hub type is probably the easiest to install as it keeps everything else stock from the forks back, but power is usually less than the rear hub and mid drive setups, and there's the question of traction when going uphill since your weight would be more towards the rear instead of over the front wheel.  Rear hubs are generally better than front, they are cheaper than mid drive units, and they can be the most powerful of all three options. However, I had this idea that I wanted to be able to shift gears and this would have required me to retrofit a derailleur, which I wasn't keen to do. So I decided to go with the mid drive unit. I bought one of the most common on the market, a 750W Bafang BBS02 that I combined with a 48V Hailong battery.

The build was relatively straightforward, with the hardest part being drilling holes in the frame to mount the battery. My bike didn't come with the normal threaded inserts for a bottle cage. If it had, I would have been able to mount the battery to those. Instead, I had to drill my own holes and use rivnuts or nutserts to mount the battery. Otherwise, the installation process wasn't too difficult, and it helped that Bafang provided a link to a YouTube video explaining the process. In general, installation involves removing the crankset and bottom bracket from the bike and installing the Bafang unit. There is a speed sensor mounted to the rear wheel, a shift sensor, and new brake levers that include sensors to cut power to the motor when braking. The controller/display unit mounts on the handlebars, along with a thumb throttle and control buttons. Had I not had to wait for the delivery of the shift sensor (not included in what I originally bought), I could have completed the process in a day.

I rode E-Bike V1.0 for roughly 18 months before an overseas move prompted me to convert the bike back into its non-electric form. During the 18 months of riding, three things stood out:

1. Speed: Man, this thing was fast. On a level surface, I could get up to 50 kph (32 mph) without any problem. Going downhill, I once reached close to 70 kph (44 mph). Technically, I should have registered this as an electric moped and put a license plate on the back. I could have gone even faster if I swapped out the rear cog and front chainring to create a more aggressive ratio. As it stood, I had a 46T chainring and a 19T rear cog, but this was good enough; I didn't need to go any faster, especially due to items 2 and 3 on this list . . . 
2. Brakes: This bike came with V-brakes and I thought that would be OK. When it was not an e-bike, there were absolutely fine, but when trying to slow down from 50 kph, I found myself wishing that I had disc brakes. No matter how I adjusted the brakes, they just weren't good enough.
3. Suspension (or lack thereof): Again, when it was not an e-bike, I didn't really mind bumps and imperfections in the road or trail surfaces. But when flying along at 50 kph, I felt every crack and every ripple. I started looking at potentially retrofitting suspension forks, but I never got around to doing it.

Otherwise, I was fairly happy with E-Bike V1.0, but I didn't ride it that often. Cycling the 16 km to my office on a non-electric bike would normally take about 45 minutes and I would inevitably be a sweaty mess. This was OK as my office had a shower and I kept a spare set of trousers and work shoes there. On my e-bike, this would take 35 minutes or less and I would be as dry as a bone. This was useful if I had an early morning meeting, but I would feel like I had cheated myself out of a workout. Plus, I eventually had my Aprilia and NIU to ride to the office as well, so the e-bike gathered dust in my garage.

After moving to London, I eventually decided to build E-Bike V2.0. Noting the lessons from E-Bike V1.0, I searched the secondhand market for a "hybrid" bike with suspension forks and hydraulic disc brakes. I found a B'Twin Riverside 900 that fit the bill. It was in decent shape and just needed some mug guards, a rear rack, and a couple of odds and ends. The installation was more or less the same as E-Bike V1.0, minus the brake levers. Since this had hydraulic disc brakes, I needed to keep the existing levers. I opted not to install the brake lever sensors, mainly because I couldn't figure out a good mounting point. Otherwise, it was a straightforward and quick process, especially since I knew what I was doing this time.

Riding around the less than perfect roads around London, I was thankful for the suspension forks and the better brakes. However, the congested nature of London, especially during the morning and evening commute, meant that I couldn't go nearly as fast as I had regularly gone when I lived in Germany. My 21 km commute normally takes about an hour on my non-electric bicycle. With the e-bike? Same time. In fact, the first time I cycled to the office, I was one minute slower than I had been on my regular bicycle two days earlier. So if the e-bike didn't provide any time advantage and didn't really provide a real workout, what was the point? This realisation led me to listing the e-bike for sale after only a month, and it sold not long after.