Monday, May 23, 2011

Ebike Update

It has been about a week since I posted on my electric bicycle, and I have learned some more about the kit in that time.

I am firmly within the 'best battery capacity' part of a SLA battery, so the highest range possible is going to be right now. I set out yesterday to see exactly how far I could go on a single battery, with me assisting pedaling 100% of the time. The route was over paved streets and flat bike paths. There were a total of 11 starts from a dead stop. I also had another battery in the rack (a total of 2) so this was a real world test of range. I was able to do 8.1 miles until the battery was dead. That is with me pedaling 100% of the time at a comfortable pace, except on the starts, when I pedaled as hard as I could to gain speed.

Just like before, there is a marked dropoff in power almost exactly at the 5 mile mark. At this mark, the battery actually shows 100% charge. At mile 7, I have to do about 80% of the work pedaling the bike to maintain top speed, and the battery shows 2/3rds full at this point, and 1/3rd full at about 7.5 miles. Just before mile 8, the charge indicator shows red, and after about 500 more feet, the motor cuts off. The charge indicator is definitely not calibrated correctly, but once you are used to riding the bike, the seat of your pants will do a much better job with estimating the battery life than the charge indicator will.

During the week, I also found that 2 of the 4 bolts on the adjuster for rack length came out. Both used plastic locking nuts to keep the nuts from coming off, but both came off anyway. I replaced them with new bolts, lock washers, and thread locker.

I also had a very crappy experience with a pothole. It was about 4 inches across and 2 inches deep with the lip about 30 degrees off from the direction of travel; so it was very small. When the rear wheel hit it, the pothole knocked the rear wheel out of alignment by about 1/2 inch at the top of the wheel. This made the tire rub on the side of the battery 100% of the time when the bike was in motion. I was worried something like this was going to happen at some point because there is very little spare clearance between the rack and the tire. I pulled the battery on that side and put it on the other side then went home where I had my tools.

Still, all in all, I like the kit.

Monday, May 16, 2011

Electric Bicycle Conversion

One of the things I miss is being able to ride a bike all over the place. As a kid, I used to go everywhere on a bike. The freedom to go as far as you wanted to pedal was liberating. But now that I am older (and lazier) I tend to go everywhere in a car. This gets expensive quickly, and isn't as much fun. There's not really a chance to take in the scenery.

Since I am a geek, and enjoy building my own solutions to problems, I was going to try building a motor powered bike. However, when pricing out parts for a proposed solution, I came to an unfortunately very high price for something that wouldn't have a guaranteed chance of working. While I am all for a challenge, I have enough of those right now. To solve my problem, there are several kits out there that can give the satisfaction of building your own, while still being on the inexpensive side. The kits range from a low of about $300 to well over $2k depending on lots of factors.

I read lots of reviews and did a bunch of research before buying a kit. Since I was looking for more proof of concept than anything else, I decided upon the low end of the spectrum, a $300 kit from Currie. This kit comes with a motor and rim combo, one 10 amp/hr 24v lead-acid battery, controller, and the associated hardware. The kit is designed to be put on a non-suspension 26" mountain bike or a comfort/hybrid (front suspension only) bike. The shipping weight is nearly 50 lbs, so it adds some serious weight to the bike. It also puts out 450 watts, which is 3/5ths of a horsepower, and a published range of 11 miles. A normal human can generate spikes of 1.2hp or sustain about 1/10th of a hp, so 3/5ths is a large gain. This seemed like a kit that wouldn't go any faster than my maximum speed pedaling, and be able to help on those long trips at the expense of extra weight. It seemed like a good tradeoff.

I ordered the kit from Amazon. It arrived looking like it was run over by the UPS truck. The box was in shambles, and it looked like it had been repacked several times on it's journey. But a quick glance inside appeared to show everything in working order. I found out later that wasn't the case.

After I laid out everything on my workbench, I started making an additional parts list. The Currie kit requires you to use the 7 speed gear (freewheel), tube and tire from the existing bike that the kit is going on. Personally, I think this is a big oversight by Currie as a freewheel is only about $10 while the tool to take off a freewheel is $8 plus a chain whip which is $11. The tools are not needed to put the flywheel back on. So the difference to a consumer is actually $-9. A freewheel should have been included instead of requiring one be taken off the donor rim.


I was planning on putting the kit on a full suspension mountain bike, which I knew would require some modifications. I picked up the GMC Topkick mountain bike, but when I received it I found that the description and pics on Amazon is incorrect. the Topkick has front and rear disks, not just front, so the Currie kit will not work. I could have retrofitted rim brakes on the rear, but when pricing it out, it would have cost almost 1/2 of what another bike would cost. Since I had already assembled the Topkick and tossed the shipping box, I decided against returning it, and would keep the Topkick as a regular bike.

That left me without a suspension bike to put the kit on. Since I was aiming for proof of concept, I decided to put the kit on my old Huffy mountain bike that I used in college. That bike had seen more abuse than anything you could think of. I rode it all over the place to get to/from class and around town. This included going down stairs and off ledges that were a few feet tall. The rims were beat to heck and bent, and the frame was very 'tweaked', but it still road decently, though steering was more of a suggestion than anything else.

Since I knew this wasn't going to be the final bike the kit was going to be on, I decided against taking the freewheel, tube and tire off the huffy. I ordered a 7 speed freewheel, chain whip and freewheel socket (the last 2 were 'just in case') from Amazon to put on the Currie kit, and a cheap tire and tube from a trip to Wal-Mart. The freewheel spun on with no problem with no tool (the freewheel tightens when you ride, so it doesn’t have to be tightened with a tool) and the tire and tube were no problem. Fitting the new rear wheel assembly on the huffy was no problem either, even though I went from a 5 speed freewheel (original on the huffy) to a 7 speed freewheel.





When I went to start assembling the kit is when I ran into the first thing that Currie needs to improve. The instructions are very bad, as in they are not complete. They state to 'remove the old freewheel'. Ok, how do I do that? I had no idea what a freewheel was until I downloaded the instruction PDF before purchase and looked up a bunch of the terms online. Then, I had to find instructions on how to remove the freewheel. It was annoying. Same thing with all the bolts and connections. The instructions assume you know exactly what you are doing, and say things like 'bolt the rack to the bike' instead of what I would expect, such as, 'use the 10mm diameter 30mm long screws to attach the bike rack to the mounting points by the rear sprocket/freewheel'. The instructions were ambiguous, and incomplete. If you haven't worked on bikes for a long time, and done your own maintenance, expect the kit to take a while to assemble, or have a shop do it for you.

When I put the rear wheel on the Huffy I ran into the first mechanical problem. The Currie rim was bent. It was still usable, but it was off by a good 1/4 inch at one point. This meant it rubbed on the brakes when the rim passed the brake shoe every revolution. I would have to take it to a bike shop to be straightened after testing. Once the wheel was on, I mounted the battery and controller holder, which takes the shape of a rear rack. This is where I ran into another problem, though this time with the Huffy itself. The rack mount holes in the Huffy were not tapped for bolts. I tapped them for the bolts that came with the Currie kit using a Craftsman tap. There was enough room to do this with the rear wheel still on the bike. Once those bolts were in, I attached the down bars to the rear frame under the seat, and the rack was now completed.

the next step was to attach the throttle. Here I ran into another problem, the throttle was also cracked. It was obvious at this point that I would need some replacement parts from Amazon. I disassembled the throttle controller, and re-soldered on 2 of the 4 LED's that indicated the power level in the battery so I could still test, and fixed 2 broken traces on the cracked PCB. I then mounted the controller to the right hand handlebar. When routing the wire back tot he rear of the bike, I found that the lead wire is shorter than would be expected. The wire loom needs to be at least 6 inches longer. As it is, I had to route the throttle loom in a non-optimal location (the absolute shortest route possible) in order for the loom to have enough length to connect to the controller in the rear rack.

The final step was to connect the power feed from the bottom of the rack to the top where the controller is. This strikes me as a poor design. Why have wires on the outside of the bike, especially 2 wires? Why not use a common ground and 1 wire inside the frame of the rear rack? That will keep the wires from being in the way, reduce cost (copper is expensive) and reduce assembly time.

Now that my bike was assembled, it was time to try it out. Unfortunately, when putting the battery into the rack, I found that the case was cracked. It wasn't leaking or anything, but the case is what held the battery on the bike. I had some reservations about disassembling the case to try to fix it, so some duct tape held it together, just in case. I rode around a little without a battery to make sure that the brakes worked, and I wasn't going to run into a tree or something. Once I was happy with the handling, I put in the battery and rode around without turning on the kit. You could definitely feel the extra weight (the battery is about 17 lbs) but it wasn't as bad as I was expecting. I would estimate it at about 10% more effort with the battery on the bike).

Finally, I enabled the battery and used light power. I was extremely surprised at the amount of power that the motor put out. It accelerated only slightly slower than I could do manually. It kept up that acceleration to the kits top speed, which was almost exactly the top speed that the Huffy and I could manage when peddling as fast as possible in the highest gear on flat ground. I rode around the neighborhood for about 10 minutes, then headed back to check out connections, screws, and such.

One of the issues I had in this brief ride is that nearly any bump (such as a large crack in the sidewalk, or going down a driveway into the street) would cause the battery to pop out of the rack by about 1/4". This was enough to raise the battery off the contacts in the bottom of the rack. There are 2 locks that hold the batteries in, which I thought were for security. But they are actually required to keep the battery seated in the rack. A quick turn of the key solved that problem. At first, I thought this was an issue with the cracked case, but the replacement battery did it too.

When I was checking the bolts, I found that the 2 bolts holding the rack to the bike down by the freewheel were starting to deform. They were already deformed by about 1/4 their diameter, and were in the process of sheering off. These were obviously crap bolts. They were the only major failure of engineering that could have been VERY bad that I have found so far. A quick trip to the hardware store and some grade 5 stainless bolts solved the problem permanently.

The next day I also contacted Amazon to see if they could send me some non-broken parts. Unfortunately, they could only send me an entire new kit. I pulled the battery, controller and rim out. During this time I also was able to determine why my last kit had such a problem. The box wasn't closed. The box is a clamshell type box, and there was no closing mechanism (tape, glue, staple or other means). This box was not made to be shipped via a carrier, it was meant to go on a shipping pallet. When I sent it back to Amazon, I taped the box closed. Otherwise, they were probably only going to get parts of the kit back, with the rest lying on the floor of a fedex facility somewhere.

For the next week, I kept testing the kit to see what it could do. I rode all over on different surfaces to figure out how the kit acted. From a power standpoint, it could handle pretty much everything I threw at it. It could go up and down hills on dirt, grass or asphalt. There is one hill near my home that has a 50' rise over about 250 horizontal feet that the kit couldn't do without pedaling help, but the kit made that so much easier that doing it manually. As for range, I came up with about 5.4 miles on a nearly level gravel bike path with light pedal assistance from me. After that, the speed drops off to the point where it is faster to walk. The kit will still continue to assist when you pedal up to a total of about 7 miles though. After 7 miles, the kit is just dead weight. 7 miles is about 63% of the advertised range. I would have hoped the advertising would be closer to actual, but a 5 to 7 mile range isn't too bad. That works out to about $50 per mile for range for the kit price.

Now that I had proved out the concept, I wanted to put the kit on a full suspension mountain bike. The Huffy just wasn't going to survive, and I wanted the kit to have a suspension buffer between the road. It proved difficult to find a suspension bike that was 21 speed, durable, strong enough to hold the kit weight and myself while not being insanely expensive. In fact, I couldn't find one a bike that would work on Amazon at all. I went to a few different stores, and finally settled on the best of the worst, a Magna Excitor Terrain Bike from Target.

The Magna Excitor Terrain Bike was about $80 on sale, and the reviews that I could find were absolutely horrible. However, it had the correct dimensions that I was looking for, has a steel frame, and the majority of the reviews seemed to stem from poor assembly either at the factory or by the store. The assembly I could correct, so I went to a Target and picked one up. After getting it home, I agree that the assembly was horrible. The handle bars were not on correct, the height of the handlebars was beyond the max limit by a huge amount, and the brakes were suggestions instead of brakes. I fixed the handlebars and spent a good hour fixing the brakes. When I got those working, I put some more air in the tires, and went out on the street for some testing. After about 100 feet, the rear inner tube popped. I found out this was because the steel bead actually separated from the tire itself. O_o. I went back to target and they exchanged the tire and inner tube. To top it off, the inner tube they gave me had a hole in it. Arrg. So I bought a few extras.

After I replaced the tire and tube again, I tested out the Magna Excitor Terrain Bike some more. It seemed like I had found everything that was wrong with the bike, so I took off the rear wheel and put the powered one on. Putting it on proved much more difficult than on the huffy. The rear forks have almost no play side to side, so it is extremely difficult to get the wheel seated correctly. After taking the wheel on and off about 5 times so far, I have it down to where it takes about 30 minutes to get back on. That is way too long, but the price I pay for saving about $300 on the bike.

Once the wheel was on, I attached the rear rack. The down bars were exactly the correct length to attach to the seat post, so no problems there. Cycling the suspension showed no binding. I was expecting to have to fabricate some down bars, but so far, the stock ones have worked fine. The full suspension is much smoother than the Huffy. Time will tell on the durability though.






All in all, I am happy with the performance of the kit. The range is low, the instructions need a lot of improvement, and a freewheel should have been included. But besides that, the kit is great. I haven't put on a lot of miles so far, I will need to expand this when I put on a few hundred more miles.


Links to some of the products:
http://www.amazon.com/gp/product/B002RRBDIW - $300 - Currie Power Kit
http://www.amazon.com/gp/product/B002RR7RD2 - $120 - Extra battery
http://www.amazon.com/gp/product/B003B8JYPU - $10 - Freewheel
http://www.amazon.com/gp/product/B001B6RGXG - $7 Freewheel Socket
http://www.amazon.com/gp/product/B00165N6TA - $11 - Chain Whip
http://www.amazon.com/gp/product/B000BT4W2C - $210 Topkick bike