TexasElectric - eBike build

[TexasElectric]

Hi, my name is Adam.

I am in the early stages of building my first electric bike. I have been playing with electronic motors and batteries for many years, only in the past year or two I've gotten into circuitry and making my own batteries and super capacitors.

For my ebike I have chosen to use 32650 LiFePO4 cells. The reason for this is because of LiFePO's inherent safety over the more popular 18650 Lithium Ion cells. I purchased 80 32650 LiFePO4 cells. I plan to use 75 of the cells. Each cell weighs approx 145 grams. (75 cells x 145 grams = approx 24lbs)

Each cell is 3.2v 5500mah. 75 cells in a 5p x 15s or 15s x 5p configuration to give me 48v with 27.5mah.

Question 1:
If I connect the cells in a 5 parallel x 15 series configuration, is that the same as a 15 series x 5 parallel configuration? Is there a benefit of one configuration over the other?

Question 2:
What type of BMS (Battery management system) or PCB (Protection circuit board) do you recommend I use?

For the motor I plan to purchase a 48v 1500w direct drive rear hub motor. I plan to buy the motor and all associated accessories as a kit on eBay.

For the bicycle, I will be purchasing a used 26" mountain bike from my local bike shop. My budget for purchasing the bike is $400-$500. -just a side note, i've owned big box store bikes and currently own a high end name brand bike. There is a HUGE difference between the two, at the end of the day you get what you pay for.

I will store the battery and motor controller in a rear rack mounted bag that will sit on top of the rack.

Technical:
I am an avid cyclist, I ride weekly by myself and with a club. On my bicycle I have what's called a power meter. Bicycle power meters are very accurate and very expensive ($xxx to $xxxx). For those that don't know, a bicycle power meter is a strain gauge that measures the torsion (or flex) of the surface it's mounted on. In the case of my power meter, it's mounted on the crank arm. The power meter calculates the RPM of my pedaling cadence and the strain (or flex) of the amount of force I apply to the pedals to give me a number represented in "watts".

My bike weighs approx 14lbs, I weigh approx 150lbs. When fully loaded with water, spare tubes, pump, etc... I estimate the total weight to be approx 170lbs.

When I am riding at a steady pace of 18mph with a 5-8mph head wind, I can hold 18mph by putting out 175-225 watts.

If I estimate this eBike i'm building 'with me included' to weigh: (approx 225lbs)
- Bike 17lbs
- motor 15lbs
- motor controller 1lb
- batteries 25lbs
- me 150lbs

Using the data I have from my power meter and the estimated weight/power of my proposed eBike. I estimate I would need 250-300 watts to maintain 18mph on a relativity flat surface with a small head wind. With an estimated range of 80 miles to 95 miles? (48v x 27.5ah = 1320 watt hours / estimated watts needed to maintain 18mph 'ie. 250-300 watts')


I'll post pictures and videos as things progress.


I'm sure i'm missing a few things, but that's what i've got figured out for now... i've got to go to work.

[vskid3]

Have you been on Endless Sphere? That's the place to go if you looking for the gritty details of ebikes.

You'll want to do groups of 5 parallel cells in series with each other. Makes balancing/protection easier and I believe there are several other benefits that I can't think of right now. On a related note, "48v" LiFePO4 batteries are 16s, not 15s. Helps keep you from hitting the low voltage cutoff of the controller if you're drawing a lot of power (probably not an issue with such a big pack), gives you more watt hours, and keeps the voltage higher throughout the discharge, allowing more power.

From my experience with my ebike and a ~16lb 48v 15Ah LiFePO4 battery, I highly recommend not mounting the battery to a rear rack. All that weight plus the weight of the motor back there makes for a bike that handles like garbage. If you decide to use a rear rack anyway, I would put half of the battery on each side as low as possible and do whatever you can to make the rack stiff.

Here's my ebike's build thread: http://ecomodder.com/forum/showthrea...ike-28487.html

[jjackstone]

What type of motor? Hub or frame mount? Frame is likely more efficient but hub needs less maintenance. Do you ride in a tuck? Your power levels seem a bit high for a 14 pound bike. You might consider a recumbent bike since they can be more efficient due to less aerodynamic drag. I ride a trike and have about the same speed as you with the same average power output you mentioned. I agree with vskid about keeping the batteries as low as possible. If you're going to run a two horse motor you'll want beefier wheels(at least the drive wheel) to be able to handle the additional torque on them. If you use a hub motor then you'll need a torque arm to attach the motor to the frame so it can't spin out of the dropouts.

Check out Justin at ebikes.ca for controllers and power monitors. I haven't been there in a while but when I had a problem with one of his controllers many years ago he took care of the problem immediately. He also may be able to help you with charging and cell balancing which you need to do with Lifepo4. Good luck with your build. Hope you document it here for us.
JJ

[Grant-53]

Weight balance and aerodynamics are key to a comfortable and efficient ride. A battery box in the frame helps with both. There are a couple options for aero bars. One is to move the bar end grips inboard, about 10 in. apart. The other is to make or buy TT style aero bars. If they are 7/8 in. diameter the throttle should mount easily. A true streamlined body is not on the market yet. You can use the HPV shell design program at recumbents.com to design a shell for an upright bike too.

[TexasElectric]

Awesome feedback from everyone!

vskid3
I have heard of "endless sphere." I've been a long time reader of this forums automotive aerodynamic threads (very interesting stuff), and when I discovered that there was a place to discuss e-bikes I figured why not...

I kinda figured that 15/16 cells in series would be the ideal configuration, I just wanted a second opinion. My thinking is that each cell has a nominal voltage of 3.2 volts, but in reality when fully charged the voltage should be a tiny bit higher (ie. 0.1-0.3 volts higher). With a total of 15 cells in series, (taking into account the tiny bit higher voltage when fully charged) I should get 49.5 volts to 52.5 volts. I could be wrong about the cells being a "tiny" bit higher when fully charged, i'm just going by what I see on YouTube videos. That could be a higher voltage that lasts temporarily once the battery is disconnected from the charger.

I've also given more consideration to where I am going to mount the battery. I am thinking that the more weight I have on the rear wheel the faster the rear tire is going to wear. Plus, as you mentioned the instability/overall poor weight distribution of all the weight being in the rear.

BTW, i've been reading about your e-bike build. It looks very cool. I am glad to see that other people are also using LiFePO batteries over Li-ion. (not that there's anything wrong with anyone using Li-ion aside from explosions and fires <--- Just kidding... But seriously be safe/careful with Li-ion)

jjackstone
I will be using a rear wheel direct drive hub motor. I plan to ride the bike is a full upright position (the more comfortable I am the better).

Recumbent bike's look pretty cool, but I would prefer to stay in an upright seated position.

I will look into ebikes.ca (i'll tell them jjackstone sent me )

Grant-53
I am now working on a new plan/design to mount the batteries just below the top tube and down the seat tube (ie. the inside of the frame. not literally the inside, you know... the outside inside)

I spent about 8-10 years of my childhood and now into adulthood (now at the age of 30) riding mountain bikes. 3 years ago I got into road bikes and haven't looked back. With that being said I do prefer a narrower handlebar.

I just googled "HPV shell", that's wild. I've watched many YouTube videos on that stuff. I've got two reasons (or excuses) for using a "traditional" bicycle.
1. It's less obvious to law enforcement that I have an e-bike (should I find myself doing 30mph uphill). Also less obvious to the public/other cars on the road.
2. I have special access to the back of the building I work at. I am able to badge into the building from a service entrance that is relatively out of sight and walk into my office with the bike. (Security doesn't care what I do, but I cant go walking around cubicles with a bicycle or an HPV )

I am sure that HPV's are legal on the road (as I have heard many YouTube videos talk about), but I would prefer to be as under the law enforcement radar as possible (ie. less conspicuous) when riding on the road.

Update
I stopped by my local bike shop yesterday to test ride a few different mountain bikes. To my surprise I was able to find new name brand mountain bikes priced around $500 (within my budget). At this time I am leaning towards the Specialized RockHopper. This bike is a 29r. While I would prefer a 26" wheel, I think the benefit of a larger wheel is less power required to maintain speed? (not to mention a smoother ride).

My plan is to use the pedals to get the bike to about 10mph then ease in the electric motor to speed up to and maintain 18-20mph. And of course, assist the motor when climbing up steep hills (of which I'd only have 2 short steep hills on my regular route).

Since I plan to use this bike as a replacement for my car to go to and from work, I will be wearing my work clothes (button down shirt and khaki pants). I can always wear something else (ie. lycra) and change when I get into work, but my idea behind using an e-bike over my regular bike is that I wont be sweaty when I arrive at work.

[TexasElectric]

Step 1 - building the battery:

I watch plenty of YouTube videos of people building battery packs with 18650 Li-ion cells. I see that they use thin strips of nickel to connect them together. Then they either solder or spot weld the nickel strips to the ends of the cells.

[TexasElectric]

I have a problem with this method... As mentioned in many of the videos "if one cell fails, it can cause a problem for the whole pack." Some of these people shrug their shoulders when this happens and either chuck the whole pack or try to salvage it. It's a lot of work to salvage the pack for one cell (depending on the size of the pack). Not to mention a worst case scenario of one of those cells (specifically an 18650 Li-ion cell) going rouge and exploding/catching fire and starting a chain reaction with the other cells.

[TexasElectric]

Here's my plan...
I purchased 4 of these "D" cell battery holders on eBay. They are not tall enough to accommodate the 32650 LiFePO cells, so I will cut the holder in half (see pictures in next post).

[TexasElectric]

Sorry, I had to have 5 posts before I could upload pictures.

Picture of the 32650 cells (80 cells total)





My plan...







By using these "D" cell battery holders I am able to easily test and replace one cell should it go bad.

[vskid3]

How far are your commute and other rides you would be doing? I would go with the smallest battery that'll get you where you need to go and be able to handle powering your motor, anything bigger will just be extra weight. And why go 20MPH when you have the power to be going 25 or 30?

When I built my bike, not many options for good pre-built batteries existed. I went with my Ping LiFePO4 because they had good reviews and if something went wrong I wouldn't have a chemical fire on my hands. I also wanted enough range to be able to do all my commuting for a typical day on one charge. It's a brick and the worst part of my ebike experience has been trying to find a good way to bring its fatness with me. Fast forward to today and there are several options for 18650-based batteries that are pretty safe and better options for chargers. Right now I would go with a Mighty Mini Cube battery and a Cycle Satiator charger (which I do have) for my current needs. Less than 8lbs and I could go around 10 miles at top speed and recharge in an hour and a half.