CUDAcores89
1 W
- Joined
- Aug 28, 2016
- Messages
- 51
So far I have been enbarking on a quest to build an ebike battery that has all of the following parameters. My system is a 36v 12.5A ebike rear hub motor I purchased from ebay 4 months ago and the system has been working really great. Although the problem is my battery is built from salvaged laptop cells and while it does work, the laptop pack does not have nearly as much "punch" during acceleration as my small lipo pack I use for emergencies. Therefore I have been exploring other options and other battery technologies and chemistries to develop a seemingly unicorn battery that will somehow be able to meet all of the following parameters:
1. Low cost
2. Long range (when I need it)
3. Plenty of "punch" during acceleration
4. long lifespan (thousands of cycles)
5. Safer than HK lipo
6. Ability to use an RC hobby charger (sorry, no battery murdering systems for me)
I think I may have just found a way to do it.
First, let's talk about how I am going to solve each problem:
1. Low cost: What my plan is for now is to purchase new laptop batteries from ebay in lots of 10. I recently purchased a single laptop battery on ebay for $10, and they contained 6 Genuine Samsung 28A cells inside. I tested the cells by charging and discharging all of them with my opus BTC-3100 and sure enough, they came out to 2800mah. The same guy is selling said laptop batteries in lots of 10 for $50, so This is how I am going to keep costs low.
2. Long range: This problem is going to be solved by simply using a ton of cells. If I place 10 cells in parallel, I will achieve a capacity of 28AH maximum capacity. This will keep discharge rates under 0.5C (my hub motor draws around 12.5A most of the time) therefore extending the cycle life of the battery itself.
3. Plenty of "punch" during acceleration: How I am solving this one is actually kind of interesting. Instead of building a typical 10S battery, I am actually going to build a 12S battery. Two days ago I crached open my ebike controller and after checking the MOSFETS and input capacitors inside, it can take up to a 50v input. Therefore if I build a 12S battery pack instead of a 10S battery pack and purposefully undercharge each cell to 4.0v per cell, this will give me a fully-charged voltage of 48v fresh off the charger, just under what my controller can take. This will mean that my new battery will have 48v fresh off the charger, 43.2v nominal (vs 36v) and 36v completely flat. Therefore because the voltage is higher, I will be able to get much more acceleration while still staying under the parameters of my 36v system. This will also do another thing which I will talk about later in the next point.
4. Long Lifespan: This is an important one. I not only want the battery to be cheap, but also last thousands of cycles. As noted in point four, I am only going to charge the cells to a maximum of 4.0v. Doing this I will take a 20% hit in capacity, but that will be made up by the fact that the battery will still be 22.4AH, and I will still be able to get a good 33 miles range out of the pack.
5. Safer than HK lipo: This should be an obvious one. I currently ride my bike to and from work and it would be very cool if I could take out my battery and bring it inside every day to charge. I am much more willing to bring in an 18650 pack into into my work than a lipo, since I have personally experienced two lipo fires first hand, while I have yet to see an 18650 go up in flames. Yes, I know 18650s can still explode, but at least in my experience they are safer than lipo.
6. Ability to use an RC hobby charger: This one is very important to me. I know that there a lot of folks out there that like to use a BMS, and that's perfectly fine. But in my experience, the cheap BMSs that yo can buy from ebay or aliexpress are absolutely garbage. I spent my summer of 2016 this year fixing a few of peoples hoverboards, and the most common problem I saw was a cooked BMS. And many of the BMSs in the hoverboard packs were the exact same BMS I have seem many ebike batteries use. In order to integrate many of the safety features a BMS provides, I am going to be doing a couple of things:
over-voltage protection: I will be charging with an RC lipo hobby charger
Under-voltage protection: I will be using 2-8S cell loggers that will beep when any cell hits 3.0v. This is much better than a BMS, since some of the cheap ones either cut off at a really low voltage per cell or only cut off at a low voltage for the entire pack.
Over-current and thermal protection - This one is going to be solved by using PTC fuses. PTC fuses are basically resettable fuses that when they heat up, they go open circuit and cut the power. If my battery pack get's very hot, the PTC fuses will get hot too and cut off the pack. If I try discharging over the hold current of the PTC fuses, the PTC fuses will heat up and cut the power again.
Now let's look at the bill of materials:
20 laptop batteries (+ an extra): $110
Solid core wire: $7
200G solder: $5
PVC heat-shrink: $5
PTC fuses: $11
Electrical tape and heatshrink: $3
Total: $141
However, there is only a single disadvantage I can think of from doing this: The pack will he quite heavy. It will likely weigh around 15lbs fully constructed. Although this really isn't a huge deal to me, since The pack will be installed in a large triangle bag most of the time, so weight won't matter to me a ton.
What do you guys think. I think this could actually work. I would be able to build at the very least a much nicer pack than I have now, at a low cost, that would be user-manageable, and last thousands of cycles.
Anything I haven't thought of?
[moderator edit to fix title spelling]
1. Low cost
2. Long range (when I need it)
3. Plenty of "punch" during acceleration
4. long lifespan (thousands of cycles)
5. Safer than HK lipo
6. Ability to use an RC hobby charger (sorry, no battery murdering systems for me)
I think I may have just found a way to do it.
First, let's talk about how I am going to solve each problem:
1. Low cost: What my plan is for now is to purchase new laptop batteries from ebay in lots of 10. I recently purchased a single laptop battery on ebay for $10, and they contained 6 Genuine Samsung 28A cells inside. I tested the cells by charging and discharging all of them with my opus BTC-3100 and sure enough, they came out to 2800mah. The same guy is selling said laptop batteries in lots of 10 for $50, so This is how I am going to keep costs low.
2. Long range: This problem is going to be solved by simply using a ton of cells. If I place 10 cells in parallel, I will achieve a capacity of 28AH maximum capacity. This will keep discharge rates under 0.5C (my hub motor draws around 12.5A most of the time) therefore extending the cycle life of the battery itself.
3. Plenty of "punch" during acceleration: How I am solving this one is actually kind of interesting. Instead of building a typical 10S battery, I am actually going to build a 12S battery. Two days ago I crached open my ebike controller and after checking the MOSFETS and input capacitors inside, it can take up to a 50v input. Therefore if I build a 12S battery pack instead of a 10S battery pack and purposefully undercharge each cell to 4.0v per cell, this will give me a fully-charged voltage of 48v fresh off the charger, just under what my controller can take. This will mean that my new battery will have 48v fresh off the charger, 43.2v nominal (vs 36v) and 36v completely flat. Therefore because the voltage is higher, I will be able to get much more acceleration while still staying under the parameters of my 36v system. This will also do another thing which I will talk about later in the next point.
4. Long Lifespan: This is an important one. I not only want the battery to be cheap, but also last thousands of cycles. As noted in point four, I am only going to charge the cells to a maximum of 4.0v. Doing this I will take a 20% hit in capacity, but that will be made up by the fact that the battery will still be 22.4AH, and I will still be able to get a good 33 miles range out of the pack.
5. Safer than HK lipo: This should be an obvious one. I currently ride my bike to and from work and it would be very cool if I could take out my battery and bring it inside every day to charge. I am much more willing to bring in an 18650 pack into into my work than a lipo, since I have personally experienced two lipo fires first hand, while I have yet to see an 18650 go up in flames. Yes, I know 18650s can still explode, but at least in my experience they are safer than lipo.
6. Ability to use an RC hobby charger: This one is very important to me. I know that there a lot of folks out there that like to use a BMS, and that's perfectly fine. But in my experience, the cheap BMSs that yo can buy from ebay or aliexpress are absolutely garbage. I spent my summer of 2016 this year fixing a few of peoples hoverboards, and the most common problem I saw was a cooked BMS. And many of the BMSs in the hoverboard packs were the exact same BMS I have seem many ebike batteries use. In order to integrate many of the safety features a BMS provides, I am going to be doing a couple of things:
over-voltage protection: I will be charging with an RC lipo hobby charger
Under-voltage protection: I will be using 2-8S cell loggers that will beep when any cell hits 3.0v. This is much better than a BMS, since some of the cheap ones either cut off at a really low voltage per cell or only cut off at a low voltage for the entire pack.
Over-current and thermal protection - This one is going to be solved by using PTC fuses. PTC fuses are basically resettable fuses that when they heat up, they go open circuit and cut the power. If my battery pack get's very hot, the PTC fuses will get hot too and cut off the pack. If I try discharging over the hold current of the PTC fuses, the PTC fuses will heat up and cut the power again.
Now let's look at the bill of materials:
20 laptop batteries (+ an extra): $110
Solid core wire: $7
200G solder: $5
PVC heat-shrink: $5
PTC fuses: $11
Electrical tape and heatshrink: $3
Total: $141
However, there is only a single disadvantage I can think of from doing this: The pack will he quite heavy. It will likely weigh around 15lbs fully constructed. Although this really isn't a huge deal to me, since The pack will be installed in a large triangle bag most of the time, so weight won't matter to me a ton.
What do you guys think. I think this could actually work. I would be able to build at the very least a much nicer pack than I have now, at a low cost, that would be user-manageable, and last thousands of cycles.
Anything I haven't thought of?
[moderator edit to fix title spelling]