DIY - LFP battery packs.

[curtis]

11 gauge or 9 gauge ?????

I am doing cell recovery with modules from Battery Clearinghouse. I am building LFP packs with used A123 - 26650 cells. Looking at three 5S - 8P packs in series. One 15S - 8P pack simply takes up too much space. I can fit two on the bottom bar of the triangle and one on the top. For dual suspension , one on the top bar and two in a front or rear basket. Total battery weight should be somewhere between 20 and 25 pounds including the boxes. I know LICO or lion 18650 cells would weigh around 10 pounds or less for a 15 Ah to 20 Ah pack.

I choose LFP as 100% safer as no chance of thermal runaway / fire / explosions'. Also even used LFP at around 80% capacity should outlast brand new LICO/Lion three to five times over saving $$$$$ on battery replacement and recycling.

My first 5S 8P pack will be a little heavier than the other two as got carried away with the tape as I chose NOT to go with cell holders. I wrapped thin strips of electrical tape around top and bottom for about 1/16" space between cells. I also use red and black duck tape and am building cell banks 1S - 8P. I can cut down on the amount of tape though am currently charging 5 in parallel with a 6S - LiPo charger. My 400W Mega charger. Parallel charging is great to start out as all cell banks will be the same.

I never did learn how to spot weld and use a Weller soldering iron with a large tip bigger than a flat head screw driver. To make it work with no shorts I use 500 degree heat tape over the cell rings that separate the positive and negative. I also save the tabs when extracting from the modules.

The next step is the series connections and am using 5S active cell balancers. I tried 10 gauge and 8 gauge wire but broke apart. I bought standard 16 gauge home extension cord. I stripped off the insulation and combined three strands. I googled it and said was equal to 11 gauge. That would be 30 amps continuous. HOWEVER upon inspection compared to 10 and 12 gauge it appears like it is thicker than 10 gauge and looks like 9 gauge.

This is important as most of my e bikes have controllers < 40 amps but I do like to build dual motors that would require about 45 amps continuous.. I also have a 3 kilowatt chain drive with a 70 amp controller. At 50V it should pull 2,500W which is 2,500 / 50 = 50 amps. For normal 30 to 35 mph operation it should pull somewhere between 35 and 40 amps continuous.

That brings me back to the question of 11 gauge or 9 ? Will 50 amps work ?

If not I have 20S - LTO. Lishen 18 Ah cells capable of pushing about 800 amps but am using ring terminals so would probably melt > 70 amps continuous but should be ok for 50 < 30 seconds and 40 to 45 continuous. I could upgrade to bus bars if needed. I would rather run the LFP pack though as about 20 pounds lighter.

On the pics showing the three vertical 12 is far right with 10 gauge center and the three 16 gauge twisted together on the left. I think by twisting them together I get more surface area than the 10 gauge wire and I think it should handle 50 amps easily. I would really like to hear other opinions on this however.


Thanks.

Curtis out.

 

[curtis]

Yea. I really wanted to use one piece of stripped wire connecting 10 cells in series but cold not manage to strip a wire that long. I ended up soldering the series connections. I used liquid tape to prevent any chance of shorts. The power wires were also soldered and very stiff. To prevent breakage when hooking and unhooking I taped the wires to the sides of the battery case which was a Hyper tough plastic storage box I hacked up with a hacksaw blade .

It worked but really wanted all clear lexan 1/4" but way too expensive.at Lowes. Almost $50 as only one size. 4 by 4 feet I think. All my battery cases need to be clear and no tape on the enclosure. These are temporary until I get a deal on lexan. The LTOs as well. They are in hyper tough tool bags for now.

Doing a capacity test on those A123 cells. 5S - 8P. If I get > 10 Ah which I should I am thinking about a series connection with the 12S - 2P Headway pack. 3.3V * 17 = 56.1V. A big difference from 49.5V - 15S.

It might work with 48V hub motor controllers and brush as long as the CAP limit is 60V. Nice performance boost if it works.

I have an 800W and 1,000W hub motor controller that are 48V. One brush controller also. The 1,500W brushless controller for the 1,800W brushless motor is 38A I think. Then the 3 kilowatt brushless 70 amp controller. All could get at least 3 or 4 mph more maybe. As long as it can do the amps without too much voltage sag.

LFP tends to have a larger voltage drop under load like SLA. Lico / Lion 18650 cells less voltage drop. However Lion is not the best chemistry for better performance. The best would be my Lishen 18 Ah - LTOs. Like that old Mike Tyson add selling batteries back in the 90s "NOW YOUR PLAYING WITH POWER ".

I have a 20 Ah - 13S 18650 factory lithium ion pack in storage. 30 amp BMS and rated at 1,200W. My 1,800W brushless motor is the test motor with a 1,500W 38 amp controller.

1, My old 12S - 5P - LIFE pack - 25 mph.

2. The 20 Ah 13S Lion pack - 28 mph.


3. 20S - Lishen - LTO - 32.5 mph.


The old LIFE pack was half dead old cells from battery hookup. K2 energy military overstock I think. About 30 years old or at least 20.
Top speed 25 mph. I think they were 3.8 Ah cells but SAG was horrible.

I recycled the old LIFE. I hope these A123 cells are newer. 2.5 Ah * 8 = 20 - 30% = 14 Ah used. They state 80% capacity but I think that is exaggerated. Testing will reveal the truth.

The 12S - 2P Headway pack was tested for capacity prior to assembly. My guess is 13 to 14 Ah. I only used the best cells. Obviously the Headway are higher discharge rate at 200 amps per cell.

Maybe not.

I watched you tube videos where 26650 LFP cells 4S - 1P started a car. Cells are same as mine. I am running an 8P though. Should work well with the Headway. I wish I had a 10 Amp charger for 5S - 8P - LFP. I have a 15 amp 15S - 54.6V charger and a 10 amp 12S - 43.8V.

Using any active balancer will increase the capacity to the pack as will boost up any weak links or cell banks to avoid SAG failure. First time I did a discharge test with an active balancer. Cant do it with 12S and why I spent the extra time testing before assembly.

Still building two more 5S - 8P packs. I spent over $130 on that 54.6V charger. It will work better for long trips requiring a recharge. I would need a 10 amp - 18V charger as 3.65 * 5 = 18.25V. , required charging for 16.5V - 5S - LFP. If anyone knows of such an animal please let me know.

Thanks.

Curtis out.







even 9.

 

[HumanPerson]

I don't see any fire extinguishers @curtis lol

Stay Safe!!

 

[curtis]

No fire extinguishers required with LFP chemistry.

The downside is the A123 cells may not even be at 50% capacity as advertised. I did a capacity test 5S - 8P and was < 7 Ah. However were around 3.3V when the test started. I suspect voltage drop or self discharge as parallel charged like two days before building the pack. The discharge test was done with the 5S active balancer hooked up.

I charged with the 400W Mega charger. I own two 6S - LiPo chargers. When I recharged the pack I used the 200W Sky charger I used for discharge as it would not fully charge LIFE. However I did it different as used the fast charge function which allows a 10 amp charge. That is equal to my 43.8V dedicated LFP charger. After a full charge I continued to trickle charge at 0.5 amp. the lowest setting. I ended up with the pack around 17.7V. Doing the math I got > 3.5V per cell. When I did the capacity test starting voltage was around 16.5V.

Not sure how much more I would get re doing the capacity test. I was told thru research that 90% of capacity is 3.3V to 2V. Basically 2.5V * 8 = 20Ah - 20% = 16Ah. I am looking at about 8 Ah or 1 Ah per cell. That is totally unacceptable.

Moving forwards I am stuck with these cells. There are also about 2 dead cells for every 10 which will recharge in parallel with cells around 3.2V but self discharge back to almost 0V in about a week.

Looks like the best 24 Headway cells from Battery Hookup out of 36 get me somewhere around 10 Ah. Some cell banks maybe 12. 10 Ah is a conservative number. The A123 cells from Battery Clearing house look like 8 Ah. Maybe 9. 8 being conservative. Looks like around 6 or 7 miles of range conservatively. The only positive thing is when I build the other two 5S - 8P packs I will be able to charge at 15 amps with my dedicated 15S 54.6V LFP charger.

What that means compared to my 20 Ah - lithium ion factory 13S - 7P pack is the Lion will go 16 to 18 miles or twice as far as 15S 8P LIFE.
HOWEVER it will take about 4 hours to fully recharge with my 3 amp Lico charger vs 1 hour with my LIFE charger. It would take an extra hour of time to travel 36 miles with the 20 Ah Lion pack as I would need to recharge 3 times with the LIFE vs 4 hours with Lion.

That is why for long distance > 20 miles my 18 Ah - Lishen - LTOs rule as looking at 1 hour because I already traveled 18 miles with 16S 40V of LTO. 50V or 20S will give me 20+ miles. The downside 46 pounds of battery weight.

The bottom line is for short distance the LFP packs rule as 50% lighter than LTO. Both LFP and LTO are 100% safe. No fire extinguisher required.
Also if I could find a portable 18V charger for a single 5S - 8P LFP pack I could charge the 12S - 40V Headway pack with the 43.8V 10 amp charger and get 56V to my 48V controllers giving me an extra 3 or 4 mph in performance. Without a 5S - 10 Amp charger I could bring a second 5s 8P pack and swap them out after re charging the 12S Headway pack. That would increase total battery weight to around 30 pounds , close to what the 40V - 16S - LTOs weigh.

I guess I will get back to cell extraction. This does not mean that LFP is not as good as Lion. It means that NEW Lion is much better than USED LFP. Brand new LFP is much higher in energy density. At least 40% higher than used from Battery Hookup or Battery Clearing house.

Thanks.

Curtis out.

 

[curtis]

From what I read most 48V controllers will work up to around 60V and some a little higher.

My brushless controllers should be alright but not sure about brush. I did find a great 18.25V charger cheap and does 20 amps.

For the 27.5" dual suspension with the 1,000W 26" direct drive hub on the back I really want to put a 1,000W Unite MY1020 motor on the front.

1,000W / 48V = 20.83V * 56V = 1,166W * 2 = 2.333W. Since the heat will dissipate between two motors and controllers I should be able to do 28 to 30 mph steady cruising for miles with no issues. Also steep hills will be possible.

That will be my spring project though as I still have two chain drives to build. The 1,000W 36V Unite motor and the 3,000W brushless motor.

As long as my 1,000W controllers work at 56V the rest will be fairly simple and straightforward. For the 27.5" dual suspension I will use my speedometer app. and find the top speed at 56V. Then gear the front chain drive accordingly.

For the 12S - 2P Headway pack in series with the 5S - 8P - A123 pack I will measure voltage after about 5 miles. If the 5S - A123 pack is a lot lower in voltage than the Headway pack I will build a 5S - 10P or 12P - A123 pack to accommodate.

Thanks. Curtis out.

 

[curtis]

As the Hub Motor Turns and the LiPo Fire Burns.

I took a ride to storage approx. 3.9 miles away. I took the bike with the 750W gear reduction motor on the back and 350W Bafang up front with the single throttle. Upon arrival I hooked up the dedicated 43.8V - lFP charger. I then proceeded to fetch the three bike wheels for my next build.

It did not take long to charge to full.. Maybe 15 minutes at 10 amps. I then put the bike I rode there and rolled out my 20" BMX bike with the 1,800W brushless chain drive motor and 1,500W - 38 amp 36V - 48V controller.

I google mapped the trip home and got exactly 3.9 miles. Note the packs measured > 59V together in series. A 12S - 2P - Headway pack consisting of the 8 Ah cells rated at 200 amps each. Old , tired and used. When I did a capacity test on most of the cells they hit 2V and stopped on the 6S - LiPo charger. I had to test at 1 Amp or start at 0.5 amp and raise to 1 amp. It is why I went with 12S - 2P with the best 24 out of 36 cells from battery hookup.

I also ordered 168 - A123 cells from Battery Clearinghouse. My plan was three 5S - 8P packs in series for 15S as I paid about $133 for a dedicated 54.6V - 15 amp LFP charger. My first 5S - 8P - LFP pack in series with the 12S - 8P - A123 pack went 3.9 miles. Full throttle for the first mile and about half throttle or less the rest of the way as I wanted to make sure it made it.

Here are the results.

16.42V / 5 = 3.284V per cell bank - 5S - 8P - A123 pack.

39.6V / 12 = 3.3V per cell bank- 12S - 2P - 8 Ah Headway pack.

3.9 miles.

I did not have my speedometer app. as did not have my phone so could not measure top speed. I do not believe that top speed was > 30 mph but not sure. It did seem fast but doubt it was faster than my LTOs at 50V. ( 32.5 mph). That is because LTO is very high performance and much less voltage drop under load compared to LFP. However 3.3V * 17 or 56 working volts is probably about 20 pounds lighter than 50V of LTO.

The only battery chemistry that comes close to LTO in performance is LiPo. I had 12S - LiPo - Lithium polymer packs before and got rid of them when they puffed up. They will compete with LTO in performance but are very unsafe. Lion 18650 cells are about the same but limited by the BMS and also unsafe. I am content with running LFP and LTO for now.

If I get a big house to rent in the country which might be in my future I will have a very safe place to charge 18650 lion as well as LiPo packs from Hobby King. I already have a 13S - 7P - factory 20 Ah - 1,200W Lion pack and a 4.4 Ah 10S - 3P or 4P Larudation factory 18650 pack. I also have two 90 amp power modules from battery hookup. For now though I prefer LFP and LTO for indoor charging.

Thanks. Curtis out.

.