I am coach for the Mulberry High School Team, car #38 in Florida. I started a team in the school year 2017-18 with a loaner from Electrathon of Tampa. In the spring of 2017 I purchased a car from Jim Robinson. The next two years, my high school team won the high school champions trophy for the 2018-19 and 2019-20 seasons. This season the team is not able to participate outside of the county due to pandemic restrictions. So, I, as car owner, have been driving it in the open class. This has given me a great deal of experience and knowledge I did not have about the car. I understand much better the car, its capabilities, and its limitations. 

The point of the car though is to teach students concepts in automotive, engineering, and trouble shooting. I am looking to add a second car, #100. This car will use a Li-ion chemistry. I think having a car in the lead acid division and the alternative energy class will give my students an opportunity to learn a bunch about the different energies. So, I have been working for about a year on what options to go with. Here's what I know so far. Please feel free to correct any statement or add information not expressed. 

Which Li-ion chemistry?

I have opted for LiFePO4. Why? Safety. Lithium Iron Phosphate is not prone to thermal runaway. I feel it might be safer for high school students to be sitting right next to these packs than something else. Recently I posted this to the Facebook group DIY Battery. Someone answered that NMC batteries pack a greater punch. I had never heard of these before. Battery University defines these as Lithium Nickel Manganese Cobalt Oxide (LiNiMnCoO2). (https://batteryuniversity.com/learn/article/types_of_lithium_ion#:~:text=NMC%20is%20the%20battery%20of,expensive%20and%20in%20limited%20supply.) Yes, I have considered the typical 18650 cells typically containing Lithium Cobalt Oxide (LiCoO2). The issue comes down, I think, to being able to get energy out is a period of time, i.e. amps, the amount of current.

What size cell to use?

It seems that different chemistries have different abilities to deliver current. Some are high current and some are low current. I have been looking at Headway Cells. The 40152 cells come in 10 Ah and 15 Ah. The 38120 HP cells come in 8 Ah capacities. The smaller 18650 cells come in a range of amp hours from 2000mAh to 4000mAh. BattryHookup.com offers a 32650 cell with either 5000mAh or 6000mAh. If you are not aware, the numbers on the cells are the dimensions. 18650 are 18 mm by 65 mm. Among the differences in size and amp hours, the batteries have different "C" ratings. This is still a little fuzzy for me. It appears that the "C" rating is related to the Ah rating on the battery. For example, if the amp hour rating is 3800mAh and the battery has a 1C max discharge rating, you can discharge the battery at 3.8 amps max. I assume that any attempt to pull more than that like 5 amps will damage the battery. A battery with 8Ah like the Headway 38120 has a max discharge current of 200A or 25C.  Check out the explanation of "C" Rating at Batteries In a Flash.com.

What I don't understand about current

What I have yet to wrap my head around is how to get 30 to 40 Amps out of a 8S4P configuration of 18650 cells rated at 3.8Ah each. Wait you say. What's 8S4P? Well that is 8 cells in series in 4 parallel rows for 32 cells in total. I would need to arrange these cells in this manner to get a 25.6V pack with 3.8x4 Ah of capacity. (remember, series adds voltage and parallel adds current). Can I pull 30 to 40 amps during a race from these cells without damage. This is where I get stuck. I do not have the experience and lack an understanding of how I can race on this for 60 minutes. 

Now in the example above, I have only 389.12 Wh of capacity in the battery (25.6 V x 15.2 Ah). I would have to increase the pack to a 8S5P configuration to get 25.6V (3.2V per cell times 8) and 19Ah (3.8 * 5) to get 486Wh. With two of these packs in series I would get 51.2V and 19Ah for 972.8Wh. Remember the rules limit you to 1KWh of capacity at the start of the race. That's 1000Whs for those struggling with metric. 

Ah, but here's the rub. My YellowTop D35 lead acid batteries are rated at 48 Ah. That means I start the race with 24V (two in series) and 1152 Wh of capacity. If I use the full voltage rating of the batteries at the start of the race, about 12.8 to 13 volts, I am technically starting with a capacity of 1248Wh. Using 24V instead of 25.6V in the calculation for the Li-ion pack described above, I can go to a 8S6P pack at 22.8Ah and have 1094Wh of capacity at the start of the race with two packs in series. That's less than the lead acid capacity of 1152Wh at the start of the race. It seems to me that either the lead acid has an advantage and the alternative energy are handicapped or there is no equivalency between them. With a BMS in the middle of the battery and the load, its possible I cannot even get the 1094Wh rated hours out of the pack/configuration. 

Ultimately, what I would like to do is to have a set of cells that I could arrange in a 48 V arrangement and then turn around and have a 24 V arrangement. It would be interesting to make comparisons between two races with these different packs using the same cells and controlling for other variables (driver, day, course, etc.). I also need to know more about the amperage demands of my current car during a race. I know that compared to my high school students, my extra driver weight is costing me energy. At the end of the race, my batteries do not seem to have as much left in them as when my students drive. Often, I am just barely making it around in the final minutes. My students still seem to be keeping pace. 

I know this was a lot to post in one post, but I wanted to outline my thinking and where I am at. Any advice or assistance would be helpful. I am looking for a power analyzer that I can record or display some data at the end of the race. Anybody know a good one. I would like to see the amperage demands as the car goes around turns and on straight aways. I am thinking there might even be some coaching potential here to get better driving from students.

Thanks for the read.

The image attached is me in the car at the Lake County event where we race on a WWII air field. Less turns more go!

Todd

It's good to see other teams experimenting with newer battery technologies as well. I am the faculty advisor for the George Fox University Engineering team and we recently built our first car and used LiFePO4 cells. We have not had the opportunity to race the car yet, but testing seems to show promise. These type of cells seem very safe to use and fit in well with existing lead-acid voltages that teams are accustomed to working with.

I built a discharge tester to test the performance of the batteries at a controlled discharge rate. Something like that would be useful to you. I built it using a 48v PWM controller, dumping the power into a resistor bank. Current and voltage are monitored via an Arduino with a 4-channel 12 bit ADC board and a shunt (not as complicated as it sounds). The Arduino measures the voltage and current over time and integrates that to get the capacity. I also use the Arduino to control the PWM controller to maintain a constant current, but you can control the current manually if you don't mind sitting there while it runs.  Let me know if you want more detail on that system.

We are also using Arduino's to monitor the car's on-track performance and record it for later analysis.  We still have some details to iron out, but it is close to functional.  With all the disruption lately, it's been hard to maintain any continuity within the team.

Nick.

Hi Todd,

There is a lot to talk about in your post. You are correct in your understanding of C rates. 1C is the capacity of the battery if discharge completely in an hour. 1/20C is the capacity if discharged in 20 hours. 20C is the capacity if discharged in 3 minutes (60 minutes / 20 ). 

Almost all of the time, the battery will deliver more energy if discharged over a longer period of time (have a greater capacity). A battery rated for 100 amp-hours at the 1/20 C rate should deliver 5 amps for 20 hours (5 amps * 20 hours = 100 amp /hours). Logically, it would seem that the battery should give you 100 amp-hours at 1C (100 amps * 1 hour) but batteries do not work that way.

First the battery loses energy because it is delivering a higher current (amps). Power (in Watts) that is lost as heat equals Current (amps) squared times Resistance (Ohms) uses this equation: P= (I^2)R. The key point here is that current is an exponential relationship. Higher currents quickly increase losses. We see this as the wires and battery getting hot.

The second part of the loss is more complicated and affects lead acid batteries more than lithium ion. Basically, a higher discharge depletes the ions in the electrolyte more rapidly next to the plates. This problem will resolve itself over time as ions are pulled from less depleted areas of the electrolyte. This is why a lead acid battery will 'recover' enough to drive back to the pits if the driver waits long enough. Advanced Lead Acids thin out the plates and electrolyte and increase the surface area which helps minimize the depletion effect.

The most common way to categorize a batteries performance under different loads is Peukert's equation. Measure the battery capacity at two different discharge rates gives a reasonable approximation of how the battery will discharge between those rates.

This is why the 48 amp-hour advanced lead acid batteries will not supply 1152 Watt-hours (48 amp-hours * 24 volts) during a race. In electrathon, I know of only one team that was ever able to get more then 1,000 Watt hours out of Optimas. Generally, I would expect somewhere around 750 Watt-hours.

A 12 volt lead acid battery is in fact, six 2 volt batteries in series, so two Optima batteries that many teams run are 12S1P. When discharged in series, each cell supplies the same amount of energy. The cells never have quite the same capacity. When any individual cell is empty, that cell begins to act as a resistor and waste the energy trying to pass through it. This means that pack capacity is limited to the capacity of the smallest capacity cell times the number of cells in series. One bad cell does spoil the whole bunch!

The other way we talk about batteries is their voltage. Lead acid cells have a nominal voltage of 2 volts at 68 degrees F; but fully charged, they are around 2.14 volts at 68 degrees F; and 1.8 volts when empty.

Notice that that the temperature is specified. The capacity of the cell is affected by temperature. This is another technique to increase the capacity of a lead acid pack, heat it. Higher temperatures give higher capacities up to a point. They also might shorten the life of the battery but racers do not care.

It is pretty useful to speak of the capacity of a lead acid battery in Watt-hours at the 1C rate at 68 degrees F. This will change as the battery ages (calender  and when the moon is full. Ok, I just added the part about phases of the moon.

When we move to lithium ion. Most of my experience is with lithium ion polymer of various composition. These mostly have a nominal capacity of 3.7 volts. 2.8 volts is empty and 4.23 volts is full. If you empty the cell below 2.8 volts or charge it over 4.23 volts, it will do damage quickly. The cell will last longer if you do not approach these limits. For example, only charging the cell to 4.1 volts will give less capacity but longer life. These voltages are for the specific cells I am working with. The cells you mentioned have different limits.

Lithium cells in series need to be cared for in series. This is done by monitoring the voltage of each set in series. If the pack is 12S1P, 12 voltages need to be monitored. If the pack is 12S100P, 12 voltages need to be monitored also. Cell in parallel share the same voltage.

Lithium also has less capacity at higher amp draws but not as severe. The battery manufacturer will often post a graph of capacity at different C discharge rates. 1 C will be 100% of rated capacity and 4 C might be 97% for example.

Cells in parallel share capacity. Two cells rated for 10 amp-hours at 1 C wired in parallel will deliver 20 amp hours at 1C. Their maximum discharge rate will also be the sum. So if the limit for each cell is 3.8 amps, together their maximum is 7.6 amps.

Cells in parallel share capacity. A 5 amp-hour cell with a 10 amp-hour cell will try and give 7.5 amp-hours(they do better at sharing if their capacities are closer to equal). Cells in series are limited to the lowest cell's capacity. The 5 amp-hour in series with the 10 amp-hour will deliver 5 Amp-hours. 

In a race, the ideal pack has a capacity of 1,000 Watt-hours. That means, the power the pack could deliver the whole race is 1,000 Watts. This could be a constant 42 amps at 24 volts or 21 amps at 48 volts or 15 amps at 67 volts etc. Higher voltage means lower current and higher total capacity delivered by the pack.

Higher voltage packs downside is more cells in series that need to be close to equal capacity. It means more voltages have to be monitored.

A high level team will test the individual cell capacities and then group them so that the added together capacity of each group of paralleled cells is the same as the others in series. Then the pack is charged in series. When the first group of cells reaches the maximum intended voltage, the series charge is stopped and each paralleled group is charged to the same voltage. 

At ProEV, we wire our packs so that we can connect cheap alarm buzzers like this to warn when any series group gets to low (https://hobbyking.com/en_us/hobbykingtm-bx100-1-8s-twin-buzzer-lipo-life-li-ion-voltage-checker-black.html?queryID=&objectID=86257&indexName=hbk_live_magento_en_us_products) . They can handle 6-8 cells in series and cost around $4 each. We run 3 of them on our 18S3P pack. They are not terrible accurate but they will make noise when the pack gets low.

The proceeding paragraphs describe our Battery Management System (BMS). There are many other designs for a BMS. Read this post about testing another type of lithium pack https://electrathonamerica.activeboard.com/t66140971/proev-tests-an-18650-lithium-pack/ .

I hope this is some help.

Cliff