I've tested groups of batteries as a pack by disconnecting my motor and hooking up a heater to create a load. I'm able to control how much power is going through by doing this. Talking with Mike Hodgert and Dave Cloud, they test batteries individually to get a better more accurate read for the number of amp hours each of their batteries can put out. Starting to brainstorm how to set up a load testing rig, I drew this out. Would this work? I'd be able to track everything with my meter to track the number of amp hours and watt hours each battery would produce. Give me your input if there is something I need to change. I've got big enough cable to work with so I'm not too worried about that and a 250 and 400 amp fuse which might be a little over kill with a 100 amp kill switch from O'reily Auto parts to turn the load on and off. The battery would probably be pulling around 50 amps to start then start tapering off over the course of the hour.
The key here isn't finding out how many "amp hours" your battery can put out. Turns out amp hours is constant. Its the same. Always. The battery has a label on it and says probably 44 amp hours. The key is using the coulombs the right way. When the voltage is high, the coulombs can do more work because they are "stronger". Therefore, you need less of them to do the same amount of work later in the race when the batteries are lower in voltage. On a side and important note, controllers work most efficiently the closer they are to wide open. Thus, we need to find the middle ground. If you go full speed for the entire race, the batteries will die half way through. But if you end the race with too much voltage left, you obviously could have gone faster. Back on track... the point of draining the batteries is to find their curve when drained a certain way. Unless you run your car wide open the entire race, that testing method won't help at all. So now you have to find the middle ground of running wide open, while still staying as efficient as possible. After years of testing... we've come to the conclusion that running the motor at 21 volts for the entire race is best. This way, JUST as we cross the finish line, the batteries should be, theoretically, completely dead. We can assure this by creating the battery drain curves. First by start by finding out the average amp draw of your car at the desired speed. Now, using at tiny bit of math, we know what this number SHOULD be. Turns out the optimas have 44(?) amp hours. Well if we have 44 amp hours and a race lasts one hour... then you should be drawing 44 amps for the entire race for the battery to drain completely. Well now you know what amperage to race at. If the amps are too high, then gear it slower. If amp draw is too high, no matter how you race it, it won't last the hour. So now you start draining batteries. You will have to experiment with the motor voltage, but keep it constant. Trust me on this one. We drain our batteries at 36 amps and 10.5 volts and they will last the hour.
Now for your bench... Use a controller and a load that you can, well, change the load on. We use rails with light bulbs on them and add or subtract light bulbs to change amps. Everything else should be fine on it. Drain each battery separately and pair them by using their curves.
Recapping some stuff. Amp hours are worthless. Forget about them. Using coulombs effectively is where it is at. Use less when they are strong (high voltage, low amps) and more when they are weak (low voltage, high amps). This will keep the same power output throughout the race. Not only that, but it keeps up the momentum. Momentum is huge, especially on oval tracks. Slowing down is the worst thing to do, unless you are below the curve and going to die too soon. More side notes: Keep a constant motor voltage. Going slower but lasting the entire race is way more important than going fast but dying with 10 minutes left. Think about it. Going 30 mph at last 10 minutes while everyone else is going 10 mph, you will pass them WAY more times than them going 40 while you are still chugging along at 30. This is exactly what we do. Only faster. We run our car 222 at about 21.5 motor volts and only increase them. We start the race at 36-37 mph and end it going over 44 mph. All this non-amp-hour mumbo jumbo might seem strange, but it's why we win. Shoot me an email if you have any more questions. I don't really look at these forums that often.
I think I understand what you are trying to say but not how you are saying it.
To start with a Coulomb is equal to 1/3600 of an amp hour. http://www.mpoweruk.com/conversion_table.htm
So when you speak of Coulombs and AmpHours, you are speaking about the same things, just different units.
You wrote "Turns out amp hours is constant. Its the same. Always."
Umm...No. Amp Hours are not constant. They change with discharge rate, temperature, age of the battery, etc. Here is a thread that gives some examples: http://electrathonamerica.activeboard.com/t34589869/yellows-vs-reds/
On the other hand, I agree that Amp Hours ignore voltage and this can skew results. The solution is to measure capacity in Watt Hours. This is Current * Voltage over a set amount of time.
Your practical advice, try and run the same power for the whole race rather than the same current since voltage will drop over the race is a sound tactic. And your information about controllers being more efficient closer to full on, I also agree with. Finding the right balance is certainly the challenge!
Geez, and all I was asking about was if the load testing idea I had drawn up would work or not......
Anyway...
Eric, I know about watt hours. My meter can do both Ah and Wh no problem. I was just saying about the current draw in my original post for the sake of if something were to go up in smoke (probably not because of the big welding cables that are being used and the amount of power/current running through the system for one battery being tested).
I was always told to use watt hours as an effective means to find out how much power the batteries can pull no matter what set up. However I will check my Wh every fifteen minutes to see if I'm "on track" or not but will use current while driving since I have both numbers to work with. The watts read too fast in my opinion causing too much throttle adjustments while the amps are a bit more tame for me.
But what can I say, I've only been doing electrathon racing for seven years now so what do I know?
On the west coast most of our yellow tops will be getting getting roughly 980-1000 Wh from our packs. If using the smaller Interstate batteries by hooking three in series, the same 980-1000 Wh amount can be achieved.
Amp hour wise, the yellows translate into 40-42 Ah. The interstates have mixed results from 24-28 Ah.
I will say I've gotten more than 42 Ah from yellows on a few occasions but this was back in 2010 when I had a really good set.
I took a belated look at your drawing. Assuming the positive terminal of the battery is on the left, it looks to me like the plus and minus current sensing leads on the shunt are reversed. I have not played with a Cycle Analysis (I assume that is what CA stands for) but I would guess this would just make 'amps out' read as 'amps in'.
Since your focus is testing the battery, I would also suggest you move your leads as close as possible to the battery so you are not reading extranous resistance from connections and the fuse. So move the shunt to the other side of the fuse and put the V- wire on the negative battery terminal and the V+ wire on the positive battery terminal. That way, any voltage drop from the switch, fuse, heater, connections, wires are all on the load side.
I would keep the leads short, exception would be the heater part just because I'd rather not have something burn. And the plus and minus for the Cycle Analyst is per their instructions (might be wrong without looking again).
You are right about the positive being on the left of the battery. I've drawn Optima batteries (75/25 or D75/25 versions) for so long I know what it is. That was the main type I always used, should have marked that on my part, oops.
Anyway, I'll get some pictures of what it looks like once it's all together if anyone else is looking for some ideas. The shunt would stay in the car since the meter is already wired in instead of yanking everything out. All I would be doing would be hooking up the leads to the shunt with the heater, battery, and kill switch outside the car. The way it is currently drawn out would show the voltage of the battery when everything is hooked up and a battery is in place. Once the kill switch is switched on, the load would start and start to record amp hours and watt hours. Then run it till the voltage reads at least 10 volts and call it good.
My older cycle analyst meter would only go to 15 volts. Now with the new one, it can go all the way to 10 volts which makes it perfect for testing batteries individually.
And yes, I'll take your suggestion for swapping the shunt and fuse around. Thanks ProEV!
