Impact of "Serial Cells vs. Parallel Cells" battery configuration
#1
'sup, this thread's purpose is to give people some guidance on what kind of battery configuration to build based on their needs for speed and range.

I've been doing some testing on my own board about the effects of battery configuration on range and speed and then doing test runs and logging up the gps data and later the state of the battery as I've charged it. I've have then logged the information on a google drive sheet to display the results of each cruise. Here's a link to the spreadsheet, also select the "Gen2.1 deck" sheet from the bottom to see newer more relevant info: https://docs.google.com/spreadsheets/d/1...sp=sharing

My battery pack is built with LG INR18650 MJ1 cells with 3500 mAh nominal capacity and 10 Amp max continuous discharge current.

[Image: WtBYz2P.jpg]

I have tested 2 different configurations:
4SP5 - 20 cells - 14.8 V & 17500 mAh
6SP4 - 24 cells - 22.2 V & 14000 mAh

The older runs with the new deck were done with 19/30 gear ratio and on 4S5P, then I changed the ratio to 15/30 and then came the upgrade to 6S4P. The motor has been SK3 149kV always. So at this point I think we are going ignore gear ratio's effect on the speed and range and just focus on the battery configuration's impact on speed and range. So we are only looking at the 15/30 geared results for now.

Averages with both setups doing a benchmarking run (This a standard route that can I ride at 95% time at max speed, so no any unnecessary starts or stops):
4S5P - 117 Watts of continuous power - 16.0 km/h average speed - 7.3 Wh/km total energy spent per kilometer - 496 mAh/km capacity per cell used per kilometer
6S4P - 262 Watts of continuous power - 23.7 km/h average speed - 11.0 Wh/km total energy spent per kilometer - 496 mAh/km capacity per cell used per kilometer

So from these data points, what can we figure out that helps us to determine our needed battery configuration?

The most directly relevant data points are the Wh/km and mAh/km for each battery configuration. The mAh/km value shows you how far you can go per parallel cell and the wh/km can be used to estimate the efficiency of the board.

(mAh/km) with average of ~500 mAh/km on both batteries. I can say that I will get about 6 km of range per parallel cell without running it completely dry for longer cell life. So the 4S5P conf. with total of 20 cells will go 30 km (and has gone in testing). while the 6S4P conf. even while having 24 cells in total or 4 more than the smaller one, will only go about 24 km.

So the amount of parallel cells will quite linearly change your maximum range. AKA. with 3P if you get 15 km of range, then with 6P you will get around 30 km. Regardless of how many series cells you have.

(Wh/km) can be used to figure out total energy spent per kilometer and then figure out continuous power on flat land. So with 4S I've averaged 7.3 Wh/km and as we increase our series cells to 6S or a 50% increase. Our Wh/km also increases 50% to 11.0 Wh/km. Meanwhile an exponential increase can be observed in the continuous power figure as it's is increased by 1.5^2 from the 4S to the 6S or 117 W * 1.5^2 = 264 W.

So the amount of series cells will again quite linearly energy consumption and your top speed, but exponentially your[b] power. [/b]AKA. if you have 100 Watts at top speed on flat roads with 3S, then with 6S you would average 400 Watts on flat roads. Speed is also affected quite linearly as can be compared with the 16 km/h average speed with 4S and 6S with an increase of just under 50%.


What this means in the end is that if you want to go fast and far you need a high cell count battery pack. Otherwise you're only doing one of those two. Or both with to a lesser degree.
Reply
#2
you were going max speed in both scenarios and trading range for power or vise versa as your result but I wonder how your efficiency and range would change if you rode at a slow speed for both scenarios.  How inefficient is it to ride with a high kv and voltage yet go around slowly?
Reply
#3
(03-12-2017, 09:01 AM)hummie Wrote: you were going max speed in both scenarios and trading range for power or vise versa as your result but I wonder how your efficiency and range would change if you rode at a slow speed for both scenarios.  How inefficient is it to ride with a high kv and voltage yet go around slowly?

That is exactly what I'm going to test first when the snow finally melts and the roads dry.

I'm going to use the 6S4P, but limit my duty to a max of 66% to simulate the 4S voltage. Technically my max power should then again be at full speed on flat ground at 117 Watts, which should result in lower continuous battery current, due to the higher battery voltage.

Basically I hope to lower the energy used per kilometer to the 4S's ~7.4 Wh/km, which would then also result in lower capacity per kilometer, ~500 mAh/km -> 66% 333 mAh/km
Reply
#4
we have to wait till the snow finally dries! from what I read, with the pwm it barely makes a difference, which is surprising and makes me wonder why bother using a lower kv and lower voltage is ever a good idea
Reply
#5
(03-12-2017, 10:19 AM)hummie Wrote: we have to wait till the snow finally dries!  from what I read, with the pwm it barely makes a difference, which is surprising and makes me wonder why bother using a lower kv and lower voltage is ever a good idea

I think it will make the difference, or at least I can't come up with any reason why it wouldn't. I mean if you're not going top speed and having 95-100 % duty then you're also operating in the PWM range that "barely makes a difference". I'll admit that it might not be quite 1:1 as the math says, because we will likely have a bit more losses in the controller, but that will be revealed once this long winter would finally end.
Reply


Forum Jump: