As every year, when the first warm days of pre-summer beckon, I feel the itch of taking TW560 on a long one-day-trip.
Part of such an one-day-trip is typically charging the pack to the maximum 4.22V per cell, unlike the usual 3.8V for everyday use, and then drive the pack to the self-imposed minimum cell cut-off voltage of 3.0V.
This works very well and is no issue if the pack’s many cells are all well balanced and at the same voltage. Since my TWIKE does not have any active cell balancing or thermal management (for weight and full-pack cut off reasons), in my pack – after a full year of driving – a few cells can start drifting slightly. Which is completely normal but needs to be taken care of before fully using the pack’s capacity.
I’ve written extensively about why lithium cells need to either have active balancing or regular external balancing – it’s a very interesting rabbit hole to go down.
In the old days of my LiFePo3 pack, I used to drive the pack to capacity much more often, which saw the cells drifting quite a lot. With my current mega-pack where a 100km trip is around 25% charge, I can keep the cell charge in ‘daily mode’ for most of the year which in itself helps cell health greatly.
Back to a weekend with acceptable temperatures beckoning – time to get TW560’s pack ready for 2023’s adventures (more on that shortly).
Let’s head out to TW560 in it’s currently very green garage…
As always before we start with one of these entries:
ONLY PROCEED IF YOU KNOW WHAT YOU’RE DOING
Depending on the setup, both 230V AC and 400V DC are present inside a TWIKE. Both voltages kill silently and efficiently. You have been warned.
After opening the battery bay, I remove the AK rail support struts and access the battery pack.
There are 42 balancing ports, each one of which will be connected to my battery workstation for balancing.
This is a good moment to have a look at the AK rail – the PCB controlling all power-related, non-traction functions within the vehicle.
I locate the first connector, hook up the battery workstation and then start the charge from there.
With FMC’s software, I get to log all relevant data and get some nice visualisations and graphs while charging.
Since the differences between the stacks turns out to be minimal, charging cycle times are very similar. This means that I can set a timer on my watch to remind me to go and move the plug to the next connector.
And without further ado: how did my battery pack age and how did the cells deviate during the last year?
The overall health of the pack is still amazing. 40/42 stacks are still in pristine health, with all cells balanced down to the milivolt.
The two stacks that actually required balancing were still in the dozens of mAh away from the others – well below 5% of the total capacity.
I then took one of the stacks and performed a synthetic capacity test over one discharge and recharge cycle. Usually, i take my TWIKE on a trip for the day and get the true capacity of the pack combined with a nice trip.
As I will be taking my TWIKE on a summer trip in just 2 weeks, I will be foregoing this tradition this year and estimate the full capacity by extrapolating the energy available in the tested stack to the full pack.
I must admit that I expected a higher capacity loss 7 years in and am very happy with my investment I made 7 years ago!
| Date | Wh Capacity | Loss since last check | Total capacity loss | Range at 38.5Wh | Range at 55.5Wh |
|---|---|---|---|---|---|
| 11.08.16 | 17,722 | 460km | 319km | ||
| 01.05.17 | 17,172 | -3.10% | -3.10% | 446km | 309km |
| 10.09.18 | 16,601 | -3.32% | -6.33% | 431km | 299km |
| 16.03.19 | 15,916 | -4.12% | -10.19% | 413km | 287km |
| 17.07.19 | 15,873 | -0.27% | -10.43% | 412km | 286km |
| 28.02.21 | 15,758 | -0.72% | -11.08% | 409km | 283km |
| 22.05.22 | 15,703 | ā0.34% | -11.13% | 408km | 282km |
| 21.05.23 | 15.650 | -0.33% | -11.69% | 406km | 281km |