The battery was fully charged by 2pm! This is surprising since the solar panel points to the afternoon sun – and this is actually “visible” in the data because the 265 W panel only reached a maximum of 90 W before it scaled back due to the battery being at maximum voltage. By numerically integrating the shaded area under the charging power plot, I discovered that the battery took 135 Wh of energy to be charged. This is spot-on my estimate of how much I had flattened it!
The obvious conclusion from this successful test is that I need more cells to make my battery bigger. Back to the cell-testing regime for now…
My MPP Solar PCM 60x charge controller has a serial-port connection, but the software that “comes with it” only runs on Windows (I need something for Linux). It turns out that a number of people have already invented this wheel, and I was able to use their efforts to hack a simple python script for crude logging. …
I wasn’t sure whether the small 0.1 A charging current yesterday afternoon was due entirely to the shade on the PV panel or whether it had something to do with minimum input requirements for my PCM 60x charge controller. On paper, the 265 W solar panel should be able to charge my 500 Wh battery in two hours or less.
I left it all connected while I was at work today, and checked it this evening. The battery voltage has risen from 26.1 V last night up to 28.3 V! That’s quite close to the 28.6 V cutoff that I programmed into the charge controller, which suggests that the solar panel essentially gave me a full charge. How exciting!
Completing the first stage of my battery was a great feeling, but it has been sitting underutilised for a few days. My iSDT Q6 smart charger takes a 7-32 V DC input (nicely covering my 24 V system) so I’ve been able to power my cell testing from the battery. This bootstrapping is nice, but it has been slowly draining my battery and I’ve been reluctant to charge from the grid – the main point is to store sunlight! This weekend I’ve “plugged-in” to the sun.
A few months ago I bought six QCELLS Q.PRO-G4.1 265 solar panels on ebay for a fantastic bargain. At 265 W each these panels give me just over 1.5 kW of electricity production – way too much for my 0.5 kWh battery. For now, I’ve wired up a single panel to give a nominal 0.5C charge (2-hours full charge if the sun is bright).
Along with the solar panels I bought a second-hand MPPSolar PCM 60x charge controller. Again, this is oversized for my battery at this stage – it can put out a 60 A charging current – but I’ve dialled it down to 10 A. This is 1 A per cell in my current battery, which matches my testing regime and represents my “upper limit” in the design parameters.
For now, the solar panel is leaning against the back of the house. With only 9.23 A of short-circuit current from the single PV panel, I have taken a temporary shortcut and used 2.5mm twin-core-plus-earth (2.5 square mm cross section area for each of active, neutral, earth – normal internal wiring for household power sockets) to connect the panel to the PCM 60x charge controller. This wire is rated for 20A, and is currently routed under my house so does not have any UV or thermal load to deal with.
It took a lot of fiddly time to measure the cable distance, attach MC-4 connectors, triple-check polarity, connect into the PCM 60x terminals – and by the time I’d finished the sun was low enough to leave my PV panel perfectly in the shade!
Despite this, I was charging at 0.1 A! Everything seems to be working perfectly, and now I need to wait for some sun.
I finally have the first modular instalment of my battery complete: 70 cells in a 7s10p configuration made of seven “packs”. The pack construction has been a bit slow, but along the way I’ve built some construction jigs and optimised the process. I think I could do the next set of packs in less than half the time!
These are only half packs, because my design is to have 20-cell packs as the basic battery building block. This means that they don’t “stand up” as intended in the final design – but I don’t have the battery rack built yet anyway, so I’m happy for them to lay flat on their sides for now.
In place of a proper battery management system (BMS) I’m using a hobby charger that can do a balance charge (monitor the voltage of each of the 7 packs individually). This is why each “join” jumper between the packs needs to have a separate return wire.
In keeping with the “start cheap and grow” philosophy of this project, I picked up a second-hand 24 V inverter for $50. It has a rated output of 600W, but is a modified sine wave inverter which is not great for sensitive loads. As a proof of principle I used the 240 V power from the inverter to power the lights for these photos. With my current battery size I can’t even use the full 600 W as it would draw more than 2 A per cell (I’ve designed and tested for 1 A max).
It was a great feeling to finish the wiring harness and hit that switch. There’s a long way to go before the battery is really useful, but it is now ready to power things and let me move into a fun new stage of testing and design. I can return to cell extraction, and while I’m accumulating the next 70 good cells I can tackle the next big challenge: charging this battery from the sun!
I’ve been interrupted by a family holiday, but just before leaving on that trip I reached the first real milestone for my battery build: 70 cells. This will allow me to make half-packs (10 cells each) and produce a fully-functioning 24 volt battery prototype. In the process I will be able to optimise the techniques for pack construction.
I’m calling each set of 7 packs a “string”, so that the battery is made up of strings which are made up of packs which are made up of cells. This first string of 70 cells has 0.518 kWh of storage, which is slightly above my estimate of 1 kWh per complete 140-cell string. My initial estimates of achievable cell capacity have turned out to be reasonable.
Just a few of these cells have had their plastic cover damaged by the extraction process, and the orange sleeves are sitting on top in this photo ready to be fitted as replacements. Obviously, some care must be put into the cell arrangement within a pack for visual appeal.
Dividing the 70 cells into packs is less arbitrary, because it is vital to have the packs as balanced as possible. I devised a simple algorithm as a starting point, and it has turned out to be remarkably effective.
Sort the cells by capacity from highest to lowest.
Take the top 7 cells in this list and distribute them in order into packs 1-7.
Take the next 7 cells in this list and distribute them in reverse order into packs 7-1.
Continue down the rest of the list, filling in this “zig-zag” manner.
This means that pack 1 gets the highest capacity cell, but then only the 14th in the list. Pack 7 gets the 7th in the list, but then also the 8th. This process produced packs ranging from 20.5 Ah to 20.59 Ah, and that 4% variation can be handled by balance-charging.
Lithium chemistry cells are not a good match for 12-volt systems, because almost exactly 3.5 cells in series would be required. To get around this, and to reduce the size of copper wire needed to carry high currents, most people run 24-volt systems, or even 48-volts. Since this project needs to start small and bootstrap up to larger sizes, I’ve decided on 24 V.
This means that I need 7 “cells” in series, although of course these will not be single cells but rather large packs of cells in parallel. Many other home battery builders have opted for large packs of 80-120 cells in parallel, and this has the advantage of getting large storage capacity fairly neatly. However, it means that 7 x 80 = 560 cells are required before the 24 V battery can first be “turned on”. I want to quantise the battery into smaller packs so that I can go “live” well before reaching 560 cells.
With cell holders that allow a triangular-close-packing of the cells, I can make packs of 2×10 cells that are attractive and slim-line. Interestingly, a battery of 7 20-cell packs would have about 1 kWh of storage for realistic reclaimed cell capacities.
There’s also a maintenance advantage of using smaller packs strung together. When the battery is larger, it will be possible to remove one of the 20-cell packs from a set of parallel packs and leave the battery operating (although it will become unbalanced if left like this for long).
One decision that I initially postponed was the threshold to use for labelling a cell as “good”. I wanted to keep the storage ability as high as possible by using only cells with a high capacity, but I also wanted to be realistic and avoid “disqualifying” a big majority of the reclaimed cells. I decided to wait a bit and see what the data suggested.
After testing 20 cells I had an almost perfect dichotomy between dead cells (assigned a nominal capacity <1 mAh) and cells with capacities over 1900 mAh. The following histogram illustrates this situation, with the dead cells omitted as they are not interesting.
The histogram looked nice with 200 mAh bins, and that suggested a threshold for “good” of 1800 mAh. I still had plenty of cells to test, but adopted this as my draft categorisation to start counting how many good cells I was accumulating. Another nice thing about this threshold is that having an average cell capacity of 1930 mAh means that a 7s20p battery is almost exactly 1 kWh of storage (1.93 Ah × 3.7 V × 7 series × 20 parallel).
By the time I had tested 100 cells, the picture was surprisingly similar. There were no longer any empty bins in the histogram, but the cluster above 1800 mAh remained. The chart below shows that a second smaller cluster had started to appear above a threshold of 1000 mAh.
I was noticing cells turning up with capacity between 1000 mAh and 1800 mAh which were otherwise perfectly passing all my tests, and it seemed a shame to throw them away with the dead cells. The appearance of this second cluster led me to introduce a third category of “poor” cells, and I can think about what to do with them later.
The clustering in the cell capacity distribution is fascinating and caught me by surprise. It is still clearly visible in my histogram (currently 172 cells have been tested).
Perhaps the biggest problem with waste laptop batteries is that their usage history is unknown. It is important to run a series of tests to ensure that each cell is safe and useful. I have developed a production-line approach, and mark each cell with a “serial number” (YYMMDD## format) as I extract them from battery packs. This lets me record a range of data for each cell across each testing stage.
I read quite a lot of good ideas and put together the following test cycle:
When I open a battery and extract the cells, I measure the “voltage at extraction”. Li-Ion cells have a normal voltage range of 2.7 – 4.2 V (extremes), but many waste laptop batteries have been sitting on shelves or in boxes for a long time. It is quite common to find cells that are significantly under-voltage, and some are even at 0 V.
Cells are charged at 1.0 A on the LiPo setting of my hobby (RC planes) smart chargers, up to 4.2 V. Any cells that heat up while charging are instantly disqualified from further testing. I have not yet properly quantified “heat up”, but it is very obvious (they get hot to the touch). I record the date/time each cell finishes charging.
One of my chargers gives an internal resistance reading. I am recording this for the cells that get charged on that device.
The next major test is for self-discharge. I leave the cells sitting on the bench (in egg cartons) and measure their voltage after 24 hours, after 7 days, and after 14 days. I need cells which can hold most of their charge over this 2 week period.
The final (and perhaps most important) thing to test is capacity. I want a standard test for this, and so typically give the cells a top-up charge to let them all start from “full”. I discharge them at 1 A on my smart charger, which logs how many mAh it draws from the cell. The cut-off is set to 3.2 V. I record the capacity. 1 A is a higher discharge current than most laptop cells would experience, but it is a nice round number and lets the discharge test be a bit quicker than if I used a lower current. It also serves as an upper-bound for the design of my home battery.
For improved safety, these tests are always conducted outside on a compressed fibre-cement sheet. Cells are never left charging while I am away from the house.
It’s just over a year since I took the first steps towards this project. After thinking and reading about battery reclamation for a while, I happened upon two laptop batteries in an e-waste bin at work. Finally I had an opportunity to get hands-on!
Gingerly opening the laptop batteries gave me 10 cells, and 9 of them were at about 3.7 V. This is the nominal LiPo voltage and is also a good long-term storage voltage, so I was optimistic that they could still be in decent condition. Frustratingly my electronics things were still in packing boxes from moving house, and so I couldn’t do serious testing on these cells straight away.
It took another 6 months for me to really bite into this as a serious project, and in that time I’d moved house again and misplaced these initial cells in one of the packing boxes. I only re-found them this week, and was delighted to find that they had dropped a mere 0.01 V in the intervening 12 months. That’s definitely a “pass” for the self-discharge test 🙂