Restoring depleted battery pack using RC charger
If you're willing to break depleted battery packs into the component modules, would it be possible to use a relatively inexpensive RC charger (like this one) to slowly bring it back up to at least 85 volts so that the bike's charger will once again function?
According to the service manual and Mik's disassembly, each battery pack contains 6 modules. The front pack is a little shorter, so its modules only have 8 cells while the rear pack modules have 9 cells. So, assuming a nominal 125-148 volts for the 102 cells, each cell should normally have between 1.2v and 1.5v. If you used a standard 12VDC charger on the pack, each cell in the long pack should get 1.3v and each cell in the short pack should get 1.5v. Altogether, the total voltage should be 144 volts.
I realize that this isn't an ideal charge distribution and that it might take a while to charge all these packs at the charge rate of an RC controller (0.2-5A), but speed really isn't a major consideration when you're trying to bring a bike back to life. (I'm assumng this would be an infrequent operation for most.)
But would it work? And would it be safe?
in my case it worked. I used a semi automatic car-charger that can switch between 10A and 3A in the final phase.
With the 9cell modules it worked well the charger switched from 10 to 3 A when the module was at 80% (only estimated!)
with the 8 cell module I had to switch the charging mode manually. I charged them to 1,48V per cell for testing them
with an CBA-unit. After this Testing I recharged them. But that required "personal-watched" charging with repeating
temp. and voltage checking.
With an RC-charger where you can select the cells before each charge it should be easier, but I still highly recomend
to watch the modules when you charge them. No "sit and wait" but doing some other work, where you can have a quick
Important be sure to have good connections between charer and modules.
The link to the charger is pointing to a NiMH charger, but you seem to be discussing the use of a Lead-acid charger.
I would not recommend using a lead-acid charger, although this is exactly what I am doing while typing this...!
Like Mikemitbike wrote, hang around and do something useful while watching the charger...so I'm typing and hope it will be useful.
Right now I use a "Dual Source" Lead acid battery charger / 15V power supply that I bought for about AU$ 200.- at "Super Cheap Autos". The multi-turn trimpot governing the voltage died after a year or so, but I bought 5 new ones off the internet for a few bucks and it runs very well again. It can put 10A into a Vectrix module, like the stock charger in CP mode, but the voltage keeps dropping as the battery voltage rises.
It is much easier to use a dedicated NiMH charger, because it does it all automatically. I analyzed the Vectux battery and built the M-BMS before The Laird came up with his great "Freddy" charger. A Freddy charger, or anything else that can give you a current limited 150V supply, is much easier than the tedious work with a NiMH 14-cell charger.
But it is possible to do it. See http://endless-sphere.com/forums/viewtopic.php?f=14&t=6277&sid=1551012a53beb20fd10179e39e7dd129 for much more detail - much of it wrong, too!
The one big imperative with this approach is this:
Do not try to fully charge the modules before re-assembly into a complete battery. It takes too long, the self-discharge rate will run rings around you while you try to get it done.
The way to do it is to finish analyzing and testing of each module of 8 or 9 cells at very low SOC. Finish Empty in other words.
The empty cells should hold their approximate 1.15V per cell for many month in case you experience any delays, which you most likely will (but only if not connected to the 7mA constant-on motor controller). Half a year later they will still all be empty, but still with enough voltage to allow the stock charger to kick in.
In the worst case scenario (assuming all really bad cells were identified and replaced), you might have to apply a 10 min charge to each module before you put them back together. That will not cause much additional time delay, because the inter-module connectors need to be tightened repeatedly to avoid the loosening that was the reason behind the 2009 battery recall (that's my theory, anyway!). Tighten them to 10Nm, let them sit for a few hours, repeat. Repeat again until your torque-wrench tells you that they have not loosened again. But maybe it would be better to apply a thread-lock component and tighten the bolts to more than 10 Nm just once. But to how much torque? It would be nice if those readers who know what the battery recall was really about could explain what exactly to do, and not to do...
Anyway, summary in three words: "Bottom-balance before re-assembly"!
By now, 55min later, I have manually re-adjusted the current on the (not recommended) Dual Source charger 6 times to keep it at 10A. The (temporarily assembled 10-cell) battery has entered the flat part of the charge curve and it's getting much easier now to keep the current at 10A. Time to let the timer take over unobserved, but only with working smoke alarms and fire extinguishers in place.
Oh, and I was wearing safety glasses throughout, believe it or not!
I installed a diode protected 20A cable from each battery terminal to the "trunk" termnating in color coded Anderson connectors. I connect these to a variable voltage/constant current DC power supply. If you set it to 150V, 3A you will have the battery back to "chargeable" within a few minutes. You could also use it as decribed by Mik for balancing at C/30-C/100 i.e. 1A or even less without heating the batteries. You have to shop around on EBAY. I have seen ones that had a fixed 150V for as little as ~$60. I purchased a variable voltage/current rackmount one used for ~$500. There are a lot of others uses for it.
The thing that really appealed to me about these multi-cell NiMH RC chargers is that they have peak prediction circuitry to prevent overcharging, essentially allowing them to be run unmonitored. Also, they not only charge (fast and trickle), but also discharge, enabling battery conditioning.
The better ones can discharge at up to 20A (for six cells at least) but the majority will only allow 1A max discharge current. In reality, a 9-cell module gets discharged at only 0.7A because it cannot dissipate the energy fast enough. That's with my RT808D charger.
The little chargers can be quite noisy and you will have it running for literally hundreds of hours.
This charger will charge a lot faster (by the way if you sit on the page for a while the first time you will get offered a one time discount) ($126.99):
Charge current range: 0.05 - 20.0A
NiCd/NiMH battery cell count: 1 25 series
More info here (posted soon after it was released):
Sat May 02, 2009
My favorite charger company has made a new charger. I love the input voltage range (4.5v to 32v), I love the ability to charge at 20amps/350w, and I love that it can discharge at 600w! (to an external source of course).
It also has 16mb of data logging flash memory, and a built-in Ri meter.
The outputs are completely isolated from the source, so for charging more than 8s lipo or 10s LiFePO4, you can simply use a pair of them.
I own 4 of the previous model of this charger, and they absolutely kick-ass. They do everything I could ever ask of a charger better than any other charger I've ever owned (even ones that cost 3x more), and they are tiny! You can fit them in your back pocket!
Wow, Luke, you are right. This is a great find. Being able to do 8s is a major improvement, not to mention the much higher charge rate. The reason I like 8s is because most LiFePO4-based setups are multiples of 8 cells. What is the max charge rate at 8s? I'm assuming it is not 20A, as that would be about 600W, but I'm guessing it'll be at least 10-11A, which is pretty amazing, especially for something this size...
You can download the manual here:
You don't need to split the pack apart (beyond getting access to the terminals). Its mentioned above but I asked an expert about using multiple isolated chargers (isolated isn't necessary for a single charger) to charge an entire large pack at the same time. Here are the relevant portions of his two responses:
If you broke the pack up into, say, 4 parts, then you would need 5 wires coming out to the chargers. If you bundle the wires together and use a 5 contact connector (single piece), then I don't see it being a safety risk. You can stack Anderson connectors and pin them into a block.
...another approach would be to have multiple 2 wire charger connections coming off the pack... This is exactly how my old Scooter works with SLA batteries. I could charge with a single charger and take turns if I wanted. This also tends to help with balancing. As long as the wires are properly routed and protected (including fuses/breakers), I don't see any particular hazard.
You could easily wire it up to charge two modules at a time. And get two chargers and set one charger for 18 cells and one for 16 cells and use different matching connectors for each number of cells to make it mistake proof.
I think you could also use RC Charger(s) for periodic pack conditioning as well. As quoted from the below there is a mode designed for that purpose but it requires a 1C charge rate and is limited to 9a. But the newer larger iCharger 3010B might do it.
NiCd/NiMH forming charge
This forming charge program aims to eliminate capacity imbalance between cells in a battery.
The iCharger first charges with constant current (CC=1C) according to the user setting. When the charging
voltage reaches the peak threshold (1.48V/cell) it switches to the CV phase. In the CV phase the current
gradually falls. When the current drops to C/4 the iCharger will charge another 25% Capacity at C/10
current and then terminate the process.
If you are interested I check the manual for this unit ($229.95):
iCharger 3010B 1000W 10s Balance/Charger
A high powered charger, capable of handling up to 1000W or 30A charge rates.