Anyone who has had experience with electric vehicles will come to find out that the energy storage device (usually battery) is the crux of the vehicle. On an electric bicycle, the battery has the biggest effect on the weight and performance of the vehicle.
Here are some battery types commonly used on electric bikes:
Sealed lead acid (SLA)
– these have been around for a long time. They are reliable, but they are very heavy, and they can only handle about 200 charge/discharge cycles for typical electric vehicle (EV) use. A typical 36-volt SLA pack weighs in at almost 40 pounds. And to make them last, it is necessary to treat them very well, by always (always!) charging them immediately after use, and never (never!) running them down too low. We try to encourage customers to look at other solutions than SLA, because they will end up having to be replaced every 1-3 years, and so in the long run, they cost more. But for someone just starting out with an electric bike, they are a good option, just to see how the ebike works out, since they represent a low investment threshold (typically $100 or so for a 36 volt lead acid pack).
Nickel cadmium (Nicad)
– these have also been around for quite a while. These are substantially lighter (a little more than 60%) than lead acids, and are well suited for electric bike use. They handle high discharge rates reasonably well, they can be discharged all the way without damage (in fact, it is good for them on occasion), they handle low temperatures, they don’t need complicated battery management circuitry, and they often last up to 800 charge/discharge cycles. They cost 2-3x more than lead acid, but they last 3-4x longer, meaning that over the long haul, they are a better value. Their main downside is that they contain Cadmium, a toxic metal, so they must be recycled properly after they finally die.
Nickel metal hydride (NiMH)
– these are also a nice chemistry, which are a bit lighter than Nicad for comparable energy storage. They also handle high currents reasonably well, and do not need fancy electronic battery management circuitry. However, NiMH does not like to be fully discharged – if done so repeatedly, they will not live as long. In our experience for ebike use, the Nicads seem to last longer. Also, NiMH is quite expensive right now.
Lithium Ion (Li-Ion)
– there are actually several types of lithium ion battery. The three most common chemistries are lithium cobalt oxide, lithium manganese, and lithium iron phosphate. Any one of these can be labelled “lithium ion,” but they have very different properties. Lithium cobalt oxide is very lightweight (high power density), but has a big problem – it likes to catch on fire if short circuited or abused. So few if any people use these for electric vehicles.
Lithium manganese (Li-Mn)
is much safer, but weighs a bit more. It lasts about 500 charge/discharge cycles. This is a very common battery chemistry for electric bikes, and is quite reliable and safe in practice. It provides a nice balance of cost/weight/safety. Several of our kits, such as the eZee hub motor kit and our new Cycle9/Txed/8fun kit (watch our blog for details) include a Li-Mn battery.
– Lithium Iron Phosphate (LiFePO4) – these are the latest/greatest (and much hyped) electric vehicle battery solution. In theory, these cells will last upwards of 2,000 charge/discharge cycles. They are extremely stable/safe, and somewhat tolerant of abuse. They can handle high discharge rates, though that depends on who makes them. They are about the same weight as Lithium Manganese, i.e. about 10 lbs for a 36V/10Ah pack (common electric bike size).
We are excited about LiFePO4 technology, but we have become a bit more cautious about it. We’ve found that these batteries vary widely in their quality and their ability to output currents sufficient for electric bicycle use,
depending on who makes them
. We’ve had long-term (1 year) experience with packs from several suppliers, with greatly varying results. This picture shows what we’ve had to do to fix packs for customers, supplied by one Chinese supplier:
This situation was no fun for us (we spent a lot of time replacing cells and reprogramming the battery management system, BMS) – and it was no fun for the customer who had to wait while we did that.
There are a few reputable Chinese LiFEPO4 suppliers, and we are continuing to examine options for a Chinese-supplied LiFEPO4 pack, because they are less expensive. We have 3 packs
on the way
here from one supplier who claims to have rigorous quality assurance procedures in place. If that testing goes well, we will begin selling them as a “beta test” situation, at a reduced price (
). Those are cylinder-cell based, 10Ah 36V LiFEPO4, with an external balancing charger, and 6 month warranty provided by the same company that provides the BMC hub motors. The good thing about these is that warranty/support is US based. The other good thing is that the packs are long and narrow (3.25″x3.25″x13.5″), making them a good candidate for attaching to a bike downtube or seat tube by your favorite attachment mechanism (duct tape works, though is not pretty). One downside is that they have very thin discharge wires. I am promised by the supplier that these hold up over time in the field, but I am thinking I may replace them with larger gauge wiring at some point.
My own opinion on those is that they should work reasonably well for bike setups with discharge rates below 20A. While they are rated up to 30A, past experience tells me it is never good to push the limits on such a rating, or cell failures are a likely result. One solution to this problem for higher-current needs is to double up on capacity, i.e. get a higher capacity battery (20Ah instead of 10Ah), which can handle higher discharges. Another is to put two batteries in parallel, so each one is providing approximately half the current. I use that solution on my bike, with two 10Ah (36V) LiFEPO4 packs in parallel to provide up to 35A for my current-hungry controller. Before, I was using just a single 10Ah pack and it was clearly unhappy with 35A – I ended up replacing a failed cell. But now that I have them in parallel, things have stabilized.
So, all that said, the most field-proven LiFEPO4 batteries that can handle high currents and really do live up to the hype in most cases are Taiwanese-made cells (and US made cells, if you can get them).
However, the USA and Taiwanese-made LiFEPO4 packs (like A123 systems) tend to induce a bit of sticker shock. But, for those who can overcome that shock, they are worth it in the long haul.
One Taiwanese supplier of LiFEPO4 is Phisiang. We’ve carried some of their packs in the past, like the one pictured here:
These are nice packs, and have been very reliable. However, they have two issues:
– They are low capacity only, 7Ah for the 36V version. That’s a pretty small pack. Also, the discharge rate is limited to around 20A. For such an expensive battery, it is unfortunate that the packs can’t handle higher currents. PHET makes larger packs, but they are unwieldy and not cost effective
– The responsiveness of the company to warranty issues (we’ve only had one) is not the best. Problems get resolved, but slowly and with big shipping expenses overseas.
We can still get these batteries for customers who request them, but we aren’t stocking these anymore.
We looked around for a suitable replacement that has a similar quality, but higher capacity and better support. The result of that is we are now a LifeBatt distributor. This is a premium LiFEPO4 battery that is guaranteed for 3 years, and can handle the high currents required for electric bike use.
This battery features cells that can handle up to around 100 amps (10C rates), with a sophisticated onboard battery management system (BMS) for cell balancing and monitoring. There is a serial port on the battery for diagnostics, and a low-voltage-cutoff circuit can (should) be added for electric vehicles where there is not a low voltage cutoff in the controller.
More pricing and information will be available soon, but we’ve pictured below the 36V 10Ah pack, which is $750 plus shipping, from Taiwan.
We’d love to stop there and be done with it – the LifeBatt packs look to be really nice over the long haul. However, there are still two things to consider. First, the shape and size of the packs is not optimal for many electric vehicle configurations. Second, one of the owners of the LifeBatt company has had conflicts on public forums with various members of the electric bike community. We won’t take sides in that, but we will say that we understand that there are some people who don’t want to buy a LifeBatt for that reason.
So we are now on our way to another option. We will be building up custom packs for customers on request, using the same high quality Taiwanese made cells that go into the LifeBatt. We cannot claim that these are for electric bike use, because there are various patent battles over who owns the rights to LiFEPO4, and who can use them on e-bikes (silly IMHO, but that’s for another blog post). But we can say that we would have difficulty preventing a customer from putting one of these on his or her bike (especially if that customer doesn’t tell us about it). And we can also say that we are planning a configuration that is more amenable to strapping these packs to various kinds of metallic tubing that comes in, say, 3/4″-2″ shapes/sizes. And our price will be competitive with other packs made from similar high quality cells. Interested? We are about to begin the first batch, and it will be limited to about 4 packs total, two of which are already spoken for. Please contact us if you want to know more.
Now, finally – in the meantime, we also continue to stock other options, such as Nicads that provide a viable and lower cost alternative to LiFEPO4. We have one Nicad pack that is triangular, and fits right into a triangular bag in the main triangle of the bike frame. That’s a pretty nice configuration, for a much lower expense.
We will be announcing our new web-store soon, and all our battery options will be listed there. Or, you can