Tesla Eyes Ultracapacitor Future With Maxwell Acquisition

As reported by Bloomberg, Tesla has acquired the innovative energy storage company Maxwell Technologies for $218 Million. The move is a direct departure from Tesla’s current energy storage requirements; instead of relying on lithium battery technology, this acquisition could signal a change to capacitor technology.

The key selling point of capacitors, either of the super- or ultra- variety, is the much shorter charge and discharge rates. Where a supercapacitor can be used to weld metal by simply shorting the terminals (don’t do that, by the way), battery technology hasn’t yet caught up. You can only charge batteries at a specific rate, and you can only discharge them at a specific rate. The acquisition of an ultracapacitor manufacturer opens the possibility of these powerhouses finding their way into electric vehicles.

While there is a single problem with super- and ultra-capacitors — the sheer volume and the fact that a module of ultracaps will hold much less energy than a module of batteries of the same size — the best guess is that Tesla won’t be replacing all their batteries with caps in the short-term. Analysts think that future Teslas may feature a ‘co-battery’ of sorts, allowing for fast charging and discharging through a series of ultracapacitors, with the main energy storage in the car still being the lithium battery modules. This will be especially useful for regenerative braking, as slowing down a three thousand pound vehicle produces a lot of energy, and Tesla’s current battery technology can’t soak all of it up.

29 thoughts on “Tesla Eyes Ultracapacitor Future With Maxwell Acquisition

  1. Another big problem is that super capacitors have a finite charged lifetime- the capacitors can only be under charge for something like 6,000 hours before their capacity starts to fail.

    I was planning to replace the batteries in my motorcycles with ultra capacitors, until I read the spec sheets and found the limited lifetime ( roughly 300 days charged lifetime). A regular lead acid cell will last several years if treated well, for the same or less money.

    1. Maybe you’re confusing charge retention with lifespan? Ultracapacitors are leaky.

      Then again, so are Tesla cars. The system standby power used to be kilowatts until they made a software update that shut down most of the system when the car is “off”.

      1. There is still a limit to the charge/discharge rate, otherwise it would imply they can handle infinite currents; it’s just that the maximum charge/discharge currents are much higher compared to a battery pack of comparable size and weight.

        However, for many electric cars the limiting factor for the battery pack is total capacity, not peak power capability. Tesla uses a drivetrain with very high peak powers, compared to more “tame” electric cars, so it may still be a problem for them. Generally, the allowable charge rate is lower than the discharge rate, while we expect a car to decelerate much faster than accelerate, meaning the currents during regenerative braking are larger.

        You can’t just connect supercaps directly parallel to a battery pack and expect the supercaps to do any significant work; batteries have a very flat discharge curve, while the charge/discharge curve for supercaps is perfectly linear (for a constant current). Capacitors can only absorb or supply current when the voltage across them is changing. This means that at least one of them (the battery or the supercaps) must have a bidirectional switchmode converter, which is going to be an impressive piece of power electronics with equally impressive capacitors and inductors.

        1. Maxwell has a hybrid LioN battery with super cap with dry electrolyte. This hybrid battery has higher energy cap in KWH and Much faster charge/discharge rate. It also has a much longer life. Current Demonstrated energy capacity is equal to Tesla’s current tech at “>300wh/kg”. The potential is “>500wh/kg” (>120kwh/880lb cells) (>200kwh/880lb cells). Recharging from 20% to 80% in 10 min would be very nice and is well within reach with this tech.

        2. Of course there’s a limit. :)

          A possible use would be a quick charge feature. You stop at a charging station, charge your capacitors quickly, then drive away and the capacitor more slowly charges the battery. That way, the capacitor isn’t holding charge for long and you aren’t tied to a charging station for as long. That could get you, for argument’s sake, 20% charge. If you wanted to leave immediately, then that’s just 20% quickly. If you use it with a full charge cycle, then you only have to charge the battery up to 80% and then you can go with a “full” charge.

          1. Tesla has multiple reasons to invest in capacitors.

            The biggest elephant in the room is that Panasonic’s NCA batteries are woefully dangerous to use for their poor fire safety. Tesla is already being sued for “not doing enough” to ensure their cars don’t burn down in a serious crash, which is a minefield for Tesla because they know the batteries aren’t safe and if this turns up in court they go the way of Ford Pinto.

            Other manufacturers use LiMnO batteries which are closer to being fireproof, but the whole industry is just waiting for the safe, affordable, dense, powerful, and long lasting battery to emerge. Tesla is trying to develop NMC cells as well but they carry a fundamental compromise between power delivery, energy density, and lifespan, so they can’t make them perform to the “ludicurous” standards of acceleration and quick charge.

            But, the performance of the Tesla vehicles is already based on party tricks: ludicurous mode uses just the top 15% of charge when the voltage is highest and then downgrades (Another lawsuit against Tesla: false advertising, acceleration goes down as battery empties. Tesla won: it was in the small print) – so they’re probably figuring that they can use ultracapacitors to the same effect while actually using slower batteries for the main traction energy.

  2. My experience was, that the larger the capacity, the lower the inner resistance, the bigger the leakage current. Good for a buffer and short term storage, but bad as battery replacement.

  3. As it is, the battery pack and power conversion circuits must all be designed to meet the needs of the maximum draw. Perhaps they want to save money by having a lower current capacity battery pack augmented with the super-cap booster. At times low lower draw, the super caps charge up, and at higher than average demand, the super caps add oomph.

    1. “Must”. As it is, Tesla cars can put out more power than the batteries can deliver – except when charged to the very top. They call this “ludicurous mode”. Every time they upgraded the battery size, they boasted faster acceleration, because adding cells gives a little bit more current out at the top.

  4. I’ve been using 6 off 3500F 2.7v super caps in series in a 4 cylinder Camry sedan for 5 years, they are kept from going flat by a small SLA battery and soon with a small solar panel and Li ion pack instead. Have heaps of cold cranking capacity even at 10v, lighter than the factory arrangement and suits me as I’ve not needed to run the electrics for more than an hour when engine off ie I can happily run park lights with radio for an hour and still with more than enough starting capacity :-)

  5. guess using supercaps for capturing energy from regenerative bakes makes sense? without knowing, i assume that’ll be a bigger spike in available amperage than what the batteries can efficiently receive? Then that energy can be dumped either as soon as acceleration is needed or into the batteries at a lower rate.

    1. Indeed, I tested my engine start super caps and they handle 200A charge at 2.7v just fine and could probably do 300A, I have six in my starting setup for a front wheel drive Camry, so if I switched them to parallel they could handle 1200A in spec such as during braking, then switch back to series to gracefully dump into a Li Ion for use in later takeoff. I would rather have a separate super caps bank just for braking with ‘gentle’ graduated PWM controls and help takeoff too.

      My rear wheels could well be driven by a suitable axial flux motor or two even without an intermediate Li Ion battery bank. There is of course nothing wrong with the paradigm where other than super caps for starting the engine I could contrive an equilibrium state where the braking super caps bank drives the rear wheels for long enough to get moving from its remaining charge then when braking charges them back up where an approx 50% sweet spot charge equilibrium allows smooth driving takeoff and brake without needing an intermediate battery bank – cheaper too. Just need effort to design the changeover switch points maybe adaptive whilst also not interfering with braking in any way for urgency – even by a mS or so delay – it’s all in the pedal travel/pressure issue I expect.

  6. Mazda uses an interesting capacitor based regen system. When breaking, the alternator acts a break, storing energy in capacitors for later use. The electricity is never turned into motion again, but rather used to feed all the various electrical loads. For a period thereafter, freeing the engine from having to power the alternator.

  7. I think using supercap in car is only anecdotal. The real use is for supercharger. You don’t care to have 2 tons of supercap in charging stations to deliver the (expected) 300kW power required to charge your car in 10mn, but you do if it’s in the car. That power can be stored in supercap slowly and discharged quickly.
    Sure, when braking you’re loosing energy because battery can’t drain that energy fast enough, but supercap is not the panacea here, because:

    1) If your battery are full, the supercap internal resistance will have to dissipate the energy (and it’s a lot of energy, so hard not to burn it)

    2) If your battery is not full, you have to follow the charging curve (so still, supercap will need some way to store excess energy while the battery is slowly absorbing it)

    There’s only a small part of the charging curve where you need all power (IIRC, it’s between 20% to 70% or so), the rest of the time, you have to burn produced energy from braking, and supercap is just adding weight in that case to your car. If you think about it, for a 200 miles autonomy, the full absorbing area is only when you’ve driven 60 miles, before this, the supercap are useless (well, at least in huge quantity)

    1. I am also not sure, how big a supercap has to be to absorb the power and energy of braking. But I am sure nobody with a minuscule amount of technical understanding thinks about dissipating the energy in the internal resistance of the cap. It would be destroyed. If you need to dissipate excess energy, you use a braking resistor.
      The battery would not be directly paralled to the cap anyway, that would not really work. Probably the inverter would be switched between the battery and the caps during braking according to the states of charge of both. The system would get some more complexity.

  8. Using a bank of super-capacitors for regenerative breaking is a decent idea.
    The main advantage here is the fact that super-capacitors compared to lithium ion batteries have far lower charge discharge losses. (Most capacitors will not loose more then 1 or 2 percent of the energy dumped into them, while most batteries loses above 10%. (30% isn’t unheard off either…))

    Though, capacitors tends to have fairly severe self discharge rates compared to Lithium ion batteries. So unless that issue if ratified, capacitors aren’t really going to be the center stage of long term energy storage in vehicle applications. (having poked about, the self discharge rate is around 1mA for every 500 F. @ 25 C as an average over a 72 hour period when charged to rated voltage.)

    Other downsides of super-capacitors at current is their relatively short MTBF values.
    A glance at digi key’s offerings shows an 6000 hour @ 85 degrees C, 1F capacitor, but not in stock.
    There is also a 90 000 hour option, but that is at 25 C, so a typical summer will reduce its MTBF by a lot.
    And otherwise 2000 hours (about 83 days) @ 85 C seems like the common value currently on the market. (ranging in a wide verity of capacities.)

    So unless the capacitors are “easily” replaceable, then this isn’t really going to be more then a curiosity. (As far as battery replacements for electric cars go that is.)

    Though, regular aluminium electrolytic capacitors exist in the ~0.1-1.3 F capacities with 10-20 thousand hour MTBF values at around 85 C. Though, they likely have lower energy density. But for regenerative breaking it might be better then nothing, though flywheel energy storage is also a thing. Also, another advantage with regular aluminium electrolytic capacitors is their far higher voltage ratings and therefor no direct need to put them in series, and thereby foregoing any balancing acts otherwise needed for super-capacitors.

    At the end of it all, capacitors are not yet having the needed properties to make good batteries.
    Their MTBF values are semi horrid for large capacities, while retaining fairly large self discharge rates.

    Unless someone finds a material with a high dielectric strength, a high dielectric constant and no discernible degradation at “high” temperatures. (Kapton tape comes to mind, but it’s dielectric constant is only 3.4 while titanium oxide compositions are in the 83 to 250000+ category.)

    1. You would have to keep the caps cool and keep their charged time low anyway. Use the energy for the next acceleration or transfer it into the battery. 6000hrs is not that bad in comparison to the operating hours of a typical car over it’s lifetime. Think of an average speed of 30km/h (mostly city drive), that would be about 180.000km. If you are much slower on average you do not have to handle much braking energy anyway.
      I don’t see a flywheel as a possibility. You need a really powerful motor/generator and inverter for it. The same size as the drive components. You can not drive it directly mechanical as you want to accelerate it while you brake the car. For this you would at least need some kind of CVT transmission, but with an insane range.

      1. Yes, the capacitors would not need to be an actual long term storage for energy, and so their life expectancy would be far greater then the pure MTBF value. Since the capacitors would get used fairly infrequently.

        Though, using a non super capacitor bank to smooth out any short high current demands from the battery would also extend the battery’s life a bit. Batteries also tends to have fairly horrid efficiency when operating close to their C rating.

        But in terms of flywheels, there is actual flywheel energy storage systems employed currently in some race cars.
        https://en.wikipedia.org/wiki/Kinetic_energy_recovery_system
        All though rare, but the technology is slowly becoming a reality, and might trickle down to less enthusiast oriented cars in the future. Though the first iteration were a 24 kg flywheel system that can store about 400 kJ, or 16.66… kJ/kg, while current super capacitors reach about 32 kJ/kg according to Wikipedia, though, I can’t find much information about later iteration of the flywheel system, so comparing energy densities is hard. (Since comparing the first usable version against fairly well developed capacitors isn’t an all that fair comparison, when there is later iterations of the flywheel system in use.)

  9. I would think that it has more to do with charging than energy storage. Better buffers in the car for regen (hot batteries dont take the charge as quickly) but they wont get rid of batteries because the trade off in charge density doesnt make it worth it.

    That being said, could this be something slightly more strategic due to the 920 mil convertable bond due march 1st?

Leave a Reply to Murray Cancel reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.