Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register

Gigacapacitor?

This site may earn commission on affiliate links.
Yet another engineer who slept through the thermodynamics section of his freshman physics class.

All too common. Some years ago I was visiting the observatory of the local astronomy club, and one of the guys - an engineer - was in the warm room replacing the electric heaters. I asked him why, and he said that the new ones will be "more efficient". Geesh. I'm pretty sure even the astronomy club can't violate the laws of thermodynamics...
 
All too common. Some years ago I was visiting the observatory of the local astronomy club, and one of the guys - an engineer - was in the warm room replacing the electric heaters. I asked him why, and he said that the new ones will be "more efficient". Geesh. I'm pretty sure even the astronomy club can't violate the laws of thermodynamics...

I tried to explain this to my dad. We were talking about a system needing additoinal heating and he kept bringing up efficiency of the electrical devices. I kept telling him it didn't matter, unless he was considering steam (from NG) as a heating source.

He eventually got it, but it was rough.

I often get in arguments with him that boil down to him deciding to redefine a word (or taking a lay definition over a technical definition). The last time we did that he decided to redefine 'sound' to the 'act of hearing'.
 
He only ever posted on the forum in a two-day window, and only in this thread. I'm assuming he lost interest when no one was buying his argument.

Yes, the thread is thoroughly hijacked. I should probably rein it in or something I suppose...
 
It would be good to hone in a little further on the stated topic of this thread, and consider why ultracapacitors (or supercapacitors) have been studied so intensively for electric vehicle use: because their electrical properties are complementary to those of lithium ion batteries, not directly competitive with them. You might have noticed that your Model S has a peak motor output power of 240 or 300 or more kW when you really step on it, but that your maximum regen capability on a steep downhill is far less - 60 kW if the battery is warm and even less if it is cold. That's unlikely to be a limitation in the motor itself, since it performs as a motor to provide torque when you supply it with electricity and as a generator to provide electricity when you supply it with torque. Therefore, its ability to accept input power should be about as large as its maximum output power.

We also know that, with the newest superchargers, Model S batteries are capable of accepting 90, 120 or in some cases even 135 kW of input power, at least when the batteries are fairly empty. Why not allow the battery to accept more power during regen if it's available?

One thread I read suggests that it could be a software issue, and that Tesla didn't want to make regen too jerky if you take your foot completely off the pedal:

http://www.teslamotorsclub.com/archive/index.php/t-9614.html

That may be true, but I think the more interesting and provocative reason is that the batteries and associated power electronics may not be intended to absorb instantaneous spikes of power that high. And yet, that's exactly what ultracapacitors thrive on. They are not intended to store large amounts of electricity for long periods of time, but they are absolutely perfect for absorbing and releasing very large power spikes over short periods of time. In fact, Maxwell and others are currently selling ultracapacitor-based products to do exactly that in buses and large trucks.

The other key desirable quality is their ability to continue performing well at very low temperatures, which is a condition lithium ion batteries are not nearly as happy about. In fact, ultracapacitors last longer if operated at low temperatures. So what does that all mean?

Imagine if Tesla offered an optional upgrade of an ultracapacitor module that would fit in the supplemental storage space below the hatchback area. That space is perhaps 5-7 cubic feet. Would it be possible to use such a device to provide four key benefits?

1. Recover more power from regenerative braking without making the handling jerky, by letting the additional power flow first into the ultracapacitor, and then more slowly into the battery thereafter.
2. Allow even greater instantaneous bursts of acceleration and/or lower vehicle weight and cost and longer battery life, because the battery pack would not have to be designed to handle such large power flows. It could instead be designed to keep some electricity in the ultracapacitor for that purpose, filled slowly from the battery but drained quickly by the motor during periods of rapid acceleration.
3. Allow faster range supercharging, by serving as temporary storage for large amounts of electricity that could be fed into the battery more slowly as the battery is able to accept it.
4. Allow the car to perform more normally at very cold temperatures, rather than limiting regen capabilities until it warms up.

Here's Wikipedia's chart showing energy density and power density for various types of energy storage devices:

Supercapacitors-vs-batteries-chart.png


The latest supercapacitor designs are based on graphene, and offer even more promise:

http://www.technologyreview.com/vie...-vehicle-energy-storage-say-korean-engineers/

As more research unfolds, the possibilities for Tesla to utilize the best of lithium ion batteries' capabilities and the best of ultracapacitors' capabilities in a combined system start to look really interesting. This thread could be a good place to house a running discussion of those opportunities.
 
3. Allow faster range supercharging, by serving as temporary storage for large amounts of electricity that could be fed into the battery more slowly as the battery is able to accept it.

While the other points appear valid, I don't see how this could really work. The ultracapacitor would have to be of comparable capacity to the Lithium pack. It might help during the tapering function a little bit, because it could take up some of the slack, but it would need to have a sizable fraction of the main pack capacity for this to work. Obviously it could only dump its charge into the main pack after the Supercharging was finished.
 
Tesla planned to do a hybrid of Li-ion and a Supercapcitor, for exactly the reasons you suggested.
They came to the conclusion that with increased size of the batterypack, the 7000 cells started to have properties of a Super capacitor.
Extrem high input and output power. The max output is limited by the transformer, the Model S 85 and P85 have the exact same battery pack but the P85 model has better a transformer.
As for the regen, as you stated correctly the regen is limited to 60kw so regen doesn't feel like a fullstop attempt, again a Super capacitor wouldn't change that.
Tesla did all the math regarding Supercapcitors and came to the conclusion that existing technologies weren't worth it.
I doubt they will ever be worth it, with 135 kw input we already reached a very high charging current, and trying to going beyond 200kw would have other barriers like gridpower, physical cabel properties, and other physical limitations.
 
Plus, there are other lithium chemistries that can take much higher currents than the Tesla NCA cells, such as lithium titanate, yet still have much better energy density than capacitors. So if one were doing a hybrid pack of a high C rate cell coupled with a high energy density cell they might make more sense.
 
How do cjc9er's arguments regarding temperature (capacitors favored at low-Ts), and time (capacitors not meant to hold a charge for long periods) stand up? More specifically, does it appear that newer combinations, such as JRP3's Li-titanate, have or may have superior low-T capabilities? Is there something inherent to a capacitor that precludes it from being a good long-term storage vessel?
 
There's a good discussion about the different ways that batteries and supercapacitors behave at low temperatures here:

http://www.cellergycap.com/index.php?option=com_content&view=article&id=17&Itemid=3

All the various types of lithium batteries rely on an electrochemical process that doesn't work as quickly or efficiently at low temperatures. Supercapacitors also lose a little performance at very low temps, but the effect is not nearly as severe.

The fact that supercapacitors have lower internal resistance at most typically encountered temperatures not only helps with cold weather performance, but also improves the overall energy efficiency of the charging process, particularly if some of the stored electricity could go straight from the regen circuit or the Supercharger to the supercapacitor and back to the motor again without having to be stored in the battery. Various estimates have been published of how much of the AC energy you start with at a Supercharger is lost in various power conversion steps before it becomes DC energy into the motor, but loss estimates of 15-25% are common. This is extra energy we all pay for when charging at home, and Tesla pays for at Superchargers, so there are good financial reasons to reduce the losses.

The single most important reason to incorporate supercapacitors into Teslas in the future may turn out to be their ability to extend the functional lifetimes of costly battery packs. More info here:

http://www.google.com/url?sa=t&rct=...vRE1__xdVp630EQ&bvm=bv.62578216,d.aWM&cad=rja

It's also worth remembering that past supercapacitor designs have had serious cost and performance limitations that prevented their wider consideration in electric vehicles, but those limitations are beginning to ease in the latest formulations and designs, and especially in the new research prototypes incorporating graphene, aerogel, carbon nanotubes, and other advanced materials. At the scale Tesla is planning in the gigafactory, all sorts of things are possible in hybrid lithium ion/supercapacitor battery backs that have seemed off the table in the past. A glance at Maxwell (MXWL) Technologies' stock performance since the Tesla announcement suggests that the market may share those sentiments...
 
I'm not sure I understand your efficiency argument, power still has to be transformed from AC to DC either in the Supercharger, the on board charger, or the motor inverter, regardless of it being sent to batteries or capacitors. Plus capacitors have a high self discharge rate which hampers their actual overall efficiency. Plus their poor energy density means you are adding a lot of weight for very little storage capacity.
 
JRP3--

True, but the coulombic losses tend to be higher in batteries than they are in supercapacitors, so even though the ac-dc conversion losses and the high voltage conversion losses are the same in both cases, supercapacitors are more efficient at storing and retrieving large instantaneous pulses of DC power than batteries are. You're quite right that self discharge losses are higher in supercapacitors over long periods of time - that's why they're really just intended to store power on a time scale of seconds, minutes, or maybe an hour, as opposed to days or weeks.

If you take a look at a 30 mile graph of real time energy use in a Model S when operating in stop and go city traffic or when climbing and descending steep terrain on a two lane road where there are brief opportunities to pass other cars, it gives you a sense of the amount of power that routinely flows in both directions (from the battery to the motor and from the motor to the battery) during normal driving. That's hard on the battery pack and would tend to shorten its life over time. It's during those transients that supercapacitors would really excel, and the intended size would be far lower than that of the main battery. Not 60 to 85 kWh and 1300 pounds or more, but something closer to 1-4 kWh and maybe 40 to 150 pounds.

In my experience, most of the periods during which I'm heavily using regen are 0.1 to maybe 0.5 minutes long at the most. If a supercapacitor could absorb 100 kW of regen instead of the 60 kW the battery is currently limited to, that would be 0.1 to 0.5 minutes * 1 hr/60 minutes * 100 kW = 0.2 to 0.9 kWh of temporary storage. Conversely, repeated rapid accelerations at 240 kW for bursts of 5-10 seconds each would use up something like 4 kWh of supercapacitor storage after a full minute, which is more temporary storage you would ever need when periods of rapid acceleration are usually followed by periods of fairly rapid deceleration.

Having a few kWh of supercapacitor storage onboard instead of just 1 would also allow the car to function more normally when it's cold, and would speed up full range charges noticeably. Once the battery is mostly full at a supercharger, the last few kWh of capacity get added as slowly as 1 mile/minute in my 60. If we had 5 kWh of supercapacitor storage onboard, a 120 kW supercharger could fill it up in about 2.5 minutes (or less, if the supercharger itself had supercapacitors embedded), vs. roughly 15 minutes for the last 5 kWh of a battery pack. So that would help Tesla get greater utilization out of a scarce and increasingly time-constrained resource as more and more Tesla drivers are sharing supercharger capacity. The supercapacitors could then power the car for the first few minutes of driving right after the charge is complete, avoiding having to fill the batteries all the way to the maximum and risk shortening their lifetime.

Not a perfect solution and not necessarily cheap, but it does seem at least to be promising enough to warrant further investigation...
 
The problem is you're really talking about fringe events, and adding a good amount of weight and cost to address them. At this point I don't think the technology is even close to being practical. Maybe someday, but equally likely is battery improvements will make them unnecessary.
 
I want to add two points when discussion supercapacitors in a Model S drivetrain.

1) you must pair the capacitor bank with a dedicated power electronics unit. This unit must transform between main pack voltage (350-400V) and the capacitor bank voltage (0-500V) at rates of approx. 100kW. This adds cost, space, and cooling requirements. To the point where an equal improvement could be made by adding more cells, but arriving at a much more elegant solution.

2) the control unit would try to keep the capacitor bank filled somewhere around half of its capacity*. Because it can't anticipate your driving, it must be prepared for a full regen AND a full acceleration any time. Whereas a KERS in racing can be designed for consecutive cycles of charge-discharge.

*) A full acceleration from 0 - 130 mph within 10 seconds at 360kW equals 1kWh.