breakthrough research

On Tue, 02 Dec 2014 14:42:19 -0800, John Larkin Gave us:

NASA is making HUGE MLCC "Ultracapacitors" as "battery" storage technology, and they say it will outdo current stuff by weight and size. HUGE improvement.

I should be able to get an electric bike that goes a few hundred miles between charges, or go really fast for a short burst.

Where are the energy density figures?

It looks like the usual contentless university promotion of some minor research publication.

We'd all love to see a capacitor or a super-capacitor that offered the same kind of energy density as a petrol tank - or even a lithium-ion battery - but this probably isn't it.

--
Bill Sloman, Sydney

energy.html

Indeed. It really isn't hard to compare C * V_rated^2 / 2 of a cap with capacity * V_half_charged of a same-sized battery. You'll come away disappointed every time.

And when it happens you'll know it works because they'll be in cell phones. Possibly blowing people's buttocks off when the capacitors short, to boot.

--
Tim Wescott 
Wescott Design Services 

If you sit and run the numbers, you find:

1. Conventional type II MLCCs aren't great;
2. type II dielectrics store energy approximately linearly (after a brief quadratic toe before saturation sets in),
3. type I dielectrics are better than electrolytics, but only in high voltages. Also, they're expensive as hell.

(1) is easy to see from physical law: energy density is inversely proportional to perm(ittiv/eabil)ity. So of course high-K dielectrics are bad for energy storage. They're also bad for DC bypass, but cheap and small, so they're good enough. (Same with high-mu powdered iron cores.)

(2) is interesting; there might be a paraelectric effect underlying it, I dunno. At any rate, the common stuff doesn't just drop dead, it saturates gradually. The linear curve rises slower than the quadratic curve of C0G, so unless you have a significant advantage in dielectric strength (about

3-5x) -- which as far as I know, you don't -- there's no help here.

(3) is neat. After surveying all capacitor types, I've learned C0Gs are by all measures one of the best capacitors, period.

Also, electrolytics offer the cheapest energy density by value* (well, duh).

*Among 'fast' types, for a suitable definition of 'fast'.

And if you ever thought, "gee, what this circuit needs is a low voltage film", well wonder no more, because that's exactly where aluminum polymers fit. Polymer density and speed is very similar to film, just lower voltage and proportionally higher value.

That's also why you won't see high voltage polymers any time soon, besides the technical issues. When they do eventually show up (I presume they're possible), they'll be a little better than film, but not by too much. They'll still be expensive, too.

Tim

--
Seven Transistor Labs 
Electrical Engineering Consultation 

On a sunny day (Tue, 02 Dec 2014 14:42:19 -0800) it happened John Larkin wrote in :

Apart from the capacitor chemistry that I know nothing about, do you have to be 'Professor' to write that?

This we had discussed in 3 grade high school.

"...the researchers investigated the charging/discharging behavior of 126 resistor-capacitor (RC) combinations of 18 resistors, three ceramic capacitors, and four aluminum capacitors."

Something has to have been lost in translation, or the author of the article is clueless about circuits, because that just doesn't make sense at all.

Probably the translation was done by a person fluent in Japanese but totally clueless about electronics.

Maybe they discovered some relation and decided they didn't need to try them all. Or maybe the remaining combinations are for the next publication. ;-)

Jeroen --how did they slip that past review?-- Belleman

18 different resistors and 7 different capacitors (of which 3 are ceramic and 4 aluminium) gives 126 combinations.

Cheers

--
Syd

I think that was John's point, although in the states, I didn't learn it till 11th grade. ;-( Mikek

--
This email has been checked for viruses by Avast antivirus software. 
http://www.avast.com

Maybe physicists have discovered things about RC networks that we mechanics have missed all these years.

--
John Larkin         Highland Technology, Inc 
picosecond timing   precision measurement  

energy.html

Yup, I agree! It should be OBVIOUS that you want MINIMAL resistance in any energy storage circuit! This stuff was done storing charges on ceramic capacitors, and I couldn't even see the point of it. It seems, MAYBE, that they are using different resistor values to get a quick charge and a slow discharge. This is stuff from first-year EE classes on RC circuit analysis!

Jon

There are things that would EASILY slip past some physicists as being something really new, that would make an EE roll his eyes and groan "not this CR** again!".

I've seen a LOT of pseudo-science and just plain nonsense in the last decade. (Oh, maybe those are the same thing...)

Jon

Both!

I will grant you that 18x7 = 126 and that only Japanese researchers would be painstaking enough to enumerate every last one of them experimentally without bothering to solve the (simple) equations.

But if you allow for network topology R1,R2,C as shown in the paper the number of combinations is at least 18x17x7 - there seems to be an L there too!

The 3D coloured graph is fun but contains very little information.

I hate to agree with JL but in this case he is right on the money.

They might eventually graduate to solving second order ODEs if given a bit of encouragement and you really have to wonder about the quality of the reviewers and the journal that published this "research".

OTOH there are really weird behaviours in real world high capacitance and ultra high resistance devices that do have to be compensated by a network elsewhere and maybe this has really been lost in translation.

TBH I can't see the point of R2 at all unless it is intended to model leakage current. R2=infinity is as good as it gets.

--
Regards, 
Martin Brown

Yup, they "optimized" the circuit by using a small resistor when charging and discharging into a large resistance. That's like saying the way to make money in the stock market is to buy low and sell high.

--

Rick

All capacitors, including ultracapacitors, can store charge for short periods of time -- sooner than later, they start to discharge . A battery, once charged up, can pump out energy for long periods of time (several hours). on the other hand, can store

On a sunny day (Wed, 03 Dec 2014 11:57:02 -0600) it happened amdx wrote in :

I cheated, was building amplifier for the school band, and interested in electronics capacitors etc, so asked the math teacher in class about charging curves, exponential stuff etc, he took of, and after a long intense discussion with the rest of the class maybe asleep or simply stunned, finally apologized to the rest and things went on as usual. I had the same event happen with the physics teacher, about electron orbits in a magnetic field. We stayed after classes with a blackboard full of equations.

Those were great guys, I could not always get along with teachers, but those were really cool.

"Some of these circuits could be charged in less than 20 seconds and hold the charge for up to 40 minutes, while having relatively large capacitances of up to 100 milliFarads (mF)."

20 seconds?

Anyway, the diagrams show charging with a constant current source, so R1 and L are irrelevant (and what happens when you put a current source in series with an inductor?).

The "holding" circuit shows an open-circuit constant current source, which doesn't seem like a good idea at all.

Sylvia.

Charging a cap through a resistor isn't a good idea for efficiency. Physicists should know that.

--
John Larkin         Highland Technology, Inc 
picosecond timing   laser drivers and controllers