Using capacitors for initial current peaks w/lithium battery

[spinningmagnets]

This video is for off-grid, and running a tool from a small battery pack coupled with a capacitor bank. He specifically shows a circular saw running off of a DC-to-120V AC inverter, which has a high peak start-up amp draw, and a much lower continuous amp-draw. this sounds suspiciously like a large direct drive hubmotor on flat land, and I am now very interested.

7-capacitor balance boards $16 each, optimized for seven $12 capacitors, so $84...roughly $100 per module in parts

Ideally, a medium-sized lithium battery pack can use cells that are optimized for long range and high Ah per cell, rather than peak amps (like the 25R / HG2). This is only an issue because bicycles don't have room for a Tesla car sized battery pack. Relevant section at the 5:38 mark...

[I know capacitors are not energy-dense like chemical batteries, but...once implemented, they should last millions of cycles, yes?]

https://youtu.be/YMKojcYP7QQ?t=296

[youtube]YMKojcYP7QQ[/youtube]

 

[Chalo]

The problem with any capacitors plus batteries implementation is that a capacitor's voltage falls linearly all the way to zero as it discharges. So whatever capacitor bank you add requires its own voltage conversion just to be usable within a battery powered system. And then the power limitations of the voltage converter become the limitations of the supplementary capacitor. It's not impossible to work with, but it takes away much of the attractiveness of caps versus just using a more powerful battery.

 

[Hillhater]

Same old, same old,....
Which ever way you mix it , supercaps are still big, expensive, heavy, low capacity, with a high voltage range, and require complex management systems.
Their only benifits seem to be a very high charge rates (if you can supply the charge demand) , and a life cycle that far exceeds any logical requirement.

 

[amberwolf]

Without a conversion system like Chalo describes, the caps can only use a tiny portion of the energy they contain.

So, as a bad math example (and my math could be wrong so it might not be as bad as I'm showing, or it could be worse):

Say you're using a system that runs from 48v LVC to 58v HVC. That's a range of 10v, out of 58v. So I guess that's 17% voltage used vs the entire range.

So if you have 100F of caps, you actually only get 17F of energy out of them over the range of the pack.

That means you actually have to have 5.8 times the number of caps to get the same energy as you would if you used a converter that sucks them down to 0V.

I guess it depends on the application and volume/weight/cost limitations whether or not it would be better to use almost 6X as many capacitors as you "need", or a converter system.

 

[spinningmagnets]

Thanks for the calc's amberwolf. Lots to think about...

 

[amberwolf]

Something else I forgot to add:

Not only are you limited by the voltage range previously noted, but you are actually even more limited by the *voltage sag* range of the battery at the moment of use.

So let's say it's fully charged at 58v, and you pull enough power to cause it to sag 3v.

3v is only 30% of the 10v range previously noted, so it's only 30% of the 17F.

So you only get the energy of 5.1F of that 100F of caps, during that particular voltage sag incident.

There's other complications I can imagine to the math, as well, such as that the energy provided by the caps negates some of the voltage sag, so you aren't even getting *that* much out of the caps, so it doesn't look like there is any straightforward "simple" equation to determine how much capacitance you need (even assuming zero ESR), or to figure out exactly how well any particular capacitance will perform.

I'm sure an equation *could* be written to determine these things, but I don't know how to do it.

 

[Chalo]

Something else I forgot to add:

Not only are you limited by the voltage range previously noted, but you are actually even more limited by the *voltage sag* range of the battery at the moment of use.

.

I was gonna point that out, but I didn't think that it was even worth mentioning on account of the already bad economics of the situation.

Caps can also soak up more regen energy than batteries are able to store while it's available, but again that's a very small benefit for a rather large outlay.