Possible to stack small caps to create larger value at higher voltage?

[motolectric]

I don't know much about big caps but came across a good buy on some small caps:

from their site

"Large value .47 Farad (470.000µF) super cap is rated 5.5VDC. Size about .80”Dia. x .29”Tall"

What I need is something along the lines of a 1-2 Farad cap like this one:

Amazon.com: Rockford Fosgate CPC10-03 1 Farad 20 volt Surge Capacitor: Automotive

that is happy at 12-20 volts. The type shown at Amazon is just too large to fit into my space.

Is it possible to put a few of the small ones above in series or parallel such that I get the storage capacity that I'm looking for?

Thanks for any and all tips or advice.

M./

 

[smanches]

If you put them in parallel, you can add the capacitance, but the voltage stays the same.
If you put them in series, the voltage gets multiplied by the number of caps, but the capacitance gets divided by the number of caps.

Example.
10 x .5F@20V caps in parallel = 1 x 5F@20V
10 x .5F@20V caps in series = 1 x.05F@200V

 

[motolectric]

If you put them in parallel, you can add the capacitance, but the voltage stays the same.
If you put them in series, the voltage gets multiplied by the number of caps, but the capacitance gets divided by the number of caps.

Example.
10 x .5F@20V caps in parallel = 1 x 5F@20V
10 x .5F@20V caps in series = 1 x.05F@200V

So I wire

3 x .47F@5.5V caps in series = 1 x.156F@16.5V

then 6 of those modules in parallel

6 X .156F@16.5 = 1 .94F@16.5 V

am I right?

Thanks for the quick feedback.

M./

 

[MikeMl]

No so fast. If you put capacitors in series, you also have to tie "equilizing" bleeder resistors across each capacitor. Otherwise, the capacitor leakage current differences from capacitor to capacitor will unbalance the series voltages across the string, eventually putting the entire volage across just one capacitor, causing it to blow up!

 

[smanches]

Yes, that would do it. Although you should derate them a bit. The manufacturing tolerances are usually printed on the caps. You should be at least that amount over the voltage you need.

So if it's a +/-20% tolerance, make sure you make your "BIG cap" handle 20% more voltage, and even a little more to be on the safe side.

And do what Mike just said.

 

[motolectric]

No so fast., eventually putting the entire volage across just one capacitor, causing it to blow up!

Yeah, that won't be much fun at all.

Yes, that would do it. Although you should derate them a bit. The manufacturing tolerances are usually printed on the caps. You should be at least that amount over the voltage you need.

So if it's a +/-20% tolerance, make sure you make your "BIG cap" handle 20% more voltage, and even a little more to be on the safe side.

And do what Mike just said.

Okay, then I'm at

4 x .47F@5.5V caps in series = 1 x.1175F@22V

along with the resistors across each caps terminals

then 6 of those modules in parallel

for

6 X .1175F@22V = 1 .7F@22 V

which sounds to me as an okay way to go. Although I will have as much into my parts as buying one of the 1 F caps at Amazon I only have about 35% of the volume which makes it all a fair deal.

Final question, since my voltage regulator is hard set to limit output to 14.2 V and 120% of 14.2 is 17.04, would I really need the 4th cap in each of the modules or can I get by at the 16.5 V?

Thanks very much for the help with this.

M./

 

[Sceadwian]

It's a good idea to err on the side of caution with voltage ratings, higher is a better idea unless you simply can't afford it.
Don't forget to add your series and parallel ESR's together to see what kind of total ESR you'll end up with, too many in series and not enough in paralell and you may not get the desired results you want, depends on how fast you need to discharge the caps.

 

[motolectric]

It's a good idea to err on the side of caution with voltage ratings, higher is a better idea unless you simply can't afford it.
Don't forget to add your series and parallel ESR's together to see what kind of total ESR you'll end up with, too many in series and not enough in paralell and you may not get the desired results you want, depends on how fast you need to discharge the caps.

Yes, I'm a cautious kind of guy anyway, just trying not to do my normal overkill production.

Unfortunately, I don't know how to calculate the ESR. I want to use this cap in parallel with a motorcycle battery so I think the discharge will come/go fairly gently (vs. say a car audio system where they want a ton of current on bass transients). It's just to augment the stability of the electrical system as there are so many variables on a bike depending on the speed and type of riding. Due to space the batteries on most motorcycles are always at the limit of being undersized and this could stiffen things enough to remove the flex in the system. One company did make an add-on cap module but it was too expensive and did not make it in the market.

 

[Sceadwian]

If it's just to augment the battery ESR shouldn't be a big deal. You might want to figure out what the leakage current is though. You can test that simply with a multimeter and the capacitors attached to a battery with no load. There's a slim chance it could be high enough to drain the bikes battery when not in operation.

Charge the capacitor bank before you put the meter inline to measure current, dead shorting a large capacitor bank across the meter before it's charged will likely blow the fuse in the meter.

 

[motolectric]

If it's just to augment the battery ESR shouldn't be a big deal. You might want to figure out what the leakage current is though. You can test that simply with a multimeter and the capacitors attached to a battery with no load. There's a slim chance it could be high enough to drain the bikes battery when not in operation.

Charge the capacitor bank before you put the meter inline to measure current, dead shorting a large capacitor bank across the meter before it's charged will likely blow the fuse in the meter.

Thanks, I do keep my bike on a float charger but I guess the above is why the big caps for auto sound all come with a small circuit board and a set of instructions on how to charge/discharge them.

Could I put a diode in series to limit any drain? I know on my voltage regulator that the current sensing lead does not have a diode but the feed line does. So when I ran the sensing line direct to the battery to get an accurate reading (vs. taking it off the loom somewhere else) that it drained current.

And another question, assuming a .7 kilowatt starter and a .7 Farad cap, what would the current capability of the cap be? To start a bike with the above starter it asks for about 52 amps, could the cap flow that much current? and for how long? (If there's an online calculator to figure this out you just point me at it).

Thanks for the advice and tips.

M./

 

[MikeMl]

...
And another question, assuming a .7 kilowatt starter and a .7 Farad cap, what would the current capability of the cap be? To start a bike with the above starter it asks for about 52 amps, could the cap flow that much current? and for how long? (If there's an online calculator to figure this out you just point me at it).

C * deltaV = I * t

or deltaV = (I * t) / C

 

[Sceadwian]

Keep in mind the ESR is going to matter at those currents, when you said you were just using it as a buffer for the electrical system I asumed you meant just to subdue noise or maybe smooth out the initial current draw of lights and what not. A starter drawing that much current is a different story. I don't know how much you really need to calculate it out, put the most capacitance you can fit into the space you have available at a decent price and see what it gets you.

 

[motolectric]

Keep in mind the ESR is going to matter at those currents, when you said you were just using it as a buffer for the electrical system I asumed you meant just to subdue noise or maybe smooth out the initial current draw of lights and what not. A starter drawing that much current is a different story. I don't know how much you really need to calculate it out, put the most capacitance you can fit into the space you have available at a decent price and see what it gets you.

Thanks for all the advice. I have done more searching and have found these 4 F caps that are quite small. If I put 4 of them together they will fit into my preferred case and I should end up with a 22 V system. Since I am dealing with DC I do not know how to interpret the ESR specification. The cap box will be in parallel with the motorcycle battery and not used by itself. So I hope it will stiffen the system and allow easier starting etc. but not expect the cap box to start the bike by itself.

Package Type: Radial - Vertical Type
Capacitance: 4F
Voltage: 5.5VDC
ESR @1khz ≤ 30
Operating Temperature Range: -25 C to +70 C
Tolerance: -20% ~ 80%
Dimensions: (DxHxP) 6 x 25 x 7.2mm.

I'm assuming that I do not need to worry about heat with these caps, am I right? My case (and point of installations) pretty much rules out ventilation as it would allow ingress of moisture etc.

I guess there have been big advances in capacitors as these are very small with a big rating and the ones for car audio are the size of a spray paint can. About a 20-1 difference. Perhaps the car audio design is made for show instead of just performance for the size.

Any further advice/tips is much appreciated. I have gotten a lot of valuable information from the members of this forum and appreciate the help.

Thanks,

M./

 

[Diver300]

According to the specifications, which may be a little optimistic, the Rockford Fosgate 1 Farad Capacitor has an ESR (Equivalent Series Resistance) <1.5 mΩ.

That is really what you are adding a capacitor for. I guess that you are using it to improve the supply for an amplifier. You have a car battery at the other end of just a few metres of cable, and the capacitor is only there to provide current surges. The ESR should be low compared to the impedance of the cable for the capacitor to make any difference.

The ESR of the little capacitors is < 30 Ω. If you put 4 in series (with balancing resistors) so the capacitance is 1 F but the ESR is up to 120 Ω. That is 80,000 times as big as that of the Rockford Fosgate Capacitor, so it will be 80,000 times less effective, otherwise known as useless. I realise that the ESR may be measured at different frequencies for each capacitor but they won't vary a lot.

The little capacitors are really designed to be used like batteries for keeping memory and clocks working on devices that have mains power most of the time.

 

[spuffock]

sorry,those capacitors wont do what you want. You will get enough current to light a torch bulb if youre lucky. Its this ESR thing.

 

[motolectric]

sorry,those capacitors wont do what you want. You will get enough current to light a torch bulb if youre lucky. Its this ESR thing.

Yes, thanks for both the replies. I was able to find a site that lays out the basics and with the above comments I can see that I have been heading in the wrong direction.

They use the term "supercaps" for these small multi Farad units and so as a near layperson (with board level electrical components) I was mislead. But I was always skeptical which is why I first posted my question.

So now I am trying to find some suitable caps that will fit into my preferred little chassis box of 2X2X1.5 or so. I don't need a full 1F but that was a goal if it was possible. All I seem to see online are large auto sound caps 1F and above and the small little discs.

If anyone knows of a source for caps in the AA battery size that I can put in series to meet my goal that would be great. It may be that in order to really store some serious current you have to have the large spray paint can size and that is why there aren't any in the middle sizes, I don't know.

I'm just trying to stabilize the system on motorcycles to get better overall performance under varying conditions.

Thanks again.

M./