Supercapacitor cycle life, myth or reality?

[ZeroResistance]

I read this statement quite often on the internet

Supercapacitors can be charged and discharged millions of times and have a virtually unlimited cycle life, while batteries only have a cycle life of 500 times and higher.

But when I skim through the datasheet of a supercap, if find this:

Life Time: 1000 hours with rated voltage applied at 65°C
Life Cycles : 500,000 cycles
( 25°C) 1 cycle= Charge to WVDC
for 20s, constant voltage charging
for 10s, discharge to ½ WVDC for
20s, rest for 10s

So it seems there are a number of constraints for the "Life Cycle" parameter.
Are there any true, unlimited (read > 10 million cycle) supercapacitors?

Thanks in advance!

[FotatoPotato]

500k cycles sure is a lot!  I think what the statement "Supercapacitors can be charged and discharged millions of times and have a virtually unlimited cycle life, while batteries only have a cycle life of 500 times and higher." means is that after all those charge cycles the super cap will still contain a good portion of its original capacity whereas a battery would be dead.

But I'm not capacitor expert, just a hobbyist so I'm not 100% sure.

[ZeroResistance]

500k cycles sure is a lot!  I think what the statement "Supercapacitors can be charged and discharged millions of times and have a virtually unlimited cycle life, while batteries only have a cycle life of 500 times and higher." means is that after all those charge cycles the super cap will still contain a good portion of its original capacity whereas a battery would be dead.

But I'm not capacitor expert, just a hobbyist so I'm not 100% sure.

Agreed it seems a lot! However there a quite a bit of fine print linked to the "Life Cycle" parameter.
Eg. 1 cycle is considered when you dicharge the capacitor to 1/2 the WVDC value...
And I'm looking at it from a point of view of discharge application. Like a photo  flash.
How would a camera manufacturer ensure that the capacitor is discharged only to less than 1/2 the voltage and that too at a limited discharge current rate. But then I guess camera's won't be using super cap's due to these limitations, they might be using specially designed caps for the high discharge rate that they are subjected to.

[SeanB]

Supercap would prtobably be discharged to half the energy, so to 0.7 of the rated voltage, and also charged to 0.9 of the rated voltage, giving a big increase in lifetime. Thus they could do a million cycles, and under the same situation a lithium cell can also do 2000 cycles.

[David Hess]

Increasing the charge voltage by 0.4 volts decreases the life to 1/10th.  Increasing the temperature by 10C decreases the life to 1/2.

So life expectancy at 2.6 volts and 65C may be only 1 month while life expectancy at 2.0 volts and 35C could be 35 years ignoring other failure modes.

[CatalinaWOW]

Remember that all components, not just supercaps, have life cycle limits.  A quick google shows one paper on aluminum electrolytics with similar warnings about temperature and voltage and a warning that repeated charge/ discharge cycles will reduce lifetime.   Another paper notes that "standard" aluminum electrolytic capacitors can lose 30% of their capacitance in only 10,000 charge/discharge cycles while claiming that capacitors made for this use would lose 5% capacity after 60,000 cycles.

The 500 cycle number for rechargeable batteries is pretty typical.  Poor care or very heavy duty can reduce that by a factor of 3 or more, while extreme care can increase the life by a factor of four or more.  But compared to the supercap rating of 500,000 cycles it is nothing.  You would replace rechargeable batteries 1000 times before you had to replace your supercaps.

Don't panic over the ratings, just make sure you understand them, understand their relationship to your operating requirements and design accordingly.

[ZeroResistance]

Supercap would prtobably be discharged to half the energy, so to 0.7 of the rated voltage, and also charged to 0.9 of the rated voltage, giving a big increase in lifetime. Thus they could do a million cycles, and under the same situation a lithium cell can also do 2000 cycles.

That probably makes sense, but would'nt that limit the application of a supercap just as a backup source of supply. Rather than as a pulsed source of energy where it was meant to rule.
I mean high Specific Power is suppossed to be the big +++ of Supercaps vs the high Specific Energy of batteries. But I don't see with all these limitatons how can Supercaps acheive high Specifec Power.

And I haven't yet meantiond the high ESR that some of the Supercaps have, as high as 500mOhm for a 1F supercap. That limits the peak current to 0.9A as per the datasheet.

The 500 cycle number for rechargeable batteries is pretty typical.  Poor care or very heavy duty can reduce that by a factor of 3 or more, while extreme care can increase the life by a factor of four or more.  But compared to the supercap rating of 500,000 cycles it is nothing.  You would replace rechargeable batteries 1000 times before you had to replace your supercaps.

Ok, I probably understand what you mean to say. But when I think "CAP", I used to beleive 100% charge and 100% discharge as 1 cycle, so the datasheet shook me up a bit. So I again get back to the issue of high "Specific power" of supercaps. How do I use them where they were best meant to be used!

[Berni]

There are different designs for supercaps too, both in construction and the intended application.

Some supercaps are designed to be time keeping backup battery replacements. They tend to be pretty compact and are designed to have a minimal self discharge rate, as a consequence they tend to have pretty high ESR of even a few Ohms sometimes, while it doesn't matter since the typical discharge current would be in microamps. On the upside they can hold on to there charge for years.

Other supercaps are designed for power delivery instead. Those are designed differently and tend to have thick beefy terminals as part of there low ESR design. A typical application for these is to help a regular battery deliver much higher peak powers because even a fairly small supercap can deliver 100s of amps repetitively or as a tiny UPS to hold up voltages trough transients. In some cases i seen them used  to replace a battery but the use for that is limited to some special applications because compared to a battery the amount of power they store is abysmally small.

When it comes to pulse operation you don't usually see a lot of supercaps there. They are useful for things like spot welders where you need lots of amps at low volts, but any sort of really high energy discharge things (Pulse magnetization, oil finding at sea, railguns, coilguns etc) tend to need more volts while supercaps can only do a few volts per cap. For those applications you still see regular aluminum electrolytics or oil filled film caps for even higher voltages.

[ZeroResistance]

There are different designs for supercaps too, both in construction and the intended application.

Some supercaps are designed to be time keeping backup battery replacements. They tend to be pretty compact and are designed to have a minimal self discharge rate, as a consequence they tend to have pretty high ESR of even a few Ohms sometimes, while it doesn't matter since the typical discharge current would be in microamps. On the upside they can hold on to there charge for years.

Other supercaps are designed for power delivery instead. Those are designed differently and tend to have thick beefy terminals as part of there low ESR design. A typical application for these is to help a regular battery deliver much higher peak powers because even a fairly small supercap can deliver 100s of amps repetitively or as a tiny UPS to hold up voltages trough transients. In some cases i seen them used  to replace a battery but the use for that is limited to some special applications because compared to a battery the amount of power they store is abysmally small.

When it comes to pulse operation you don't usually see a lot of supercaps there. They are useful for things like spot welders where you need lots of amps at low volts, but any sort of really high energy discharge things (Pulse magnetization, oil finding at sea, railguns, coilguns etc) tend to need more volts while supercaps can only do a few volts per cap. For those applications you still see regular aluminum electrolytics or oil filled film caps for even higher voltages.

I like the idea of using the supercap in combo with the battery.

Regarding the pulse operation part, probably you are right that there may be a certain breed of supercaps meant for pulse power applications.
Although when I do open a Supercap datasheet the first thing I see mention is "excellent pulse power handling characteristics an very low ESR"
And when I see the product table the ESR starts for 0.5ohm for 1F ones and drops to 5mOhm for the 400F ones. And typical applications as "Camera Flash Systems".
To add to that the peak current of a 1F cap is 0.9A and around 15A for a 25F cap so a manufacturer would likely have to use a very large F cap for a flash application, possibly these are meant for LED flash only, that's my guess though.

So that begs me to ask whether camera manufactures who use Supercaps for flash, only discharge them to 50% of the energy(as SeanB pointed out earlier)  so that would extend their life.

[Zero999]

As mentioned above the voltage and temperature are the main factors which limit a capacitor's lifetime.

I doubt there's a hard limit on the number of charge/discharge cycles for capacitors. What will be more important, will be the peak charge/discharge current, rather than the number of cycles. If the capacitor is charged and discharged at a moderate current, which doesn't cause much heating, then the ambient temperature and maximum voltage will limit the life, more than the number of cycles.

If the number of charge/discharge cycles of capacitors was that limited to something like 500k cycles, then one wouldn't expect much life from a capacitor used on a rectifier, which is being charged/discharged a bit at 100Hz.

[ZeroResistance]

Supercap would prtobably be discharged to half the energy, so to 0.7 of the rated voltage, and also charged to 0.9 of the rated voltage, giving a big increase in lifetime. Thus they could do a million cycles, and under the same situation a lithium cell can also do 2000 cycles.

Now that's an amazingly profound insight. SeanB.
I guess that clears things up a bit 

[Rerouter]

Having had to deal with ab automotive product built that uses a super capacitor as its RTC backup instead of literally any normal coin cell,

I have to say when the tempaeratures or voltages get high, they die stupidly fast,

the device in question was running it at 5.2V, the enclosure was a sealed ABS box mounted inside a bus, so 55+ C was not uncommon,

Referring to the datasheet, life expectancy under those conditions, 500 Hours,....

It puffed up, and leaked after about 700 Hours. then hard shorted soon after.

[spec]

I would only use supercaps for low current back up for memories RTCs and the like. The reason is that the servicing guys hate them because they are always failing and that has discouraged their use. The other thing is, unlike a silver oxide and lithium batteries, the terminal voltage is exponential on discharge.

Just my 2p worth