Experiments with supercaps - part 2

/part 1/

Self Discharge

One of the biggest problems of the supercapacitors is their self discharge rate. I charged one of them and measured it's changes for a couple of days. The capacitors is D-size SAMWHA Green-Cap 2.7V 500F.
Here are the results:

2.4228.12.2013 18:00146480.3
2.3428.12.2013 19:00136975.1
2.2228.12.2013 22:00123267.6
2.0429.12.2013 09:00104057.1
1.9929.12.2013 15:0099054.3
1.9629.12.2013 18:0096052.7
1.8430.12.2013 18:0084646.4
1.6602.01.2014 17:0068937.8
1.5605.01.2014 10:0060833.4
1.4708.01.2014 22:0054029.6
1.3912.01.2014 13:4248326.5

You can see that @2.7V the capacitor is charged @100%. I managed to charge it only to 80%. The last 20% are the slowest :). In the beginning the voltage drop is very significant! After ~1.8V the graphics are more linear. We lost half of the charge after only 2 days! 
So for 2 weeks we drop from 80% - 26%.
I may conclude that, the supercapacitors need something to keep the charge. Otherwise it is lost.
I don't have any Li-Ion supercapacitors for test. They should have low self discharge.

The capacitor:

Now! I expected that, but take a closer look to these two capacitors:

Now to tell you the facts:
- 1 & 2 look absolutely the same, even the blue color, markings, numbers, everything is the same except the pins
- 1 & 2 price was the same

 Capacitor 1 facts:
- I ordered it few days ago from another online shop
- It looks absolutely new
- It has 4 pins (no documentation, but as far as I know the 2 extra pins are just to keep it stable on the electronics board)
- It charged for almost the same time as number 2, which means the capacity should be the same
- HOWEVER, the discharge time is drasticaly low! It drops to 1 Volt for less than 24h. I have 2 of them and I tested both of them. I recently noticed people complaining about that on some online shops forums.

Capacitor 2 facts:
- The first that I ordered
- It looks worn-out (that was mentioned on the shop's web site)
- It has 2 pins
- and now I remember it was written "low leakage" in the offer, that is the capacitor that I measured with the graphic above. I thought discharging time is bad, but it seems it could be worse :)

My advice is to look for "low leakage" phrase. You cannot be 100% sure. I think the number of pins does not matter and very often you recieve something different from the picture on the online shop (in my case it was with 2 pins, but I received with 4 . upgrade!? :) )

I will definetely order more of the first type, however the web site moves from domain to domain. I hope they don't have legal problems. That is why I do not mention links.


  1. I could not find on shamwa web site these capacitors. I bought 24 of them, they do not seem real. They are 300F at best, the leakage is anywhere between 10 to 150 mA at 2.7V charge they seem like factory rejects re-branded. 10mA leak disqualifies them as a valid electronics component, the leak should not be more than 1.5mA !!!!! There is no capacitor in the world with that leakage specified in the data sheet. The last 20% as you say it will not EVER charge at 2.7V due to probably in your case 50-60mA leakage.
    If anyone is able to find these caps on the official website please let me know. Mines are black marked EDLC(DB) and are unusable. There is no application that I can think of where huge esr caps with monstrous leakage can be used. Due to ESR their "response time" is very long (they charge in 2-10 hours) so wind generators, energy recovery, voltage stabilizers and sound systems are out ... huge leakage so storage systems are out ...
    But again if you can find on the official website something let me know. I have just lost 100USD on these pieces of s**t

  2. Thank you for your comment. It was waiting for moderation, that is why it did not appear instantly. How did you measure the leakage current? I am not very good at electronics, but want to experiment. As I said, I ordered twice and only the ones with "low leakage" were OK. Also I noticed that charging them several times and keeping them charged for some time, somehow "wakes" them up, and they stay charged longer, just staying on the shelf. At this very moment I measure 6 capacitors which I haven;t touched for at least 6-8 months. They read - 0.4V, 0.76V, 0.93V, 0.64V, 0.83V, 0.00V - the last one is one with 4 pins that I described above. I don't have black ones, I don't know is there any difference between blue and black. But my caps are good enough to play with my small projects. I am sure these are some sort of rubbish, and you have to be lucky to have several good ones, but I cannot afford to buy new ones - you can find on farnell, digikey websites. Probably Maxell are good!

  3. Also I think this is the link where I bought the last ones - http://www.for-sale-china.com/Fala-super-capacitor-2.7V-500F-original-new-low-leakage-p-2572?filter_name=22352496639
    There were a site www.noweshop.com and www.shop-from-china.com, but I see the first one exists no more, and the second is not working.

  4. http://www.offerany.com/p-22352496639-South-Korea-three-and-SAMWHA-super-farad-capacitor-2.7V-500F-original-new-low-leakage.html


  5. hi again,
    samwha has it's own website:
    These caps are nowhere to be found on their web page. However, the similar design is marked 360F....
    To measure the leakage is easy if you have a smart power source. Select 2.7V and after about 3-10 hours (no 2 of these caps are the same) the circuit (the cap's) voltage stabilizes at let's say 2.548V while the amperage drawn from source it's 52mAmps. That's the leakage. Now these caps are only for play. They are useless for energy recovery, wind projects, sound systems because of the high esr (at least 4 ohms) and that is why the long charging time. Because at 4 or 5 Ohms the current left when the difference is 0.2V goes to 50mA....
    About the price, 5 USD for a cap like this is HUGE considering that at 14-15 USD you can get a real capacitor with a 1.5mA leakage !!! and negligible internal resistance (it charges/discharges instantly and keeps the charge for months)
    a real cap:

    ... read the technical data and be amazed ESR DC:10mΩ(max) ! about 400 times lower than these fakes, Short Circuit Current:270.0(A) ... and so on ...

    Hint: caps on screws are good, caps on snap-on are bad (high esr, otherwise the contacts would melt at 15Amps)
    Maxell are overpriced

    the comment as google account does not work, the page is defective, so only as anonymous...

  6. Thanks for the info. It killed my hopes but now i know to buy from Digikey or some other trusted electronics store.
    It seems the glass solar panels from China work for a long time. However, I'm afraid to buy new ones from eBay, now. Some smaller epoxy or pet laminated ones don't work and others are still doing great after a few years of occasional tinkering. One blisters "instantly" but still works. And i believe the pet ones are better (if they don't short out in the meantime).

    I hope you found better supercaps!