Assistance with Super Capacitors

Discussion in 'The Projects Forum' started by Sinbaad, Sep 9, 2011.

  1. Sinbaad

    Thread Starter New Member

    Sep 4, 2011
    2
    0
    Hi,

    I am currently working with fuel cells and I am using super capacitors for making a buffer system. The reason I need a buffer system is because fuel cells purges every 10 seconds and momentarily there is no power output from the fuel cells.

    Currently my setup is very simple (please see the attached schematic) and I have not faced any problems with this circuit yet, however I am wondering if there would be any need to add resistors at the output or input to limit current? I want a simple reliable system which gaurantees to work everytime.

    Specs of fuel cells and the capacitors are as follows:
    Fuel cell output = 8.5V @ 8amp
    6 x Super Capacitors (rated at 2.3V and 50F)
    1 x Blocking Diode (rated at 45V & 16amp)

    P.S. I am mechanical engineer and harldy know anything about electronics, so kindly explain me in simple words.
     
  2. tgotwalt1158

    Member

    Feb 28, 2011
    111
    18
    I think you have to connect in this configuration (check modified circuit), since your source (fuel cells) output is 8.5V. Capacitor adds their capacity when connected in parallel not in series! But this could be done only if you are using FA or FS Series Super Capacitors which have voltage tolerance range up to 11.0 V, otherwise maintain your existing configuration.
     
    Last edited: Sep 9, 2011
    Sinbaad likes this.
  3. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    The supercapacitors are rated for 2.3v; they would be damaged if 8.5v were to be applied across them.

    A minimum of 4 in series are needed in order to meet the output voltage requirement of the cell. 4 x 2.3v= 9.2v.

    Placing capacitors in series means that the capacitance will be reduced.
    Two equal caps in series = 1/2 of the rated capacitance; 25F
    Capacitance in series is calculated like resistance in parallel.
    Ctot = 1/((1/C1)+(1/C2)+...(1/Cn))
    Having six in series results in 8.333...F.

    Using the diode is simple, but you'll have a voltage drop across it which is a loss in efficiency.
     
    Sinbaad likes this.
  4. hgmjr

    Moderator

    Jan 28, 2005
    9,030
    214
    Unfortunately, the supercaps can only handle 2.3V maximum so you have to put them in series or they will be damaged by the 8.5V from the fuel cell.

    hgmjr
     
    Sinbaad likes this.
  5. wayneh

    Expert

    Sep 9, 2010
    12,093
    3,033
    What is the nature of the load?

    I doubt you need current-limiting resistors unless the load is capable of drawing an enormous current - capable of acting like a short - that might put your fuel cell or wiring at risk. In other words, find your least current-capable component and make sure that weak link is protected. If things are working reliably now, a fuse might be a better approach than a resistor.
     
    Sinbaad likes this.
  6. hgmjr

    Moderator

    Jan 28, 2005
    9,030
    214
    You neglected to indicate the polarity of the supercaps as well as that of the fuel cell. It is critical that the correct polarities be selected.

    A significant challenge when using supercaps in series as you have done is taking measures to keep the caps balanced. The term "balance" in this context refers to the act of maintaining the voltage across each supercap so that each of the caps sees the same voltage during the charge cycle. It is important to avoid having any one cap be subjected to more than its 2.3V maximum rating. Even exceeding the 2.3V by a small amount can degrade the cap in an irreversible manner.

    Another thing to watch out for is mechanical stress at the interface between the body of the supercap and the lead. If lead bending is required then it is a good idea to use a pair of needle-nose pliers to support the lead near the body of the cap to minimize stress at the interface. If the interface is damaged there is a high probability that the supercap will develop a leak. The chemical inside is not toxic but it is caustic. If it gets onto surrounding components is can cause damage. My experience has been that over time the chemical will change the value of some of the components with potential bad consequences.

    The information on supercaps at www.batteryuniversity.com is a good source of dos and don'ts.

    hgmjr
     
    Last edited: Sep 9, 2011
    Sinbaad likes this.
  7. ErnieM

    AAC Fanatic!

    Apr 24, 2011
    7,386
    1,605
    The general scheme looks fine, assuming you have polarities and mechanical connections correct.

    What would worry me is there is no limiting between the fuel cell and the supercap. The cap looks like a mighty dead short when the cap is discharged, and that may damage eventually your cell.

    I would suggest adding some limiting between the cell and the caps:

    [​IMG]
     
    Sinbaad likes this.
  8. Sinbaad

    Thread Starter New Member

    Sep 4, 2011
    2
    0
    Thanks for replying everyone..

    Yes I cannot connect them in parallel as they are rated 2.3Volts each, hence I have connected 6 in series (13.8Volts). Also, the open circuit voltage of the fuel cells reaches approximately 12volts. I know the capacitance will decrease in series but for my needs 8.33 Farad is enough.

    E = 1/2C*V^2 = 1/2 (8.33)*(8.5^2) = 300J and my load draws approximately 50 to 60W (J/s) hence theoretically this setup should sustain for about 5 seconds.

    My load is basically a small remote control truck, at its maximum power requirements it consumes about 50W to 60W.

    I think I should add some resistors to limit the current from the fuel cells to the supercaps, however the only problem I see in that is that the charging time might increases. Right now as supercaps acts like a huge load, but I can see the fuel cells can handle it when they are fully discharged.

    To add a resistor should I just use V=IR and calculate what resistance I require?
     
  9. Hi-Z

    Member

    Jul 31, 2011
    157
    17
    Actually, it's a bit of a tricky situation here vis-a-vis the avoidance of excessive currents:

    At power-up, there's potential for very large charging current, even with a series resistor.

    If there's a series resistor, we have the situation where the supercaps are being discharged at a constant current, then topped up via a resistor. There's not too much fuel-cell excess current capability to charge the supercaps in the 10 second on-period, so I would imagine there would be a progressive "droop" of the supercap voltage in the course of several 10-second cycles (I may be wrong here though).

    Even with a constant-current arrangement for charging the supercaps, 10 seconds may not be enough - it all depends on the time taken to purge. You can use the following formula to determine the drop in supercap voltage following a purge:

    V - t*I/C

    where V is the voltage drop, t is the purge time in seconds, I is the load current in Amps, and C is the capacitance in Farads.

    If the purge time is about a second, then you'll get a drop of about 1V. To make this up in 10 seconds you'd need a charging current of 0.8A - just about OK in terms of fuel-cell current capabilities (assuming a max load current of 7.5A or so).

    Hopefully, the purge won't take too long, in which case you'll need a circuit for a constant-current charging system. This could be pretty simple, I think: ErnieM's current-limiting resistor would be replaced by a pnp power transistor, and you could probably get by with just a couple of resistors in the base circuit to crudely define the charging current. I'm in a rush, as usual, and I've never posted an image to a forum before anyway, so if you need more details, this would have to follow at a later date.

    A constant-current means of controlling the charging would of course have big advantages at power-up too.

    Just one more thing: capacitor matching has been mentioned. If your supercaps are +/-10% tolerance, then you should be OK with 6 of them. However, I would imagine another factor here, in terms of protecting against unequal voltage sharing, might be leakage current mismatch. You can guard against this by putting a fairly high value resistor (say 10k ohms) across each supercap - easy and cheap to do.
     
  10. Hi-Z

    Member

    Jul 31, 2011
    157
    17
    A couple of points:

    I've just noticed that you're anticipating that the supercaps will hold over for 5 seconds. Is that the time taken for purging? If so, I think you'll need more capacitance, as you need to consider the voltage presented to the motor (at present, it looks like it will sag dramatically over 5 seconds).

    As regards limiting the current, I suspect the fuel cell may already do a decent job of this - after all, it's not like a battery; I would imagine that you're putting in hydrogen and oxygen at a given rate, so power output will tend to be constant. That being the case, I suspect your original circuit is OK, though it might be a good idea to have a look at how fast the supercaps charge up from power-up (from which we can infer peak current).
     
Loading...