Replacing The Lead In A Motorcycle Battery With Supercaps

[Raphael] has a motorcycle he’s constantly working on, and for him that means replacing the battery occasionally. Tired of the lead-acid batteries that have been used for 100 years now, he took a look at some of the alternatives, namely lithium and the much cooler supercapacitor option. A trip to the local electronics distributor, and [Raphael] had a new supercapacitor battery for his bike, and hopefully he’ll never need to buy another chunk of lead again.

The battery pack is built from six 2.7V, 350F caps, a few connectors, and a handful of diodes. These are lashed together with rubber bands to form a 16V, 58F capacitor that makes for a great stand-in for a chunk of lead or a potentially puffy lithium battery.

[Raphael] put up a walkthrough video of his battery pack where he shows off the enclosure – an old, empty lead acid cell. He also goes through the back current protection and his method of balancing the supercaps with a few diodes.

99 thoughts on “Replacing The Lead In A Motorcycle Battery With Supercaps

    1. Its not, I believe, the fact that a very polluting construction and usually limited on power reserves and supplying accessories. Not to mention the dead weight of a ” lead” ACID corrosive power source that even when sealed , during a wreck, can disfigure and even kill…..think of the fact that a lithium ion battery powered electric or hybrid car produces more pollution in construction and shipping of all parts than a normal gas car over a lifetime, then compare to old lead batteries. If I could strap 16 volts of safe and light caps under the bike plastic or Volvo bonnet and cruise on…..I’m in.

  1. How long do motorcycle batteries usually last? A battery in my car usually lasts about seven years, which should mean maybe one battery replacement somewhere mid-to late in the average life of a car. I know motorcycles are different.

    1. Seven years?!

      I guess the triple digit temperatures in Arizona must be why I get 2 1/2 years tops. Sadly, the electrolyte in those super caps would dry out pretty quickly.

      At least I can top off the electrolyte in a lead acid cell…

      1. Seven years and still going. Every now and then I take it off and put it on the bench power supply (set to 14.4V), and I do take it inside when I know I won’t be using my bike for a while. I don’t have a garage, though, so the bike sits outside come hell or high water with the battery in it. This has included some fairly harsh winters, during which I’ve stubbornly kept using it.

        You can see that the battery is old because the voltage sags more than it used to when the starter’s running (lights get a lot dimmer), but it still starts the engine every single time.

        Bike is a Kawasaki ER-6F/650R, battery is the stock sealed one.

    2. My cheap-ass battery has lasted 4+ years of casual use now. Some friends say theirs lasts only a year or two. My fancy “long life” replacement battery has been sitting in its original box for over 2 years. YMMV.

      Motorcycle batteries are very cheap though.

  2. Supercapacitors are definitely better candidates for replacing the batteries on motorcycles than on cars, since you can more easily push start a motorcycle. If I did the math correctly, this one has about 15 CCA, which isn’t bad at all.

    1. Try push starting a “real” Harley-Davidson!

      There is still no substitute for a lead acid battery in a bike and I replace mine every 3 – 5 years. This guy has other problems contributing to his battery demise and his super caps are just a patch.

          1. Your response tells me you don’t get that it’s a joke, possibly due to the Harley-branded stick up your butt.

            To put it more nicely, don’t get so butt-hurt over a silly joke. I’m sure it wasn’t a personal attack against you, and if you took it as personal as it seems, then maybe there’s some truth to the joke that adds a bit of a sting?

          2. Sure he does.
            Modern Harley’s as well as most modern bikes today can not be push started with a totally flat batter since they all use electronic fuel injection.

          3. Oh, I understand the joke. But a lot of folks who don’t know any better really believe “the joke”. And BTW, I’m not a Harley guy, I’m a motorcycle guy. My oldest daily rider was a 1949 Indian and I currently ride a 1976 BMW R75/6. The Indian leaked or more correctly dripped oil when you parked it. And just like the Harley’s of the era, if it didn’t you better find out why because your drive chain was not getting proper lubrication!!! In other words the oil “leaks” were intentional.

            And yes, I own a Harley-Davidson too.

        1. I hear the main reason the Brits never got into computer design was because they couldn’t figure out how to make them leak oil….
          Heard that joke from a guy that rebuilt Brit cars.
          Not a Harley rider myself. I actually own a 73 triumph bonneville. My honda crotch rocket’s battery never lasts me more then a year or two, so I’ve been looking at doing this myself.

          1. I’ve owned my share of Japanese motorcycles. At least through the 70’s and 80’s their charging systems were marginal at best. The generator or alternator would over charge at road speed, plus in my opinion the batteries were always too small. The end result is you usually cooked out the electrolyte. The regulators on these were shunt style, basically a heat sink mounted zener diode that was supposed to shunt excess current past the battery. The solution has always been to buy a better after market regulator that would do a better job of clamping the charge voltage to a level that does not kill the battery.

          2. “I hear the main reason the Brits never got into computer design was because they couldn’t figure out how to make them leak oil…”

            While entertaining, this is at odds with the fact that the computer in everyone’s pocket is of British design.

    1. So far the only things I have seen measured in Metric in the US is engines and bullets (I am not counting the small print metric found under the “standard” measurements on packaging). I think the metric system is easier to use, and wish that we would join the rest of the world and switch.

        1. …and tire size, but not wheel size for some reason. To wit, a 225/45R17 is 225mm wide, has a sidewall that’s 45% of the 225mm tread width (per sidewall), and mounts to a wheel 17 *inches* in diameter.

      1. Bullets? For realsies? Why then, in South Africa do we measure bullet mass, case mass and powder mass in grain? Also most calibres are measured in fractions of inches, but that is just a naming thing I guess. The grain measurement is closer to Imperial than metric to me since 7000gr = 1lb.

  3. What’s the point of this ?

    It’s not like supercaps are eternal or indestructible.

    They especially don’t like being charged for extended time periods.

    So, what makes this better than say Li-Ion or LiPO solution ?

        1. It looks like the combination of charge and high temperature is bad for endurance. At 25 deg C, they’ll last 10 years, but at 65 deg C, lifetime is only 1000 hours. Under the hood in a car is a really bad place to use these things.

          1. Then I guess it’s a good thing they are on a bike in moderate temperatures, with air cooling, instead of sweltering under a hood on a car, with no airflow.

    1. The discharge after the initial charge is terrible, and gets worse with farad size of the individual caps. I have a bank of 6 3000 farad Maxwell ultracaps that I did some pretty extensive testing with. Initial charge leaked down to 12.63 volts in 19 hours (from ~14v initial charge), but only leaked down .8 volts when recharged from 12.63v to 14.4v in about 23 hours.

      The last time I put a charge on it, it hit about 15 volts; that was on August 30th. I just went out and checked the voltage on it, and it’s resting at 12.12v; I should note, however, that I forgot to turn the volt meter that I have on the enclosure off for several weeks, and I’ve pulled some power out of it using one of the charging ballast halogen lights to show it to a couple of people.

  4. In addition to expressed concerns about the impact of heat & the design-purpose of these “supercaps” I’m concerned about safety.
    The sudden-discharge of a big “supercap”, e.g. by accidental shorting, can be extremely dangerous.
    I like the outside-the-box creativity.
    How about an improved old-school jump-start mechanism & reserve a small deep-cycle battery just for the electronics?

      1. I once dropped a piece of binding wire (~3mm dia) across a lead acid battery’s terminals. I liquefied in an instant onto my arm. Now I am scared shitless of accidentally dropping a spanner across the terminals when I am working on a car. Lead acid batteries are really scary sometimes.

    1. To summarize your post. “why push the limits when you can use existing tech”

      Is that not what HaD is for. Thermal insulation exists as do cooler places in a car.
      I ma probably wrong but dose a car not drain power then recharge is a “wave” like manor. I don’t think after start you could just use relays to disconnect a battery from the circuit with out stalling the car. meaning the caps are not always being charge as the car runs.
      Basic common scene should be used around all dangerous devices including capacitors that can discharge hight amps and car batteries that can be very painful.

        1. Technically, that would be filtering that you are referring to, something that can be achieved in electronics just as easily. Also, any issue the capacitors have with charge cycles could easily be dealt with by making a custom charge controller.

          1. I don’t know that it can be dealt with “easily” with electronics. We’re talking about potentially very large currents, which means a lot of energy that has to be stored somewhere.

          2. Then Trui, how is it that you came here? I’m curious because that is the bases of how a car works. I can unplug a battery from the car while its running and it will if (no issues are present) continue to run. This guy obviously has issues beyond power source if he can’t keep a simple battery charged and is trying to build a high power cap to start it up. Infact I would say it is dangerous because he probably has no idea what he is doing if he hasn’t realized the genuine problem and source. Oh by the way it can be “easily” dealt with using electronics, as mentioned “FILTERING” -DrainBramage. Go brush up on it.

          3. Have you actually tried unplugging the battery in a running car, and see if everything still works as before ? And don’t forget to turn on the lights, fans, electric heaters, radio, etc. And sure, you can make a filter, but the filter would have to involve a large capacitor to store the energy, so you’re not actually getting rid of the capacitor at all. Well, you could replace the capacitor with a big inductor, but that’s the same idea.

          4. A normal car alternator has no filtering. It generates 3 phase AC which is rectified and fed to the battery. The battery acts as a low pass filter. If you disconnect the battery while the engine is running (old trick to test the alternator) you will feed dc with a huge amount of ripple to your electronics which can easily fry them.

          5. @John: “Old trick” indeed. I’d do something like that on my 32 year old truck with barely any sensitive electronics on it, as it’s about as simple as a vehicle electrical system can be. But on a modern car (even my “old” 2000 Corolla) I wouldn’t dare. It’s a good way to fry any DC circuit, including the car’s ECM.

          6. @ John: Most bridge rectifiers in alternators also do have a cap stuck to them for filtering. Its just very small since its rectified 3 phase AC. Fun side note they are a great source of high powered rectifiers for cheap. The ones I’ve tested from ACDelco alts can handle 120v at like 80+ amps.

  5. The reason that people replace batteries with supercaps on bikes is weight savings. The standard bike battery weighs between 10 and 15 pounds, and it’s all weight that rides high up on the bike.

    For instance, most dual-sport motorcycles are heavier than their dirt-only variants. Every pound on a dirtbike counts, especially when you have to drag the thing out of a swamp, again.

    It’s also nice if you want a “clean” look on a custom bike, no big battery hanging down.

    1. Yeah, but it’s not at all clear that supercaps DO offer a weight savings vs. high-current Li-ion of comparable sizes.

      Consider this build, a 6s pack of 2.7V 350F capacitors, vs. a 4s pack of A123 LiFePO cells, specifically 26650M1B:

      W/R/T physical size, the 350F capacitors are 33mm diameter x 62mm long; the A123 cells are 26mm dia x 65mm long, but there’s 4 instead of 6, so significantly less volume.

      Each cap weighs 60g (x6 = 360g); over the 1.1-volt interval from 2.7V (x6 = 16.2V) to 1.6V (x6 = 9.6V) they deliver 107mAh; their maximum continuous current is 34A, with absolute maximum 170A (not sure for how long, but at that current it will drop below 10V in about 2 seconds).

      Each LiFePO4 cell weighs 76g (x4 = 304g); from 3.6V (x4 = 14.4V) to 2.4V (x4 = 9.6v) they deliver over 2000mAh; their maximum continuous current is 50A, with maximum pulse discharge 120A (10 seconds).

      Li-ion is lighter, smaller, has way more energy, and more usable power for any time over 3 seconds. To me, 120A sounds high for such a small engine, but I really don’t know much about motorcycles — anyway, if cranking does take more than 120A, the caps’ “better” rating of 170A pulse current may seem good, but you better hope it turns over first time, every time.

  6. For those of us who grew up in cold country, what happens if you supplement the lead-acid system with something like this, perhaps on a thermal switch? A -20F or colder night will often leave you barely cranking without a battery heater. While I understand that the self-discharge would limit its utility over a long period, for a daily driver having a reliable “super start” in the morning would be a wonderful thing.

  7. putting the diode across the battery is good because the alternator has diodes to convert the ac to dc.

    the internal diodes could fail and short such a way to dump ac to the battery.

    also ac output can blow the voltage regulator and it can cause the battery to over charge or may not charge at all.

    the voltage regulator should have current limiting built in but the fuse is good so if you connect the battery in reverse you would blow the fuse.

    if you have a local scrap yard you can recycle the lead from the battery.

    1. If that blow out or lack of charge was the issue then it would be all over the news.. and if it wasn’t then it would be a short in your wires and you should go back and look into it. Have had several bikes over the years never once ran into a problem needing special patching or caps to power them.

  8. A very similar setup starts my car engine, its been in the car for around a year. The caps deliver the cranking current and a small dirt cheap lead battery provides the Ampshour, otherwise I have to push or jumpstart the car if I forget the lights on for more than a minute.

    1. Ummm, yeah so that puts you into the categories mentions above. If you can’t have a lector ice in for more than a minute without you battery being discharged then there is a problem with h YOUR battery. Not a problem with lead acid batteries in general! It means you need a new battery. Not to buy loads of super caps and hobble something together.

        1. Exactly. The caps alone start the engine but they are not enough for normal use, however with the small battery its quite good. I can remove the battery to save some weight if I want.
          I also tried a very dead battery that alone couldnt start the engine, and with the caps it started right away and I still have several Ah. I think that with a cap bank you can increase your battery life for several years, so I was trying to gather some evidence about that

          1. OK…
            lets think about some numbers.

            if you have a car with a decent sized engine, (lets say a 2 litre or more, (that’s 122 Cubic inch for americans).

            you’re going to want a battery that will provide about 150CCA in order to start the car. and provide these for say 1 second (which is actually pretty short time for starting a car… most “larger engine” cars will start with a position of having a 400CCA (or more) capable battery to allow for degradation.

            150A at 8 volts (allowing for huge amounts of voltage sag) for 1 second
            (E = Voltage x current x time)
            suggests 8v x 150 x 1 = 1200 Joules required to start an engine. -this is the minimum energy required to start the engine.

            350F caps (as in this article that your solution is “very similar to” provides:)
            Q = CV (Charge = Capacitance in F x voltage)
            6 x 350 F capacitors in series gives a capacitance of 58.3 F
            Q = (58.3 x 12) = 699.6c
            and E = 1/2 QV
            So the energy out of your 6 super caps is only 349.8 joules less than a third of what is actually required to start the car. (and I didn’t allow for any voltage sag there, if you do then you get 235.2 joules, so super cap solution looks even more stupid at only slightly over 1/6th of the power required to actually start the car!)

            which means if you do actually leave your lights on for any length of time.
            (where a single 55w bulb will pull 4 A (and you’ve got two). you’ll discharge your bike battery pretty fast (this is the biggest capacity bike battery I could find on this UK auto site provides 28 Amp Hours (over 10 hours) e.g 2.8 amp draw.

            So with the best bike battery you’re drawing too many amps leaving the lights on for any length of time and shortening the battery life, then you’re cranking what I assume is a large engine, and cooking it that little bit more after you’re vastly inadequate super caps are discharged, (which is before the engine turns over, after that they just increase the drain on the bike battery that actually starts the engine! (but because the super caps only provide power for the first ~ half a second (or less) than the total 1 second crank time, you burn up half the capacity of the bike battery running the starter!

            And that bike battery costs £70 and weights 8.6KG

            or you could buy this battery,
            which is cheaper, and has better capacity, (Ah) and more CCA so would turn your engine over a bit better anyway!

            Worse, because you’ve got it under the hood of your car, you shorted the life of the caps -so as well as not really providing the energy required and taxing your cheap battery (meaning you replace it more often) but you’ll also end up with knackered super caps too.

            perhaps the most crazy thing about this whole situation is as follows:
            energy density of super capacitors ~1 – 30Whr/Kg
            (7th paragaph on this page

            Energy density of lead acid
            30 – Wh/Kg (

            So you’ve not actually saved weight at all! you actually have less charge available, rely on your small battery to do a large amount of the grunt of starting your engine.

            Of course Lithium, or nickel batteries DO have better density numbers, but can’t supply the large CCA required to actually start an engine.

            As I said earlier.
            you had a problem, which was likely that your battery was knackered.
            your solution was a convoluted system of expensive super capacitors and replacing small lead acid batteries… wherein your misapplication of technology saved neither weight (from an energy density perspective) nor cost, gave you an in-noticeable real weight benefit and reduced the usability of your vehicle…

            I suspect your principal reason for having weight reduction is that you want your car to go faster? – based on the numbers above I would therefore expect that you had a sub 2 litre engine the weight of a 2 litre car is going to start at around a ton, and have a round 100HP at a start (we’re talking average consumer cars here) the increase is pretty pathetic, 100 HP and 1 tonne is 100 HP per tonne, reducing your cars weight from 1000Kg to 995KG gives you a HP/tonne figure of 100.5BHP per tonne.

            We can take this to extremes and lets say a 1 litre saxo (60Ci) with say 60BHP, and a weight of about 800Kg,
            Your BHP per tonne was 75BHP/Tonne…
            Now it is 5Kg lighter congratulations your car is now 75.4BHP per tonne. that’s about the lightest European cars on the market, if you;re running something American they would tend to have more power and more weight, so your 5Kg drop is even less noticeable.

            the “cost” of this that the your car is unusable with the engine off, you can’t have lights or radio on for any length of time whilst waiting anywhere.

            so as I said earlier, your solution makes no sense regardless of what lens you try to filter it through.

            the maths prove that for cars replacing batteries with super capacitors of this type, and at this time, is an expensive and inefficient way to kill a battery that may actually be more expensive and not that noticeably lighter than the battery that is stock in your car…

            Basically, you wasted your time and ruined some of the use-ability of your car. as we now start to get into winter and your car gets a bit harder to start in the mornings I can only recommend that you undo your work.

            in the case of a motorbike which may require less power to turn over a smaller engine this makes much more sense… looking at a GSXR-1000 and a 12 year old one (say that’s old enough to start thinking that you know better than the designers at the time)
            170KG and 143BHP (that’s 841BHP per tonne)
            changing that to 165 and 143BHP is 866BHP per tonne,
            also as mentioned you reduce high up weight as an added benefit. (and a bike is easier to push when your super cap plan fails if the vehicle doesn’t start on the first turn of the key (as that’s all the power you have in the caps!)

            Basically, with a bike it makes a tiny bike more sense -but only because a bike will have a lower starting requirement! ad weighs so little to start with! with a car it makes no sense.

            but still doesn’t escape the really obvious statement that is made above:
            if something is broken (bad alternator means charging current is low, bad regulator means baking the battery is a problem, old battery with most electrolyte dried out means starting is poor, the most obvious and most practical thing is to fix the underlying problem!

  9. Not sure why I can’t reply directly to Dan… Too many quotes maybe? Anywho:

    Here’s some real-world information regarding starting a car on caps (6×3000 farad Maxwell boostcaps; 500 effective bank farads) vs battery (which is a Braille B2317, listed as ~>1100 cranking amps).

    Car in question is a 2009 Mazdaspeed 3 (sport model, no navigation or aftermarket sound system, or anything like that), which is a 2.26L (2.3L listed; ~138 cubic inch) turbocharged direct injected 4 cylinder. I did two different methods of testing the same start-ups:

    1. Logged battery voltage and RPM using my tuning device (0.1 second interval)
    2. Watched a volt meter connected directly to the battery terminals, for both cap bank and battery tests (really terrible interval, probably between 0.1 and 0.3 seconds)

    Note that the battery voltage in the log file might be pulled from the ODB2 port, as the voltages don’t quite line up between the external tester and the logged value.

    All the logs, the build process and pictures can be found here:

    Note that (free) registration might be required to view the attached pictures; if that’s the case, and there is interest, I can re-host the images off-forum for viewing. Also, that thread also contains links to videos of testing, bank leak down testing, and testing on big block Ford/Chevy trucks as well as my little Mazda.

    My testing actually produced some interesting (and somewhat demoralizing) results:

    Battery test 1 (cold engine start):
    Cold start in 0.62 seconds (0 to 1094 RPM; log values are 0, 114, 114, 174, 249, 672, 1094)
    Voltage dropped from 11.71v down to 10.02v at 0.1 seconds, then down to 9.31v at 0.2 seconds, followed by peak dip of 8.95v at 0.3 seconds; peak voltage drop is 2.76 volts.

    Note: The logged time interval changed slightly, probably due to key event, which is why it’s not a perfect .7 or .6 seconds.

    Battery test 2 (hot engine start):
    Hot start in 0.4 seconds (0 to 1104 RPM; log values are 0, 114, 243, 640, 1104)
    Voltage dropped from 12.02v down to 9.57v at 0.1 seconds, followed by peak dip of 9.53v at 0.2 seconds; peak voltage drop is 2.45 volts.

    Cap bank test 1 (hot engine start; 13.6 bank voltage):
    Hot start in 0.3 seconds (0 to 1006 RPM; log values are 0, 114, 528, 1006) per log
    Per the volt meter: 1.11 volts lost, netting a cap bank usage of 559 amps
    Per the log file, the voltage drop (which took place in 0.1 seconds) totals 1.44 volts or 720 amps.

    Cap bank test 2 (hot engine start; 12.2 bank voltage):
    Again, hot start in 0.3 seconds (0 to 1007 RPM; log values are 0, 114, 508, 1007) per log
    Per the volt meter: 1.12 volts lost, netting a cap bank usage of 560 amps
    Per the log file, the peak voltage drop (which took place in 0.2 seconds) totals 1.35 volts or 675 amps.

    The truth in amperage consumption is probably somewhere in between the two types of measurement, across both tests (most likely).

    I also did complete shutdown draw testing and found that my car EATS power at a rate that exceeds 250 milliamps (this is with the key removed from ignition, no lights or anything on, just the car sitting parked as it would be on a day I don’t drive it). This means I’d have to start it every 5-7 hours (assuming I let it drop to 9-10 volts before starting) if powered by the cap bank alone.

    With that out of the way, there is another reason why someone would want to run a capacitor bank vs a battery, other than to increase speed. Weight is a huge factor in fuel consumption from a dead stop, and every pound adds up. I’m pretty sure that there are more than a few people over on Ecomodder that have lightweight batteries to help with MPG, as well as all the other tricks they do (thinner oils, harder tire compound, aero, etc). Weight aside, since a small cap bank charges so fast, there’s less time spent actually charging it, and thus less drag from the alt while running.

    That said, there’s generally not that much that needs to be or even can be running on a bike, which in turn makes it much more viable for battery replacement than a regular car does, especially one as power hungry as mine. You could, however, leverage a small(er) cap bank with a larger amp hour (but physically smaller) deep cycle battery to have your proverbial cake and eat it too; higher amp hour rating, big cranking amps, and maybe even a dash of weight savings.

      1. I think the reply didn’t work as the reply was too long I’ll break it up… first positive or helpful things, then the whole bunch of naysaying…

        6 x 3000 is a lot different (not very similar) to 6x 350, so there likely IS enough power in the caps to start the car. it gives you more cranking amps, and the ability to have more than one go at cranking if the engine doesn’t start first time…
        -have you actually measured the caps discharge rate, if a 6x3000F bank can’t keep a couple of 55w Bulbs on for more than 1 minute then I suspect that you bought these caps from China?

        When you say that you did a shut down test, how soon did you do the shut down test after removing the key?
        I know that my car has computers that stay active for a couple of minutes, after removing the key (hence why people advise to wait after turning off the car before disconnecting the battery etc). -so to measure you’d need to have had the car off and hood up for a while, opening any doors or opening the hood is going to wake the BSI computer, (which will need a couple of minutes to go back to sleep again.) – I’d be surprised if your car is drawing 1/4 amp that big capacity bike battery could only provide 24Ah, (so your car wouldn’t move after being sat for a long weekend!)

        I suspect that this idea of super caps for cranking and a battery to trickle charge the caps when the engine is off could work, – but a better backup battery would be Li-ion.
        The old laptop battery sat in my desk draw at work feels like it weights maybe 1/2Kg (maybe less), and provides 4.2Ah @11.1v – 11.1 is close enough to 12, 6 of those makes the 24Ah of the motorcycle battery and only weighs 3KG (about 1/3rd that battery linked earlier) which is a much more significant weight saving. -and it’ll last about 3 times as long as a lead acid (500 – 1000 charge cycles instead of 300 – 500 charge cycles of lead acid) -also with Lithium it’s easy enough to make a 14.4V battery pack – which should be near what your alternator kicks out. (but Li-Ion costs 4x as much as lead, there is certainly no cost saving immediately… but the weight saving is still what I would call insignificant compared to the total weight)… (i guess that huge initial investment that likely never pays off in weight save/economy is possibly why automakers don’t install this OEM?)

        What is most peculiar is that the car you have was made either in or after 2009, it’s got to be five years or less old. (that site linked earlier offer batteries with a five year guarantee -with free replacement). it is surprising that you battery hasn’t lasted five years -it really should have!
        Anyway, I think that you sort of hit the nail on the head when you said. “when my car is off the power draw if 250mA.

        that’s high.

        which brings us back to the earlier statement.
        “I think that your car is probably broken, you ought to look into fixing that”

        Wait a minute, I went to the forum you linked. your car eats a battery each and every year?!

        once again, your car, has an electrical problem. replacing one battery technology with another will not fix that problem.

        Failing that you should just go spend less than you spent on super caps buying a decent branded battery (something like Bosch) who offer five year guarantees – you could have had your battery replaced for free!

        1. I bought the big bank to do testing with; I knew that worst case it would be enough to start the car at least once, but the Lasersaber guy (on youtube) is who inspired me to do this testing (again, for my project car which is still being built). I think every vehicle will have a different startup performance (power consumption wise) so not every vehicle will be a good candidate for battery to capacitor conversion.

          My shutdown test was done by shutting the car down, having everything closed/locked except for the hood. Over the course of 30 minutes (this was logged/discussed in the thread I linked), bank voltage dropped from 13.77 to 13.05. I don’t know what was drawing power, but it was drawing it consistently and there wasn’t anything on; it was sitting exactly as it is right now, not having been started since yesterday.

          I didn’t post this in the thread I linked, but I figured out what was causing an early “death” in those small batteries I use (and still only use them, no capacitors). The ECU commands (for whatever reason) a spike charge at 14.1 volts for only a few seconds on a cold start; everywhere else, it only commands a charge at 13.7v or 13.8v, so the battery probably isn’t getting fully charged every time. If I take the battery out and hook it up to a *good* battery charger when it starts to bog on cold starts (usually once a year), I can put a repair charge on the battery and get another couple years out of it.

          I don’t really think there’s an electrical system issue with the vehicle, I think that’s just how it’s designed. When I first got the car (brand new), the battery it had in it lasted about a week before it was replaced under warranty (I bought the showroom floor model that had been sitting for quite a while, I’m sure).

          As for lion battery backup, it would probably be wise to have some sort of current limit in place (both for charge and discharge) so that you can get a better cycle life out of the lion pack; that’s the one thing that worries me about using lithium batteries in the car along side a cap bank is exceeding the charge rate and hurting the pack before getting a good useable life out of it.

          As for replacing the battery, my car his heavily modified and I’m using what amounts to a larger motorcycle battery; the spec sheet says it’s a 23 amp hour battery with a full charge voltage of 13.2, but I’ve never had one actually sit past 12.5-12.6 volts or so (even after coming off the battery charger). The warranty is only 1 year replacement, 2 year pro rated, and considering my application, I’m not worried about Braille replacing it, though I am tempted to look into a lithium battery when both of the Brailles I have now kick the bucket, maybe something in the 50 amp hour range or more should be feasible for the custom bracket I have already.

      2. and now for the naysaying…

        I think that by and large I’m still in the camp that says. right now at this point in time this is nothing more than an expensive trick. the caps aren’t good enough to replace the battery all together, a very energy dense battery (such as Li-ion) might make a better backup/trickle charge source for the caps battery than the bike battery, (since it’s more energy dense – around twice if not more) therefore you can get much better weight gains, but overall you’re actually spending between 4 and ten times as much for a hybrid cap/Li-ion solution. for a probably less than 0.1MPG saving, -though if your caps aren’t properly sized you’re going to try to draw too much from a li-ion backup battery trying to crank the engine and cook it -maybe with spectacular results!

        I’m still sceptical about this weight saving. the batteries I posted last time shows the bike battery to be about 9Kg and the car battery to be 11Kg, that’s not very much difference…
        you still need to add what is likely at least another kilo of capacitors which robs some of the glory of that 2Kg weight loss…

        A quick search suggests a 10% reduction in weight adds a 6% increase in economy.

        I don’t know if your Mazdaspeed3 is the 2009 variant of the 2011 variant, so I’ll put both sets of numbers – from carfolio (and I’ll be generous and more than double the weight saving to 5KG…)
        2009 variant is 1450KG with a power of 260BHP
        that’s 179.3BHP per ton
        MPG is listed as 25.9MPG (using US gallons)
        5KG represents a 0.3% weight saving (179.9BHP per ton) which may equate to a 0.2% fuel saving. or a 0.05 MPG increase

        the 2011 variant is 1488KG and 263BHP
        MPG is listed as 18 MPG US
        That’s 176.7 BHP per tonne
        5KG weight loss is: a 0.3% drop in weight an increase to 177.3 BHP per tonne, and should equate to a 0.2% increase in fuel – again a 0.03 MPG increase. (there are a few thousandths difference obviously)

        Lets say you’re driving, 50,000 miles per year (which probably a huge over estimate -that’s 4x the average)
        then you save:
        2009 variant: 50000 / 25.9 = 1930 Gal used (over the year)
        2009 variant with weight loss: 50000 / 25.95 = 1926 Gal used
        2011 variant: 50000 / 18 = 2777 Gal used
        2011 variant with weight loss: 50000 / 18.03 = 2773 Gal used

        i.e you can save a few gallons per year, even at $10 dollars per gallon does that even cover the cost of the caps? ($40 per year) or replacement of the smaller bike battery that you’re stressing to an early death?

        essentially if I more than double the weight saving, more than double the cost of fuel, probably quadruple your yearly mileage then you can get a $40 saving?

        Realistically you probably travel the ‘average’ 12,000 miles, that cuts $40 in quarters to $10, and gas is about $4 – $5 per gallon, so you’re likely saving about $5 a year, but your weight saving is actually about 2.5KG on the battery not 5KG, when you swap out a big Lead battery for a smaller one so about $2.5 is saved, except you add weight in super caps that’s a lot of work and expensive components that you’re killing fast in an inhospitable environment for probably less than two bucks per year saving. – that doesn’t even buy the beer you’re using to congratulate yourself for a job well done!

        Digikey quotes a price of $60.40 each for these 3000F 2.7v caps.
        so you spent $362.4 on six caps, which will return no significant weight advantage, no significant power to weight ratio advantage, no significant gas reduction/economy advantage. and likely decrease the life of the backup battery you have to keep the caps charged when parked? – so you need to buy this more often.

        On the average 12k per year $5 per gallon ($2 per year saving) it’ll take just over 180 years to pay back (those caps will have long since died inside that engine bay! -according to the datasheet the maximum operating temperature is 60degrees, maximum storage is 70)

        Even with quadrupled average yearly mileage, doubled gas price, doubled weight saving, so you get to save $40 per year that’s still over 9 years, (at which point those caps are probably dead, and the car has done at least 450,000 miles! -i.e the car is probably dead too).

        actually now that I took the time to read a data sheet for the Maxwell 3000F super caps these are half a kilo each (510g)
        so OEM Lead acid = ~12KG
        replacement lead acid = ~9.5KG + 3.5KG in super caps, =~13KG + heavy gauge wire and connectors!
        so your net saving is actually -$1 per year – it costs MORE to use caps + backup battery than just a battery!

        I wonder if the same hyper-millers who are buying these are also buying magnets to align the fuel and vanes to put into their air intake too??!

        I’m even more sceptical about the claim that a capacitor bank will give better fuel economy than a lead acid due to shorter charge time and less alternator drag. -to put it another way. the lead acid will draw slowly over a long period, the cap bank will draw large and fast. – stressing your alternative and making it break sooner. but over all both draw the same amount, and that still needs to be produced via the traditional gas… -conservation of energy laws would suggest that you’re not getting a free charge for those caps any more than you’re paying extra for the lead. -ie. your alternator must still provide the correct amount of energy for the car to crank, the same Energy measured in joules must be provided by the alternator regardless of the battery type… it doesn’t matter if you charge at the start of the journey or using the journey, that electrical energy is still recovered from the potenial energy in the gasoline, -in fact I’ll go as far as to guess that super caps actually reduce efficiency as pulling that larger current will produce more energy wasted as heat inside the wiring loom. -unless you regulate that charging current down to somewhere near what a lead acid uses -in which case you see no benefit on the alternator run times (i.e same drag during the journey), and have additional losses in your regulation circuit to consider – i.e more wasted energy = more fuel!

        the argument of drag from the alternator works in one respect.
        i.e if you want to start the car and then have ALL the power available – i.e we’re back to go faster/race mods and not caring about fuel… but once again the saving is small (very very small).

        1. My Mazdaspeed is a 2009, purchased towards the end of 2008; it’s actually been over 6 years since I’ve purchased it, and it’s heavily modified (closer to 450 brake horsepower). When I’m not using fuel to make that kind of power, it is more fuel efficient than stock (despite running a ~29% mix of ethanol), not that I’m terribly worried about that; it’s all about the smiles per gallon for me. Weight savings and fuel economy don’t apply to my vehicle; again, this was all just a test for my project car which is still in the process of being built and is thus an unwritten book. The only reason I have a smaller battery in the Mazda in the first place was to make room for the turbocharger I have in it currently, and as it turns out I didn’t even need to do that. Bit late now, though. Also, I work from home, so in the 6 years I’ve owned the car, I’ve put less than 10k miles on it per year; this includes a trip to another state, and multiple trips to the state capital which exceed 300 miles round trip or so (once a month I go up there for work).

          For the record, when I tested this setup, I didn’t use any kind of extra wiring at all; I simply used some extra aluminum I had laying around, drilled holes in it, and then attached the terminals from the spare Braille I had directly to the bank making it a complete plug and play affair. Cost was a dollar or two for the screws/nuts, and I’m sure the terminals can be purchased somewhere if required.

          I don’t think charging a small (properly sized for the application) capacitor bank will cause an alternator to die extremely fast; usually its heat that kills an alternator, and being at full draw for 10-15 seconds is unlikely to cause that much issue in my opinion (especially considering I’ve tested this big bank against an older vehicle with older parts and had no issue).

          I totally understand your skepticism regarding alt drag on batteries vs caps and even share some of your thoughts regarding the subject. The only way to know for sure, though, is to test it (which isn’t really something I’m prepared or equipped to do, unfortunately).

        2. Those were lengthy replies…
          Dan, my car is a 2002 1.3L 16V fiat palio. It starts every time with 6x350F. And it starts right away, no hesitation.
          I wanted to avoid the cranking load on the battery, we have freezing temps every year. The goal was to make the battery last longer and sell some packs to the local drag racing fellows.

          Without alternator It cranks once, and maybe twice, but no more, thats where you need the battery. Also, after 20 or so seconds with the low lights turned on it will get under 9v and it wont start (again, if you don’t use a battery).

          I don’t care much if you don’t believe me, I only remember that my car starts on supercaps when I open the hood once in a while. Its a regular car that (hopefully) won’t need a new battery for many many years.

          1. I believe that it works right now. -with a car in good condition.

            but you can’t get away from the fact that an engine needs a given power to start, and whilst your caps may be (just about) providing that, on a day when it doesn’t start first time or takes a little while to crank you’re going to be out of luck!

            the point isn’t that you can’t or shouldn’t do this, more that you actually reduce your starting power, and increase the weight, and increase the cost.

            Whether or not you believe physics and maths is kind of besides the point.

            the manufacturer spec’d a battery with 400 CCA (four times more than used for workings before) for a reason other than shit’s a giggles, I doubt it’s because they have shares in lead mines either.

        3. The setup has plenty of power for this engine, the module weigh 600gr so its not heavier, and it is more expensive, although its getting cheaper every day. Today you get a 10F cap for $0.48 and I expect to see cheaper and bigger caps.

          I used to work in automotive stuff, I’m not a newbie here.

          1. Then you have the smallest pack available?

            the manufacturer doesn’t rate that for automotive use, as it’s too small. the maximum available amps is 170 amps, (bear in mind that your automotive battery is rated to provide 400Amps for an longer bursts as 30 seconds) – even the bike battery linked to earlier was rated for 200CCA.

            You need to get to at least the 16v large (made up of 6x 3000F) before the manufacturer will back automotive use.

            do you see the point I’ve been getting at?
            replacing a car battery with a cap-pack is not the intended use.
            even for a motorbike as in this article it is not providing the power that the manufacturer specified, nor is it an application that the cap-pack manufacturer supports.
            when you get to a pack big enough for automotive use, (and bearing in mind that you need a supplementary battery then all these ideas about taking away weight etc are false because you must keep the original battery…)

            The manufacturer claims that their cap packs are useful in situation where long term standby use of the battery without engine may drain the battery. – in this case it would appear that they are specifically targeting long distance truck drivers who will have lights tvs and a kettle being powered for hours at a time without the engine running.

            And for cars running in extreme cold, – a kind of cold that most of the English speaking world just never sees! -and if you happen to live in Siberia where winters get really cold (-60*C) then bear in mind that this is colder than the minimum storage temperature, and well below the use temperature of these packs!)
            even running at -20*C the ESR of the pack is at 125% so the maximum power available is 80% of rated specs… (now your too small pack is really too small).

            The manufacturer doesn’t claim this as a place where it’s useful, but I will -start/stop short run drives (like a delivery driver) in these circumstances there may not be time to charge a battery after starting before the next stop, and then current hungry start. -a cap-pack may increase battery life here.

            A battery should last 4 – 6 years, so we’ll call that 5 years, and costs something like $100.
            The super cap-pack you’re using costs $150 (based on a quick search that may not be the cheapest).

            Essentially if your solutions doubles the battery life you get ten years on 1 battery, AND you reach the end of your caps 10yr life (according to the datasheet) then you throw it all away and start again. you’ve spend $250 dollars -that’s IF you can double the life of a battery (and I don’t believe that you can!) also you’ve added weight with your super caps that supplement the battery -albeit only 1/2 kilo)
            sadly with your under rated caps you probably won’t double the battery life, and probably will shorten the cap-pack life.

            you may, if you’re really really lucky double the battery life with the much larger cap-pack that is rated for automotive use, but that’s a lot more expensive and a lot heavier. -this is all largely dependant on usage profiles anyway.
            in fact then -let’s say that you do the extraordinarily unlikely and double the battery life.
            you spend $100 on a battery and $600 on the large cap-pack for 10 years use…

            My solution is to buy 1 battery and wait five years then buy another battery.
            at the end of ten years just like you I’ve got to buy a new battery. but I’ve only spend $200. – additionally I’ve not been buying extra gas to haul around extra mass unnecessarily. (I call that win-win)

            the conversation thus far has centred around replacing batteries with capacitors alone, or reducing the size of a main battery and having it supplement cap-packs. (which is not what you’re doing anyway)

            In your solution, you are using a capacitor pack that is not intended nor rated for automotive use, that cannot supply the complete starting current necessary, and you’re mostly just using your battery to start. you stuff the capacitors underneath the hood in a place where temperatures will exceed the storage requirements of the capacitors. Basically you don’t significantly extend your car battery life, and you do significantly shorten your capacitor pack life.

            Essentially what you have done is stupid. -the temperature storage rating are there for a reason, storing too hot increases the chance of punch through in the dielectric. results to this is essentially a short, in this case it’ll be a short across the capacitor and across the battery, both of which may have explosive effects.

            oddly enough the replies are long because the proof of what i’m saying is detailed -I know that’s not usual internet argument rules, but I find it significantly better than just saying I’m right and you’re wrong. in this case I’m right, you;re wrong and that’s why.
            if you find an error in what I’m saying feel free to point it out so that I can also improve.

            I don’t care if you used to work in an automotive stuff -your internet claimed CV/resume doesn’t change the laws of physics.

          2. Dan:
            Not the intended use…Isn’t that part of the point of Hackaday in general? As for long term use, well, I think NitrousNRG is in a good position to test this for us; perhaps the results on his vehicle (which may be an outlier compared to other makes/models) will show profitability regarding how many batteries he would have gone through if not for the pack.

          3. Yes, sure, Hackaday is often about new uses for existing tech, or re-engineering, and it’s very cool.

            Perhaps you misunderstood, the reason these are not rated for use is that, (if you look at the datasheet) the continuous current of engine cranking is more than these devices should supply. The datasheet says 170Amp absolute maximum current, you said that you measured both hot and cold starts at more than twice that.
            So these do work, for a bit, but will likely not extend battery life significantly and fail early. (and as said above, cost more

            When I said not the intended use I meant it like how pop tarts are not intended for construction. it’s not that you can’t insert the odd pop tart here an there in leiu of bricks, the solution will work and be both warm and weather tight at first, but you expect that the solution will not last as long as a cheaper, easier more conventional solution.

  10. “The battery pack is built from six 2.7V, 350F caps, a few connectors, and a handful of diodes. These are lashed together with rubber bands to form a 16V, 58F capacitor”


    Correct: it is a series circuit! 6x
    2.7V = 16.2V 350F

    1. Unfortunately, that’s not how series capacitance is calculated. You basically lose half the rating per additional cap, but retain the overall amp storage of a single cap (at a higher voltage, of course). 16.2v @ 58.333F = ~945 amps, while 2.7v @ 350F = 945 amps.

      1. Yeah, unfortunately that’s not how capacitors work either, the capacitance doesn’t half for each addition capacitor added, the carrier formula is
        1/Ctotal = 1/C1 + 1/C2 + … + 1/Cn

        More unfortunately That’s also not how ampres work.
        Charge in capacitors is measured in coulombs,

        Where the charge is Q = capacitance (F) x voltage V

        350F * 2.7 = 945 coulombs
        But if the 6 350F capacitors are in series then 58 F x 16.2v = 945 coulombs.

        It might seem strange that 6 capacitors are storing the same charge as 1, but it’s not because energy stored is (w) = charge (Q) x V

        So in this case the total energy stored in a single cap is 945 x 2.7 = 2551.5 joules
        The energy stored in these six is 945 x 16.2 15,309 joules.

        Of course if you only needed 2.7v you’d put the 6 capacitors in parallel, summing them to a 2100 capacitor, with 5670 coulombs if charge and 15,309 joules if energy…

        1. My bad, I had the math wrong. Normally I use spreadsheets for this. If coulombs are an amp per second per volt then it’s not incorrect to say that a 2.7v 350 farad capacitor will “hold” 945 amps, is it?. When you string them in series, you’re trading amp storage for voltage range, no?

          Everything I’ve read about batteries and capacitors interchanges the term “amp” with coulombs when discussing ratings, unless the term “amp” is supplied with a suffix like “amp-hour.”

          1. Weirdly, not exactly.

            For the same reason you can’t take the 2.7v X 350Fand call it 940A. By the same token you can’t take that 950a multiply by the 2.7 and call it 2551 watts of power,

            The reason is that watts are a measure of work done, e.g 2amp flowing in a 12v circuit means 24watts of power is available.
            But that doesn’t describe the energy used.
            If you ran that circuit constantly for an hour you could express it as 0.025KWh, or you could express it as 86,400 joules of energy having been used in the hour. (As a joule is 1watt per second).

            But that’s because you express in a time domain also.
            You wouldn’t express it as 24amp. Or 24amp per hour.

            That’s why it makes sense (in some way) to express batteries (and capacitors) in amp hours (describing the amount of hours a battery can provide 1amp for, whilst staying above a set threshold voltage. (So in your equation that 950A would take the voltage down to zero… Because the amp hour relate to a specific test, known loads and specific voltage thresholds, and crucially includes the time domain.

            Given the steep discharge curve of a capacitor however, amp hours expressed by the proper test, is likely to be very low. On the other hand another metric for measuring batteries Cca is consistently high on capacitors.

            Basically, I know what you meant but was being pedantic. The units used do express different things.

          1. 945 Amps at what load? For how long? 945 Amps at a dead short would be pretty useless, on the other hand 945 Amps at 12 Volts across 1 Ohm for an hour would be huge!

          2. Thargrav, the internal resistance of supercaps is very low, so it’s possible to explode things with current alone (even at low voltages). For the sake of argument, let’s say 945 amps over the course of 10 seconds.

          3. OK, then let’s assume 945 Amps over 10 seconds. That would equal 2.625 Ampere Hours. A typical car battery would be about 120 Ampere Hours. And assuming the average output voltage was 11 volts through its entire discharge, the battery would be able to deliver about 1320 Watt Hours. Or another way to look at this is – put a 1 Ohm load on that battery and you’ll draw 11 Amps for about 11 hours.

            My point is Ampere Hours and Watt Hours are both measures of work and blindly quoting 945 Amps means nothing. So, please convert your 945 Amps into something meaningful. A typical car starter will draw between 300 and 500 Amps and in most cases will draw your car battery down to about 9 Volts because of your battery’s internal resistance. How long will your 945 Amps spin a car starter with a 300 – 500 Amp draw?

          4. That wasn’t the point of my original post. A 2.7v 350 farad cap can supply 945 amps as it drains from fully charged to empty, as can a 16.2v 58.333 farad cap bank. A farad is an amp second per volt, so the scale you’re looking for is already stated.

            You’re also changing what I said. I said 945 amps over 10 seconds, meaning from full charge to empty as stated previously. You’re assuming it would hold voltage and current for longer but the reality is the bank would be dead after 10 seconds and a total draw of 945 amps (time irrelevant; could be 945 amps in a second or 945 amps in an hour).

            The *ACTUAL* amp hour rating of this size bank is 0.26 AH (assuming 16.2v drained to 0v and a bank of 58.333 farad total capacity). I’m not really sure how you came up with 2.6 amp hours for the bank when knowing that a 1v 58 farads wouldn’t supply an amp-minute, so best top of the head estimate for a bank this size would be 16.2 amp minutes (actual is less due to rounding farad rating up to 60 for ease of maths).

            When I tested my 500 farad bank on my car, I found that starting it required 1.1v from the bank, suggesting total amperage consumption of the vehicle for that start was ~550 amps (time reference not needed, as I was only looking for a rough estimate of power consumption for the vehicle on an “average” start). THAT’s what I mean when I say 945 amps.

          5. Enki, ” 945 amps as it drains from fully charged to empty” is not a measure of energy. Amperage is a rate, it’s an instantaneous measurement. It’s like asking how far it is to the next town, and somebody saying “30 miles per hour”.

            You need to work in Joules or Watt-hours or some unit of energy, not a unit of power.

          6. Enki,

            The point I was heading for before we started talking past each other is:

            Although this is a interesting project and you can start a car with a bank of super caps, they really aren’t a practical replacement for a battery. The most obvious reason is what do you do when your car doesn’t start on the first try and your super caps are dead? Car batteries manage this with their reserve capacity and super caps have no reserve capacity. Also, with most modern fuel injected cars all the electronics plus the fuel pump kick on the moment you turn the key on. Wait too long, seconds too long, and the car won’t start off a bank of super caps. Then most modern cars including my Mercedes and my Wife’s Jeep keep the headlights on for a short while after you turn off the car. That would be another super cap killer.

            I did state back in 2014 when this article was first posted that a hybrid would be a great compromise. You should be able to set up a bank of super caps across a small lawn and garden battery and get the best of both worlds. The super caps would provide the surge current you need to start the car. The battery would supply the lower current loads and would recharge the super cap bank in case you needed a second start.

          7. Alright then, this again. Saying amp implies the rate does it not? Isn’t the definition of an amp one coulomb per second? In that case, 945 amps in one second from fully charged to fully discharged is still an accurate statement for a cap bank, right?

            I’m well aware of the limitations of cap banks, but this thread references a motorcycle, which as a class of vehicles tend to be much easier to push start than the four wheeled variety. Having to push start a bike (likely once a month) that’s only driven every so often in exchange for likely never having to buy another battery again seems like a good trade-off to me, at least, especially considering there are numerous very easy ways to make sure the caps don’t fully discharge in the interim.

            I will give you credit though. You got me with your finely crafted feeler posts that made it look like you were asking genuine questions. Good troll sir; that will teach me to try and be helpful on the internet.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.