[AzCal Retred]

On my trail bike project I replaced the battery with a Mallory 8.9 uF 16V - 20V cap with no ill effects. This has been suggested by Adrian II, Paul W, and many others. Walmart sells these for about $14 online, "free delivery" for orders of $35 or more. They come with ( #10? ) screws on their terminals and attach conveniently to the battery leads, on my machine at any rate. A little work with a paint-on coating such as "Star-Brite Liquid Electrical Tape" and you'd have a submarine, or at least a waterproof "battery".

If you set up a cap in parallel with the battery and fused the battery lead, you could pull the fuse and start your machine with a dead battery "Cap Only" like in Olden Tymes. This assumes you still have points & coil ignition, of course.

One of the drivers here was to stop the continuous battery killing 16V overcharging. Both of my 1999 Bullets routinely hit 16V delivered to the battery. The cap I picked had a 20V upper range and was twice as big as suggested, on the "If some's good, more's better" philosophy. It actually holds a charge after you shut it off and the points aren't closed. It takes maybe 20 seconds to bleed down across an analog voltmeter I installed in the casquette.

The next stop will be to graft in a $20 Roomba Nickle Metal Hydride battery on my street bike. These operate from 14V - 22V, so the Bullet charging system will have met it's match...no more acid dripping onto the cases or frame. The battery for street use is nicer after dark, should that happen.

[Adrian II]

I ran a 250 BSA C15 (converted to 12V) with the old Lucas 2MC capacitor quite successfully, though it did work better with a nice new Lucas RM21 in place of the old RM13.

Hitchcock's sell a high-output replacement for the 4 wire Indian stator actually made by Lucas, which might boost low-speed output, especially if you use a Lucas rotor with it.

A.

[axman88]

On my trail bike project I replaced the battery with a Mallory 8.9 uF 16V - 20V cap with no ill effects.

The one in your picture looks like an 8.9 mF, (millifarads), 8900 uF (microfarads) which is more suitably sized.  You probably know but I'll post anyway for others, there's 1000 uF in 1 mF, and 1000 mF in one Farad.  Seems like electrical engineers mainly use uF units, which is 1x10-6 F.

From what I understand a suitably sized cap would run from around 1500uF in a small dirt bike up to something like 18,000uF in a bike with more power.   The cap is there to give the charging system something to work into, to smooth the ripple.  I've been thinking of trying a similar trick with my C5, which is going to be a bit trickier, because I need to power up the fuel pump , power relay, and ECU as well as the ignition.

I understand that it is possible to replace chemical batteries with capacitors, provided they are these new fangled "Supercapacitors", which don't self discharge as readily.  Here's a guy starting his pickup with a bank of these.  https://www.youtube.com/watch?v=ewdtTUrTx74  I'm not interested this strategy, personally, I've got a kick starter.

Had my mind on capacitors lately, because I've been fooling with some old Speedotron flash power packs. 

Are you running the headlight on AC, if you kept the headlight, that is?

[AzCal Retred]

Axman88 - A Farad is a lot of capacitance. The mallory unit I have is indeed 8.9 microfarad (uF); 0.0000089 F. It is working just fine to excite the ignition coil and store enough charge from the Reg/Rec to do it again.
The headlight is AC, this bike came with a 4 wire alternator from the second half of 1999 transition year.
The amount of stored energy needed only depends on the coil requirements. The Bullet coil puts out enough voltage to reliably jump a 0.025" gap under maximum combustion pressures obtained with mild cam timing, a 6000 RPM threshold and a 6.5/1 CR. The same lash up would work on a 50cc or 1000cc engine with similar combustion pressures.
Smoothing DC ripple is a complete topic in itself, involving a complex interplay between capacitance, resistance & supplied  Henries from a "smoothing inductor". Again, quite a study in it's own right.
I think it's way easier to use a good lithium battery to run your C5. Batteries maintain a relatively fixed output voltage under discharge, and their short circuit currents are limited by internal resistance & chemical constraints. A BIG capacitor can discharge it's energy almost instantly, analogous to the difference between a sledgehammer blow and 4600 FPS rifle bullet impact. The "muzzle energy" in foot pounds may be the same, but the rate of application ( "impulse"; force per unit of time ) makes the high speed bullet massively more destructive. A lithium battery might catch on fire, but dropping a penny onto the contacts of a big Ultracap might blow a hole in your motorcycle frame. Check out the specs on this UltraCap "Battery":
https://www.techbatterysolutions.com/xs-power-sb630-1200-supercapacitor-battery-module-4000-watt-630-farad/?utm_source=googleshopping&utm_medium=cse&utm_medium=cse&utm_source=google&utm_term=SB630-1200&gclid=CjwKCAjwgOGCBhAlEiwA7FUXko9zWQw8tSOU-C_u5yg2nbWp0VG19qJBKjamOvt2mcptEhjsd45svRoCbCoQAvD_BwE
Max Amps are rated at 15,500. In the world where Zimmemr and I came from that is real Utility grade fault current, and can do savage amounts of destruction & mayhem in a very brief span of time. Currents like that turn a piece of 4/0 copper into an expanding cloud of copper vapor and molten bb's. Tread carefully and MAKE NO MISTAKES.

https://en.wikipedia.org/wiki/Farad
1 mF (millifarad, one thousandth (10−3) of a farad) = 0.001 F = 1000 μF = 1000000 nF
1 μF (microfarad, one millionth (10−6) of a farad) = 0.000 001 F = 1000 nF = 1000000 pF
1 nF (nanofarad, one billionth (10−9) of a farad) = 0.001 μF = 1000 pF
1 pF (picofarad, one trillionth (10−12) of a farad)

https://en.wikipedia.org/wiki/Supercapacitor
Unlike ordinary capacitors, supercapacitors do not use the conventional solid dielectric, but rather, they use electrostatic double-layer capacitance and electrochemical pseudocapacitance,[4] both of which contribute to the total capacitance of the capacitor, with a few differences:
Electrostatic double-layer capacitors (EDLCs) use carbon electrodes or derivatives with much higher electrostatic double-layer capacitance than electrochemical pseudocapacitance, achieving separation of charge in a Helmholtz double layer at the interface between the surface of a conductive electrode and an electrolyte. The separation of charge is of the order of a few ångströms (0.3–0.8 nm), much smaller than in a conventional capacitor.
Electrochemical pseudocapacitors use metal oxide or conducting polymer electrodes with a high amount of electrochemical pseudocapacitance additional to the double-layer capacitance. Pseudocapacitance is achieved by Faradaic electron charge-transfer with redox reactions, intercalation or electrosorption.
Hybrid capacitors, such as the lithium-ion capacitor, use electrodes with differing characteristics: one exhibiting mostly electrostatic capacitance and the other mostly electrochemical capacitance.

Pretty cool buzzword collection, eh? The propellerheads have their own language, far beyond us mere mortals...

xxx

[Paul W]

I have to declare innocence here!

Unfortunately I know nothing about running with a capacitor instead of a battery and unless I was possessed by some spirit or other (and these days I seldom drink alcohol other than one very small beer, on doctor’s orders), I haven’t got involved in such discussions.

As far as I’m aware, “old” 350 needs a battery in the system because it has TCI ignition, rather than points.

[Adrian II]

The classic iron barrel Enfield India 350 always had points ignition (apart from VERY early ones with magdynos), your Electra 350 was a slightly modernized version which only appeared in the early 2000s.

A.

[AzCal Retred]

Paul W - sorry for inadvertently dragging you into the Cap/Bat issue! Must have been an "Implanted False Memory". 

Axman88 was apparently looking to expand the envelope and use Ultracaps on a machine that has "modern" electronics. Most of those electronics appreciate a stable operating voltage. Caps are a "linear" device and so charge is proportional to voltage, which drops as charge is withdrawn. The "Ultracap battery" link I posted may possibly use internal electronics to achieve this, but with a max discharge amp rating of 15,500 Amperes it seems unlikely. A great tech, but maybe not the most appropriate application. A lithium battery is fully developed tech which supplies a high charge density battery at minimal ( for current tech ) weight.

The Mallory cap is one of a series sold through Walmart. There are several others that have greater capacity but operate at 25-30 V. I picked the one I did because that caps operating range (16V - 20V ) approximated the Bullets Reg/Rec normal charging range of 14V - 16V. That means the actual storage capacity closely approximates the label value. The slightly higher versions will lose a bit, but would be more resistant to failure as the max charging voltages would only be about 60% of design. If this one fails I'll just go up a voltage level, but I think I'm good.

These readily available Ni-MH Roomba batteries look like a great option for kick-start machines that need to run a battery. Only about $25 and they have a small footprint, about the size of an ignition coil. They're nominally 14.5V, the Bullet charges at 16V. The Roomba charger operates at 22V and the manual says you can park it there indefinitely until used, so the 16V from the Bullet should be a non-issue. No more boiled dry, "cooked"  battery.
https://www.amazon.com/3500mAh-Replacement-Battery-iRobot-Cleaner/dp/B07D2ZJMK7/ref=asc_df_B07D2ZJMK7/?tag=hyprod-20&linkCode=df0&hvadid=242004741748&hvpos=&hvnetw=g&hvrand=13375567095247646803&hvpone=&hvptwo=&hvqmt=&hvdev=c&hvdvcmdl=&hvlocint=&hvlocphy=9031849&hvtargid=pla-456427272930&psc=1

[Paul W]

Adrian, yes I’m very much aware most iron barrels had points. I actually meant to type “my “old” 350”. I was reinforcing my point to AzCal about my particular bike needing a battery to run, rather than me trying to experiment with something that works very well already.

AzCal, no worries 😉.

[axman88]

Axman88 - A Farad is a lot of capacitance. The mallory unit I have is indeed 8.9 microfarad (uF); 0.0000089 F.

The Mallory CGS892U016R2C, like shown in the picture you attached is rated at 8900 uF.  Thats an electrolytic 1.38 diameter x 2.12 tall.  More info here:  https://www.cde.com/resources/catalogs/CGS.pdf

8.9uF would be a tiny thing.  I agree a Farad is a lot of capacitance.  Most of the commercially available "battery eliminators" offered commercially are at least 8000 uF.

This Lucas is 10,000 uF:   https://www.classicmotorcyclespares.com/index.php/default/capacitor-lucas-2mc-type-energy-transfer-54170009-54483156.html

Oregon is offering one they claim is rated 35,000 uF:  http://www.oregonmotorcycleparts.com/BEC.html

Most sellers aren't disclosing the capacitance, OR the voltage rating.  After all, that's the JuJu that you are paying extra for.

https://bikerdirect.com/electrical-battery/batteries-covers/battery-hardware/battery-eliminator-for-magneto-models?_vsrefdom=adwords&gclid=EAIaIQobChMI8vjL5PzG7wIV3TizAB2ZNwZeEAQYBSABEgJaavD_BwE

https://www.hughshandbuilt.com/product/battery-eliminator-capacitor/

https://www.summitracing.com/parts/alm-151308?seid=srese1&gclid=EAIaIQobChMI8vjL5PzG7wIV3TizAB2ZNwZeEAQYDSABEgJRvfD_BwE

The capacitor is mainly about eliminating the worst of the ripple from the output of the rectifier.  If you are retaining a battery in the system, the capacitance of the battery will far outweigh any additional coming from the cap.  Generally speaking, nobody intends to use these capacitors to start their bike with an electric starter.  Starting energy comes from your leg, and electricity to run comes from the generator / alternator.

Having power for starting is not my intention either.  I'm more interested in a temporary measure to keep me riding with a battery that is no longer capable of doing its job.

Most of the time the battery eliminator, i.e. capacitor route is taken by folks who want to eliminate the battery for weight (racing) or cosmetic (chopper) reasons. 

Arguably, one could trim more weight, and get more performance by eliminating the alternator, with it's torque drag and heavy stator and rotor, and use a battery in total loss if this is allowed in one's racing class.

[AzCal Retred]

Right you are! 8.900 uF is apparently equal to 8,900 uF (comma, not period). Two countries separated by a common language?

The C5 electronics like a relatively constant voltage source. If the charging system when kicking it over pumps it up faster than the electronics bleed it down, it might be a winner.

The 8.900 uF cap is working great for my points bike as the draw is intermittent.

The operating voltage is the key here, that'll give you the rated output. The Max is just that, the voltage limitations of the internal material.

Parallel caps for capacity. A pair of the 740,000 CGS744U016X5R Mallory units would give you nearly 1 1/2 Farad, quite a lot. It would also be at about 14 - 16 volts, a useful number.

Using a cap in parallel with a battery is more about surge capability, as the cap will discharge into a load before the battery as it has less internal resistance. It prevents voltage sags, a useful trait.

With a cap only the voltage regulation will flail about more than with a battery. My casquette mounted analog meter windmills between 12V - 18V on cap only. If your big cap allows for a start up with your electronic ignition, you could use a 20A plug in auto-style fuse for starting isolation, then just plug it back in once it was running.

[nonfiction]

I probably should have just used the capacitor on my dirtbike project. But I got impatient and also wanted badly to be quit of the ugly regulator and rectifier, so I opted for the spendy Boyer Bransden Power Box. It's worked perfectly, though I left the tail light circuit disconnected when I did the swap. Always something. Neat writeup, ACR!

[axman88]

The 8.900 uF cap is working great for my points bike as the draw is intermittent.

Using a cap in parallel with a battery is more about surge capability, as the cap will discharge into a load before the battery as it has less internal resistance. It prevents voltage sags, a useful trait.

With a cap only the voltage regulation will flail about more than with a battery. My casquette mounted analog meter windmills between 12V - 18V on cap only.

I'm not understanding the value of putting a capacitor in parallel with a usable battery?  As I figure it, your 8900uF cap, if it starts at 13V and discharges down to 8VDC, is releasing less than 1/2 of a Joule of energy.  0.50 watt-seconds at 12volts is the equivalent of delivering 40 milliamps for a second, or 1 amp for 1/24 of a second.   It's not much usable power.  The cap, even if it were 50,000 uF, is a tiny amount of energy capacity, relative to the battery.

The battery, on the other hand, is like a massive capacitor.  Discharged to half of its theoretical capacity, a 12V 7Ah battery would have delivered something like 151,000 Joules of energy.  At 12Volts, that would be like a 2000 Farad capacitor.  I don't see the benefit of putting a  .009 Farad cap in parallel with that massively larger capacitance?

Maybe it does help keep the regulator calm, which is the point of having the cap there as a "battery eliminator".  How much did the regulator bounce without the capacitor present?  Didn't you suspect the regulator of misbehaving, on account of it overcharging the battery in the first place?  I recall you discussing this, but don't recall the details.  Can you point me to the thread(s), where adding the cap was recommended?

The C5 electronics like a relatively constant voltage source. If the charging system when kicking it over pumps it up faster than the electronics bleed it down, it might be a winner.

If your big cap allows for a start up with your electronic ignition, you could use a 20A plug in auto-style fuse for starting isolation, then just plug it back in once it was running.

My idea is quite a bit different.  I'm thinking to start the bike using a few alkaline cells, like 4 AAs, through one of the cheap and tiny, but remarkably efficient DC/DC BOOST converters that are ubiquitous these days.  This gadget would attach to my SAE battery tender port, and needs to generate enough energy to activate the fuel pump, power relay, ECU, and ignition.   It can deliver that energy at it's leisure, over a minute or more.  The capacitor is there to act as a 12V accumulator to start the bike, then keep the alternator / rect / reg happy once the engine is running.  Once the bike is running the battery pack with the DC/DC Boost can be disconnected and go into my pocket.  I think to play this game, the battery would need to be disconnected.  Otherwise it would suck up all the power I can hope to deliver with a pitiful handful of dry cells.

The gadget is a kind of, as Dave Letterman used to say, a "stupid human trick", but, if my battery goes south, I much prefer to purchase my replacement over the internet, and in the meantime, I can ride, albeit carefully, probably turning off my headlight when the engine idles, or perhaps while revving my engine like the Harley guys in that South Park episode, "The F Word".  https://www.youtube.com/watch?v=ipDmsxQVxIM

If it's a wacky idea, it's probably best to hear that now.  I know you guys won't spare the truth to save my feelings from being hurt.

[AzCal Retred]

Capacitors are analogous to an accumulator on a hydraulic system. A battery is closer to an elevated pool connected to the load through smallish piping.  The cap/accumulator rapidly compensates for pressure fluctuations in a "solid" piping system. The battery/pool provides a steady flow of fluid at a relatively constant head pressure until the pool is empty. A pressure/voltage drop happens if demand exceeds piping system flow ability.

The alternators are permanent magnet units and function like positive displacement (PD) pumps in a hydraulic system. The alternator feeds the reg/rec. The reg/rec puts out pulsating DC at starting speeds to the energy storage device. The storage device, battery or cap, has to provide appropriate power to the ignition system. In your case it needs to run the paraphernalia associated with the EFI system. It has to provide this power long enough for the engine to start.

in my application the cap provides a small amount of time displaced power for the ignition coil, as there is no guarantee that the alternator DC pulsation will happen at the right time. The DC from the reg/rec "pumps up the accumulator" which stores it until the contact points close and current flow begins in the coil. Everything else is Bernoulli physics, fluid flow & gravity, just as God intended. 

In your application the alternator/PD pump normally supplies power to "pump up the power pool". The battery dampens these pulses by virtue of its chemical process. The battery charges and discharges much more slowly than a cap due to internal resistance and its chemistry. The battery/pool supplies juice to the appropriate devices.

A bad battery is like a pool with a large hole in the wall, the alternator/pump can't refill it and head pressure available from it is wrong.

Without a battery or cap the charging system can't store energy, everything would have to be phased perfectly, like in a magneto.

You need to put a meter on your machine and see what the actual power draw is for your fuel pump, power relay, ECU, and ignition. Your "Pony Battery" and interconnecting cabling just needs to be sized to accommodate this load. The alternator will cover all of the lighting loads as long as the engine is running. I would either fuse it or put a small circuit breaker in line with it to cover those times you forget to shut off the lights before stopping the engine.

1) Measure your actual load
2) Bootleg in a "Pony Battery" sized to cover this minimum amperage, maybe plan on 150% capacity to provide a bit of wiggle room.
3) the alternator will keep this charged if you leave it attached. The Pony Battery will try to run the headlights if you forget.
4) Drill - Engine ON, Lights on; Lights OFF, Engine off. Fuses are good!

https://www.tractorbynet.com/forums/threads/what-is-a-pony-motor.16114/#:~:text=A%20pony%20motor%20is%20a%20small%204%20cylinder(on%20John,and%20they%20deisel%20started%20up.

[axman88]

Using a cap in parallel with a battery is more about surge capability, as the cap will discharge into a load before the battery as it has less internal resistance.

One would think so, but, ...

The data sheet for your Mallory 8900uF says its Equivalent Series Resistance rating is 0.034 ohm at 120 hz. and gives a multiplier rating for higher frequency of around 1.25 for frequencies in the range that would be produced by an engine turning at Bullet engine speeds coupled to a 3 coil alternator.  So, something like 42 milliohms at alternator frequencies.  There are much lower ESR caps available, even in the same series from the same manufacturer, you can get down to a third or less of what yours has.

I haven't tested your battery, and maybe it's got quite high internal resistance, But this guy Jörgs measured a few of his motorcycle batteries and published measurements of 35 and 26 milliohms for two different batteries in good health.   This seems consistent with some other published results I've found.  https://www.schweizerschrauber.ch/mot/batt.html#:~:text=A%20healthy%2C%20fully%20charged%2012,is%20drawn%20from%20the%20battery.

On the other hand, a bad battery with a defective cell gave Jörgs a measured internal resistance of 100 milliohms, which is about what a D sized alkaline cell will measure.   http://www.learningaboutelectronics.com/Articles/Battery-internal-resistance

A healthy battery should have lower internal resistance than that particular electrolytic capacitor and 100,000 times the capacitance.  I'm having a hard time understanding why a parallel cap would help prevent the regulator from overcharging the battery?

My scheme has NO battery in the system, once the bike is running.  The battery is only there for starting, and has very limited ability to produce power.  The cap could produce decent amperage, based on it's size and ESR, but, it doesn't hold much energy, I think the pulse would be too short to trip a fuse.  The C5's 3 phase alternator will be a huge plus in producing steady power, there's something like 9 coil pairs in the stator.  If the voltage sags too low, for whatever reason, the ECU will likely protest by shutting down operation. 

As you suggest, knowing the draw of the various components is critical to sizing the components.  I'm thinking that the fuel pump draw will be the difficult obstacle.  It's gonna be tough to get 3 or 4 seconds of multiple amps of 12V power out of what I'm envisioning.

One of the first cars I owned was a Toyota Celica with a fuel pump buried in the gas tank that had gone intermittent.  I kept a hammer in the trunk, the access cover was left loose, and knew just where to bang to get it going, but eventually I got more sophisticated.  I made a hole in the air filter housing with little cover hinged from a single rivet that I could rotate to the side such that I could put a shot of gasoline straight into the carb.  Once the engine started, the vibration and increased system voltage would convince the failing fuel pump to join in the fun.  I drove it that way until it was stolen, lucky thief!  It was recovered, with the ignition lock pulled out.  I drove it for another half a year using a screwdriver as the ignition key, until it got T-boned.  No way that heap was getting a new fuel pump out of me!   I still have the screwdriver though.

I have too much respect for my C5 to Frankenbike it like that.

[Nitrowing]

Hold on a minute!
Is all of this because the reg/rec is providing too much voltage and cooking the battery?