Dan has not answered yet,
Still typing away on my phone...
Dan, do you notice any loss when you "move" your energy from one battery to another?
Yes, there are always losses in these circumstances. There are also losses when charging a battery (source -> storage) and even some when drawing on the resources of a battery (some types also have circuitry to prevent full drain, so available capacity may be as little as 85% of calculated maximum...this seems to be a common limit to prevent damage to the pile).
The amount of loss can vary depending on battery type, load, circuitry, and a number of other variables. From what I've seen in manufacturer literature, generally somewhere around 71% net efficiency is considered to be pretty good, as a thumbnail figure, but it varies.
Is there any noticeable difference between charging a device directly from your dyno hub, versus charging a "buffer" battery from the dyno hub then charging your device at night from the buffer battery.
Well...yes, but this is really three questions and not two.
1) Yes, charging a battery direct from a dynohub via charger *may* be most efficient.
2) If the buffer battery is a pass-through type and the capacities and demand are matched, charging efficiency at the gadget may come close to direct charging from a dyno source, or it may be less efficient. All things being equal, ganged charging through a buffer battery could reasonably be expected to be less efficient than direct charging. Part of it depends on whether the pass-through buffer battery is fully charged or not and how often it is called into use (say with frequent stops or crawling uphill at slow speed. In these circumstances, the buffer battery will also need to be charged and the source may be unable to maintain equilibrium, so a net overall loss may occur).
A key determinate here is use. If you just want to use the buffer battery to *power* a device intermittently (gap-filling when dynopower would otherwise be too little), that is different from using the buffer battery to power *and* charge a device. Sometimes the device may draw so much power in use it cannot also be charged. Conversely, a device may draw too much current to be operated solely by the dynohub, and one must be content with charging the device's battery while it is powered-down. Sometimes, a buffer battery only works in the system as a way to slow the rate of overall drain, thus extending the service life of the gadget. I sometimes do this when using my GPS for all-day track-making. I then make up the loss by charging overnight from my solar accumulator batteries or from my large mains-rechargeable pile. This way, I don't start the day with an energy deficit that would only become worse.
3) When I charge an item while in camp at night, the storage battery is a "buffer battery" of sorts, but not by intent. The source/storage battery is really an accumulator battery and instead of passing current through and being ready to step in as a gap-filler on a charging system, it instead has been used as an electrical reservoir to accumulate energy to either draw upon directly to power a device while at rest or to recharge another battery. Without knowing the particular setup and combination in use, here again 71% efficiency in energy transfer seems to be considered a "good" figure, possibly about the best one can hope for in practical usage. Kinda depends on how one defines "efficiency". it most often is taken to mean energy lost when going from a source to a storage battery.
To put a twist on an old phrase, "You can never be too rich or too thin or have enough power in reserve". having more than enough energy at your disposal makes losses much less worrisome or problematic.
Like you, I am fascinated by the combination of electronics and self-generating power while biking. When I was a teenager, I had a bottle generator that rubbed on the side of tire, powering a regular (non-LED) headlight.
This brings back memories. When my father and I would tour together, we used Union 9814 bottle dynos with rollers set to use the tire sidewalls. Drag from their plain bushings alone was so high we called them "drag brakes" and "retro-rockets". Efficienty was also low and light output from the incandescent bulbs was miserable. Their one virtue was relative reliability compared to the competition at the time, though I always thought Soubitez was better and Lucifer and Miller bottle dynos were much better but unobtainable for me. They were nowhere to be found in US bike shops in those pre-Interweb days.
It was a lot of work, and not much light.
Agreed! Then, what light there was, became less as the filament coated the bulb globe and got dimmer and dimmer. Or blew on fast descents. This was how I came to make my first voltage limiter, back in the day.
When I got my first battery-powered light, it was like "night and day". But what I really wanted was a universal power and connection system on the bike: one larger battery that powered everything on the bike (lights, computers, etc) using standardized connectors. It seems that we are converging on something like that, with USB being the standardized connector. I still would like a single larger battery that could power everything. It could be charged off the mains if necessary, and topped up from the dyno hub. I wonder if a single larger battery would be more efficient (or less) than smaller (AAA or AA) batteries in each device.
I share your desire, but here is where we run into trouble: The USB standard is fine, but the connectors aren't weatherproof, and I have a big quibble with that. Practically, the touring cyclist must shield the USB connection by placing it under cover, usually in a handlebar bag.
Actually, I do have a problem with the other part of USB, and that is the limited voltage and current supplied by the USB 2.0 standard, a nominal 5.0vdc @ ~500mA. It works, but it can double or quadruple the charging time of high-draw gadgets designed to "quick-charge" on mains power using ~1.5A-2.0A of current. If you have the time for charging and select your gadgets carefully, you can do well. However, things could be even nicer if we had that 12v SON dynohub that operates at roughly twice the current. I think we may get there eventually, via the need to power/charge high-draw/high-drain devices that are growing ever more power hungry.
Ultimately, the limiting factor is the human powering the lot. If drag is increased too much, it will begin to take away from the riding experience and become "too much work". Right now, we're doing about the best we can with what we have in terms of 6v/3w dynohubs. Shutter Precision claims higher efficiency with less drag on their dynamos, and of course dynamos that produce less power like the SON Deluxe have less drag than the SON28 Klassik or Neu, but the difference is pretty slight in real-world use. I think the next breakthrough will come in terms of greater circuit efficiency in gadgets to be charged and in better technology. I'd really like to see practical, small fuel cells come to market. That would solve many problems and perhaps completely obviate the need for dynocharging. Battery technology is holding back the entire electronics industry. If we only had more efficient power sources.... Some visionaries in the battery industry have wondered if supercapacitors might someday take the place of batteries, but that idea has problems when scaled.
My Anker mega-storage pile comes closer to your goal, but it is not bike-(re)chargeable and requires mains power to fill. This is the Achilles heel of most large storage batteries when combined with on-bike charging: They simply require more power to charge than a dyno charging system can supply. Too, many such batteries have charging "trigger" circuits and won't even allow charging unless the supply meets requirements.
One thing to keep in mind: A "big" battery is rarely a single battery. Instead, it is most often comprised of a number of smaller cells that are ganged to supply the needed power. You can see a bit more how this works in my background discussion of the Tout Terrain/Cinq5 SmartPower II, here:
http://www.thorncycles.co.uk/forums/index.php?topic=10387.0 An aside: If you have time some rainy Sunday afternoon, take apart a PP3 9-volt battery. I do this sometimes to use the 6 AAAA (four "A's" @ 1.5v each) batteries for projects. They are an LR61 size, similar to the LR8D425 AAAA cells used in medical equipment. Most Alkaline 9-volt internal cells are round. The carbon-zinc and a few Alkaline internal cells are rectangular. Their usefulness depends on how you want to repurpose them.
It is entirely possible to charge what amounts to a "big" battery by individually charging the smaller cells that compromise one of your own making. I've done so, with reasonable success and little practicality. Charge enough AA cells and then put them in holders wired appropriately with a controller, and you end up with a battery that would be too large to charge if you tried to do so direct from a dynohub. Charge the components separately and you can manage, but it will take time. Depending on demand, it might not be practical to try. It wasn't for me. I did the math in advance, then went ahead and tried anyway. It certainly "worked" and I had a "big battery" sufficient to power a netbook through it's charging socket, but it took forever to accumulate enough power in pieces to do so and it weighed a lot and wasn't small. That's when the phone began to look like a Much Better Idea and I went that direction. You pondered...
I wonder if a single larger battery would be more efficient (or less) than smaller (AAA or AA) batteries in each device.
...and the answer lies above. *Charging efficiency* appears slightly better with fewer aggregate losses if mass charging a lot of ganged cells compared to individual ones (according to battery company data I recall seeing), but then you run into the problem of a battery too large to charge by bikie means. There's lots of other issues related to rate of charge and the charge-characteristics of different kinds of batteries, also.
All the best,
Dan. (...who can usually be seen charging ahead...
)
