Hello everyone:
I started to post a comment to an existing 18650 thread, but as I thought about this topic, I realized how darned confusing these cells can be. I thus thought it might be helpful to post a summary of what I’ve found as some basic info for the person who is (like me recently) just learning about lithium ion.
I know that a lot of folks are very familiar with the following information, but it represents some details that were not what I might have expected and that can be very helpful in selecting the proper 18650 for various uses. There are a bewildering array of formulations out there now, and it is definitely worthwhile to consider things other than just the mAhr capacity.
If those who are more familiar with this topic find that I am mistaken in any of this, please do not hesitate to point it out:
I’ve been interested in LED flashlights for quite a while. I hadn’t followed them closely for a year or two, but got bit by the bug with some of the new, small-and-bright options (everything from the Fenix PD32UE to the Nitecore EA4&EC25, currently researching some of the EDC-sized “convoys”). I am also currently being tormented by the need for a 3x XM-L light. I can blame a friend who asked my advice on a couple lights . . . knowing that my information was stale, I decided I should refresh myself before suggesting anything for his needs. Of course, the bug was just sitting there waiting to bite!
As part of getting “re-interested”, I also did some research on 18650s and found some non-obvious things to consider. Most notably, in the case of a 1x 18650 light that does not have any boost circuitry, the humble Sanyo “UR18650FM”, at a “mere” 2600 mAhr can actually be a superior choice to the higher capacity (and usually more expensive) 3100 to 3400 mAhr cells. As I understand it, the vast majority of lights designed to run on one 18650 do NOT have any voltage boosting circuitry. Those that can also use a pair of disposable lithium 123A cells (or rechargeable RCR123As) will usually hold a flat lumen output over most of the battery cycle because there is plenty of voltage to work with. But with a single 18650, a decreasing lumen output may be observed as the 18650 discharges and the voltage drops. With the low forward voltage of the XM-L LED, this may not be a problem with lower output settings, but may be quite significant when the light is running on a maximum “turbo” or “burst” setting. I’ve definitely seen this with the Fenix PD32UE and the Nitecore EC25, as examples.
The important thing to note is the qualification to the statement that the Sanyo may give “better” performance. This statement assumes that the performance feature of most interest is maintaining a higher lumen output for as much of the battery cycle as possible. If a person wants longest runtime and is not as concerned with maintaining lumen output (this could be the case in a power failure or even a camp-out), then the higher capacity cells (Panasonic 3100 or 3400) may be the better choice. Although, since these cells are rated based on a discharge to 2.5 V, one may not be able to use all of this capacity with some lights (some lights may start strobing or otherwise warn of a low battery at 2.75 volts or higher).
As a practical matter, these details might not be too big of a deal for everyday use because shallow discharges are better, anyways, and the cells will often get recharged before they get drawn down very far. In such cases, performance differences may not that noticeable. Perhaps that is the most important thing of all that I have figured out - with good quality Sanyos, Panasonics, and Samsungs now available for really decent prices, a guy doesn’t need to get TOO worked up! Buy good quality cells, go protected (unless and until one thoroughly understands the details of LiIon operation, where unprotected is acceptable, how different formulations may vary in their degree of hazard, and how to handle them) and don’t get too obsessed with the “ideal” cell.
But if a person is interested in the best choice for maintaining higher lumen output in one-cell lights with cells that are going to be run to a modest discharge or deeper, performance details can be pretty significant.
The easiest way to see this is to go to HKJ’s battery comparator at Battery test-review 18650 comparator.
If one compares, for instance, the Panasonic NCR18650B (3400 mAhr) to the Sanyo 2600 mAhr UR18650FM (listed as “Sanyo 18650 2600 mAhr (red)), the Panasonic will definitely provide a lot more total energy (even if one only goes down to 2.8 V as HKJ did with these tests). At 3 amps (probably at or above the upper limit of what most 1x 18650 / 1x XM-L lights would draw at max), the Panasonic will deliver ~3100 mAhr while the Sanyo will deliver only ~2350.
But the Sanyo will maintain a lot higher voltage for a significant portion of the cycle. For the first amp-hour of discharge, they are pretty close in voltage. But the Sanyo will be able to deliver 1750 mAhr before falling below 3.5 volts. The Panasonic, on the other hand, will fall below 3.5 volts at ~1350 mAhr.
The voltage differential is even more pronounced if one compares the Sanyo to a Panasonic 3100. HKJ doesn’t list a bare 3100, but I believe that 3100 mAhr cells like the AW, Orbtronic, and others use the Panasonic 3100 cell and could be used for comparison. I did test a Fenix PD32UE (via ceiling bounce with a simple light meter) and found that a 2600 mAhr Sanyo did indeed maintain output on the “burst” setting a lot better than a Panasonic 3100. I don’t have the numbers immediately available, but could provide them up if anyone is interested.
In my case, for EDC of something like the Fenix PD32UE (or most other 1x 18650 lights), the Sanyo is a better choice. I would rarely discharge a cell DEEPLY, but may discharge it to a moderate degree before getting back to a charger. In this state, the higher voltage curve of the Sanyo will maintain higher “burst mode” lumen output if I need it.
One note: HKJ’s site lists two “Sanyo 2600” cells. The one that is listed with the “bv” in the title shows a lot poorer performance. But if you look back to the individual test, this cell is actually listed as a 2800 mAhr Sanyo UR18650ZT which is sold by eFest with added protection. This cell is actually one designed for a 4.3 volt charge, but it was only charged to a standard 4.2 volts for testing. If eFest was selling it as a general purpose cell intended to be charged to only 4.2 volts, representing it as a “2600” is actually appropriate - but also confusing. The other “Sanyo 18650 2600mAh (Red)” (no “bv” in the listing) is an actual UR18650FM.
That brings up another point that can easily fool a person. Before digging into things, I almost ordered the Sanyo UR18650ZTs instead - after all, they are 2800 mAhr instead of 2600 and about the same price (at Fasttech). But the 2800 mAhr rating is based on a 4.3 volt charge. With a 4.2 volt charge, these 2800s will actually deliver a little LESS total energy and a little LOWER voltage profile than the 2600 mAhr UR18650FM (compare the “bv” cell to the other Sanyo 2600 in the comparator)!
I believe that there are other cells, like the Samsung ICR18650-26F, that also maintain higher voltage during discharge. But please note that I have not seen detailed discharge voltage curves for them or tested them myself. As with the Sanyos, the Samsung 2600, being a 4.2 volt cell, may actually provide superior capacity and/or voltage than the 2800 and 3000 mAhr Samsung varieties if one will only be using a standard 4.2 volt charger. As with the Sanyo 2800, these higher-capacity Samsungs are designed for a higher max charge voltage (see http://www.samsungsdi.com/battery/cylindrical-rechargeable-battery.jsp for a summary of max. charge voltages for the Samsung cells).
If I am thinking correctly, the same considerations that apply to a 1x 18650 light will also apply to those multi-XM-L lights that use multiple 18650s in PARALLEL (and that do not use any “boost” circuitry). Again, a cell that holds better voltage will maintain better lumen output over the course of a cycle.
Anyways, if there is one most important point to note for the person new to lithium ion batteries (other than SAFETY), it is probably the one I mentioned up front. Knowing these details can be helpful for optimizing cell choice for particular needs. But these modern cells, if used appropriately and treated well, will all provide good service, so it is important to not get TOO worried about the details. Before finding out this information, I DID order a couple of 3100 Panasonics for my PD32UE. They are not “ideal” for output maintenance, but they work just FINE for all practical purposes . . .
I wound up digging around and scratching my head a fair bit in discovering this information (although that is attributable to what may be called “slow uptake”), so I thought I’d throw it out to help possibly short-cut the process for others in the same boat.
Thanks for the forum, I’ve learned (and am continuing to learn) a lot.
Tim
