This post by AW shows
17670 discharge curves. It looks to me that the AW cells show comparable performance. The Sanyo cells may have a small advantage in capacity, but it's not 2100mAh vs 1600mAh. The AW cell may hold voltage under load a little better.
Thank you for digging out that discharge graph! That is what I was looking for but could not find. Yes, judging from the graph, it looks that under a one amp load (the bulb draws 800mA) they are quite close.
I have read somewhere (i.e. this is purely hear-say, or read-say), that the 16650s on the market are actually pretty old stock, and as such may have lower than stated capacity.
http://www.candlepowerforums.com/vb/showthread.php?329628-Test-Review-of-Intl-outdoor-Sanyo-16650-2100mAh-(Red)&p=3829744&viewfull=1#post3829744
Yes, Jason did mention that in another thread. However, I prefer to look strictly at what the discharge curves indicate, and as Justin Case states, there is very little difference in terms of real world capacity, though purely by physical dimensions, the Sanyo is certainly a bit slimmer, lending it an advantage in fitting in bodies not specifically designed for the 17670 form factor, say, a popular flashlight brand focused mainly on CR123A cells.
Why aren't 16650's more popular? They seem like a perfect replacement, especially for E-series with their smaller inner diameters. Also the Sanyos state 2200mAh, which is more than AW's 1600.
16650 cells qua "16650 cells" aren't more popular in a literal sense because when one states "16650 cells" it is not as if one is stating "AA" battery, which has as its referent a widely known battery type that is not restricted to any one manufacturer or dealer. To the contrary, when one states "16650 cells" one is referring almost implicitly to the cells made by Sanyo for a highly specific application sold by one particular dealer. As others have pointed out in countless other threads, by far most of the R&D is in the 18650 form factor for portable electronic/computing devices: us flashoholics are just riding the wave of far more popular, and far more profitable enterprises.
They would be more popular if new fresh cells were more available. Currently the only 16650's floating around on the China market (that I am aware of) are about 2 year old reject cells from a pre-production batch. Secondly, to get the full capacity out of the cell, you would need a special charger that charges at 4.30v.
This is the one thing that concerns me. I know you stated previously that you evaluated these cells for the specific purpose of seeing if they were worthy of stocking. If the cells truly are that old, then the main concern is a building internal resistance. While I am no Silver Fox, correct me if I am wrong, but as lithium cobalt cells age, they suffer an increase in internal resistance, which is no flashoholic's friend.
Therefore, the chief concern with these cells is perhaps one of safety, and not one of ultimate capacity.
In my opinion, the nature of this thread here splits into two separate topics, which, though certainly highly intertwined, are not, in fact, one and the same. The first question revolves around theory, while the second revolves around more practical matters:
1) The first question is, "How do AW 17670 cells compare to Sanyo 16650 cells?"
The answer to this question has two smaller components, namely:
a) At 4.20 volts charge, the Sanyo has a smaller advantage in capacity. However, because the cell is not being charged fully to 4.30 volts, it should have a much higher cycle life.
b) Charged to 4.30 volts, the cell clearly has a higher capacity.
2) The second question is, "Since there is only one place to get these cells currently, the question almost automatically reverts to one of 'Is the current crop of available cells worth purchasing from, if, in fact, it is true that they are several years old already, and not fresh?' "
The answer to this is less clear. The cells reviewed by HKJ have good capacity, including under load. However, a contributing member who buys and sells many batteries has reviewed these cells, and has come to the belief that given that they are several years old already, they will have higher internal resistance.
For many of us, the chief contributing factor in a given purchase is cost. There is no doubt that the current crop of available Sanyo 16650 cells are cheaper than the similarly sized AW 17670 cells. In their favor, they are also smaller, meaning they are perhaps more versatile, able to fit in a larger variety of flashlight tubes. They are available both protected and unprotected.
I am not an expert in the mechanics of cell construction. I do know that AW cells are widely touted as having many safety features.
I work neither for the sole dealer of Sanyo 16650 cells nor do I sell AW cells nor am I affiliated with the entity who evaluated Sanyo 16650 for potential relabeling and resale. In the vernacular, I do not have a horse in this race.
For a long time, I have wanted a quality incandescent, and loving the smaller size of the Surefire E2E, traded for one last year. I purchased a high-CRI dropin for it, and though I found the workmanship admirable, realized I would really like to have an incandescent to hike with. I have a modded Shiningbeam L-Mini II with XP-E (unknown efficiency bin) 5C tint that is my favorite LED light for the outdoors, and I wanted a small incandescent to complement it.
The salient question driving this thread is "Which will give me the longest, brightest, most consistent runtime with the Lumens Factory HO-E1R dropin, the AW 17670 cell, or the Sanyo 16650 cell?" Above, I teased out that though the question can be interpreted to mean the Sanyo cell in a theoretically brand new, factory fresh state, the question almost automatically becomes "Which will give me the longest, brightest, most consistent runtime with the Lumens Factory HO-E1R dropin, a fresh AW 17670 cell, or a Sanyo 16650 cell, available from one place only, and apparently a couple years old?"
The answer seems to be that strictly in terms of available capacity at an 800mA load there is little difference between the two. The key factors swaying one's decision, therefore, come down to cost, availability, physical size, and potential safety concerns directly related to the fact that these cells are a couple years old. This is not a criticism of the cell size or Sanyo brand, simply a matter of dealing with aging lithium cobalt cells.
Given the information as presented, I do not see it as my position to prescribe to the community which cell one should choose, since in terms of capacity alone there appears to be little difference. Which cell you choose, then, will be based on other factors, and whether those factors are significant to you.
At some point in the future, I would love to see a comparison of a straight-from-the-factory Sanyo 16650 and a straight-from-the-factory AW 17670 cell. However, until that occurs, I feel that there is little more to discuss now that a discharge curve for the AW 17670 has been disinterred, and we can see from the perspective of capacity under the relevant 800mA load of the Lumens Factory HO-E1R cell, there is little difference between the two cells.
—and with plugged sinuses, a drippy nose, and a clogged left ear, I'm calling it a night.