UPDATE May 16, 2013: The CR123A testing results in this thread have been superseded by a new 2013 comparison posted here. Please refer to that thread for more up to date info.
I thought I would provide a quick overview comparison of some of the popular battery brands in the CR123A and AA formats.
This comparison is not meant to compete with SilverFox's outstanding 1xCR123A Shoot Out thread. I don't have the equipment or anywhere near the range of experience that Tom has on the topic of batteries. :bow: This thread is just meant to give you a quick overview of my recent experience with a few brands.
For this comparison, I have chosen to use the Olight T10/T15 on max output. In my case, I am using the exact same head for these tests, just on the 1xCR123A (T10) or 1xAA (T15) battery tube. I have chosen this light as it has very flat regulation on a full cell, followed by a rather quirky step-down pattern as the light falls out of regulation (i.e. I thought it make for an interesting visual comparison).
UPDATE JUNE 30, 2009: For additional comparisons of a limited number of popular batteries in the Quark 123, LiteFlux LF3XT, and Nitecore EX10, please see my: CR123A Comparison Review: 4Sevens, Titanium Innovations, Tenergy, Surefire, Duracell
All batteries were purchased directly from Battery Junction, Battery Station, CFR lights, Lighthound, 4sevens, or my local Surefire distributor.
CR123A Shoot Out
To start, here is an overview of various brands of CR123A. In each case, I have used the most recent batch of cells in my possession (i.e. all bought new within the last 1-3 months,tested within one week of each other). Note that I have done multiple runs of batteries from the same lot, and see virtually no difference in runtime patterns (i.e. each lot is highly internally consistent). As such, a representative trace is shown for each battery lot.
DO NOT make a big deal of the variation you are seeing between the Made in the USA brands - in most cases, this is simply due to the variation in date of manufacture (and corresponding loss of capacity). Scroll down for a comparison of the effect of age on cells.
This fact underscores the difficulty in doing comparison battery testing. Depending on turn-over, the actual age of cells purchased can vary significantly - even from major distributors. As you can see above, for cells purchased in the last 1-3 months, I had manufacture dates that ranged from 4-17 months old! :huh: Typically, most cells are ~6-12 months old at the time of purchase, with a median age of ~9 months.
China Brands
The main take-home message from the graph above is that the Tenergy-brand cells were considerable under-performers compared to the USA brands, and the Titanium Innovations were surprisingly good performers compared to the USA brands.
Unfortunately, the China brands do not give manufacture codes, so I don't know the exact age of these cells (both bought new for this testing). But in my experience, Powerizer brand cells perform very similarly to the Tenergy cells shown above. :sigh:
UPDATE June 17, 2009: I have now tested 3 separate lots of Titanium Innovations batteries, and have revised the graph above with the run closest to the "average" performance. I am seeing a bit more variability here than I do with the USA brands, but all three lots perform at least as well as USA #2 (and one lot actually had ~4% better performance than USA #1).
As a general rule, I typically recommend the Made-in-the-USA cells as I have much greater confidence in their consistency and quality control (especially in regards to the presence and performance of PTC protection circuits). As always, please practice safe battery handling and use. :candle: But I am impressed by the performance of the Titanium cells (which also have a good track record here).
Made-in-the-USA Brands
I originally started doing my CR123A runtime testing on Energizer cells, and quickly switched to Duracell and Surefires exclusively once I realized there was a slight performance advantage to the Surefire/Duracell cells - even after controlling for the effect of age.
Based on my tests of various lots over the years, I have come to the general conclusion that the USA cells shown above fall into two general categories based on performance:
USA #1: Duracell, Surefire, Panasonic
USA #2: Energizer, Rayovac, Battery Station
The difference between these groups is fairly small (i.e. <4% in my testing!), but seems to be consistent. Note that I make no claim as to who actually manufactures which cells - simply that based on max output performance in single-cell lights, this is how my data shakes down after controlling for age effects.
From the above graphs, I have measured area-under-curve (AOC, I can't spell) to estimate overall capacity. Note that this is a completely arbitrary and relative measure - I have no way to actually measure mAh. But it gives you some numbers to compare.
UPDATE June 17, 2009: I have revised the table above with the average of the 3 recent lots of Titanium Innovation batteries. Performance ranged from ~5250-5600 AOC in my testing.
Effects of Age
Having bought most of my cells in small batches (to insure I always had fresh cells for testing), I have had to opportunity to compare capacity changes over time. Although I haven't kept samples of too many older lots, I do have enough on hand to give you a quick comparison of age effects in my current T10 test bed.
As you can see, there is a clear pattern of loss of performance/capacity over time (all batteries tested within a week of each other in the above graphs). This is why I cautioned earlier about not reading too much into that first graph showing a lot of variation between USA cells - what you were really looking at was variability in the date of manufacture.
I have more data than shown above, but frankly I don't think I have sufficient data to accurately quantify the average rate of loss of capacity over time. Based on a rough approximation over the first 18 months or so, it appears to average out to ~0.25% per month, or 3% per year. But again, that is just a very rough estimate - I would need a larger data set to state with any degree of confidence.
UPDATE June 30, 2009: As MorePower has pointed out, there is potential confound here. Because these runs are done with batteries from different lots manufactured at different times, it is possible that some of the "age effects" I'm seeing are in fact due to process improvements in manufacturing over time, leading to greater capacity on newer cells. To determine that, one would need to test the same lot at varying times after manufacture in a fully calibrated setup. Unfortunately, neither a typical flashlight (where the emitter can suffer age effects) nor my home-made lightbox (which is not calibrated) would allow me to do that.
Preliminary CR123A Conclusion:
I don't think the ~4% difference between USA brands matters a great deal for most users, nor does the potential age/manufacturing difference of ~4.5% loss at up to 18 months. I have stuck with a high turnover of Duracell/Surefire lots to insure max performance for comparison runtime testing, but that's hardly necessary for the average user.
Personally, I would just stick with whatever USA cells you can get the best deal on, but don't buy too many cells in advance of your need. And again, take all these numbers with a big grain of salt, as they based solely on single-cell flashlight tests.
I am impressed by the Titanium Innovations performance (made in China cells) - I had one batch that consistently outperformed even the USA#1 group. I have also placed an order for 4sevens new in-house brand (also made in China), and will add those results to this mini-review.
Otherwise, I personally prefer to pass on the other China-made cells due to their lower capacity (and potential QC issues, especially in regards to safety and functioning PTC circuits).
For more information of the differences between individual brands/cells under different loads, I suggest you check out SilverFox's seminal 1xCR123A Shoot Out thread. I can only hold a :candle: to the master.
UPDATE June 26, 2009: I've done some comparison testing of the 4Sevens CR123A cells, and the results are interesting. I've posted this analysis in its own mini-review here in the batteries sub-forum:
CR123A Comparison Review: 4Sevens, Titanium Innovations, Tenergy, Surefire, Duracell
1AA Shoot Out
As expected, Energizer e2 Lithium (L91) is a MUCH better choice than alkaline for running the T15 on Max. Note that the Powerizer brand version of the L91 had lower total capacity and reduced max output. Again, I recommend you stick with the Energizer cells.
For alkaline, note that there was absolutely no difference between plain Duracell alkalines and Duracell Ultra alkalines. :shrug: I'd stick with the lower cost ones.
Of interest to me was the poor performance of the Duracell nickle oxyhydroxide NiOx cells (branded under the "PowerPIX" label). Duracell markets these as a competitor for Energizer lithiums in digital cameras, claiming "up to 2X more digital pictures" over standard alkalines. Yet in terms of total capacity, my T15 test shows they actually have lower AOC performance. :thinking:
So what gives? I suspect Duracell is referring to the fact that alkalines can quickly get exhausted under a high load (e.g. rapidly taking a number of digital pics, especially with a flash). The NiOx cells do actually give greater initial output on the T15, so I'm guessing they are better able to keep snapping pics when placed under this sort of rapid high-drain load. But I think we have something a "tortoise and a hare" issue here - the NiOX may be faster out of the gate, but plain old alkaline seems to win at the endurance race (at least based on the simple T15 max output test here).
FYI, this is consistent with the Duracell FAQ on NiOx, which points out that the NiOx chemistry is designed to release more power in high drain applications such as digital cameras, but that there is no performance advantage versus regular or premium alkaline in other devices. For flashlights, they specifically mention that "Flashlights may burn brighter, and motors may run faster but for shorter durations than with alkaline because of the higher operating voltage of PowerPix." In other words, skip 'em.
Preliminary AA Conclusion:
For max output on flashlights, I think the choice is clear - stick with Energizer lithiums. :thumbsup: But standard alkalines are still fine for lower output modes. I'm just not too sure if there is any real advantage of "premium" alkalines over the plain jane ones.
UPDATE July 9, 2009: I received an interesting question in my other CR123A shoot-out thread. Apparently, there is some concern that CR123A auto-discharge rates may accelerate after even the briefest use. To test this this idea, I compared two Energizer cells from the same batch (manufactured March 2007). One was left untouched in original packaging for two years ("Unused"), and one has been one of my "tester" cells for the last one year or so ("Lightly Used"). As a tester, I have been using it to quickly test the max output/throw on new lights as they arrive (i.e. click on Max, take a reading, turn off). I would estimate that I have used up the equivalent of ~1.5mins of continuous runtime, scattered over a one year period.
As you can see, there is no difference in time to 50%, and only a slight reduction in total capacity as measured by area under the curve. Looks pretty convincing to me that Energizers don't suffer any ill effects from "breaking the seal". :twothumbs
UPDATE May 16, 2013: The CR123A testing results in this thread have been superseded by a new 2013 comparison posted here. Please refer to that thread for more up to date info.
I thought I would provide a quick overview comparison of some of the popular battery brands in the CR123A and AA formats.
This comparison is not meant to compete with SilverFox's outstanding 1xCR123A Shoot Out thread. I don't have the equipment or anywhere near the range of experience that Tom has on the topic of batteries. :bow: This thread is just meant to give you a quick overview of my recent experience with a few brands.
For this comparison, I have chosen to use the Olight T10/T15 on max output. In my case, I am using the exact same head for these tests, just on the 1xCR123A (T10) or 1xAA (T15) battery tube. I have chosen this light as it has very flat regulation on a full cell, followed by a rather quirky step-down pattern as the light falls out of regulation (i.e. I thought it make for an interesting visual comparison).
UPDATE JUNE 30, 2009: For additional comparisons of a limited number of popular batteries in the Quark 123, LiteFlux LF3XT, and Nitecore EX10, please see my: CR123A Comparison Review: 4Sevens, Titanium Innovations, Tenergy, Surefire, Duracell
All batteries were purchased directly from Battery Junction, Battery Station, CFR lights, Lighthound, 4sevens, or my local Surefire distributor.
CR123A Shoot Out
To start, here is an overview of various brands of CR123A. In each case, I have used the most recent batch of cells in my possession (i.e. all bought new within the last 1-3 months,tested within one week of each other). Note that I have done multiple runs of batteries from the same lot, and see virtually no difference in runtime patterns (i.e. each lot is highly internally consistent). As such, a representative trace is shown for each battery lot.
DO NOT make a big deal of the variation you are seeing between the Made in the USA brands - in most cases, this is simply due to the variation in date of manufacture (and corresponding loss of capacity). Scroll down for a comparison of the effect of age on cells. This fact underscores the difficulty in doing comparison battery testing. Depending on turn-over, the actual age of cells purchased can vary significantly - even from major distributors. As you can see above, for cells purchased in the last 1-3 months, I had manufacture dates that ranged from 4-17 months old! :huh: Typically, most cells are ~6-12 months old at the time of purchase, with a median age of ~9 months.
China Brands
The main take-home message from the graph above is that the Tenergy-brand cells were considerable under-performers compared to the USA brands, and the Titanium Innovations were surprisingly good performers compared to the USA brands.
Unfortunately, the China brands do not give manufacture codes, so I don't know the exact age of these cells (both bought new for this testing). But in my experience, Powerizer brand cells perform very similarly to the Tenergy cells shown above. :sigh:
UPDATE June 17, 2009: I have now tested 3 separate lots of Titanium Innovations batteries, and have revised the graph above with the run closest to the "average" performance. I am seeing a bit more variability here than I do with the USA brands, but all three lots perform at least as well as USA #2 (and one lot actually had ~4% better performance than USA #1).
As a general rule, I typically recommend the Made-in-the-USA cells as I have much greater confidence in their consistency and quality control (especially in regards to the presence and performance of PTC protection circuits). As always, please practice safe battery handling and use. :candle: But I am impressed by the performance of the Titanium cells (which also have a good track record here).
Made-in-the-USA Brands
I originally started doing my CR123A runtime testing on Energizer cells, and quickly switched to Duracell and Surefires exclusively once I realized there was a slight performance advantage to the Surefire/Duracell cells - even after controlling for the effect of age.
Based on my tests of various lots over the years, I have come to the general conclusion that the USA cells shown above fall into two general categories based on performance:
USA #1: Duracell, Surefire, Panasonic
USA #2: Energizer, Rayovac, Battery Station
The difference between these groups is fairly small (i.e. <4% in my testing!), but seems to be consistent. Note that I make no claim as to who actually manufactures which cells - simply that based on max output performance in single-cell lights, this is how my data shakes down after controlling for age effects.
From the above graphs, I have measured area-under-curve (AOC, I can't spell
) to estimate overall capacity. Note that this is a completely arbitrary and relative measure - I have no way to actually measure mAh. But it gives you some numbers to compare.
UPDATE June 17, 2009: I have revised the table above with the average of the 3 recent lots of Titanium Innovation batteries. Performance ranged from ~5250-5600 AOC in my testing.
Effects of Age
Having bought most of my cells in small batches (to insure I always had fresh cells for testing), I have had to opportunity to compare capacity changes over time. Although I haven't kept samples of too many older lots, I do have enough on hand to give you a quick comparison of age effects in my current T10 test bed.
As you can see, there is a clear pattern of loss of performance/capacity over time (all batteries tested within a week of each other in the above graphs). This is why I cautioned earlier about not reading too much into that first graph showing a lot of variation between USA cells - what you were really looking at was variability in the date of manufacture.
I have more data than shown above, but frankly I don't think I have sufficient data to accurately quantify the average rate of loss of capacity over time. Based on a rough approximation over the first 18 months or so, it appears to average out to ~0.25% per month, or 3% per year. But again, that is just a very rough estimate - I would need a larger data set to state with any degree of confidence.
UPDATE June 30, 2009: As MorePower has pointed out, there is potential confound here. Because these runs are done with batteries from different lots manufactured at different times, it is possible that some of the "age effects" I'm seeing are in fact due to process improvements in manufacturing over time, leading to greater capacity on newer cells. To determine that, one would need to test the same lot at varying times after manufacture in a fully calibrated setup. Unfortunately, neither a typical flashlight (where the emitter can suffer age effects) nor my home-made lightbox (which is not calibrated) would allow me to do that.
Preliminary CR123A Conclusion:
I don't think the ~4% difference between USA brands matters a great deal for most users, nor does the potential age/manufacturing difference of ~4.5% loss at up to 18 months. I have stuck with a high turnover of Duracell/Surefire lots to insure max performance for comparison runtime testing, but that's hardly necessary for the average user.
Personally, I would just stick with whatever USA cells you can get the best deal on, but don't buy too many cells in advance of your need. And again, take all these numbers with a big grain of salt, as they based solely on single-cell flashlight tests.
I am impressed by the Titanium Innovations performance (made in China cells) - I had one batch that consistently outperformed even the USA#1 group. I have also placed an order for 4sevens new in-house brand (also made in China), and will add those results to this mini-review.
Otherwise, I personally prefer to pass on the other China-made cells due to their lower capacity (and potential QC issues, especially in regards to safety and functioning PTC circuits). For more information of the differences between individual brands/cells under different loads, I suggest you check out SilverFox's seminal 1xCR123A Shoot Out thread. I can only hold a :candle: to the master.
UPDATE June 26, 2009: I've done some comparison testing of the 4Sevens CR123A cells, and the results are interesting. I've posted this analysis in its own mini-review here in the batteries sub-forum:
CR123A Comparison Review: 4Sevens, Titanium Innovations, Tenergy, Surefire, Duracell
1AA Shoot Out
As expected, Energizer e2 Lithium (L91) is a MUCH better choice than alkaline for running the T15 on Max. Note that the Powerizer brand version of the L91 had lower total capacity and reduced max output. Again, I recommend you stick with the Energizer cells.
For alkaline, note that there was absolutely no difference between plain Duracell alkalines and Duracell Ultra alkalines. :shrug: I'd stick with the lower cost ones.
Of interest to me was the poor performance of the Duracell nickle oxyhydroxide NiOx cells (branded under the "PowerPIX" label). Duracell markets these as a competitor for Energizer lithiums in digital cameras, claiming "up to 2X more digital pictures" over standard alkalines. Yet in terms of total capacity, my T15 test shows they actually have lower AOC performance. :thinking:
So what gives? I suspect Duracell is referring to the fact that alkalines can quickly get exhausted under a high load (e.g. rapidly taking a number of digital pics, especially with a flash). The NiOx cells do actually give greater initial output on the T15, so I'm guessing they are better able to keep snapping pics when placed under this sort of rapid high-drain load. But I think we have something a "tortoise and a hare" issue here - the NiOX may be faster out of the gate, but plain old alkaline seems to win at the endurance race (at least based on the simple T15 max output test here).
FYI, this is consistent with the Duracell FAQ on NiOx, which points out that the NiOx chemistry is designed to release more power in high drain applications such as digital cameras, but that there is no performance advantage versus regular or premium alkaline in other devices. For flashlights, they specifically mention that "Flashlights may burn brighter, and motors may run faster but for shorter durations than with alkaline because of the higher operating voltage of PowerPix." In other words, skip 'em.
Preliminary AA Conclusion:
For max output on flashlights, I think the choice is clear - stick with Energizer lithiums. :thumbsup: But standard alkalines are still fine for lower output modes. I'm just not too sure if there is any real advantage of "premium" alkalines over the plain jane ones.
UPDATE July 9, 2009: I received an interesting question in my other CR123A shoot-out thread. Apparently, there is some concern that CR123A auto-discharge rates may accelerate after even the briefest use. To test this this idea, I compared two Energizer cells from the same batch (manufactured March 2007). One was left untouched in original packaging for two years ("Unused"), and one has been one of my "tester" cells for the last one year or so ("Lightly Used"). As a tester, I have been using it to quickly test the max output/throw on new lights as they arrive (i.e. click on Max, take a reading, turn off). I would estimate that I have used up the equivalent of ~1.5mins of continuous runtime, scattered over a one year period.
As you can see, there is no difference in time to 50%, and only a slight reduction in total capacity as measured by area under the curve. Looks pretty convincing to me that Energizers don't suffer any ill effects from "breaking the seal". :twothumbs
UPDATE May 16, 2013: The CR123A testing results in this thread have been superseded by a new 2013 comparison posted here. Please refer to that thread for more up to date info.
Last edited:
