Torchy had a test of these on his website, they were close to the AW IMR 2000mAh.. so cant really be good performers. Maybe for the first few seconds..
But for the XML2 it seems 4.35V cells seem to be the best.
I'm really not sure how to correlate these graph results to actual usage on XM-L2/SinkPAD setups (yes, they are all mod'ed but they are still flashlights and XM-L2's are already in production lights, maybe SinkPAD's some day). Also, how do you determine internal resistance of a battery based on these graph results? QWell, one batt may be better for the frst 500 mah, another past that point -- very confusing, which is really better? I can only go by the first couple of minutes because that's how I and most others test lumens/throw - fresh off the charger, 30 seconds, then throw.
SinkPAD's, I believe, change everything, more so than the XM-L2 - can't compare to a high draw stock light like the HD 2010 because the voltage demand (Vf) is different.
NightCrawl - Judging by the charts, there is only one battery tested that is 4.35v (LG 18650 D1 3000mAh), and it's results are mixed, not across the board better - first 0.5 Ah, the Samsung is much better, then the LG is a little better after that. Are there other 4.35v options, maybe better? What do you need to charge them?
Internal resistance is calculated via the voltage drop between two different loads. I dont know exactly tho. To calculate it, you subtract the two voltage readings and divide them by the difference of the load. (V1-V2)/(L2-L1).
Well, the Samsung is better for the first 0.5Ah at 5A.. but lets assume a normally high driven light at 3-3.5A. LG would have an advantage there. You basically need to look at the curve until 3.7V (should be the Vf of XML2 at 3A).
There are other 4.35V cells from Sanyo and Samsung, but I doubt they would be better than the LG. To charge them you need a hobby charger (4.3V) or a cottonpicker (4.35V).
Thanks HKJ! Stupid me - didn't see you actually list the calculated internal resistance for each battery tested on that last line! The batteries I'm looking at are the Pana PD's, Pana CH's, Samsung INR's, Sanyo UR18650FM 2600, - think the AW IMR 2000's are out of the running. Like the tests you did for the AW IMR's, the 1600 (0.05 ohms) were lower resistance than the 2000's (0.09 ohms) - think the same is going on with the Samsung INR 1500's vs 2000's.
But of course I wish you had test results on the Pana PD 2900's and Pana CH 2250's, then could really see what's going on in this current field (still reasonably prices batts somewhat). The Pana PD's are of course a great option with their apparent low resistance, 2900 mAh runtimes, and relative low cost ($16/pair at FT).
The Samsung 1500's I have are INR, not IMR. Bought from FastTech "as is", obviously pulled from a pack because of fastening marks that I filed down.
Here's test results on a mod'ed XinTD C8 - XM-L2/U2, SinkPAD, 3.85A Nanjg driver, extra copper, 22 gauge silver tinned teflon coated wire, sanded/polished mating surfaces on the SinkPAD and top of pill. In order of results, 12 tests performed, all batteries are unprotected, freshly charged, first # is @start, 2nd # is @30 secs, in lumens:
Sanyo 2600 #1, 4.20v, 3.74A: 1190 - 1135
Pana 3400 #1, 4.19v, 3.56A: 1156 - 1111
Sam. 1500 #1, 4.19v, 3.85A: 1264 - 1213
Pana 2900 #1, 4.18v, 3.44A: 1128 - 1088
AW IMR 2000, 4.18v, 3.84A: 1203 - 1145
Sam. 2000 #1, 4.20v, 3.85A: 1258 - 1230
Pana PD #1, 4.19v, 3.85A: 1230 - 1183
Sanyo 2600 #2, 4.20v, 3.77A: 1190 - 1135
Sam. 1500 #2, 4.20v, 3.85A: 1237 - 1186
Pana 3400 #2, 4.18v, 3.54A: 1139 - 1094
Sam. 2000 #2, 4.20v, 3.85A: 1254 - 1227
Pana PD #2, 4.21v, 3.85A: 1244 - 1183
The "#1" and "#2" designates unique batteries because some tests were repeated on the same make/model. All tests had 3-5 minutes in between, accept the last which had a gap of 17 mins.
It does look like the winner is the Samsung INR 2000, better than the Samsung INR 1500 and Panasonic PD's. Of course this is only for the first 30 seconds...
So NightCrawl, I do believe you are correct here about the 1500 not being an improvement over the 2000. I think earlier tests I did contradicted that, but these tests were done now with more control (voltage and amp measurements, all batts off the charger, etc.).
Oh, thanks for the Pana CH and PD refs daveam and NightCrawl - looks like this Samsung INR 2000 beats those batteries as well when comparing the curves for 5 amps on: http://lygte-info.dk/review/batteries2012/Common18650comparator.php. It also beats the AW IMR 1600 and 2000's, wow! Wonder if there's anything out there better than this battery right now at any price...
I still like the PD's though for the greater runtime at 2900 mAh, with the understanding you are losing a little on the output. They also cost about the same. Think I'll be quoting measurements taken on the Samsungs while using the PD's .
INR, IMR.. you know I meant high current batteries. ;) All these names are moot imo.
My guess is that the 15R and 20R are pretty similar, just with less capacity. That goes pretty well with your readings because the 15R sags more than the 20R.
Oh btw, the Sanyo UR18650RX might beat the crap out of all the other cells but its got a MOQ of 8 for 8$/each on alibaba..
Yes - really, really close to the 20R in performance at high amps on single cell XM-L2 builds using 7135 drivers at 3.85A and higher. In some lights, I've seen no difference at all. I don't have a scientific way of comparing, but the single cell XM-L2 high amp lights will be an excellent indicator. I've gotten in a bunch of 20Q's now, latest buys from FastTech, best price of course.
HJK, thanks for all you do. I often refer to your cell discharge comparator before making my purchase decisions. I just have to laugh at some of the answers people are given when asked which is the best cell to buy… thanks to you, the choice is more obvious. Please keep up the good work. A lot of people will echo my sentiments.