DX  » Community » Forums
«See all threads (in Product Forum)
Product Forum

Capacity test and comparison graph

  • sheepish Top 10 Forum Poster Sunday, June 8, 2014 12:56 PM Reply
    What a nice discharge curve these Sanyo cells have!

    img src: http://i58.tinypic.com/2na8hex.gif

    The graph compares the latest cells of each type I have tested, along with approximate delivery date. Since then the Trustfire 2400 has changed at least its label and the Sanyo 2600 has changed to having its protection circuit on its positive end.

    Sanyo 2600 cells, sku.121447.
    Trustfire 3000 cells, sku.120476.
    Trustfire 2400 cells, sku.20392.

    For single cell lights run at high current, since a Cree XM-L is rated to need about 3.35V at 3A, the capacity down to around 3.4V gives an idea of when the torch will drop out of regulation or drop to a lower mode. The Sanyo cells are clearly ideal for that situation with about 2310mAh/8400mWh, while the Trustfire cells have only about 760mAh/2690mWh (TF3000) and 1000mAh/3630mWh (TF2400).

    Down to 3V, the Sanyo 2600 cells have 13% higher capacity and 22% higher energy than the Trustfire 3000. For about half the time the Sanyo cells are holding 0.3V or more higher voltage than the Trustfire 3000 cells.
    Sanyo 2600 cells; capacity 2448 mAh, energy 8836 mWh.
    Trustfire 3000 cells; capacity 2175 mAh, energy 7245 mWh.
    Trustfire 2400 cells; capacity 2052 mAh, energy 7058 mWh.

    What useful capacity the Trustfire cells have will depend on how they are being used. Testing down to 2.55V (or whatever) is interesting for seeing the shape of the discharge curve but the single cell lights I have will start complaining (flashing) long before that and will give next to no light at that low a voltage. It's just not usable capacity.

    Charger used for the Sanyo cells was sku.248026 (sku.289179 is version 2 of this charger, and does 7.2V/8.4V batteries as well). Voltage before discharge was 4.18V (cell A) and 4.16V (cell B), so they could have been a little more fully charged. I'm not going to worry about the difference, since this is how I'll be charging them in the real world; they will still perform extremely well. Discharge started about quarter of an hour after charging finished.

    Previous graph: http://club.dx.com/forums/forums.dx/threadid.1042763?page=19#1211807.

    Edit - took out a redundant "down to 3V".
    Remembering 30 years.
    Posts(24949) | Reviews | Tip post

    post edited by sheepish on 6/8/2014 at 1:00 PM
  • gasbag11 Top 10 Forum Poster Sunday, June 8, 2014 7:26 PM Reply
    Thanks sheepish. Its nice to know that these are still good quality cells. I have to admit I had my doubts when they lowered the price to $8.36 per pair.
    NO! I don't know why DX screwed up the site so badly.
    Ask them.
    Posts(12568) | Reviews | Tip post

  • sheepish Top 10 Forum Poster Sunday, June 8, 2014 7:43 PM Reply
    Thank you again for letting me know they were available at that great price. The difference in voltage is even more than I was expecting, and because I was using a constant resistance for the discharge, the Sanyo cells were delivering more current and still holding a significantly higher voltage. They really perform.
    Remembering 30 years.
    Posts(24949) | Reviews | Tip post

  • gasbag11 Top 10 Forum Poster Sunday, June 8, 2014 7:59 PM Reply
    No problem. :)

    The ironic thing is, now that its too late, I'm sure everyone wishes they had ordered more. :(
    NO! I don't know why DX screwed up the site so badly.
    Ask them.
    Posts(12568) | Reviews | Tip post

  • lagman67 Monday, June 9, 2014 12:14 AM Reply
    You should always include the discharge current in your graphs, as it is an important data.
    Thanks a lot for your review/test. These batteries are impressive...
    How many did you buy? Are all of them able to deliver ~5A? I'm trying to determine if I was very unlucky or if the QC is not good because one of my cell had a defective protection chip. In the end I had to remove that chip as it caused self discharge! read here.
    Posts(99) | Reviews | Tip post

  • sheepish Top 10 Forum Poster Monday, June 9, 2014 8:17 AM Reply
    The test used a constant resistance of 3.3 ohms (measured), with manual recording of voltage under load. Hence the current for these cells started at 1.2A, and at 3.3V was 1.0A (V=IR). Energizer calls 3.3 ohms an industry standard test for portable lighting (ignoring duty cycle, which really doesn't have much affect on modern rechargeable technology). It's a good approximation of an incandescent torch bulb, so it could be argued that it's not as relevant now.* Main advantage is that it's easy to do and it results in a Goldilocks test - it doesn't take too long but still produces a respectable current. It tends to even out differences between cells rather than emphasise them because better batteries produce higher voltage, therefore higher current, therefore slightly reduced capacity.

    Because of testing inaccuracies I suggest exercising some caution comparing results between any two different testing regimes anyway. The results on this graph should be pretty consistent with each other, but there are still differences. For example, the Sanyo cells had a different charger.

    I bought 4 of them. How do you suggest I test for >5A? I do not have any 2/3 ohm, 24W resistors. I note a certain other site sells Sanyo UR18650RX 2000mAh Li-ion cells, said to be good for a 22A continuous discharge.

    * It does allow new results to be directly compared with tests from decades ago, which is nice. For a more appropriate modern testing regime, I guess a constant current test would be best for assessing the suitability of cells for single cell LED torches, many of which have AMC7135 drivers, and a constant power test for cells' suitability for multiple cell LED torches, which typically accept a wide input voltage and have a constant current output. The latter test would ignore driver efficiency changes.
    Remembering 30 years.
    Posts(24949) | Reviews | Tip post

  • lagman67 Monday, June 9, 2014 6:46 PM Reply
    About the 5A: I was just asking if they *can* deliver that many amps without the overcurrent triggering. So a simple test is to place the cell in a power hungry flashlight and see if it works. On of my cells triggers at 2.5A...
    If you plan on making many graphs, instead of doing it manually which is quite tiresome, you could get a DC electronic load. Or cheaper (and less accurate) get a hobby charger with PC data logging like a icharger 206B.
    Posts(99) | Reviews | Tip post

  • sheepish Top 10 Forum Poster Monday, June 9, 2014 9:35 PM Reply
    http://lygte-info.dk/info/Batteries2012Info%20UK.html has a wealth of information and tests. Unlike him I don't actually test many batteries, so I'm not even thinking of data loggers. A good bench power supply would be nice, but the one he pictures is US$654 plus shipping. Not going to happen any time soon. The DC electronic load he uses has a manufacturer's suggested retail price of US$1095. I appreciate the suggestion, but did you know they cost that much?

    All 4 cells work fine in a nominally 2.8A torch, a self-assembled Ultrafire C2; 2.66A - 2.72A measured. They are a bit long for it, though, at a full 70mm. I can tighten it up if I really try hard, but I'll probably dig out another glow ring tomorrow to fill the gap.
    Remembering 30 years.
    Posts(24949) | Reviews | Tip post

  • lagman67 Monday, June 9, 2014 9:45 PM Reply
    No I didn't knew his devices are that expensive. but then again, you can get an Icharger for about 100-150USD.
    That was just a suggestion...
    My cells, which have the PCB on the positive side are only 68mm long. Review here: http://budgetlightforum.com/node/31393#comment-582665
    Posts(99) | Reviews | Tip post

    post edited by lagman67 on 6/9/2014 at 9:47 PM
  • sheepish Top 10 Forum Poster Monday, June 23, 2014 6:36 AM Reply
    I used a pair of these Sony 2600mAh cells in a bike light over the weekend. (No, I wasn't biking.) They lasted longer than I expected, and significantly longer than any of the Trustfire cells have, but because they hold their voltage so well I didn't notice any low voltage warning before they died. With the Trustfire cells there is an extended period where their voltage is low enough to make the light's battery indicator show cyan or red. (Indeed, it's sometimes not long after starting to use them that the indicator shows cyan.) The Sanyo cells were still showing as green the last time I checked, not too long before they suddenly died as the protection circuit kicked in. These cells have great performance.
    Remembering 30 years.
    Posts(24949) | Reviews | Tip post

Page 1 of 2
Go to Page:
«Reply to this thread (in Product Forum)
Product Forum



* 50 characters max


Please note: HTML codes are not allowed anywhere on this page (otherwise you will see an error).

Please note that DealExtreme Forums are not a sales or product support board. While we do constantly participate in this forum, please contact us via support ticket for a guaranteed fast response. We make every effort to make the quickest replies.

DX Everywhere