Suggestion on 26650 battery for low current draw & long lasting

24 posts / 0 new
Last post
nexaen
Offline
Last seen: 9 months 1 week ago
Joined: 01/15/2018 - 07:46
Posts: 6
Suggestion on 26650 battery for low current draw & long lasting

Hello,
I would like to have your suggestions on battery for the following project.
I’m planning to run a micro-controller and a sensor on battery for a long period of time (more than a year). I’ve set microcontroller to go into sleep mode and wake up for 6 seconds every 15 minutes to power on the sensor and read a value and go back again to sleep mode till the next round. All the project is fed through one power source (a battery pack that I’m using, more explanation below).
In Sleep mode, total current draw is at 1.15mA (measured for whole the electronics including sensors and …), and while it wakes up it is at 12mA.
I’m using a voltage regulator to feed the project with 5V, voltage regulators usually are in need of at around 1V higher than output. (to summarize this sentence, I mean that minimum voltage for my electronics should be more than 6 V)

I’m thinking to use 26650 KEEPPOWER 5200mAH (protected) in 2S3P configuration, which is 6 batteries with nominal voltage of 7.4V and capacity of 15300mAH (minimum margin, considering battery might be 5100mAH).
I calculated that, with this configuration, I can get at least 14 months (considering 20% discharge safety).

Now I would like to know your suggestions, would you recommend me this battery? or do you have any other recommendation?
Is 26650 KEEPPOWER 5200mAH (protected) performing well? Is the capacity really 5200mAH? Are there better types of 26650-large capacity batteries? , without considering budget limit?

Thanks in advance
nexaen

Edited by: nexaen on 01/15/2018 - 11:04
mrheosuper
Online
Last seen: 1 min 9 sec ago
Joined: 09/30/2016 - 12:44
Posts: 1225
Location: Ho Chi Minh City, Vietnam

i would recommend 18650 over anything, they have higher energy density
21700 batteries look promising, but quite hard to find

Forgot my pen

nexaen
Offline
Last seen: 9 months 1 week ago
Joined: 01/15/2018 - 07:46
Posts: 6

mrheosuper wrote:
i would recommend 18650 over anything, they have higher energy density
21700 batteries look promising, but quite hard to find

Thanks mrheosuper for reply. I couldn’t find 18650 batteries in large capacities, around 5AH.
By saying “higher energy density”, do you mean compard to battery size and volume? If so, I have no problem in fitting 6× 26650 batteries in the housing of my project.
wstrachan
Offline
Last seen: 3 hours 29 min ago
Joined: 02/19/2016 - 02:30
Posts: 379

The Liitokala 5000 mah cyan and black cells tested extremely well. They deliver far more amperage than you are asking for but the pricing is very good.
https://www.gearbest.com/batteries/pp_663613.html?wid=21&lkid=11244387
there is a code on:
https://couponsfromchina.com/2017/06/19/ultimate-flashlight-coupons-deal...

every once in a while you can find codes from Trevi and Fin for under $10 a set of 2.

light addict....yeah I've got the bug.

Haikelite MT07S
Imalent DT70
Klarus XT12GT
Olight javelin mx3x
thrunite tn32 ut
acebeam k60
many, many others...

Henk4U2
Henk4U2's picture
Offline
Last seen: 4 hours 6 min ago
Joined: 02/13/2014 - 17:52
Posts: 2169
Location: The heart of the Netherlands (GMT+1)

I think you mean a 3P2S configuration. I would go for 18650 cell’s. Which gives 3*3500mAh = 10500 mAh.
If that is not enough (considering the voltage drop while depleting) think about a 4P2S configuration.
That is a lot of cell’s. Ever thought about a solar panel?

I always think long and hard before I say something really stupid.

nexaen
Offline
Last seen: 9 months 1 week ago
Joined: 01/15/2018 - 07:46
Posts: 6

@wstrachan, thank you.

Henk4U2 wrote:
I think you mean a 3P2S configuration. I would go for 18650 cell’s. Which gives 3*3500mAh = 10500 mAh.
If that is not enough (considering the voltage drop while depleting) think about a 4P2S configuration.
That is a lot of cell’s. Ever thought about a solar panel?

Yes 3P2S, thanks, it was a typo, I edited it. 10500mAH would give me around 10 months. Economic-wise, 18650 batteries are usually the same price as 26650 and sometimes even higher (looking on Amazon) and have lower capacity. I can go for 4P2S, but the cost for batteries would increase compared to 3P2S of 26650; and still battery capacity would not be as high as 26650. I’m quite interested to know why 18650 batteries are suggested over 26650? Is it related to their depletion curve (voltage drop)? Does it mean that 26650 cannot withstand such long time?

Yes, thanks for mentioning. I have thought about solar panel already, and I’m going to use it as well. But, for some places that there is no access to sun light, I have to rely on batteries only. For other places, I can use solar panels.

mrheosuper
Online
Last seen: 1 min 9 sec ago
Joined: 09/30/2016 - 12:44
Posts: 1225
Location: Ho Chi Minh City, Vietnam

no, it’s all about energy/volume, the 26650 is 2 times bigger( more than 2 times, since v=r^2*pi*h), but the capacity is only about 1.5 times higher than highest capacity of 18650(3500 mah vs 5500 mah)

Forgot my pen

wstrachan
Offline
Last seen: 3 hours 29 min ago
Joined: 02/19/2016 - 02:30
Posts: 379

I would be very careful buying 26650s from Amazon unless they are name brand (I see Imalent there often) and still no guarantee of authenticity. Most cells advertised there have ridiculous claims of mAh and deliver a very poor product.

light addict....yeah I've got the bug.

Haikelite MT07S
Imalent DT70
Klarus XT12GT
Olight javelin mx3x
thrunite tn32 ut
acebeam k60
many, many others...

Mike C
Mike C's picture
Offline
Last seen: 2 months 3 weeks ago
Joined: 01/22/2014 - 08:03
Posts: 2009
Location: Sweden

As far as I know the Shockli 5500 mAh are the dominating 26650 cells on pretty much all aspects: http://budgetlightforum.com/node/53998

nexaen
Offline
Last seen: 9 months 1 week ago
Joined: 01/15/2018 - 07:46
Posts: 6

@mrheosuper, @wstrachan, @Mike C,
Thank you all.

in 18650 range, which one would you recommend me that has a large capacity and is performing well? Do you recommend Panasonic NCR18650B Protected 3400mAh?

Mike C
Mike C's picture
Offline
Last seen: 2 months 3 weeks ago
Joined: 01/22/2014 - 08:03
Posts: 2009
Location: Sweden

For 18650: Samsung, Sony and Panasonic are all good. Have a look in this list at Mountain Electronics: http://www.mtnelectronics.com/index.php?route=product/category&path=59_88
He doesn’t sell crappy cells, so any 3500 mAh cell he has listed will be a decent one.

nexaen
Offline
Last seen: 9 months 1 week ago
Joined: 01/15/2018 - 07:46
Posts: 6

Mike C wrote:
For 18650 Samsung, Sony and Panasonic are all good. Have a look in this list at Mountain Electronics: http://www.mtnelectronics.com/index.php?route=product/category&path=59_88
He doesn’t sell crappy cells, so any 3500 mAh cell he has listed will be a decent one.

Thanks a lot Mike C, and all other people who replied and helped me. Such an active and great community
Delta_V
Offline
Last seen: 11 hours 20 min ago
Joined: 09/21/2017 - 15:25
Posts: 207

If you are in the US, I’d recommend buying batteries from Illumn, IMR Batteries, or Liionwholesale. Amazon has issues with counterfeits from 3rd party sellers due to their awful inventory mixing (one shady seller uses counterfeits, and they can get mixed in with other sellers’ items) that affects a bunch of items – lithium ion batteries, SD cards, and basically any item that has a standardized form factor. These sellers are more reliable.

18650 batteries are recommended because they take up ~48% of the volume of a 26650, but the best (3500 mAh) have about 64% of the capacity of the best 26650 (5500 mAh). This means you can fit more capacity in the same amount of space.

The Sanyo NCR18650GA, Samsung INR18650-35E, Sony US18650VC7, and LG INR18650MJ1 are all very similar in capacity (~3500 mAh). A “protected” cell adds an extra circuit that protects the cell against over-discharge and a few other things that wouldn’t be relevant in this application. Those circuits are added by another company, such as KeepPower. Like this is the Sanyo NCR18650GA with a protection circuit added and rebranded by KeepPower.

If you want to search for a protected version of those cells, try searching “protected insert model name” without the brand.

Otherwise, if total size isn’t a huge concern, a lower capacity cell like the Samsung INR18650-30Q (3000 mAh) might provide more mAh per dollar spent. But you would need more total cells to reach the same capacity.

edit: Mountain Electronics and Orbtronic are other good sellers.

double edit: If you need the protection circuit, and max current doesn’t seem to be an issue at all for this application, the Protected Panasonic NCR18650B might be the best bang for the buck. Under $7 each (plus shipping) at Liion Wholesale when you factor in the wholesale discount. Looks like the protected NCR18650GA usually goes for at least $8.50.

texas shooter
Offline
Last seen: 14 hours 48 min ago
Joined: 08/26/2012 - 02:14
Posts: 1084
Location: Texas

Samsung, Sanyo and Panasonic in 18650 would be my only choices, Sony lean more to high out put over high capacity. Not only do you have higher energy density, but much better quality. Your cell are going to be a closer match to each other and have probably a much lower self discharge rate.

Delta_V
Offline
Last seen: 11 hours 20 min ago
Joined: 09/21/2017 - 15:25
Posts: 207

Does anyone know the typical parasitic drain caused by the protection circuit of a protected battery? The protection circuit itself is an electronic circuit, and thus is going to place a certain (hopefully incredibly small) drain on the battery, but I haven’t been able to find any measurements of this.

In the flashlight world, we are normally concerned with a drain in the hundred of milliamps to multiple amp range causing a low voltage situation in hours or minutes, thus making this parasitic drain irrelevant. But in this application, where the load is only a few milliamps, a small parasitic drain could be an issue.

nexaen
Offline
Last seen: 9 months 1 week ago
Joined: 01/15/2018 - 07:46
Posts: 6

@Delta_V, @texas shooter,

Thanks you guys for tips.
I’m not in the US, and I’m not sure if those sellers can ship to Denmark. I have to ask them or find a local shop, else I have to look on Amazon.

Btw, Can I use protected batteries with solar cells directly (without using a charging board like TP4056 ), since they have a built-in protection circuit already? Samsung INR18650-35E 3500mAh Protected Button Top has protection for charge/discharge cutoff voltage.

Delta_V
Offline
Last seen: 11 hours 20 min ago
Joined: 09/21/2017 - 15:25
Posts: 207

Check out Nkon

I think they are the preferred battery place for people in the EU. Somebody can correct me if I’m wrong.

I don’t know a whole lot about charging via solar cells, but I don’t think I’d put a battery in direct connection with a power source. Even if it’s a protected battery, I still wouldn’t want that little protection circuit to be the only line of defense against a potentially dangerous situation.

Mike C
Mike C's picture
Offline
Last seen: 2 months 3 weeks ago
Joined: 01/22/2014 - 08:03
Posts: 2009
Location: Sweden

Nkon = Thumbs Up

Agro
Online
Last seen: 19 sec ago
Joined: 05/14/2017 - 11:16
Posts: 2757
Location: Ślōnsk

3*18650 is way less efficient in terms of capacity/volume than 26650. I did some calculations that included 1.5 mm thick battery tube and Shockli 26650 gets roughly 0.14 mAh/mm^3 while 3*18650 only 0.11 mAh/mm^3.
For comparison 1*18650 gets 0.15 mAh/mm^3, just like 21700.
In this calculation I use the real capacity measured by HKJ on low currents: 5900 mAh (5500 mAh is specced and measured at high current).
And I used the best real capacity of 18650 cells: 3400 mAh.

Shockli is most likely the top dog, though I can’t help but mention that there were no independent tests other than that of HKJ….and the HKJ one was made on Shockli-supplied cell which may have been sherry-picked.

There is suspicion that Liitokala 26650-55A may be the same cell as Shockli, but there’s no confirmation.
There are also 6000 mAh Keeppower UH2660. Frankly, I consider them the safest bet, but I don’t know any source for them….

I ordered Shockli recently and if you want the best I suggest that you do the same. Or wait for my test. Wink

EDIT: This helped me spot errors in my spreadsheet, I corrected the numbers above.

Henk4U2
Henk4U2's picture
Offline
Last seen: 4 hours 6 min ago
Joined: 02/13/2014 - 17:52
Posts: 2169
Location: The heart of the Netherlands (GMT+1)

Delta_V wrote:
Does anyone know the typical parasitic drain caused by the protection circuit of a protected battery? The protection circuit itself is an electronic circuit, and thus is going to place a certain (hopefully incredibly small) drain on the battery, but I haven’t been able to find any measurements of this.

In the flashlight world, we are normally concerned with a drain in the hundred of milliamps to multiple amp range causing a low voltage situation in hours or minutes, thus making this parasitic drain irrelevant. But in this application, where the load is only a few milliamps, a small parasitic drain could be an issue.

There has been a limited inquiry of (middle) long term discharge of batteries in this forum, in this thread .

I always think long and hard before I say something really stupid.

Delta_V
Offline
Last seen: 11 hours 20 min ago
Joined: 09/21/2017 - 15:25
Posts: 207

That’s because the 3×18650 configuration has to account for the empty space between cells, while a single 26650 only has the volume of the cell. If you look at a 6×26650 configuration like the OP mentioned, where the 26650s also have empty space between them, and compare it to a 9x or 10×18660 arrangement, and I suspect the 18650s will come out ahead.

edit: back of napkin math gives 0.166 mAh/mm^3 for 9 18650s arranged in a grid configuration. 6 26650s in a 2×3 arrangement yields 0.125 mAh/mm^3. These is in a grid arrangement, which is less space efficient than staggered cells, but that would involve more complicated math that I don’t feel like trying to do on my phone. Regardless, the percentage gained from staggering should be similar regardless of size.

double edit: This is of course dependent on how the OP would arrange the cells in their application. It is certainly possible that fewer, larger cells would be more practical than a larger number of smaller ones. For instance, if the batteries are laid horizontally and vertical “depth” isn’t a huge concern, that could create a situation where 26650 cells are more efficient than the equivalent number of 18650s.

robk
Offline
Last seen: 2 weeks 4 days ago
Joined: 07/12/2013 - 15:51
Posts: 215
Location: New Hampshire

I would go for 26650, either of the cells recommended. You might inquire of Henrik (HKJ) what degree of normal parasitic capacity loss those cells could be anticipated to experience over that period, and I wonder whether the voltage regulator’s losses might throw off purely theoretical energy usage expectations. If you can measure the energy usage in operation and at the temperatures you expect to encounter you could sleep more easily.

Just for giggles you might consider Eneloop NiMH cells, if only for their stability. Nah, it’s probably a bad idea, but you could possible skip the voltage regulator…

Agro
Online
Last seen: 19 sec ago
Joined: 05/14/2017 - 11:16
Posts: 2757
Location: Ślōnsk

I didn’t read carefully enough…indeed, 3*26650 ain’t good. 7*18650 is smaller and has more capacity.

Barkuti
Barkuti's picture
Offline
Last seen: 2 hours 47 min ago
Joined: 02/19/2014 - 14:46
Posts: 2890
Location: Alhama de Murcia, Spain

First of all, nexaen requires a minimum of 6V, so 3V is the per cell cutoff point. Buck converters are cheap as chips but you are going linear, you must have cooked this well and determined your setup consumes less power this way, isn't it?

Average power consumption: P = (12 × 6/900) + (1.15 × 894/900) = 1.222333333333mA

Let's say 1.25mA, or 30mA/day. 

As Delta_V inquires, the parasitic drain of the BMS also needs to be borne in mind. A 100μA drain means ≈2.5mA/day.

Are there any space or weight constraints? If not, buy quality cells providing you with the highest energy for your money. I say quality because dubious quality cells do not last long. Samsung cells are known for their top class life cycle, but nowadays there are cells like the aforementioned PLB 5000/5500mAh which are designed for the vehicle market and also with long cycle life in mind.

The LiitoKala Lii-50A/26650-50A (a PLB 5000mAh rewrap) can be purchased at ludicrously low prices in AliExpress with a little search. Also, they've recently launched the yellow wrapped 21700 Lii-40A cell, which I believe it is made by some chinese OEM and is likely to have a real capacity above 3800mAh. Its price is really affordable, check this: https://www.aliexpress.com/wholesale?SearchText=Lii-40A

One last important thing: higher voltages cause undue stress on li-ion cells. If you want your battery pack to live forever, I recommend drastically reducing the maximum charge voltage. See BU-808: How to Prolong Lithium-based Batteries at Battery University. Also, Orbtronic provides a Samsung 35E datasheet where reducing charging voltage down to 4V/cell is recommended for UPS applications.

If, for example, you're willing to bet on these newer inexpensive cells my recommendation is to oversize your pack by up to 50% and rest easy. Heck, with an 8-pack of the Lii-40A cells your maximum capacity will be above 15Ah (2S4P config), so even if you were to only charge them a bit above 3.9V/cell you'd still enjoy at least ⅔ of maximum capacity, this means 10+Ah. This means you're still guaranteed above 300 days of battery life with this level of partial charging.

Of course fully charge the cells if so you wish, but you'll degrade them much harder this way. I believe it is better to go with a bigger pack and ensure maximum life. If at a later date you decide to dismantle the stuff, your cells will still be in excellent condition. 

 

Cheers