Kobalt 40V packs - doubling the cells questions

Announcement - Help us fight the BOTS! Please report all spam including stuff in your inbox!


New member
Jun 17, 2021
New member here...this is a bit off of your usual topics but you guys really know your 18650 stuff so here goes...

The 2.0 and 2.5 AH kobalt 40V batteries use a 10s configuration with 10 empty spaces in parallel. The 4.0 and 5.0 AH versions have the extra 10 cells installed. It appears to be relatively simple process to add 10 matching cells to double the capacity of the small packs. Given that a new 5.0 AH pack retails for $150 I would love to upgrade my old 2.0 and 2.5 packs. We'll ignore that I have already spent $130 on a battery charger/tester, spot welder and a dead 80V 20 cell pack that yielded zero usable cells. On the bright side, I have lots of cells to practice spot welding on.

My original plan was to harvest old cells from old packs on the theory that they would be same or similar to what I have. Also looking at using new cells. So how close do the old and new cells have to match? For example, I'm trying to match a Sanyo UR18650RX (~2000 mA) and UR18650NSX (~2500 mA) both rated for about 20A max....and neither one appears to be readily available unless ordering 50+ from overseas. The Samsung 25R appears to be similar (2500 mA 20A) and readily available (and very much like the 20R in the dead 80V pack). Would installing the new samsungs in a pack of old sanyos be a problem? Big problem like going boom or little problem like killing the older weaker cells if the pack is pushed to 100% discharge. I assume the pack performance will be limited by the weaker (weakest?) batteries and some of the capacity of the new batteries will be unusable.

Similarly, any problems with replacing a Samsung 13Q (1300 mA 15A) with the 25R? Craftsman 19.2V 5-cell pack....

I recently picked up the 40V mower and leaf blower and plan to swap out the packs well before the packs die.




Feb 25, 2020
New member, you forgot the introduction post where we can all say to you "Welcome to lithium addiction etc etc" 😅
...welcome dprice!

Oh, I like your question, I had a similar one some time ago. What I learnt is this, it's a sort of guideline to have a good quality battery.

Mixing old and new cells: what I found is that my older cells, harvested from laptop batteries, have a higher IR (5-8 years old, decent capacity, surely can be used). Newer cells have a lower IR and a slightly better capacity. I wouldn't mix them because you'd have an unbalanced pack where some cells work more (more stress) than others.

Mixing capacities: I'm going for a 300mAh frame, cells from 2000 to 2300mAh. I see people mixing cells in a 500mAh range. A well-known strategy is to round-robin pick cells to make packs: if you have 10 cells @2000mAh, 10 @2150mAh and 10 @2300 you could make the pack picking one 2000, one 2150 and one 2300, then again a 2000, a 2150 and a 2300, and so on, and compose the pack. What counts is to make balanced packs, all of the same total capacity.
And the parameter to keep into consideration is IR; @Wolf (and others, too, sorry if I only remember Wolf talking about IR!) made very detailed studies (you can find them in the forum with a bit of patience) and wisely suggests to keep IR of cells in a 15mOhm range. I'm going for a slightly wider range of 25mOhm, I'll be testing all summer and I'll share the outcome.

Mixing chemistries: now this is the good bit. Experience from our addicted members shows how mixed chemistries can be a problem. You can find the post here. So I'll try to avoid this by checking every single cell type's datasheet, I won't mix power cells with energy cells. The best solution here would be to have all the same model/make cells, which is not my case.


Jan 7, 2017
Welcome, and you've come to the right place! (y)

To add to italianuser, please make sure to use the correct terminology
...trying to match a Sanyo UR18650RX (~2000 mA) and UR18650NSX (~2500 mA
This is "mAh", not "mA". The former is capacity, the latter is current, ie millAmp-hour and mill-Amp respectively
Capacity is always in Ah or mAh (1Ah = 1000mAh)
Also note you'll see a "C" designation. Such as a charge rating of 2C, 1C, 5C. The "C" rating is the capacity divided by 1000. So a 2C rating on a 2500mAh capacity cell would be a 2 * (2500/1000) = 5A. 5C of the same cell would be 25A

Capacities are kinda like water tanks connected in parallel. In almost impossible to have 1 tank have a higher level of water in it than another tank. It is still possible, for example, if one pipe is 1/2" and the others are 1". You can obviously pull more from the larger pipe. But once you stop pulling, they'll equalize, rather quickly.
However, with batteries, it is a little more stressful on a cell if it's capacity is pulled down while the others still have some in reserve. With that said, it would take a lot of abuse to make it an issue.
In other words, if the packs you currently have are 2500mAh capacities, and you drop 2000mAh cells in, it won't be that big of an issue, overall. The total capacity of the pack just won't be as high as you were looking for.

Now, what's a bigger issue is mixing output currents. If a cell is rated for 15A, and you pair it with a 30A, then you really are limited to only pulling a max of 30A, which would be 15A per cell. You could pull 45A, and the lower rated cell will give its all, but it actuality, it'll be pushing out a little bit more than 15A. It'll probably be closer to 23A, or thereabouts.
This will stress out that cell far more than other cell, and shorten its cycle life.

In the end, you want to get capacities close, but not necessary to be example. There can be some variances. However, the C ratings should be pretty close for this type of application. I state this because in high load applications the discharge rate is going to be high, which is what powertool cells are designed for. Don't mix in laptop cells for instance.
In a low discharge application, ie you have something like 100p cells and the load is 50A, then each cell would only get 500mAh (0.5A) load. At this point, mixing C ratings wouldn't make that big of a deal.


New member
Jun 17, 2021
Thanks for the advice! I picked up a working Kobalt 80V 2.5 Ah pack for my second batch of donor cells. Turns out all were good (Samsung 25R) and only 1 tested below 2500 mAh at 2470. So rather than mixing older and newer I just stuck all 20 donor cells into a bad 40V pack to turn it into a 5.0 Ah pack. I just finished it up this morning and it seems to be working - 4 lights. Now I have to mow the yard to put it to the real test.

Now I just have to find more dead battery packs to fix!