Home Creators Posts Import Register Favorites Logout

Content

Oct. 13, 2026: 21700 table updated, LOTS of changes! For BAK 65E and Linkdata 65P I added a note that cycle life is short at higher current levels. It might be short at all levels (compared to other cells), I’ll update the table later if so.

**************************************

Here are the latest tables.

Files

Comments

Geoff R.

15 and 20 amps on the 18650's are identical.

Battery Mooch

Yup, check the Ratings & E-Scores table. 🙂 With high-performance cells there’s often just not enough IR to make a difference for such a small change in current.

Eric Nelson

I know I've mentioned it before and we've talked about the hurdles and issues there would be doing so, but adding a price vector to these charts would be extremely useful to some. i.e. you might get a 10% performance increase from one cell to the next, but at a 100% cost increase, etc. EVE has a lot of powerful contenders at very good relative prices, that would factor into large pack builders decisions as a few bucks per cell really adds up with 100+ cell packs. Would just be annoying as you'd have to keep updating the charts every time a price change or sale happens as well as do the initial research of finding the best price.

Battery Mooch

Unfortunately there’s just no way I can visit all the possible vendors (dozens) for every cell (almost 100 of them) every couple of days to update pricing and availability. It would more than a full time job. Even worse, many Alibaba vendors will list a cell (and its price) without even having the cells in stock. They just want you to contact them so they can try to sell you something they do have in stock. And then there’s the problem of differently wrapped cells or new versions appearing (and not tested yet)…price them too? If I don’t, some will be mad. If I do price them but say they’re not tested then some will be mad. While I agree it would be great to somehow incorporate pricing it only takes a few minutes for someone to check pricing/availability of the couple different cells they’re considering. I’m afraid that checking process will have to stay as a job for the buyers of the cells…sorry.

Battery Mooch

I forgot to mention a big one…the rage thrown at me if a price goes up before I can update the table and they get surprised after making a decision based on my table prices. Or if a sale goes by and I don’t catch it in time. Folks get angry…really angry…at that kind of stuff. I get enough abuse already without that added on. 😂

Kavya Joshi

Is VAPCELL F63 really good? I mean, considering all the big players like MoliCel EVE LG SAMSUNG which are amongst the top cell manufacturers.... Had never heard of VAPCELL before. Whats your honest take of them?

Battery Mooch

Vapcell “rewraps” cells from the big manufacturers so they can choose very good performing cells. Vapcell is a small company and doesn’t manufacture any cells. The F63 is a great performing low power cell that is a rewrap but we don’t know what cell Vapcell is using.

Battery Mooch

Contact me at batterymooch@yahoo.com and we can talk further. 🙂 What company is this?

izzeww

When is the 50PL CCC getting added to the recommended list? Assuming you're doing self discharge testing right now.

Battery Mooch

And needing to wait for any feedback from vendors and other buyers of that cell. I should ask again.

Cody T

Thanks for the work man, been referencing your data for many years. Cheers!

Joshua S

they sent out a few that were impressive before the real deal, but seems like they ripped a lot of people off, or at least got there hopes up. I always wait till they're tested and finalized before buying. Is there a way for people to get there money back on cells like this when they don't disclose proper values or mislead. I've heard of pre production cells that discharge to fatal levels, thought they would have fixed this before final product, but they are almost worse I'm hearing. Yikes

Battery Mooch

It wasn’t defective cells IMO, it was lower grade cells being sold by uncaring distributors. Presumably because all the higher grade cells were sent to the big customers. But any of the cells we buy can be higher or lower grades at any time, we just have to stick with known trusted vendors who will replace low grade cells or refund the purchase. Check with your vendor for a refund,

Battery Mooch

I haven’t posted the full test report yet so too soon to judge where it might be on the table. Will do so soon though.

Corné

Lets say we run a tabless cell (RS50) at lower powers, 5A. How will that influence the cycle life / capacity?

Battery Mooch

The lower the current (charge or discharge), and the smaller the voltage range, the longer the cell will last. I don’t have hard numbers to give you but there won’t be a fundamental change between, let’s say, running at 5A vs 20A (depending on how long they’re run) as long as the cell is never overcharged or overdischarged. Anything that keeps the cell always below about 45°C will extend cell life. Anything over about 60°C will speed up cell aging.

Battery Mooch

Capacity won’t change for different discharge current levels (over short time periods) but run time will drop due to increasing voltage sag at higher current levels. The “capacity fade” seen as cells age depends on how hard they are used though (charging, discharging, storage) so the gentler they are used the slower the capacity will drop over time. They also lose some capacity while just sitting in storage too (“calendar aging”) but not nearly as much as when being used.

Corné

Appreciate the answers. Thanks for clearing things up.

Francisco Javier Durán

More general LiIon questions :) But looking the 10-40-100A shootout the 50XG cells need a bit of heat to deliver more Ah. At 40A they are similar to RS50 but at 10A they are lower. It´s a interesting matter. Because energy cells would deliver more energy at that level or still less? But maybe the powercell would have relatively longer cyclelife than the energy cell because of the lower "stress"?

Battery Mooch

You can compare cells at several different levels using the Ratings & E-Scores table but there is a particular power range for each cell where it performs best. Assuming equal true cycle life ratings the cell that is stressed less will last longer. But less stress on a 200 cycle cell versus more stress on a 1,000 cycle cell would probably not result in the 200 cycle cell lasting longer. The true ratings and degree of,abuse (or lack thereof) would have a big effect on this.

viper blah

just completely ignore the 50pl its trash lol

Battery Mooch

It’s just not in the top 5. 🙂 But almost all of the top performers are about the same. Most you wouldn’t notice a difference between.

BOT

Thank you for these performance-based rankings. But wouldn't it be interesting to have a safety score as well? Unless performance is directly linked to manufacturing quality and quality controls. But I doubt it.

Francisco Javier Durán

Hi Mooch, I just read on the top of the tables "don´t use or share after..." Do you mean the tables are 6 months true? In the past I thought you write "don´t share before..." because of intellectual property :) Cheers!

Felix

It's interesting to see that the EVE 35V (rated 3500mAh) deliver less Wh than the EVE 30PL (rated 3000mAh) both at 10A... So the high mAh rating of the 35V should only be expected to be seen at low currents. Has the 30PL been tested at low currents?

Battery Mooch

You’re right, they’re not linked and safety test results would be great. Unfortunately there’s no way we can do that. It takes a huge number of cells of multiple grades tested in many different ways and would take months for each model of cell. All of the cells we use pass UN 38.3 safety testing and several are also UL and/or CCC tested too. Whatever degree of safety they have, or lack depending on your view, is pretty much the same for all of them. The huge customers for these cells would never accept an “unsafe” cell with a high failure rate or that was dangerous to use. And the manufacturers can’t afford a bad reputation in the brutally competitive battery market.

Battery Mooch

I don’t remember ever using “don’t share before…” at the top for any of the tables (for IP or patron-only reasons). It’s always been “after”. I ask this because some data could have been made obsolete due to retesting or a flood of low grade cells. It also helps users of the table to be more aware of the new cells being added to the table.

Battery Mooch

I don’t remember if the 30PL made it to the Top 10 for the 1A Shootout but, yes, for the ultra-high capacity cells that extra capacity is only going to be available at about an amp or so. The extra voltage sag of these cells means that as the current rises their effective deliverable capacity quickly starts dropping. It can often help to look at percentage of the rating. A 35V at 10A is running hard, at its rating. That creates a lot of voltage sag which really shortens run time. A 30PL is running at less than 1/4th of its rating at 10A and so is just cruising along, barely getting warm. It will have much less voltage sag and that will result in much longer run time even against higher capacity cells.

Rain

For a non-tabless cell, the P30B is really awesome. I’m sure I’ve asked this before but is the reason Molicel isn’t moving towards tabless as quickly a lot more complicated than it might seem at the surface?

Battery Mooch

I don’t know. It costs a fortune to switch over to tabless and it might not offer enough benefit to be worth the cost for their target customers. Raw performance is near the bottom of the list for most customers, it’s just not needed. Safety, consistency, low defect rate, thermal stability…those things mean a lot and just can’t be found (yet) in the China tabless cells. Moli is also dealing with that factory fire and that might have sucked all the money out of new cell development for a year OR the rebuilt factory could be a perfect excuse to finally jump into tabless. 🙂We’ll see what happens.

Rain

What makes you think the China tabless cells are less consistent, safe, stable, and have higher defect rates? Are they not a 'long-term' choice?

Battery Mooch

Upon re-reading my post, I was a bit too harsh. But those tabless cells we are all drooling over are new tech, unproven over the long term. Sure, some of the China manufacturers have been around for a long time. But we won’t know anything about these cells for at least a couple of years. Until then many companies will stick with known high reliability cells like Molicels. I just realized that another reason why Moli hasn’t gone tabless yet might be because they could be 100% booked making existing cells. They could have the money and desire but need to get new lines built and validated before they can add on tabless cells, which takes years.

Rain

Makes sense. Probably mainly a financial and long-term problem. What's undeniable though is Molicel's lead on fast charging. 871Wh/L and 5C in the P60C is just insane. I don't think many cells would come close in the product of that, even the really low energy density ones, and have the same level of reliability.

Rain

Crazy how close the performance is in the top 4 21700s for 10A and above. We're talking differences of less than 2.5%. Also, will you add the 30XG to the 18650 table, and are you interested in samples of BAK 30D and Highstar 30PS?

Battery Mooch

Agreed…thermal stability, cycle life, price, and availability might be the deciding factors. Yes, yes, and yes. 🙂

Satori80

What ever happened to 20700s? Are they gone or just very very niche?

Battery Mooch

Just very niche, a tiny market. But some are still around. Might be old stock though.

Tripp

Can someone help me choose a mod? I am thinking Hyperion DNA 100c with a singe 21700 OR Thelema DNA250c with 2x 18650

Battery Mooch

assuming cost differences aren’t a problem then IMO it’s going to be personal preference more than anything else, unless a search brings up some reliability issues anyone might have had. Dual 18650 can give you more run time but the mod is a different size/shape and that’s a very personal thing.

Tripp

Great thank you. Yes, I plan on running a 30mm tank and just realized that the Thelema would have a few mm overhang. Plus some other things, I’m going with the Hyperion. That said, are there any disadvantages (other than cost) for buying the Tenpower 50XG? I simply want the best performing device possible. I’ll buy two of them to swap between if needed for duration🤣

Battery Mooch

What power levels will you be running at for a single 21700 mod?

Tripp

Mostly ~50W but I’d like to experiment more with the 75W range. Thank you for being so helpful and quick to reply btw!

Battery Mooch

👍 75W is about 25A worst case and that’s a great level for the 50XG, RS50, or 50PL (CCC). All three will perform essentially the same at that level so choose based on price and vendor reputation IMO. But the 50XG is a great choice.

Battery Mooch

Sorry…50XG disadvantages…not aware of any yet but that stuff typically comes out over time as more and more people use them (like the large number of low grade pre-production 50PL’s being sold). Right now it looks good for all three of those cells.

Link

Hey Mooch, I am noticing a discrepancy for the Ampace JP30P1 when cross-referencing the 18650 Recommended Batteries Table with the Master 18650 Ratings & E-Scores table. The Ampace JP30P1 essentially gets robbed at 20 A (it should be 10.0 Wh instead of 9.7 Wh); therefore, it should be in second place. Hope this helps!

Battery Mooch

Arrgghhh….you’re right! I used the 30A Wh number for the 20A column for that cell in the Recommended table. It doesn’t make any practical difference but it’s a big jump visually in placement so it definitely has to get fixed. Thank you for catching that!

Link

Glad to help. Good work with that quick fix 👍

Joshua S

It's garbage, burns itself out and inconsistent. They mislead lots of people, if u didn't get the pre-realase ones your sitting at like #17 out of #20. Trade em in for the rs50's.

Francisco Javier Durán

Shi.... did not saw that coming.. better than 50Xg and the rest... waiting for the 100A shootout ;)

Battery Mooch

No difference between their actual performance though, only a 2.5%-3.2% run time difference between the top five and the top three are identical considering the 1°C-2°C possible temp shift between tests.

Francisco Javier Durán

OK, good point. I wanted to see the temperature for example on the 40A discharge but the cell isn´t still on the graph? Have a nice sunday!!

Thathip Rujirananthabut

Has the 18650 chart not been updated to include the Amprius SA110? Because the E-scores say that it is the best battery for 5A and 10A

Battery Mooch

The Recommended Batteries table? Being updated soon. It uses the E-Scores though so nothing new being revealed in that table. It’s just a less technical way to see what the top five “best” cells are if the E-Scores are too technical.

Eric Nelson

interesting to see a 6250mah rated cell outperform a 6500mah rated cell.

Battery Mooch

I agree, it can get crazy…especially when mixing energy and power cells. A 4Ah power 18650 can run for longer than a 6Ah energy 21700 if the power levels are moderate to high. I’ve been trying to come up with a way to show how capacity ratings are not a great way to choose a cell (unless at a couple of amps or lower).

Eric Nelson

Yea that's true, I didn't notice the SA112 is actually a power cell, my use case is max efficiency at low draw, so I really only look at the first two columns of your charts lol.

Cainan Kitts

The only issue I've ever had with your tables is that I use my batteries at 40w and can never decide if I should get the best performing 30w or 60w, lol. What would you recommend at 40w currently?

Battery Mooch

40W with 10A-rated energy cells is pushing them very hard (over their rating) and you’ll get a lot of voltage sag. Always use any cell at about 75% of its rating or lower to get the best performance and long cell life. In your case, use my recommendations in the 60W column.

Cainan Kitts

Thank you so much mooch. Not just for replying, but for what you do for the community as well. Cheers!

nabeel sakran

I would like to get feedback on the Tenpower 35HE. I plan to use it at continuous discharge of 7A or 25W. What would be the max temperature ? Is that too aggressive ?

nabeel sakran

I would like to get feedback on the Sinowatt 35HP. I plan to use it at continuous discharge of 7A or 25W. What would be the max temperature? Is that too aggressive ?

Battery Mooch

Both of those cells will perform about the same and can be run at 7A continuously as long as the pack is cooled. Keep them below 60°C maximum to help them last a long time. Never let them get above 70°C!

Robert

I fly RC aircraft using brushless motors and still haven’t found any Li-ion packs that can match the power to weight ratio of using Li-po packs.

Battery Mooch

Agreed…round cells offer easier cooling, greater physical protection, much longer life, but tough to match the raw performance of good LCO-chemistry (“LiPo”) cells. Which round cells have you tried?

Robert

I tried Molicel P26A cells.

Battery Mooch

Cells like the 30XG and 30PL (18650) and especially the 21700’s like the RS50, JP50P1, and 50XG are noticeably better performers than the P26A. Might be worth trying one of them out in a pack. 🙂

PikeKing53

I have Vapcell K25 as a pair and tbf, i get a get day and half out of them in my Boro, i believe them to be Sony rewraps so i think they do sonys batts a lot. Good rated batteries anyway, my local angling shop insist on using them.

PikeKing53

Mooch - My search for the longest life 18650 continues, im looking for a battery for a single battery device, i vape mtl / rdl between 12 - 25 watt yet all the main brands i fail to get a full day, most of their batts rated around 20A, 2500mah. i need to find a battery with around 3500mah at least but for safety reasons im struggling to find one around 15A, there is a 10A but i dont want to be too close to the max as its only a 1 batt mod.

Battery Mooch

There are no 3500mAh 18650’s rated at 15A. You can try the 12A 4000mAh Amprius SA110 though.

Battery Mooch

I haven’t tested them yet, not available. Hopefully soon.