How trustworthy are Sofirn batteries?

147 posts / 0 new
Last post
ChrisGarrett
ChrisGarrett's picture
Offline
Last seen: 15 hours 6 min ago
Joined: 02/12/2014 - 22:03
Posts: 594
Location: Miami, Florida

JasonWW wrote:
Brainsick67 wrote:
They are good to me..tested like 4 of the 3000mAh ones, and they ALL came to 3,000mAh or a tad more.

Capacity is just one aspect of the battery. The black 18650 3000mah have always been accurate for capacity. It seems to be a bit lacking as far as internal resistance and voltage sag, though. It looks like the 30Q definitely beats it in those regards.

Back when DB tested them the performance was so close to a 30Q he thought that’s what they were, just rewrapped. Their amp draw and output matched the 30Q exactly. The one tested above shows a bit below the 30Q at least as far as a 5A load is concerned. I’m wondering if Sofirn switched to a different battery or a different chemical mix in the last 6 months or so.

And also longevity. Just about any cell can perform well for the first 10 cycles. Come back after 30, 50, even 100 cycles and see what’s up?

Chris

JasonWW
JasonWW's picture
Offline
Last seen: 1 hour 15 min ago
Joined: 10/22/2016 - 11:41
Posts: 11782
Location: Houston Texas

All cells loose capacity the more you recharge it. It varies from cell to cell. This info is usually found in the battery documentation.

Here’s the info on the Samsung 35E as an example.

Quote:
7.9 Cycle Life
Each cycle is an interval between the charge (charge current 1,020mA) with 100mA cut-off and the discharge (discharge current 3,400mA) with 2.65V cut-off. Capacity after 500cycles.

Capacity ≥ 2,010mAh (60% of Standard Capacity)

So you lose 40% of the capacity after 500 recharge cycles. Even the batteries I charge the most are still only once a week. It would take me 9.6 years to get to this.

I don’t really pay this much attention. If I wear out a battery I’ll replace it. I’m sure we will have all kinds of better batteries in the coming years. The batteries that I very seldom use and recharge might last me decades. I guess that’s a good thing? I don’t really care. Just my opinion.

Texas Ace Lumen Tube calibrated with Maukka lights

Click this to go to signature links.

ChrisGarrett
ChrisGarrett's picture
Offline
Last seen: 15 hours 6 min ago
Joined: 02/12/2014 - 22:03
Posts: 594
Location: Miami, Florida

JasonWW wrote:
All cells loose capacity the more you recharge it. It varies from cell to cell. This info is usually found in the battery documentation.

Here’s the info on the Samsung 35E as an example.

Quote:
7.9 Cycle Life
Each cycle is an interval between the charge (charge current 1,020mA) with 100mA cut-off and the discharge (discharge current 3,400mA) with 2.65V cut-off. Capacity after 500cycles.

Capacity ≥ 2,010mAh (60% of Standard Capacity)

So you lose 40% of the capacity after 500 recharge cycles. Even the batteries I charge the most are still only once a week. It would take me 9.6 years to get to this.

I don’t really pay this much attention. If I wear out a battery I’ll replace it. I’m sure we will have all kinds of better batteries in the coming years. The batteries that I very seldom use and recharge might last me decades. I guess that’s a good thing? I don’t really care. Just my opinion.

I’m not talking about proper lifespan, from a good cell, I’m talking about inferior cells getting nowhere near that 500, 300, or 100 cycles that the Big 5 test to.

I could rattle off well regarded cells that I’ve used over the past 7 years, Soshine being one brand, that died quick (<1 year/30 cycle) deaths with proper handling and charging.

Sure, I’m only out a few bucks and I don’t care about cells dying from a long, hard life, but there are cells out there that initially test well, that have no longevity within their stated parameters, so that’s a component to be considered and unfortunately people like HKJ and Mooch can’t go back 2 years later and retest their ‘patients,’ so we’re left with a big ‘?’ for anything other than the Big 5, who have a proven track record.

Personally, I treat cells/batteries like I do gun my barrels, in that the more I shoot, the closer I am to having to rebarrel the weapon, so I’m not some miser looking to stretch things out for decades.

Chris

RobertB
RobertB's picture
Offline
Last seen: 10 hours 8 min ago
Joined: 12/18/2015 - 17:49
Posts: 3335
Location: USA, Michigan
JasonWW wrote:
RobertB wrote:
SKV89 wrote:
JasonWW tested the 21700 Sofirn in the C8F and it performs well. I think I recall someone said the Sofirn 3000mah or 2800mah battery might be a rewrapped Samsung/Sony.

I have 4 of the Sofirn 21700’s and although they seem to be decent cells, they under perform. Fully charged, I put one in my FF E07 and turbo mode didn’t seem that great. Put a Samsung 30Q in, and turbo was noticeably brighter. In other words, they can’t handle the amp draw. Tried with all 4 of the 21700’s, same result. They act like a 10A cell. Ordered some Samsung 30T’s for the E07 instead.


Some actual numbers might help.

They blew away the 30Q in my C8F. 11.5A and 3330lm compared to the 30Q’s 9A and 2900lm. The 30T performed about the same as the Sofirn 4000mah with 11.5A and 1160lm. (The 30T should show its extra muscle at higher amp draws)

I just don’t see how it’s possible a 30Q could actually outperform the Sofirn 21700.

What is the amperage range of the E07?

Also, what is a “10A cell”?

Well, you won’t get any actual numbers, just my eyesight. In the Fireflies E07, the Samsung 30Q and 25r’s, blew away the Sofirn 21700. Very noticeable brightness difference in Turbo mode.

I believe the E07 draws around 20A. I also think you already know what I mean by “10A cell”.

Lexel
Lexel's picture
Offline
Last seen: 3 hours 38 min ago
Joined: 11/01/2016 - 08:00
Posts: 5475
Location: Germany

JasonWW wrote:
All cells loose capacity the more you recharge it. It varies from cell to cell. This info is usually found in the battery documentation.

Here’s the info on the Samsung 35E as an example.

Quote:
7.9 Cycle Life
Each cycle is an interval between the charge (charge current 1,020mA) with 100mA cut-off and the discharge (discharge current 3,400mA) with 2.65V cut-off. Capacity after 500cycles.

Capacity ≥ 2,010mAh (60% of Standard Capacity)

So you lose 40% of the capacity after 500 recharge cycles. Even the batteries I charge the most are still only once a week. It would take me 9.6 years to get to this.

I don’t really pay this much attention. If I wear out a battery I’ll replace it. I’m sure we will have all kinds of better batteries in the coming years. The batteries that I very seldom use and recharge might last me decades. I guess that’s a good thing? I don’t really care. Just my opinion.

The life cycle also depends a lot on the current you draw from a cell
at 3C it ages a lot quicker than the test 0.2 or 0.5C

so you can expect batteries in triples and quads to age a lot fater per cycle than a conservative light with just 1.5A LED current

JasonWW
JasonWW's picture
Offline
Last seen: 1 hour 15 min ago
Joined: 10/22/2016 - 11:41
Posts: 11782
Location: Houston Texas

The 20A range is higher than any 3v lights I currently have (except the S43S, but it’s 18650 only). Maybe they perform differently at these higher levels and the Sofirn cell falls flat on its face? I dont know, as I can’t test it. I know that at lower levels it whips the 30Q. I can only hypothesize that it does the same at higher levels, but who knows.

As far as I’m aware of there is no such thing as a “10A cell”. Are you using it as a slang word? Since both cells are capable of well over 30A, you must not be refering to their upper limit. Are you refering to their max continous output that is under 80°C? (Some cells use 70°C as their limit)
The 30Q is rated by HKJ for 15A continous. Are you thinking the Sofirn 4000mah 21700 would exceed 80°C if discharged at more than 10A continous? It’s possible. I don’t have a way to test this.

Continous amperage ratings are kind of a meaningless spec. It gives you a little insight about the battery, but that is all. None of our flashlights can run continuously at these high amperages. I don’t know of anyone who categories batteries by this spec. Maybe Vapers do? I don’t.

Texas Ace Lumen Tube calibrated with Maukka lights

Click this to go to signature links.

ChrisGarrett
ChrisGarrett's picture
Offline
Last seen: 15 hours 6 min ago
Joined: 02/12/2014 - 22:03
Posts: 594
Location: Miami, Florida

JasonWW wrote:

Continous amperage ratings are kind of a meaningless spec. It gives you a little insight about the battery, but that is all. None of our flashlights can run continuously at these high amperages. I don’t know of anyone who categories batteries by this spec. Maybe Vapers do? I don’t.

It’s not meaningless. It’s not meaningless because for ‘pulse discharge’ tests, there is a specific duty cycle to be adhered to and those can be all over the map.

5 seconds on, 30 seconds off, or is it 7 seconds on and 45 seconds off?

That might be problematic for somebody in the middle of a Vape Bender, don’t you think?

Hell, even my stock D4 can handle 20+ seconds on Turbo, before stepping down. Who’s going to remember 60A Pulse at 7 seconds on and 30 seconds off?

It’s just easier to use the continuous number and stick below that, for any given light’s max current draw.

Chris

JasonWW
JasonWW's picture
Offline
Last seen: 1 hour 15 min ago
Joined: 10/22/2016 - 11:41
Posts: 11782
Location: Houston Texas

You lost me Chris. I don’t know anything about vaping.

Texas Ace Lumen Tube calibrated with Maukka lights

Click this to go to signature links.

Lexel
Lexel's picture
Offline
Last seen: 3 hours 38 min ago
Joined: 11/01/2016 - 08:00
Posts: 5475
Location: Germany

How much the cell sags is absolutely consistent with continuous A rating, because both are caused by the internal resistance

But on some applications like DD a single Nichia219C you dont want a cell with low internal resistance

ChrisGarrett
ChrisGarrett's picture
Offline
Last seen: 15 hours 6 min ago
Joined: 02/12/2014 - 22:03
Posts: 594
Location: Miami, Florida
JasonWW wrote:
You lost me Chris. I don’t know anything about vaping.

Most flashlight applications are benign for li-ion cells, it’s the sub-ohm vapers who are cutting edge vis-a-vis li-ions and what they can do and what punishment they can take. They torture cells and need to know the continuous numbers, not the pulse, so that they can build/plan accordingly.

Chris

hank
hank's picture
Offline
Last seen: 2 hours 57 min ago
Joined: 09/04/2011 - 21:52
Posts: 8151
Location: Berkeley, California

And the batteries get revenge when tortured by vapers:

https://www.google.com/search?client=firefox-b-1-d&q=vape+battery+explosion

The Denver TV channel (first in the search result) https://www.youtube.com/watch?v=MROw8k9CQIs says they have a database of over 300 vape battery explosions.

ChrisGarrett
ChrisGarrett's picture
Offline
Last seen: 15 hours 6 min ago
Joined: 02/12/2014 - 22:03
Posts: 594
Location: Miami, Florida

hank wrote:
And the batteries get revenge when tortured by vapers:

https://www.google.com/search?client=firefox-b-1-d&q=vape+battery+explosion

The Denver TV channel (first in the search result) https://www.youtube.com/watch?v=MROw8k9CQIs says they have a database of over 300 vape battery explosions.

I’m well aware of the problems and most of them are most probably due to shoddy electronics, general manufacture, or just recycled ChiCom crap.

I’m mostly referring to ‘educated’ vapers being on the cutting edge of li-ion usage, which used to be the sole provenance of flashlight geeks.

Chris

Funtastic
Funtastic's picture
Offline
Last seen: 5 hours 5 min ago
Joined: 06/26/2014 - 02:14
Posts: 370
Location: New Zealand

I’ll charge up two more Sofirn 3000mAh cells and compare against the 30Q, it’ll give us a better idea.

The one I tested reached 60°C at 10A down to 2.75V. Since 60°C is the max operating temp I’d rate it at 10A max continuous.

Keep in mind when looking at this 10A test my CBA has a voltage drop of 0.05V. For my 30Q test, the battery clamp had a bad connection causing the dips at the top.

The Sofirn cell isn’t looking so good at this current. Still not bad for a cheap bundled cell.

Chart2

The below is tests of the Sofirn again at 0.6A, 1A, 2.5A, 5A, 8A, 10A. AT 5A there’s a 0.02 voltage drop & 10A 0.05V.

Chart3

Texas Ace Lumen Tube calibrated with maukka lights

JasonWW
JasonWW's picture
Offline
Last seen: 1 hour 15 min ago
Joined: 10/22/2016 - 11:41
Posts: 11782
Location: Houston Texas

Funtastic, would you be able test the Sofirn 3000mah against a 30Q in a FET powered flashlight? Assuming you have a suitable light and a clamp style ammeter. This is how DB tested his Sofirn battery a ways back.

I believe the amp draw he saw was about the same between the two cells. He also has an integrating sphere and measured about the same lumens. I don’t know if you have an integrating sphere, but just a comparison of the amp draw would be interesting. I’ll try and dig up his original post about it.

Texas Ace Lumen Tube calibrated with Maukka lights

Click this to go to signature links.

JasonWW
JasonWW's picture
Offline
Last seen: 1 hour 15 min ago
Joined: 10/22/2016 - 11:41
Posts: 11782
Location: Houston Texas

Well, it seems my memory was a little off. It wasn’t the black 3000mah Sofirn 18650 that DB had, it was the one before that when they had the red labels. So my suspicion that Sofirn is using a different battery might be correct. Judging by this red ones performance, maybe they should go back to it if possible. It seems great!

Below is an excerpt from his review on the Sofirn C8F from Dec 2017.

DB Custom wrote:
They even supplied a charger and a cell, this time an unprotected cell (I know, the label says it’s protected but testing and questions to Tracy proved out they made an error with the printing on this label) I unwrapped it and it appears to be a Samsung 30Q, but it does better than the 30Q’s I have. Bested only by Sony in the form of the VTC4 and VTC5A in my tests.

This might also be a good time to point out that Sofirn has already done the spring bypasses for us! Yes, 22ga wire bypasses already installed at the factory! Nice touch!

Tests? Yep, I tested the light with their cell and with a couple of others to see where they all stood…

……………Sofirn ………………………30Q……………………Sony VTC4
1. 0.552 Lumens……………0.552 Lumens…………0.587 Lumens
2. 37.50 Lumens……………22.29 Lumens…………22.81 Lumens
3. 83.15 Lumens……………..82.8 Lumens…………84.53 Lumens
4. 890 Lumens…………….872.85 Lumens…………931.5 Lumens
5. 1338.6 Lumens….…1314.45 Lumens………1400.7 Lumens
6. 3015.3 Lumens…….3018.75 Lumens………3125.7 Lumens
….2756.66 at 30 sec……….2691 at 30 sec

Texas Ace Lumen Tube calibrated with Maukka lights

Click this to go to signature links.

Funtastic
Funtastic's picture
Offline
Last seen: 5 hours 5 min ago
Joined: 06/26/2014 - 02:14
Posts: 370
Location: New Zealand

Unfortunately I don’t have a clamp meter.

I do have a TA lumen tube calibrated with maukka’s lights. I’ve already tested them with the Sofirn Q8

30Q

5020 – On
4720 – 30 Sec

Sofirn

4730 – On
4440 – 30 Sec

GA

4820 – On
4460 – 30 Sec

I do have a C8F I can test to see what single cell perform is like. From memory it’s much lower. Thinking it was around 2500 and the 30Q 3100

Must be different than DB Custom’s but it still looks the same if I unwrap it.

Texas Ace Lumen Tube calibrated with maukka lights

JasonWW
JasonWW's picture
Offline
Last seen: 1 hour 15 min ago
Joined: 10/22/2016 - 11:41
Posts: 11782
Location: Houston Texas

There’s no need for further tests. It seems obvious that Sofirn is using a different battery compared to a year ago.

Texas Ace Lumen Tube calibrated with Maukka lights

Click this to go to signature links.

teacher
teacher's picture
Offline
Last seen: 3 months 1 week ago
Joined: 02/23/2016 - 19:04
Posts: 9596
Location: NE & SW Alabama

That’s a shame too. I had been thinking all along they were still on par with the 30Q. Facepalm
Oh well they probably saved 10 cents, but most all of us would have been willing to pay at least 50 cents more and them keep using the better battery.

Why is that concept apparently so hard to understand in China? . Smile

You never know how a horse will pull until you hook him up to a heavy load./"Bear" Bryant 

 .................................. "Slow is Smooth, Smooth is Fast" ...................................

       Texas Lumens Flashlights / M4D M4X Deals : sign up - save $$$$  

         Rudeness Level _ mΩ _ {width:70%} _ LightWiki _ LED Tint Chart  

      Xlamp size chart _ BatteryU _ Flashaholic? Need Professional Help???            TheOriginal _ TAB _ LightSearch _ BatterySearch _ 14500's _ DiCal 

 

                                             

pennzy
pennzy's picture
Online
Last seen: 43 sec ago
Joined: 12/10/2017 - 19:45
Posts: 1401
Location: United States , Pa.

My SP33 loses turbo faster with the Sofirn battery compared to the Shockli for what it is worth.

ChrisGarrett
ChrisGarrett's picture
Offline
Last seen: 15 hours 6 min ago
Joined: 02/12/2014 - 22:03
Posts: 594
Location: Miami, Florida

That red Sofrin cell has the exact looking wrapper scheme as some XXX-Fire knockoffs that HKJ has tested in the past.

Too funny.

Chris

JasonWW
JasonWW's picture
Offline
Last seen: 1 hour 15 min ago
Joined: 10/22/2016 - 11:41
Posts: 11782
Location: Houston Texas

pennzy wrote:
My SP33 loses turbo faster with the Sofirn battery compared to the Shockli for what it is worth.

That’s to be expected – assuming you are comparing the Shockli 5500mah (5750mah actual) to the 5000mah Sofirn. There should not be too big of a difference, but some.

How are you measuring and how big a difference are you seeing?

Texas Ace Lumen Tube calibrated with Maukka lights

Click this to go to signature links.

BOMBAY
Offline
Last seen: 20 min 10 sec ago
Joined: 04/24/2016 - 10:07
Posts: 151
Location: Polska

my sofirn 21700 4000mAh battery gives more power in C8F and FF PL47 than sony VTC6 and samsung 30Q

RobertB
RobertB's picture
Offline
Last seen: 10 hours 8 min ago
Joined: 12/18/2015 - 17:49
Posts: 3335
Location: USA, Michigan
BOMBAY wrote:
my sofirn 21700 4000mAh battery gives more power in C8F and FF PL47 than sony VTC6 and samsung 30Q

I must have 4 defective 21700’s then. Ran all four through discharge/charge cycles just to make sure. I get the exact opposite on the Fireflies E07

Samsung 30Q’s, and 25r’s purchased through Liionwholesale, not Aliexpress Liitokala stores

pennzy
pennzy's picture
Online
Last seen: 43 sec ago
Joined: 12/10/2017 - 19:45
Posts: 1401
Location: United States , Pa.

JasonWW wrote:
pennzy wrote:
My SP33 loses turbo faster with the Sofirn battery compared to the Shockli for what it is worth.

That’s to be expected – assuming you are comparing the Shockli 5500mah (5750mah actual) to the 5000mah Sofirn. There should not be too big of a difference, but some.

How are you measuring and how big a difference are you seeing?


Sorry, no measurements, but enough of a difference to notice in walks down the driveway. Both are 5500 cells. IR is about twice the Shockli if you can believe the charger.
JasonWW
JasonWW's picture
Offline
Last seen: 1 hour 15 min ago
Joined: 10/22/2016 - 11:41
Posts: 11782
Location: Houston Texas

pennzy wrote:
JasonWW wrote:
pennzy wrote:
My SP33 loses turbo faster with the Sofirn battery compared to the Shockli for what it is worth.

That’s to be expected – assuming you are comparing the Shockli 5500mah (5750mah actual) to the 5000mah Sofirn. There should not be too big of a difference, but some.

How are you measuring and how big a difference are you seeing?


Sorry, no measurements, but enough of a difference to notice in walks down the driveway. Both are 5500 cells. IR is about twice the Shockli if you can believe the charger.

I contacted Sofirns battery store (E-Battery store) and they told me they don’t carry a 5500mah 26650 even though there are tons of pictures and such. They said they are all 5000mah. This is quite confusing. I may need to ask Tracy at Sofirn if she knows what the situation is.

Tracy has told me their 26650 does not pull a lot of amps with a FET driver. In the SP70 prototype (70.2, FET driver and 2 x batteries) their 26650 performed about the same as a pair of Panasonic NCR18650B. That’s an old cell from 2012. So we know they aren’t good for high drain usage.

The Shockli 5500mah is quite new and is a top rate battery. I would expect it to be better than the Sofirn 26650, but not by much as far as the SP33 is concerned. It might have higher voltage sag causing the SP33 to loose turbo sooner, maybe?

The question is, is there a big enough difference in performance to justify the higher cost of the Shockli 5500mah?

Texas Ace Lumen Tube calibrated with Maukka lights

Click this to go to signature links.

JasonWW
JasonWW's picture
Offline
Last seen: 1 hour 15 min ago
Joined: 10/22/2016 - 11:41
Posts: 11782
Location: Houston Texas

BOMBAY wrote:
my sofirn 21700 4000mAh battery gives more power in C8F and FF PL47 than sony VTC6 and samsung 30Q

Your forgetting that the E07 pulls about 20A while your C8F and PL47 pull about 10A, roughly. I got the same results as you.

What RobertB is saying is that at the higher amp draws (the 20A range), the Sofirn 4000mah seems to be having difficulty while the 30Q keeps pulling strong. It seems strange to me, but I suppose it’s possible.

Some scientific testing of the Sofirn 4000mah is needed.

Texas Ace Lumen Tube calibrated with Maukka lights

Click this to go to signature links.

hank
hank's picture
Offline
Last seen: 2 hours 57 min ago
Joined: 09/04/2011 - 21:52
Posts: 8151
Location: Berkeley, California
Quote:
they probably saved 10 cents

https://www.google.com/search?client=firefox-b-1-d&q=chinese+%22good+eno...

Quote:
… start-up companies developing and releasing new products and services don’t necessarily need to aim for perfection to make inroads against established players. The phrase can be similarly applied to middle-market players in China that have been able to steal a march on incumbents by developing and releasing good-enough products that are displacing premium ones.

The Battle for China’s Good-Enough Market – Harvard Business Review
https://hbr.org/2007/09/the-battle-for-chinas-good-enough-market

It’s hard to unlearn the “good enough” attitude toward corner-cutting quality, once it gets established.
https://www.businessinsider.com/china-bridge-collapses-2012-8

I worry about China’s “good enough” nuclear power plant expansion.

JasonWW
JasonWW's picture
Offline
Last seen: 1 hour 15 min ago
Joined: 10/22/2016 - 11:41
Posts: 11782
Location: Houston Texas

The whole thing about switching to save $0.10 is just a theory. They may have had to switch to a different battery for other reasons.

Texas Ace Lumen Tube calibrated with Maukka lights

Click this to go to signature links.

hank
hank's picture
Offline
Last seen: 2 hours 57 min ago
Joined: 09/04/2011 - 21:52
Posts: 8151
Location: Berkeley, California

Or they’re getting their batteries from a subcontractor who cut a corner and Sofirn didn’t assess the change.

teacher
teacher's picture
Offline
Last seen: 3 months 1 week ago
Joined: 02/23/2016 - 19:04
Posts: 9596
Location: NE & SW Alabama

hank wrote:

I worry about China’s “good enough” nuclear power plant expansion.
Yep, that could have a real bad outcome….. Facepalm

You never know how a horse will pull until you hook him up to a heavy load./"Bear" Bryant 

 .................................. "Slow is Smooth, Smooth is Fast" ...................................

       Texas Lumens Flashlights / M4D M4X Deals : sign up - save $$$$  

         Rudeness Level _ mΩ _ {width:70%} _ LightWiki _ LED Tint Chart  

      Xlamp size chart _ BatteryU _ Flashaholic? Need Professional Help???            TheOriginal _ TAB _ LightSearch _ BatterySearch _ 14500's _ DiCal 

 

                                             

Pages