Should I test every single Cell I received and other doubts/questions after Leaking Cell

[EastFalcon]

Fuck ? ...

I read a thread also here stating that Docanpower and Amy were reliable sellers.

I was in contact with Beki from Docanpower though.

Makes me really wish I went with an off the shelf solution now ??

I thought everyone knew about Docan, I searched for docan when looking to buy and the 3rd result was that beware Docan thread in the results. Avoided them and went with 18650batterystore instead.

The other 2 companies that have grade A cells are ezeal and sun fun kits. The rest are just selling grade b with scams to make them appear grade A.

 

[silverstone]

I thought everyone knew about Docan, I searched for docan when looking to buy and the 3rd result was that beware Docan thread in the results. Avoided them and went with 18650batterystore instead.

The other 2 companies that have grade A cells are ezeal and sun fun kits. The rest are just selling grade b with scams to make them appear grade A.

But isn't grade A also a bit of a legend? Aren't they supposed to be dedicated to automotive, no way for end consumers to buy them directly, and would cost like 200 USD a ach as opposed to 100 USD each?

And how would I know? The threads here that I found on Diysolarforum said to contact Amy from Docanpower and that they were reputable sellers. It was also reccomended by a friend I had...

Maybe I should go with Gobelpower or whatever their name is. Andy reviews about them seem to be quite positive. Assuming I build another battery.

I didn't finish building it and am already regretting it ?

 

[silverstone]

Peel the black cover off that cell, clean gently with rubbing alcohol see if the background of the QR code looks different than the other cell. If it does than you have been scammed by Docan.

What do you mean exactly by black cover and background of QR code looking different than the other cell?

 

[Alkaline]

you have been here for a while, are you trolling?

EVE identifies Grade B cells using new marking Method

Yesterday I got a new information from my supplier. I was supposed to order EVE LF230 (230Ah) cells from them. But it seems there has been a change on how EVE is sending out rejected cells (Tested to be lower than Automotive Grade) for auction to the Chinese market (grey). According to the new...

diysolarforum.com

There very much is a grade A and grade B reject. Since about March 2022 cells made by EVE are marked with a B. Resllers get these grade B cells, they sand off the B and relaser them. However, this process isn't perfect so what happens is you can see sanding marks or sometimes the background of the QR codes looks noticeably different the rest of the cell.

It has nothing to do with automotive grade. Cells can be certified to any specs, used for auto, use for enterprise storage, they just have different test. Auto cells will probably have a greater cell crush strength and maybe higher voltage under loads. While enterprise storage will focus on longer usage life. You can check the data sheet of each cell it will tell you what tests it had done.

Another big myth: Cells that fail testing are automatically now considered storage grade. No they are not. No one know why specifically a manufacturer calls a cell grade B, they don't disclose this information so any smart alec coming up with theories on why it is is full of it.

Many times it can be very hard to tell the difference between grade B and grade A, capacity is not a valid test, I have seen grade B cells have higher capacity than grade A cells of the same model type. Most people say it has to do with self discharge rate.

But bottom line, whatever the reason maybe, any seller that goes and alter qr codes should be avoided. That is fraud.

 

[silverstone]

you have been here for a while, are you trolling?

EVE identifies Grade B cells using new marking Method

Yesterday I got a new information from my supplier. I was supposed to order EVE LF230 (230Ah) cells from them. But it seems there has been a change on how EVE is sending out rejected cells (Tested to be lower than Automotive Grade) for auction to the Chinese market (grey). According to the new...

diysolarforum.com

There very much is a grade A and grade B reject. Since about March 2022 cells made by EVE are marked with a B. Resllers get these grade B cells, they sand off the B and relaser them. However, this process isn't perfect so what happens is you can see sanding marks or sometimes the background of the QR codes looks noticeably different the rest of the cell.

It has nothing to do with automotive grade. Cells can be certified to any specs, used for auto, use for enterprise storage, they just have different test. Auto cells will probably have a greater cell crush strength and maybe higher voltage under loads. While enterprise storage will focus on longer usage life. You can check the data sheet of each cell it will tell you what tests it had done.

Another big myth: Cells that fail testing are automatically now considered storage grade. No they are not. No one know why specifically a manufacturer calls a cell grade B, they don't disclose this information so any smart alec coming up with theories on why it is is full of it.

Many times it can be very hard to tell the difference between grade B and grade A, capacity is not a valid test, I have seen grade B cells have higher capacity than grade A cells of the same model type. Most people say it has to do with self discharge rate.

But bottom line, whatever the reason maybe, any seller that goes and alter qr codes should be avoided. That is fraud.

I am not trolling ?.

And I agree with you about fraud, if they are counterfeiting them.

But specifically when I bought from docanpower they clearly stated that the cells are slighly below capacity. And if I recall correctly they also stated grade b.

I think ordered September 2022 and march 2023 or somewhere along these lines.

When I checked nkon it was much more expensive, that's why I went with docan.

It's so frustrating. This fucking battery it's taking all my energy, which isn't much to begin with. Now this. I wonder if I should thrash these cells away and just order new ones then ?. In the other thread about scammers, @upnorthandpersonal said to just put them in the battery though.

 

[upnorthandpersonal]

I think you need to build another 100 for that goal. Like a 40 foot container.

There is almost no sun in Denmark already during winter. Cannot imagine how it's like k Finland ?. Fully sunny during summer midnight (no sleep) , fully dark during day at winter (no power) ? It's similar here in Denmark but I don't think it's to *that* extent

The trick is to have a 2 to 3 kWh per day electricity requirement only.

In the other thread about scammers, @upnorthandpersonal said to just put them in the battery though.

That's what I do. I stopped worrying about the 'grade' of a battery long ago. My first pack with these cells (I was one of the first on this forum to get them) were definitely not 'grade A'. I'm now entering my 5th year with them powering my house 24/7 without issues.

 

[Alkaline]

I am not trolling ?.

And I agree with you about fraud, if they are counterfeiting them.

But specifically when I bought from docanpower they clearly stated that the cells are slighly below capacity. And if I recall correctly they also stated grade b.

I think ordered September 2022 and march 2023 or somewhere along these lines.

When I checked nkon it was much more expensive, that's why I went with docan.

It's so frustrating. This fucking battery it's taking all my energy, which isn't much to begin with. Now this. I wonder if I should thrash these cells away and just order new ones then ?. In the other thread about scammers, @upnorthandpersonal said to just put them in the battery though.

Well Sorry,

I'm seeing a very much parallels between your experience with mine. I built 3x 8s battery packs about 2-3 years ago 8 cells form company called xuba and 8 from basen, finally 8 from Luyuan.

I have the same tester as you. When I finally broke out the tester I was pretty pissed, 258Ah out of 280 cell after just maybe like 10-15 cycles.

What is going on here? Just got ZKE-40

how exactly do you do this? I just took my tester out attached the v-sense cables to the screw and used 2 buss bars on each terminal with the clamp. Verify the voltage as recorded by the tester matches your reference dvom. Verify the current as recorded by the tester matches your reference...

diysolarforum.com

Also that wasn't the main issue, its that on my xuba pack 2 cells would always dip in voltage any time you draw over 70 amps and it would trigger low voltage cut-off on my overkill solar bms.

basen was not as bad but did have 1 low cell that required baby sitting this was alleviated by using an active balancer (I just left it plugged in whole time).

Luyuan was the only set of cells that ran fine, no issues, stayed in balance, you really didn't even need an active balancer for those. I paid 197/cell delivered to East Texas, a lot but better than. Now you can get these for under $150 delivered in the US.

I never had success with grade B, you always would have a few cells that were pain in the butt.

I would not use that leaky cell. Just accept as cost of DIY and move on.

 

[silverstone]

Well Sorry,

I'm seeing a very much parallels between your experience with mine. I built 3x 8s battery packs about 2-3 years ago 8 cells form company called xuba and 8 from basen, finally 8 from Luyuan.

I have the same tester as you. When I finally broke out the tester I was pretty pissed, 258Ah out of 280 cell after just maybe like 10-15 cycles.

What is going on here? Just got ZKE-40

how exactly do you do this? I just took my tester out attached the v-sense cables to the screw and used 2 buss bars on each terminal with the clamp. Verify the voltage as recorded by the tester matches your reference dvom. Verify the current as recorded by the tester matches your reference...

diysolarforum.com

Also that wasn't the main issue, its that on my xuba pack 2 cells would always dip in voltage any time you draw over 70 amps and it would trigger low voltage cut-off on my overkill solar bms.

basen was not as bad but did have 1 low cell that required baby sitting this was alleviated by using an active balancer (I just left it plugged in whole time).

Luyuan was the only set of cells that ran fine, no issues, stayed in balance, you really didn't even need an active balancer for those. I paid 197/cell delivered to East Texas, a lot but better than. Now you can get these for under $150 delivered in the US.

I never had success with grade B, you always would have a few cells that were pain in the butt.

I would not use that leaky cell. Just accept as cost of DIY and move on.

Thanks for sharing your experience.

Yeah I'm skeptical about the leaky one even though do can claims it's just "looks" . Probably too late for PayPal dispute. Maybe threaten them to spread the word of how scanners they are and make them close their business, but apart from that not much I can do ?.

It's it's "just" the one cell that is bad I could live with it. If it's all other 32 as well then no ?.

Right now the idle consumption of the house is sitting around 400-800 W. So you could argue that in the worst case scenario I should be able to cover those base loads, without any peaks. That would be 10-20 A shared across 2 strings. Knock on wood, they should be able to deliver even that.

It would really suck if I cannot do more & charge fast etc, but if they can at least do that it would be good.

 

[silverstone]

Well Sorry,

I'm seeing a very much parallels between your experience with mine. I built 3x 8s battery packs about 2-3 years ago 8 cells form company called xuba and 8 from basen, finally 8 from Luyuan.

I have the same tester as you. When I finally broke out the tester I was pretty pissed, 258Ah out of 280 cell after just maybe like 10-15 cycles.

What is going on here? Just got ZKE-40

how exactly do you do this? I just took my tester out attached the v-sense cables to the screw and used 2 buss bars on each terminal with the clamp. Verify the voltage as recorded by the tester matches your reference dvom. Verify the current as recorded by the tester matches your reference...

diysolarforum.com

Also that wasn't the main issue, its that on my xuba pack 2 cells would always dip in voltage any time you draw over 70 amps and it would trigger low voltage cut-off on my overkill solar bms.

basen was not as bad but did have 1 low cell that required baby sitting this was alleviated by using an active balancer (I just left it plugged in whole time).

Luyuan was the only set of cells that ran fine, no issues, stayed in balance, you really didn't even need an active balancer for those. I paid 197/cell delivered to East Texas, a lot but better than. Now you can get these for under $150 delivered in the US.

I never had success with grade B, you always would have a few cells that were pain in the butt.

I would not use that leaky cell. Just accept as cost of DIY and move on.

By the way your 70 A issue causing voltage drop triggering BMS disconnect... Are you talking about a dynamic peak / transient of 70 A (e. G. Inrush current of a transformer or start up of a motor without soft-starter)? Or do you mean a static / constant 70 A load?

I believe LiFePO4 cells have a similar behavior as lithium titanate cells. The dynamic resistance and the static resistance are definitively not the same.

I would expect a dynamic phenomenon to trigger an undervoltage disconnect of the cells were not good. I think I also saw some pictures of the cell voltage when a peak current is delivered/requested to the battery. The voltage drop is quite a thing.

But for static load I wouldn't expect much voltage drop.

Not sure about Lifepo4, but I know lithium titanate internal resistance varied a lot as a function of temperature and SOC. I'm talking about a factor 1000 x or so.

But I guess you weren't talking about battery behavior in the cold of winter, right?

 

[Alkaline]

By the way your 70 A issue causing voltage drop triggering BMS disconnect... Are you talking about a dynamic peak / transient of 70 A (e. G. Inrush current of a transformer or start up of a motor without soft-starter)? Or do you mean a static / constant 70 A load?

I believe LiFePO4 cells have a similar behavior as lithium titanate cells. The dynamic resistance and the static resistance are definitively not the same.

I would expect a dynamic phenomenon to trigger an undervoltage disconnect of the cells were not good. I think I also saw some pictures of the cell voltage when a peak current is delivered/requested to the battery. The voltage drop is quite a thing.

But for static load I wouldn't expect much voltage drop.

Not sure about Lifepo4, but I know lithium titanate internal resistance varied a lot as a function of temperature and SOC. I'm talking about a factor 1000 x or so.

But I guess you weren't talking about battery behavior in the cold of winter, right?

I was using MPP solar LV2424 a 24v inverter, it was hooked up to a heater and lamps. When I would draw appx 1500 watts my cells 5 and 7 would drop voltage to like 3.08 while the other cells were like 3.22. I cleaned, sanded smooth, top balance, bottom balanced, replaced bus bars everything did not matter those 2 cells were POS and never worked when you had any significant load.

According to the XaoXing app for the overkill solar BMS it was drawing about 66-69 amps at 1500 watts.

Cells 5 and 7 would always be the first to say full charged (they were not...) and also the first say, ohh no we are dead.

I hated them...?

Basically to keep those 2 clown cells happy you could not draw more than 45 amps.

 

[silverstone]

I was using MPP solar LV2424 a 24v inverter, it was hooked up to a heater and lamps. When I would draw appx 1500 watts my cells 5 and 7 would drop voltage to like 3.08 while the other cells were like 3.22. I cleaned, sanded smooth, top balance, bottom balanced, replaced bus bars everything did not matter those 2 cells were POS and never worked when you had any significant load.

According to the XaoXing app for the overkill solar BMS it was drawing about 66-69 amps at 1500 watts.

Cells 5 and 7 would always be the first to say full charged (they were not...) and also the first say, ohh no we are dead.

I hated them...?

Basically to keep those 2 clown cells happy you could not draw more than 45 amps.

Well with heater and lamps, so static resistive load, it's Kind of the best case scenario actually for a test (unless a bunch of leds, that's another topic).

So yeah, definitively bad, sorry to hear that ?. The internal resistance was definitively too high for those 2 cells. Did you at least manage to get it working with say 80% capacity at reduced current, or it was a complete no-go?

My lazy approach would be tomorrow to hook up the latest big wires hopefully to the battery. Then try to put the self clinching nut in the busbar with the arbor press, so that I can connect the balance leads.

Fast charge to say 90% SOC the 16 cells in series. Then for the last bit, charge very slowly at say 1A-2A up to 16*3.65V and let the active balancer of the jk bms do it's job. Then connect the inverter and get this show started within the next 7-14 days.

 

[Alkaline]

I sold all 3 of the batt packs for $500/each for xuba / basen and $700 for the luyuan.

Will be doing a 16s build soon, already have REPT 280 grade A cells, just need to finalize an enclosure probably looking at the white eel box and I have the sungold copy of the LV6548 for the inverter.

 

[silverstone]

Small update ... So I torqued the cells tight. I cannot pull out the spacers now at least and I felt some resistance in the wrench. I didn't squeeze the hell out of them though.

I just hope it's NOT too tight. Remember, these cells are at approx 50% SOC ...

 

[Daddy Tanuki]

I don't think I have the budget for ordering so many more right now...

Who does ?

But then again capacity of the pack will be dictated by the lowest capacity cell, so whichever will cause the whole pack to hit overvoltage or undervoltage limit. As long as you don't hit those you should be fine. Or are you throwing the lower capacity cell away?

no i will use the "lower capacity cells" in a different project probably for my camper.

My problem is that I'm not really saving a lot of money compared to off the shelf solution nowadays. And I already used several months on it

I go back and forth on this, with ordering directly from china shipping and customs might as well be free (250 USD to ship 32 to yokohama, and then about another 250 USD in import taxes.) yet all the premade server rack types I have seen get a different rate for shipping and import here....

 

[silverstone]

Who does ?

no i will use the "lower capacity cells" in a different project probably for my camper.



I go back and forth on this, with ordering directly from china shipping and customs might as well be free (250 USD to ship 32 to yokohama, and then about another 250 USD in import taxes.) yet all the premade server rack types I have seen get a different rate for shipping and import here....

Well I'm already thinking about the next battery without even having finished this one .

I already have 30 kWh of capacity (or whatever it is in the end ...) / 11 kW of power so I'm NOT power limited already I'd say. Yes there can be edge cases (during maintenance etc) but even if I have heat pump running, stove etc, I don't think I'll go above 6kW right now ...

So the biggest cost-outs for the next batteries might be:
- (Wooden box instead of metal ? Not too keen on this one I'd say)
- Go with 35mm2 INTERNAL wiring instead of 50mm2 or ... -> easier to bend
- Go with 16mm2 INTERNAL wiring and downsize to either 63A or 2x63A in parallel -> MUCH easier to bend
- Avoid all custom crimped connectors Amphenol RT360 (right now I probably have 60 EUR for each 4-23 pins connectors and have 5 of them ...)
- No precharge circuit in each battery. -> Mostly space saving
- No contactor in each battery -> BIG SAVING
- Just 1-2 extra Output breakers for maintenance and tinkering. Well these are 63A so not that expensive to be honest ... I'll have a 100A "head" breaker anyway (already ordered)

Cost-In for "mental sanity":
- Use WAGO 285-135 Terminal Blocks for Busbar_BMS_Neg, Busbar_Bat_Net and Busbar_Bat_Pos. You cannot imagine all the cursing I had while trying to fit all cables to the Negative terminal (output of the BMS). No **** space. This will do for up to 35mm2 and 125A.

This would be more "bare-bone" version of the first battery, mainly to save time, a bit of cost and a lot of frustration. Smoke Detector + Trip unit on the breaker will be "last line of defense, so I'll keep that just to be safe.

The lack of precharge circuit is NOT optimal in my opinion, but I can probably precharge on the side of the inverters. Or have another battery connected.

The smaller wire cross-section in comparison to the current 35-50 mm2 will cause some current unbalance, so I'll probably need to tell the inverter to go from 200A to 250A only so that things are kept under control. Or just leave it at 200A. No need to fully charge the battery in 3h in summer time anyway. And it will NOT happen in automn/spring that it charges so fast to begin with ...

Given the relatively high cost of cells I decided to invest in wiring breakers etc, but to be honest I think I would have built 3 x 15kWh batteries if I only used 16mm2 and 63A Breaker. And still reached more or less the same target for power. 200 EUR just for the 50mm2 internal cable is not something I'll do again (also because WAGO 285-135 Terminal block accepts only up to 35mm2 ).

I guess you call that learning ...

 

[silverstone]

Small update: Battery 01 seems to work ok-ish. Battery 02 seems to still be quite unbalanced, with Cell 01 being quite low in voltage and Cell 16 always hitting the Overvoltage disconnect, in case the Inverter Voltage is set too high (in particular the float voltage). Problem being that if the float voltage is set too low, the cells aren't getting charged enough. Damn . I don't want to open up this battery and replace the cell, that's an absolute nightmare ... Should I consider a NEEY active balancer (4A vs JK 2A Active Balancer) ? Or possibly add a bleeding resistor on the cell that has tendency to always go high ?

 

[MisterSandals]

with Cell 01 being quite low in voltage and Cell 16 always hitting the Overvoltage disconnect, in case the Inverter Voltage is set too high (in particular the float voltage). Problem being that if the float voltage is set too low, the cells aren't getting charged enough.

What is the battery voltage and individual cell voltages when you hit over volt?
What is your top charge voltage and your float voltage?

 

[silverstone]

What is the battery voltage and individual cell voltages when you hit over volt?
What is your top charge voltage and your float voltage?

I disconnect at 3.60 V/cell. I set the recovery voltage to 3.45 V to prevent too many disconnect-reconnect and give some time to balance a bit (not much, maybe 15 minutes).

Float voltage 53.8 V is OK. 53.9 V might work. 54.0 V causes cell 16 to Overvolt.

Absorbtion / Equalising Voltages I think are 54.8-55.2 V.

53.6 V float & 54.8 V absorbtion work without issue, but I think I don't fully charge the cells with these conservative settings.

 

[MisterSandals]

Equalising Voltages I think are 54.8-55.2 V.

First off, you need to have equalization turned off (or eq time set to 0) for lithium.

For float, i prefer to set the a few hundredths of a volt below where my cells settle after a half hour. This reduces thrashing of micro cycling: charge, settle, charge , settle...

What are the cell voltages of the other cells when #16 hits 3.6V over volt? This is important to know how unbalanced your cells are and to formulate options moving forward.

 

[silverstone]

First off, you need to have equalization turned off (or eq time set to 0) for lithium.

For float, i prefer to set the a few hundredths of a volt below where my cells settle after a half hour. This reduces thrashing of micro cycling: charge, settle, charge , settle...

What are the cell voltages of the other cells when #16 hits 3.6V over volt? This is important to know how unbalanced your cells are and to formulate options moving forward.

Not sure how I could turn equalizing off..
I have equalizing voltage, equalizing days and equalizing hours in my inverter settings. Maybe I should set the same as float then?

Not sure if I remember correctly for the unbalance but quite a bit below... Mostly 3.43V I guess, some maybe 3.45V ? Cell 01 being 3.40V actually.

Actually the two battery packs are quite different in terms of voltage... Like 0.5V-1.0V difference with battery 02 always being higher (54.8V VS 53.8V for instance). I thought it could be some voltage measurement tolerance (JK bms HW v10 vs v11), but maybe it's all due to this cell 16 being always a bit high... That being said it shoots up all of a sudden, most of the time they are all around 3.45V excet cell 01 which is 3.40V.