diy solar

diy solar

Our NEW GO-TO LiFePO4 vendor...09.20.21

Bought 16 of the LF280K cells plus their JBD 7S-20S-200A BMS through Jenny from the Houston warehouse.

All cells look pristine, all were within 7 mV of each other, near 3.3v. Everything is now up and working.
The JBD BMS has no trouble keeping them in balance after I first wired them in parallel and did a top balance.

Here's some documentation on the above items:
The Android ap here is working great with the JBD BMS via bluetooth: https://jiabaidabms.com/pages/download-files

You can get cells more cheaply if shipped to you direct from China, though I did not want to wait months for them to arrive
though shipping times should be getting more reasonable if the pandemic continues to ease.
Shipping from China, Jenny may be able to personally inspect the cells for you before they go out.
My understanding is that product shipped from China that is found defective can be returned to the Houston warehouse.

Almost all reports I see in this forum indicate that product from this supplier is good.
Keep in mind that those few who do receive damaged goods are more inclined to speak up here.
Every vendor will occasionally ship something that should not have gone out, the test is in how they respond.
I made a video recording as the boxes were opened so if there was damage it would be clear they arrived that way.

It would be very easy for an end user to bloat the cells during an initial top balance
if the cells were left too long above 3.45V.

(Edited some of the URL's 10 minutes after posting)
 
Last edited:
This is my first step away from lead acid for off grid household power, so correct me if I'm wrong.
> Why would you think that?
It took me about a week or so to top balance the parallel set of 16 batteries at 15 Amps from a CC-CV supply.
My understanding is that these batteries can be fully charged if left at 3.45 volts for long periods.
Leaving them for a few days at 3.625 would overcharge them, this is why they should be
top balanced in stages, to 3.4v, then occasionally checked as it goes to 3.45, then watched carefully to 3.5 and 3.6.
Overcharging them can cause them to bloat, as could an inaccurate max voltage setting.

I don't know how high the voltage can be or for how long till significant bloat occurs on these particular cells.
Will do my best not to find out.
I suspect the next dozen responses will have a dozen different opinions on this.

Regarding QR codes: I downloaded the ap, but my cell phone camera did not do any good
with these QR codes. Print is small and apparently not the number of characters expected.
May need to review that youtube tutorial.

Regarding the JBD xiaoxiang android bluetooth ap I am using, I can't seem to set parameters
such as "Nominal Capacity", which is frozen at 200AH.
"Charge Overcurrent and "Discharge Overcurrent" are both 300A, which is fine with me.
But "Hardware Overcurrent Protection" is too low at 100A, I bought a 200A BMS for a reason.
The Overkill manual warns against changing the parameter (assuming I could).
 
Finally succeeded in typing in a QR code manually:
EVE Power, Cell, LiFePO4, LF280K, 280Ah, 26/11/2021, City: Jingmen, Factory line: 83, Factory task: 24, Number of cell: 1354
 
Tried a different phone with better camera, caught all 16 QR codes with some trouble.
All were made last December and first half of January, except one from late November.
Perhaps some spec didn't quite make it into somebody's A+ list.
So far I am very happy, don't see anything to complain about.

These were ordered on June 1, arrived via UPS parcel service from Houston on June 7.
We're in a remote corner of Oregon, that may have added a day or two.
Packed two to a box with lots of styrofoam, 26 lbs each box, extremely well protected.
Apparently they might also arrive via an LTL (less than truck) service, a truck with a lift-gate.

manufacturers product_types cell_types model_codes capacity production_date production_city factory_product_line factory_task_order number_of_cell_produced_that_day original_qr_content
EVE Power Cell LiFePO4 LF280K 280Ah 14/1/2022 Jingmen 84 38 8856 04QCB76843800JC1E0008856
EVE Power Cell LiFePO4 LF280K 280Ah 16/12/2021 Jingmen 84 69 5142 04QCB76846900JBCG0005142
EVE Power Cell LiFePO4 LF280K 280Ah 26/11/2021 Jingmen 83 24 1354 04QCB76832400JBBT0001354
EVE Power Cell LiFePO4 LF280K 280Ah 15/12/2021 Jingmen 84 68 6098 04QCB76846800JBCF0006098
EVE Power Cell LiFePO4 LF280K 280Ah 15/1/2022 Jingmen 83 39 1188 04QCB76833900JC1F0001188
EVE Power Cell LiFePO4 LF280K 280Ah 14/1/2022 Jingmen 84 38 8974 04QCB76843800JC1E0008974
EVE Power Cell LiFePO4 LF280K 280Ah 16/12/2021 Jingmen 83 68 1595 04QCB76836800JBCG0001595
EVE Power Cell LiFePO4 LF280K 280Ah 6/1/2022 Jingmen 84 25 6896 04QCB76842500JC160006896
EVE Power Cell LiFePO4 LF280K 280Ah 7/1/2022 Jingmen 84 25 6212 04QCB76842500JC170006212
EVE Power Cell LiFePO4 LF280K 280Ah 10/12/2021 Jingmen 84 66 6278 04QCB76846600JBCA0006278
EVE Power Cell LiFePO4 LF280K 280Ah 9/1/2022 Jingmen 83 35 362 04QCB76833500JC190000362
EVE Power Cell LiFePO4 LF280K 280Ah 29/12/2021 Jingmen 83 61 371 04QCB76836100JBCX0000371
EVE Power Cell LiFePO4 LF280K 280Ah 4/12/2021 Jingmen 83 64 629 04QCB76836400JBC40000629
EVE Power Cell LiFePO4 LF280K 280Ah 15/1/2022 Jingmen 84 39 5426 04QCB76843900JC1F0005426
EVE Power Cell LiFePO4 LF280K 280Ah 26/12/2021 Jingmen 84 19 6830 04QCB76841900JBCT0006830
EVE Power Cell LiFePO4 LF280K 280Ah 15/1/2022 Jingmen 83 38 143 04QCB76833800JC1F0000143
 
It took me about a week or so to top balance the parallel set of 16 batteries at 15 Amps from a CC-CV supply.
My understanding is that these batteries can be fully charged if left at 3.45 volts for long periods.
Leaving them for a few days at 3.625 would overcharge them, this is why they should be
top balanced in stages, to 3.4v, then occasionally checked as it goes to 3.45, then watched carefully to 3.5 and 3.6.
Overcharging them can cause them to bloat, as could an inaccurate max voltage setting.

I'm not sure where you are getting your information, but my experience with 48 EVE cells says:
3.45v on a cell is 98 to 99% full.
3.65v on a cell will not cause bloating. Left there for extended periods, it will cause degradation, don't leave them at 100% charge for extended periods (i.e. more than a day or two).
The "stages" theory of top balancing is just silly. Charge them up as quickly as possible. Best way to do this is to use a BMS and charge a pack until high voltage disconnect (you did test that works on your BMS first, right?). Then you can parallel and finish the top balance.



I don't know how high the voltage can be or for how long till significant bloat occurs on these particular cells.

As long as your measurement is accurate, EVE cells will NOT bloat at 3.65v.

Regarding QR codes: I downloaded the ap, but my cell phone camera did not do any good
with these QR codes. Print is small and apparently not the number of characters expected.
May need to review that youtube tutorial.

Regarding the JBD xiaoxiang android bluetooth ap I am using, I can't seem to set parameters
such as "Nominal Capacity", which is frozen at 200AH.
"Charge Overcurrent and "Discharge Overcurrent" are both 300A, which is fine with me.
But "Hardware Overcurrent Protection" is too low at 100A, I bought a 200A BMS for a reason.
The Overkill manual warns against changing the parameter (assuming I could).

Older versions of the app have problems with the newer BMS. Even then, some settings need to be changed using the PC software.
If you are using Apple, you have to pay for a version that is out of date, and that will allow you to change settings.

 
John, thanks for the reply!
I'll try the PC software next, see if I can clean up the BMS settings.
Eventually would prefer this on a chromebook under linux: https://github.com/sshoecraft/jbdtool

I am very curious if my BMS will cut out if the discharge load exceeds the
"Hardware Overcurrent Protection" of 100A even though the "Discharge Overcurrent" is at 300A
I'm worried that my Magnum PT-100 PV charger (and MS4448PAE inverter)
will blow due to a voltage spike when the BMS drops out with some obscure fault when charging at 100A.
Maybe best to add a series string of 4 car batteries in parallel with the LiFePO4 bank and BMS
to absorb such a spike?

Like I say, lots of opinions out there on how to top balance.
My BMS shows all cells in balance (verified with a good DVM) so I did something right.
For now, I'll stick with slow and conservative.

Haven't seen much said on exactly what happens if you leave it for a day at 3.65v,
just plenty of warnings not to do so (and others who don't seem to think it's an issue).
Does Eve say how long that 3.45v is on the cell?
My guess is the cell is at their 98 or 99% the minute it arrives at 3.45v, and that after many hours
it will be at 100%. But the cell is at 100% the moment the voltage reaches 3.65v.
This will vary a bit depending on the charging current used and temperatiure.

Spent a week balancing 16 cells in parallel using one of these for monitoring with a DVM.
$12usd: https://www.amazon.com/Aceirmc-Converter-Adjustable-Regulator-Protection/dp/B0823MM1DV
Worked well enough but voltage setting was a bit touchy hence my mention of inaccuracies.
Calibrated the current by measuring mV across a foot of 14awg copper house wiring, 2.58 ohms/1000ft.
Also used 14awg to wire the cells in parallel, but had 30mv of drop with the currents involved so would advise
using the busbars instead.

Using four red 80x30mm die springs in a compression box, likely another case of being overly conservative.
Here's a good source: https://www.amazon.com/gp/product/B07GSNCYNK
The cells want 300kgf, so 75kg per spring, spring constant is 7.03kg/mm, so spring deflection is 75/7.03=10.67mm
Put roofing felt (tarpaper) between cells to make certain the metal cases didn't short through the thin plastic covers.
Rods are rebar with 1/2" bolts welded to one end (what I had on hand), a hole drilled in the other for
a washer and a nail. 3/4" plywood endplates, but still needs channel iron braces (drilled for the rods)
to keep it from bending. I'm building double wide, planning to eventually go 2P16S. For a single row of cells,
could get by with the yellow 80x30mm die springs at 1.8kg/mm. (Edit: Best to use all red springs. See post #884.)

Used carbon grease on the bus bars to ensure good contact, but wouldn't do it again.
Messy stuff, whenever I work on the batteries now I have to spend 15 minutes
cleaning up my hands and anything they touched.
 
Last edited:
I am very curious if my BMS will cut out if the discharge load exceeds the
"Hardware Overcurrent Protection" of 100A even though the "Discharge Overcurrent" is at 300A
I'm worried that my Magnum PT-100 PV charger (and MS4448PAE inverter)
will blow due to a voltage spike when the BMS drops out with some obscure fault when charging at 100A.
Maybe best to add a series string of 4 car batteries in parallel with the LiFePO4 bank and BMS
to absorb such a spike?
I can answer this in regards to the inverter, though I use a different SCC. My DALY BMS all shutdown several time due to a mistaken settings on my SCC. (Morningstar TS MPPT-60 x four) The MS4448PAE showed a spike when the BMS's shut down.

In my case one BMS shutdown, then a second one then the last shut down and the inverter and the BMK recorded a 75 volt spike. (you could see this as the incoming amps dropped with each BMS shutting down.)
This happened several times as the inverter tried to power the house straight off of the panels without the batteries in line. they literally fluttered off and on 30 or 40 times in the case of one afternoon while we were gone from the cabin.

This did no damage to the inverter. (this was not the inverter taken out by the lightening strike, it was the replacement unit.)

obviously I would not recommend doing this, but mine survived the above three times in one week before we got back and I discovered the issue and corrected it.

YMMV.
 
Ken, Thanks for the data point. Indeed, "YMMV". Lots of variables here, including how much current the charge controller is sourcing when the fault occurs, how fast the electronics in your charge controller can react to reduce current flow, how large the caps are in the charge controller and inverter.
Would be nice to have a simple, cheap, small, generic, and reliable cure for this. Perhaps a high voltage sense that turns on a bunch of FET's to divert power to a dump load? Should be momentary, so not a lot of energy needs to be dissipated in that dump load, and the FET's could be relatively wimpy. I'm not keen to keep a lead acid battery in parallel with the LiFePO4 bank due to cost, size, and maintenance issues.
 
Ken, Thanks for the data point. Indeed, "YMMV". Lots of variables here, including how much current the charge controller is sourcing when the fault occurs, how fast the electronics in your charge controller can react to reduce current flow, how large the caps are in the charge controller and inverter.
Would be nice to have a simple, cheap, small, generic, and reliable cure for this. Perhaps a high voltage sense that turns on a bunch of FET's to divert power to a dump load? Should be momentary, so not a lot of energy needs to be dissipated in that dump load, and the FET's could be relatively wimpy. I'm not keen to keep a lead acid battery in parallel with the LiFePO4 bank due to cost, size, and maintenance issues.
Well it worried me immensely when it happened but after I figured out the settings i no longer have that particular issue. basically with 4 solar controllers I had programmed three of them properly but the last one I somehow did not save the program changes. this was fine until I added my new panels to the mix and that controller ended up with 2.4kw on it vice the 1.2kw it started with. it was charging so fast that it would start the above issues at about 11:00 am as it would have charged the batteries by then, and went into trying to overcharge which the BMS's would then drop out to protect the packs. (thank you sweet baby Jesus.) once I figured out that the one controller was not matching the others and reprogrammed it to match the others problem was solved.
 
Good to hear that you found a cure for your problem. But there are many potential faults that could cause the BMS to cut out the battery, perhaps just a bad crimp job on a sense wire to one of the cells. (I crimped and then soldered, but there's 100 other potential points of failure.) Have seen several reports in the forum of blown inverters and/or chargers from such a spike, would be odd if nobody is selling a good solution to this.
I asked Magnum/Sensata via email (once, should pester them more), no response yet. Asked Backwoods Solar from whom I bought the Magnum gear, their response was that I should always throw the PV breaker before disconnecting the battery.
 
I asked Magnum/Sensata via email (once, should pester them more), no response yet.
Yeah when I bought my first Magnum about 12 years ago they were the best thing since sliced bread. would answer quickly if you had an issue they would remedy it. then they got bought out by Sensata and its been a downhill slide for them.

Honest to god if i had not already invested in them I would jump to another 1st tier maker. but I have two of the MSPAE 4448's a MS2012 (thats 12 years old and still chugging along) along with all of the accessory modules . Both of the 4448's I bought after Sensata bought them out. When the 1st one went down from a lightening strike I immediately ordered another and at the same time asked them about returning it for repairs (not under warranty either) and got led around for about 6 weeks before I lost my cool and ripped several of them new backsides via email and via phone. I have not contacted them since but I am not that happy with their service anymore. the units still seem pretty robust but their after service has decidedly went downhill.
 
Yeah when I bought my first Magnum about 12 years ago they were the best thing since sliced bread. would answer quickly if you had an issue they would remedy it. then they got bought out by Sensata and its been a downhill slide for them.

Honest to god if i had not already invested in them I would jump to another 1st tier maker. but I have two of the MSPAE 4448's a MS2012 (thats 12 years old and still chugging along) along with all of the accessory modules . Both of the 4448's I bought after Sensata bought them out. When the 1st one went down from a lightening strike I immediately ordered another and at the same time asked them about returning it for repairs (not under warranty either) and got led around for about 6 weeks before I lost my cool and ripped several of them new backsides via email and via phone. I have not contacted them since but I am not that happy with their service anymore. the units still seem pretty robust but their after service has decidedly went downhill.

Bought the MS4448PAE in 2013, the PT-100 in 2016, both units have been rock solid. Was still seeing a quick and complete response to questions last time something came up in 2020.
Regarding your problem, the three BMS's shut down one by one. Outcome may have been different if they all shut down at once, causing a much bigger spike.
 
Hi all.
For the ones in Europe, it seems Docan has a warehouse in Poland now ?.
We can have cells in less than a week.
Nice
 
I meant to be adding to this thread but have been busy with other things. Anyway, I ordered 16x 280's from Jenny out of China on May 26th and received them July 6th and 7th. They all look really nice (no bulging, dings, or dents) and are the same voltage (3.33). I wish I had asked to buy some extra buss bars so I could top balance them all together but oh well. We'll see how they test out but overall I am satisfied with Docan and Jenny. Just wanted to share my experience.
 
You can use 10g wire with a couple of lugs for top balancing - it's all low current anyway.
It can take a LONG time, so most will connect in series with the bms and charge them to 90-95%, then do the top balance. You'll probly still have to do some fine tuning.
 
That first buy of 16 LF280K cells plus the JBD 200A BMS from Houston have been working great.
I have now placed a buy of 16 additional LF280K cells plus 16 extra busbars to be shipped from China.
The following describes how my buy process went.
Should be the same for any customer in the continental US.
Prices haven't changed for nearly 6 months now.

Each cell comes with one busbar, I bought 16 additional busbars at $0.75 each in the second buy,
needed to connect all 32 cells in a 2p16s configuration of two rows.
The busbars are sized for a single row of cells, using those busbars on two rows of LF280K cells
leaves a one inch gap between the rows. The older version of the EVE 280AH cell has the terminals
placed differently, and would have no gap between rows.

Each cell shipped from Houston costs $149.00 and the JBD 200A BMS costs $145.00, shipping included.
The price for cells shipped from China is lower at $110.00, but shipping is not included in that price.
Whether you have them shipped from Houston or China, you can work with Jenny Wu: jennywu896@gmail.com
or Amy Zheng: amy@docanpower.com who I believe are both based in China and both employees of Docan.
You could buy through the Houston website at www.powerwholesale.net without a salesperson,
but the price is the same and you then don't have someone at Docan to answer questions and
advocate for your interests. I believe they can inspect cells that ship from China (but not Houston).

Alibaba charges a fee of 2.04% on these transactions whether shipping from Houston or China.
You could pay for the merchandise, any shipping fees, plus the Alibaba fee with a wire transfer.
Using a credit card is safer but incurs an extra 3.00% charge from Alibaba to cover credit card fees.
Buying direct from Alibaba with my Mastercard failed, so instead bought through a paypal account
and paid paypal with that same credit card, still just an extra 3.00% charge.


First buy of 16x EVE LF280K cells plus the JBD 200A BMS, ordered May 31 2022, shipped from Houston:
Total value of merchandise: 16*149.00 + 145.00 = $2529.00
Shipping from Houston via UPS: $0
Alibaba transaction fee: $51.60 (2.04% of $2529.00)
Additional fee for using a credit card: $77.16 (3.00% of $2529.00+51.60)
Total cost as billed to my credit card: 2529.00+51.60+77.16 = $2657.76

Second buy of 16x EVE LF280K cells plus 16 extra busbars, ordered on July 20, 2022, shipped from China:
Total value of merchandise: 16*110.00 + 16*0.75 = $1772.00
Shipping fee from China, arrives via UPS: $276.00
Alibaba transaction fee: $41.80 (2.04% of $1772.00+276.00)
Additional fee for using a credit card: $62.49 (3.00% of 1772.00+276.00+41.80)
Total cost as billed to my credit card: 1772.00+276.00+41.80+62.49 = $2152.29


Total cost of just the 16 EVE LF280K cells shipped from Houston on a credit card including fees
should be around 16*149*1.0204*1.03 = $2505.61
Total cost of just the 16 EVE LF280K cells shipped from China on a credit card including fees
should be around (16*110+276)*1.0204*1.03 = $2139.86
So buying from China saves 2505.61 - 2139.86 = $365.75 on 16 cells
but takes a month or two to arrive instead of a week or less from Houston.
Regardless whether you buy from China or Houston, report any problems to your
salesperson (Jenny or Amy), and any defective product is shipped to Houston.

Well, that's how I understood the transaction, let me know of any errors in the above.

Edit: I'm in the state of Oregon, which has no sales tax.
 
Last edited:
Just John wrote: The "stages" theory of top balancing is just silly. Charge them up as quickly as possible. Best way to do this is to use a BMS and charge a pack until high voltage disconnect (you did test that works on your BMS first, right?). Then you can parallel and finish the top balance.

Browneye wrote: It can take a LONG time, so most will connect in series with the bms and charge them to 90-95%, then do the top balance. You'll probly still have to do some fine tuning.

To which I say: I wouldn't trust several thousand dollars worth of new cells to a new BMS. To thoroughly verify the BMS would require verifying that it cuts out when any one of my 16 cells hits the max voltage. What's more, given how opaque the manuals and software tools are, I wouldn't trust that I even understand my BMS on a first attempt at use. I am far more confident in a cheap CC-CV power supply that I can easily test for proper function. Also, when the BMS contactor cuts out to shut down charging when that first cell hits 3.6 Volts, will the resultant voltage spike blow my $1000 charger? I'm quite fine with spending a week waiting for the voltage to arrive to 3.4v, then watch it closely as I step it higher. Having browsed the forum some, my impression is that most do an initial parallel top balance just like I did. But clearly opinions vary, on this and seemingly every other tidbit of advice on LiFePO4's.

Unless the cells arrive pretty much top balanced already, doing a series balance as described by Just John and Browneye only gets you to where the first cell is topped out. You then have to take it out and repeat for each of the other cells. Wiring in parallel seems far more straightforward. On this next set of 16 cells, I will parallel balance using runs of 10awg copper down the cells, going down one side of the screw terminals then looping back on the other side of those same screw terminals to halve the resistance even further and keep things mechanically balanced.

However, if you are doing many somewhat matched cells and know your equipment, doing an initial series charge could be the best way to go. The JBD BMS I have only does passive balancing at around 200ma. If you have a BMS with active balancing at several Amps it might fine to let the BMS handle the entire initial top balance, but with a reduced charging current after the BMS contactor first cuts out due to a high cell voltage.

My first batch of 16 cells were very close in voltage, all within 10mv. If they aren't that closely matched, might be best to first wire them up in parallel
with 12 inch lengths of 14awg wire till they all even out. At 2.58 ohms per 1000 ft, the two lengths of 12 inch wire between two adjacent cells would have a total of 0.00258*2 Ohms resistance. An assumed worst case difference of 0.5 Volts between cells would cause a current of (0.5 Volts)/(2*0.00258 Ohms) = 97 Amps and each 12 inch wire would be dissipating half of (0.5 Volts * 97 Amps), or about 24 Watts.

Once the cell voltages agree within 10mV, then wire them all in parallel with that 10awg wire as suggested by Browneye. The EVE datasheet says my LF280K cells have an internal resistance of around 0.00025 Ohms, which should keep currents from a 10mV difference well below 100 Amps. The low resistance of 10awg copper wire should keep all cells very close to the same voltage even when driving the array of cells with 20 Amps from my cheap CC-CV supply.: https://www.amazon.com/Aceirmc-Converter-Adjustable-Regulator-Protection/dp/B0823MM1DV

(The bottom three paragraphs were severely edited 8 hours after first posting)
 
Last edited:
At this point it may be a bit redundant, but since I was only able to make this purchase because others shared their experiences, I thought I would add my purchase experience with Jenny at Docan to the thread.

I have purchased two batches of 36 EVE 304ah cells from Docan.

The first was ordered on November 18th and paid with Pay Pal. No payment issues. Jenny told me that these were not in stock in the U.S. and would have to be shipped from China, so I should expect to wait at least a couple months. I emailed on January 19th, and she responded that they were in route to the U.S. Another month had almost past, so I planned to email again on February 19th, however when I arrived home from work on February 17th, a bunch of boxes were in front of my garage door. Happy Day! I'm glad my UPS driver was conscientious enough to bring the boxes up right by the door in front of my pickup, because I had no idea they were coming and they could have been sitting there for a while. So lack of communication on the expected arrival date was the only hiccup in this transaction. No damage. Did the cursory voltage check and weighed each cell to compare with the specification, and everything looked good.

The second was ordered on April 4th. I tried to pay with Pay Pal again, but Alibaba would not let me pay the entire transaction with Pay Pal. I went through several days of song and dance with Alibaba, using different credit cards, calling the credit card customer service to see if they were blocking the transaction, sending photo I.D. to Alibaba, and making some smaller Alibaba purchases with the card I wanted to use. Finally, with no explanation, they accepted the card they had previously declined. This order also had to be shipped from China. This time Jenny had emailed me with the approximate delivery date of June 5th to 10th. They arrived on June 5th, and all was good once again.

Once I heard about the LiFePoQR app, I scanned the codes on all of the battery cells to get the manufacture dates.
The first batch, ordered on Nov 18, 2021, consisted mostly of cells manufactured during the last week of October 2021 or the first week of November 2021. There was an exception with two of the cells that were manufactured on June 27, 2021.
The second batch, ordered on April 4, 2022, consisted of cells manufactured in the first week of March 2022, with the exception of one cell which was manufactured on February 12, 2022.

All things considered, I am pleased with this purchase. No egregiously old cells and no obvious issues.

After they have sat in my basement for a few months, I have finally put together one 4S battery, which I plan to use it with an iCharger X8 to capacity test some of the other cells. I ran one cycle on the battery, and it appears to have capacity in excess of 304ah, but I have some uncertainty over the exact number, so I'll leave those questions for a different post.
 
Last edited:
All things considered, I am pleased with this purchase. No egregiously old cells and no obvious issues.

Good to hear things went well with your two buys. Will be interesting to see how long my second buy takes to arrive from China. Hopefully covid becomes less an issue and things start moving faster. Yes, mine were just dropped off by UPS too, no signature required.
 
Just John wrote: The "stages" theory of top balancing is just silly. Charge them up as quickly as possible. Best way to do this is to use a BMS and charge a pack until high voltage disconnect (you did test that works on your BMS first, right?). Then you can parallel and finish the top balance.

Browneye wrote: It can take a LONG time, so most will connect in series with the bms and charge them to 90-95%, then do the top balance. You'll probly still have to do some fine tuning.

To which I say: I wouldn't trust several thousand dollars worth of new cells to a new BMS. >snip

You're supposed to monitor the process when you do this, not just leave it to its own, so you don't fry "thousands of dollars worth of new cells."

At the end you contradict yourself and recommend a series charge to get the ball rolling. Actually, the old guard recommend you charge to like 3.4 or 3.5 to near zero current, THEN wire parallel and finish the top balance to zero current at 3.65V.

My point was that 8 x 280A cells could take an interminable length of time to parallel top balance with five or ten amps. I had 4 x 230 and it took about four days with a 30V-10A bench supply. The first mistake I made was thinking it could be done in stair-step stages, something I saw recommended that made sense to me, and so you could monitor progress and not over-volt them. So I started at like 3.5V charging. But when the cells get close to full then current tapers way off, I was sitting at like a couple of amps, for hours - a day. So I restarted them at the full 3.65V and they charged all the way up to zero current. It took so long I thought something was wrong.
 
> At the end you contradict yourself

IF you know your equipment well, go ahead and try variations on the series charge
to do an initial top balance, I won't argue if that's what you want to do.
It can make sense if you have lots of cells and/or not much time.
Perhaps finish the cells off using that series connection with a 10A charge current
using a BMS capable of several Amps of active balance as I had suggested.
You only have to wire up the cells once. Seems ideal.

Do you know if there is a voltage spike when the BMS contactor cuts out due
to your 100A charger not reacting fast enough to reduce the current down to 0A?
Several instances of this blowing out chargers and/or inverters here on the forum.
I consider the BMS contactor to be a good fail safe to have, but not exactly a tool
to depend on.

Telling everybody with untried BMS devices they don't fully understand
to just go for it seems not quite so ideal.
Most of us punters are better off doing a parallel charge from the start.
A week spent charging them is not much of a burden if the cells take months to arrive.

You are right, a parallel charge at 3.4 volts tails off considerably once the cells reach 3.35 or so.
Most of this is due to the resistance in the wires, most of that in how you wire up the cells
(even if using 10awg wire), and some inside the power supply (after the point it senses the voltage).
A little bit would be due to internal resistance of the cells, but that should be a fraction of a mOhm.
Yes, I did cheat a bit on my recommendation, bumped the supply up to 3.5V after seeing it crawl
along toward 3.4V for a day or two. But once up at a solid 3.4V, it's just a few hours up to 3.5V
and then minutes to 3.6V, so monitor closely.

If you don't think the cells will get damaged if held at 3.65V for long periods, then just
set your supply to 3.65 and check it once every 12 hours during the parallel top balance.
If somewhat more cautious like me, don't do that.
 
(First post...)
I am pretty fully researched at this point and ready to start acquiring system components.
I see Docan and Jenny, I've seen the issues of cell grades and I'm down to just shopping price and a reliable transaction.
I came across Gobel and their $105 for the EVE 280ah K model shipped from China. What I found interesting is in the link to their site describing what I understand to be that ALL of the supplier cells are what are known as "B" grade and that the the true "perfect" cells are all absorbed directly through EVE's big business contracts. Note their last bold paragraph below, they describe the ratings as:

"Cell Grade:

In our system, Grade HSEV>LSEV>ESS.

Grade HSEV are the best quality cells, which are sold in EVE official shop. HSEV cells have strict matching standard, so price is pretty high.

EVE sell ESS cells to distributor such as Gobel Power. These cells are the second best quality cells you can get in the market.

Gobel Power do some tests and group better matching cells toghter, these are LSEV cells. LSEV cells have better matching standards than ESS cells.

If not all, 99% distributors have only ESS grade cells, no matter how they call these cells, Grade A+ or Grade A, they are ESS cells.

To discriminate HSEV cells from ESS cells, EVE starts to print B Mark on ESS cells, these cells are same as Intact QR cells, The only difference is QR.

Why EVE do it? In previous months, some distributors claim their cells are HSEV cells, but actually they are ESS cells. At that moment, Both HSEV and ESS cells have Intact QR, the only difference is HSEV cells have official testing report (Excel or Paper format), but testing report is easy to counterfeit, so EVE starts to print B Mark on all ESS cells."


If "ESS" or "B" cells are what we get from all of these distributors, so be it, I figure if they are undamaged, fresh and meet capacity and performance.
My questions come to this:
Anybody have transactions with Gobel?
Have "we" purchased a whole bunch of "ESS" cells represented as "A" cells?
At $105 + shipping, if these are indeed the same as the best of what "we" have been getting, it appears something just under $4000 for 32 of these all said and done. Am I seeing a best price story or a lesser quality story here?

I quoted what I thought mattered, but here's the link:
 
At $105 + shipping, if these are indeed the same as the best of what "we" have been getting, it appears something just under $4000 for 32 of these all said and done. Am I seeing a best price story or a lesser quality story here?
I mean it's $5 cheaper than Docan who many people (incl. myself) have bought from and been happy with. Worth saving $5 a battery? Up to you man.
 
I mean it's $5 cheaper than Docan who many people (incl. myself) have bought from and been happy with. Worth saving $5 a battery? Up to you man.
I show $4768 delivered from Docan.
From Gobel, I'm doing some easy math of $3360 + a scientific wild ass guess of $550 in shipping = $3910
I haven't run it on the calculator but it looks more than $5 per cell ;) , hence my inquiry.
 
I show $4768 delivered from Docan.
From Gobel, I'm doing some easy math of $3360 + a scientific wild ass guess of $550 in shipping = $3910
I haven't run it on the calculator but it looks more than $5 per cell ;) , hence my inquiry.
The price from Docan shipped from Houston for 32 LF32K cells is indeed $149*32 = $4768,
plus a 2.04% alibaba fee plus a 3% fee if you use a credit card. Plus sales tax if any.
The price from Docan shipped direct from China is $110*32+shipping+2.04%+3%.
See my post of July 21 for more details. I suspect that comes out quite close to Gobel's.

There is no legal definition of A+, A, or B, vendors are free to claim whatever they wish.
As you say it doesn't much matter, all I want is something that works for off-grid power.
Who I buy from comes down to vendor reputation. Some people thought they were getting
"A" cells from alibaba vendors and wound up with obviously used or damaged cells.

How things like air compressors and automotive jump starters get sold here in the US can be
a bit creative too, the specs many vendors claim violate the laws of physics.
There seems to be no legal consequence.

Yes, I can believe electric vehicle makers get to cherry pick the best cells when buying
several million. When only buying a dozen or two, we pay double that $110 price if
we want the same grade of cells since it's so much harder to deal with so many different
(and often bumbling) customers. What I got from Docan is new, unblemished, and works well.
But other vendors have figured out that they can sell used cells since it is so much hassle
to return defective product.

As an example of slightly mismatched cells, here's what I see with my 32 cell bank at 54 and 56 volts.
At 54v: 3.367 3.382 3.373 3.365 3.386 3.365 3.359 3.378 3.368 3.368 3.367 3.367 3.377 3.366 3.380 3.376
At 56v: 3.508 3.480 3.482 3.523 3.473 3.521 3.530 3.483 3.492 3.496 3.509 3.504 3.482 3.500 3.477 3.483
Note that at 54v, cell #5 is the highest of the bunch at 3.386v and cell #7 is lowest at 3.359v.
But at 56v, cell #5 is lowest at 3.473v and cell #7 is highest at 3.530v.
Currents were very low when these readings were taken, so not a matter of internal resistance.
I would guess this is due to slightly different chemistries of the cells, and this flipping around
will cause my passive BMS balancer to waste a few watt-hours. I'm ok with that, would much rather
have twice the cells for the same money than save a handful of watt hours. If you care, an active BMS
wouldn't waste much power at all, as it moves the power from one cell to another instead of burning it up
in a resistor.

Cells 5 and 6 were built on Dec 10 of 2021 and Jan 6 of 2022, respectively. From the QR codes:
manufacturers product_types cell_types model_codes capacity production_date production_city factory_product_line factory_task_order number_of_cell_produced_that_day original_qr_content
#5: EVE Power Cell LiFePO4 LF280K 280Ah 10/12/2021 Jingmen 84 66 6278 04QCB76846600JBCA0006278
#6: EVE Power Cell LiFePO4 LF280K 280Ah 6/1/2022 Jingmen 84 25 6896 04QCB76842500JC160006896
 
wild ass guess of $550 in shipping
Exactly. A wild guess. Why not get an actual shipped quote from Gobel before you make your claims?

I mean it may be cheaper, I have no idea - but I'm confused how you would think think two suppliers from China will have wildly different shipping rates? They will both ship the same way.
 
Exactly. A wild guess. Why not get an actual shipped quote from Gobel before you make your claims?

I mean it may be cheaper, I have no idea - but I'm confused how you would think think two suppliers from China will have wildly different shipping rates? They will both ship the same way.

Given the $4768 price he quoted, I assume he is buying 32 cells.
My buy of 16 cells from Docan shipped from China was $110 each plus $276 for shipping,
so his figure of $550 is about right.

What he missed was that the Docan price of $149/cell is the price shipped from Houston,
arriving in a week. For $110/cell, Docan can ship to you from China, plus $276/16=$17.25
for shipping, so 127.25 per cell to make a fair comparison of the $149 price from Huston.
Of course, shipping from China will take 1 to 3 months.

Also missing in his calculations are the 2.04% alibaba fee and then the 3.00% fee for credit card use.
That's on top of the prices quoted by Docan, and likely Gobel as well, doesn't matter if it
is shipped from Houston or China. See my July 21 post here for more details.

A quick search of this forum for "Gobel" finds only a handful of references, mostly in regard
to their early QR decoder. They may be ok but I don't see anybody who has bought from them.

There have been vendors asking far less than $110.00 per cell.
When you see a price of $50/cell, best to assume it is used or damaged.
 
Given the $4768 price he quoted, I assume he is buying 32 cells.
My buy of 16 cells from Docan shipped from China was $110 each plus $276 for shipping,
so his figure of $550 is about right.

What he missed was that the Docan price of $149/cell is the price shipped from Houston,
arriving in a week. For $110/cell, Docan can ship to you from China, plus $276/16=$17.25
for shipping, so 127.25 per cell to make a fair comparison of the $149 price from Huston.
Of course, shipping from China will take 1 to 3 months.

Also missing in his calculations are the 2.04% alibaba fee and then the 3.00% fee for credit card use.
That's on top of the prices quoted by Docan, and likely Gobel as well, doesn't matter if it
is shipped from Houston or China. See my July 21 post here for more details.

A quick search of this forum for "Gobel" finds only a handful of references, mostly in regard
to their early QR decoder. They may be ok but I don't see anybody who has bought from them.

There have been vendors asking far less than $110.00 per cell.
When you see a price of $50/cell, best to assume it is used or damaged.
Ok, so if the big picture is coming in now, Docan has the big advantage, if true, of being able to return defective shipments to Houston and Gobel, it would seem, is a one way trip, take it or... haggle/beg for some adjustment. Also, all other things the same, it is a $5 difference.

On the Docan site I was not able to get or see the $110 price for shipped from China units, is that an email to Jenny Wu to get that $110?
Also, does anybody feel confident or know that bad cells can be returned to their USA warehouse?
 
This is the sixth 12 volt 280 ah battery I have done with cells from Docan, all of them balanced well. Here is a screenshot after a 72 hour top balance, 25 amp draw to inverter low voltage cut off at 11 volts and then full charge, I never saw any balancing during charging.7750AB8E-FC0C-4ED0-93BE-2EAA5347FCDE.png
 
Ok, so if the big picture is coming in now, Docan has the big advantage, if true, of being able to return defective shipments to Houston and Gobel, it would seem, is a one way trip, take it or... haggle/beg for some adjustment. Also, all other things the same, it is a $5 difference.

On the Docan site I was not able to get or see the $110 price for shipped from China units, is that an email to Jenny Wu to get that $110?
Also, does anybody feel confident or know that bad cells can be returned to their USA warehouse?
Whether you have them shipped from Houston or China, you can work with Jenny Wu: jennywu896@gmail.com
or Amy Zheng: amy@docanpower.com who I believe are both based in China and both employees of Docan.
Yes, you get the Docan price for shipping from China by sending email to your sales person.
The price for shipping from Houston is on their website at http://www.powerwholesale.net/
Jenny and Amy can help you start an order either from Houston or shipped from China.

The primary Docan advantage is that this thread has a fair number of people reporting success when buying from Docan. (Also one or two reports of trouble, but that seems rare and it isn't always be clear where the fault lies.) There have been vendors where some people got good product, and then they started shipping obviously used or damaged items. Or vendors that always ship banged up used cells after claiming they are shipping new A+ cells. Another Docan advantage is that they have the Houston warehouse, shipments within the continental US typically arrive at your door within a week of the order. Ordering from Docan and having them ship to you from China is cheaper, but it takes a couple months for the cells to arrive, this is comparable with most other alibaba vendors as very few have a warehouse in the US. (I vaguely recall seeing one other that is copying Docan's strategy).
Gobel may prove to be a perfectly fine vendor. At least the price is not suspiciously low. They did good work for the community by making the QR decoder available. But I don't see any reports in a quick search of the entire diysolarforum where somebody bought from them, perhaps I wasn't thorough enough. You could be our test case, and save $5/cell in the process.
It is my understanding that a faulty cell shipped from China by Docan can be shipped back to Houston instead of China if you wish, but you should ask your sales person for confirmation before assuming that is true. There may well be plenty of haggling involved to return cells to Docan. I made a video of my unpacking process so I would have some vague proof that they arrived damaged, but I have had no issues with any of my Docan cells.

Edit: Docan is especially interesting if you are within driving distance of Houston. You can inspect your cells when you pick them up. Or at least stand outside in the Texas heat and picket the warehouse if they do send bad cells.
 
I did try to look a Docan's website before I purchased, but didn't really find very helpful. In the U.S., the expectation is that you should be able to go to a website, see a product description and a price, figure out your shipping cost and place an order without human interaction.

When dealing with Chinese companies, even on web based commerce sites like Alibaba, they use the website kind of like brick and mortar stores use a guy standing outside of the store spinning a sign. The websites main purpose is to get you to contact them. Their web store information tends to be riddled with errors and contradictions, and is often not up-to-date. The only way to complete a transaction is to text chat with a live person online.

Also, they are often using translation software, or have limited english language skills, so don't use complex sentences and avoid compound sentences. Each sentence should have one subject, one verb, and one meaning or one question. It makes communication go much more smoothly. Of course all slang and regional terms, similes and metaphors, and any English specific idioms should be avoided.

An example of something I thought would be clear, but wasn't, occurred when Jenny contacted me to say the battery cells were being shipped. I replied "Thanks for the update". The word update was confusing, and she had to ask what I was trying to tell her. (I think she might have separated the words 'up' and 'date' which causes it to make no sense. I replied using the word 'information' instead of 'update' and then she understood.
 
Here's another thread about buying from Docan:

TLDR:
Mostly good experiences with Docan, sets them apart from almost any other vendor folks in this forum have tried.
Buying spendy A+ cells directly from EVE can be problematic, they weren't packed as well as from Docan (one report).
Can wire money to a Chinese bank avoiding alibaba and credit card fees, but zero protection so must trust your vendor.
There are Docan sales people other than Jenny and Amy that get a good reviews.
Docan prices and fee structure have changed a bit over the life of that thread, one person reported
that buying from Houston cost the same as having cells shipped from China, definitely not true now.
I don't really know about buying other than through Jenny and paying the alibaba and credit card fees,
am curious if anyone has a better way than what I posted in this thread on July 21.

Summary: I'm happy with my buy.
 
The word update was confusing,

My guess is that Jenny thought you might want something about the order updated.
So she followed through to see if further action was expected
English is tricky.

I learned enough Mandarin Chinese to be sort of conversational some 30 years ago, it's now long gone.
Fully sympathize with with anyone doing high dollar business transactions in other than their mother tongue,
dealing with all kinds of people with all sorts of expectations. For example, assuring them they will get
their cells when the ship has been sitting off the port of LA for a month due to the epidemic.
 
18650batterystore
Same or better price
Us stock
Eve welded dual helicoiled terminals
Shipped in crates if you buy enough.
My experience was free shipping plus 5% off the web price for 48pcs.
No I haven't yet tested them, inverters not up and running.
 
This is the first page I come up with for Prismatic cells from18650batterystore. Sad Besides the limited variety they are all sold out.

 
Web reviews are admittedly not reliable, but a google for "review of 18650batterystore.com" doesn't look promising. Here is one of the more alarming hits: https://www.facebook.com/18650BatteryStore/reviews Note that most of the good reviews on that webpage are suspiciously short and generic. On the other hand, this 2 year old reddit page suggests they are legit with some happy customers of vaping batteries (also mentions some other US based battery vendors) As littleharbor2 has stated, they currently have zero stock of their two kinds of prisimatic lifepo4 cells. Current prices are higher than Docan shipping from Houston. On the plus side, sounds like Maitake received his 48 cells at a good price, and they look new.
 
Back
Top