Quality batteries from EEL? Testing the EVE LF304 B-Grade cells. 

Thanks a lot for your feedback and kind words.

Thanks 👍

Thank you!

Yes, you're right. The thing forgot to mention is the BMS which was connected when we tested with the shunt. That consumes some energy as well. Temperature difference and different equipment also plays a role of course..

@@OffGridGarageAustralia The BMS might impact the Wh reading, but the Ah reading should be the same. Of course, different equipment and changes in temperature will affect both Ah and Wh reading.

I agree with you (upnorthandpersonal). So the difference in Ah between the pack (303.8) vs the cells (>307) must be in how it gets counted? So either the BMS and/or the VE shunt is wrong or both - or the EBC-A40L is wrong? I guess (hope) the VE shunt is the most precise unit here? Would be nice to calibrate the EBC-A40L or test how far off it might be at least. I would disregard cell temperatures for this test as the ambient seems similar and the low C-rate. But maybe the difference in load could give some difference, I'm not sure - again too low C-rate to bother I guess.

@@tx2gnd AH difference (~304Ah vs 307Ah) is a result of: 1. Different discharge current (higher inside a pack (discharged with different load) - higher loss inside a cell - a little bit of capacity is unavaiable at higher current). 2. Charging to less than full in a pack. 3. Measurement inaccuracy

@@volodumurkalunyak4651 I think your point number 1 can be dismissed due to the low C-rate in both cases? Regarding your second point, cell 11 was charged to at least 3.624V while in the pack and discharged to 2.495V. While tested standalone, it was charged to 3.65V and discharged to 2.5V. This is too close so there shouldn't be a >3Ah difference in my opinion. Cell 12 was charged to 3.610V while in the pack and discharged to 2.775V. Again tested from 3.65V to 2.5V standalone and providing 7.7Ah more than the pack ended up providing although not a precise nor relevant number. This cell is not as important, as it's the lowest cell that will determine the capacity (cell 11 in this example). So I would have to disagree with charging differences being the reason for the discrepancy. Your third point I agree with. The question is what tool to trust the most.

Thanks 👍 It's time for some smiles 😊

Wife approval is the hardest to get. Don't forget the links on my website to support the channel: off-grid-garage.com/diy-kits/

Thank you😊

Yes, I explained this not quite right with the capacity loss. In the first full pack test, there was also the BMS still connected and doing it's thing while in the second test, we have tested the bare cells. Secondly, we used two different methods, currents (as you say) and devices to determine the capacity. There is always a variation in doing so as someone pointed out here in the comments.

Does the stackable version come with a NEEY by default?

The holder for the NEEY is by default in the V3 Box. You can choose an option with the NEEY if you order a V3 Box. Then the box will come with BMS and NEEY. I have them in heavy use at our farm and in a full autark Installation with Wallbox for the cars. Work really great. Also without the NEEY. Would mean there is no need for an activ balancer after initial Setup.

@@antonewagen249all right. I thought they had the NEEY installed by default for some time. So, you're not running any active balancer with this BMS in your setup? What is your charge voltage?

@@OffGridGarageAustralia Moin, right, no active balancer. I have one NEEY as Backup but never need it. Charge voltage ist up to 56.7 Volt. 56.7V because there is a 14s Nissan NMC battery in parallel.

Yep, exactly. I forgot to mention that. In the first test, we still had the BMS connected while in the cell test, we only measured the bare cells. Smaller current, different temperatures, different test devices (smart shunt vs. ZKE). This all makes the difference.

Yes, absolutely. I would not buy anything else.

Soon, my friend, soon...😉

Great, thanks for your feedback.

Ma, this thing is amazing for my purposes.

Ah, yes, right, I have some screws left in this little container. I saw the holes in the PCB but totally forgot about adding them. Thanks!

Great. Yeah, it's a huge upgrade when you come from lead acid.

You mean, the cells are touching the bare metal of the case without any epoxy sheets or EVA tape?

@@OffGridGarageAustralia Each cell is covered only with its blue plastic film, but the cells are NOT separated by epoxy sheets or EVA tape! The battery cell 16# is touching battery 15#, 1# touches 14#, 2# touches 13# and so on. E.g. I measure 44V (!) between the case of battery 1# and the touching 14#. Do you think it is safe to operate cells without a separator (epoxy or EVA tape) and having 44V between them? I am looking forward to your answer. Thank you!

Mr Multiplex, thank you very much for your ongoing support!

So far, all B-grade cells have performed the exact same way as so called certified automotive grade cells or A-grade. I made several video with tests: ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-6PyANZWtKnY.html ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-hrWEGNS4jhY.html

This might have a bit of an effect on the result, but not by much. The curves show almost the same c-ratings when discharging with 0.1C and 1C. Here it was more the fact, that the first test had still the BMS connected, the temperature was different and probably mainly because I used two different testers.

There shouldn’t be any Ah losses due to internal resistance: Amps cannot get lost in a closed circuit. What is lost is energy ( in Wh) because the current is delivered at a lower voltage because of the voltage drop due to resistance*current

Time will tell... I don't think internal resistance is a big issue.

Yes, you can if everything else is disconnected. Hence I had disconnected the balance cables to the BMS at the top.

The equipment was different as well and when measuring with the shunt, the whole BMS was in between as well (I forgot to mention that), so not only the terminal connections but also the BMS have obviously losses.

What do you mean by 'they produce too much energy'?

@@OffGridGarageAustralia plenty of sun on your panels give you for instance 4kw and then at à moment batteries getting full and no consumption. Where those 4kw will go ? How panels and other components react on that ? I'm in ongrid, so if there is more than I need, it goes to grid. In off grid, it's not used but still produce. There is no issue with that ?

Yes, I forgot to mention the BMS which was in line with the shunt when we did the capacity test. That was probably the main reason, we see a difference.

Ah are not consumed. The current is the same everywhere in the circuit, because it's in series. A measurement anywhere in the circuit for Ah will give the same results and will not be affected by voltage drops or losses. Wh on the other hand will be different - that's why Ah is a useful measurements because it is independent of bad connections, losses, etc.

I can make a video about the setup again, show some of the core parameters and also upload the settings file to my website.

What does Monomer Overweight Protection mean. The dictionary definition doesn’t make sense to Lithium BMS’s. I’m setting up my Patreon right now!! 3 Spats coming your way Andy!!

I think most B Grade cells are sold by EVE before adding the terminals, with companies like QSO buying them at auction and adding their own welded terminals. These cells appear to have got past that QC process because they seem to have the EVE terminals, but have been downgraded for other reasons (bumps, sub-optimal resistance with terminals(?), other stuff) so I'm guessing they are closer to proper grade A for high discharge/charge tests than QSO etc cells.

You never know why they have been downgraded. So far, all my storage grade cells have been great and I had no real issues using them in solar storage systems.

@@OffGridGarageAustralia I read the EVE QA process which I think I got from the "LiFePoQR Battery QR Scanner" app after scanning my QSO cells. Assuming yours do have EVE terminals, they must have passed a suite of tests that are performed before adding the terminals (which is the state they are in for most auctions). They do tend to be fine for storage because the tests include things like running the cells at 2C discharge for 30s, which is not an issue for storage. Another difference I guess is that since they have added the terminals they've deliberately put a "B" on the cell QR code, to differentiate them from grade A cells, whereas all the ones using a different terminal are either grade B or even used grade B.

We have discussed the p-setup a while back but I can do another video as a refresher...

I have replaced the pot for the output voltage in this video but have not seen one for the under voltage cut off. I'll ask them if this has a permanent set in this device. ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-uPhtOUbvrXc.html

These Andersons are not for safety, they are just connectors... Blue!

@OffGridGarageAustralia Red is only for the kids and grandkids my friend.

Yep, exactly. I didn't explain this correctly. I forgot to mention that in the first test, we still had the BMS connected while in the cell test, we only measured the bare cells. Smaller current, different temperatures, different test devices (smart shunt vs. ZKE). This all makes the difference.

Man, I have the Frankenstein battery sitting here and watch the mailbox with eagle eyes. Top priority once it arrives!

That is very hard to tell. If it stops charging at 3.43V that could only mean, it has reached this voltage and is also saturated at this point. Do you have a variable power supply and can increase the voltage?

Yes. I have a buck converter cc cv 5 A max. Yesterday I loaded at 3.46v cell 15 . She accepted the charge ,so I stopped at 3.42 like the other cells . I balanced with a heltec 5A and everything seemed ok,9mV deviation .today, charging the battery with the inverter, cell 15 and back again at 3.38,while the other cells are at 3.42. I've seen all his videos since the beginning. You are a great . I probably have to go up to 3.45 and redo everything as you say in your videos

Thanks for sharing. That is pretty good quality from QSO again. Paralleling and optimising these cells is something for the moment, I guess. As they drift and get older, these pairs won't be as optimal any more in 1 year time, is my assumption, so maybe don't spend too much time with that.

@@OffGridGarageAustralia Good point. I won't get too OCD over it! Just avoiding putting two of the lower capacity ones together, for example, which will impact the whole setup. And I guess over time I can use your high/low test to check which pairs are top and bottom performers and swap out a cell from each.

@@davidpenfoldI would probably do the same and optimise it as much as I can. It will be a long term experiment for sure.

I was surprised he didn't do that, having mentioned it earlier.

We would have seen this cell peaking at the beginning when we had 70A of inrush current. Edit: just went out and tested all 4 screws for cell 11. All perfect on 7Nm...

The reason it did this was shown . It has the smallest capacity .

@@onthelake9554 That was inferred. Not all cells were tested. It probably is, but if it was a bit loose we're using circular logic to assert that it's the lowest.

@@davidpenfold Not in my opinion , Andy was correct it would have shown up with the high amp charge had it been a loose connection . And yes all cells were tested though you may not have caught it , this was the easiest way to get them to show up .

That is a very difficult situation which only you can answer. Do you plan to get a second battery at some stage? Horizontal, vertical setup? How much energy do you need? What is you solar setup like? An how about your budget? There is not much difference between 280/304/320Ah. If you're tight capacity, get the largest ones (not sure if the 320Ah still fit in the box?). I would go with B-grade batteries but compare prices first. There is no capacity difference in A or B grade. Seplos or EEL both make good boxes and both have the Seplos BMS included. Reliable batterie cell suppliers are on my website: off-grid-garage.com/batteries/

@@OffGridGarageAustralia I'm not planing on getting a second one, it's a thight fit in my RV as it is, it's going to be on the floor, laying horizontal. My solar will grow with time and budget, my roof is only 7m x 2.3m. I will probably be able to fit 5-6 460w panels. I saw EEL and Splous, the box cost like 700usd without shipping, insane price compare to the 16p I'm planning on buying which will be 1800usd~. The amount of cycles and total life expectancy of B grade are the same as A?

Epoch is the house brand for 18650 Battery Store. Apparently they have a direct working relationship with EVE and are a verified supplier of EVE cells directly from EVE, no middle man. Epoch 18650 and 21700’s are fairly popular, most because they priced their Samsung 25R clone (Epoch 25P?) at 99 cents recently and I saw on Reddit there were at least a couple people that made massive purchases of hundreds of cells. I was tempted to buy a bunch because they are great cells for making an E-bike/Scooter or practically anything with a motor since they have a 20A continuous discharge current. I kinda wish I got a few just to test AC and DC IR to see how they compare to some old’ish 25R’s from a Dewalt tool battery AFAIK 18650BS is one of the only companies in the US that you can buy genuine cells from. They were selling Grade B cells from REPT and CATL but looks like they’ve mostly stopped the B game reportedly from complaints but every review I saw on the DIYSolarForum gave rave reviews for the REPT cells specifically. Grade A LF304’s should be giving you just shy of 320Ah. 280’s should be giving you just shy of 290 although there were quite a few people talking about the K vs N models. One of them seemed to be quite disappointing as it appears they sacrificed capacity for cycle life with people testing and only getting 279-283Ah from what were supposedly Grade A cells. Grade B (when new) is often separated from the Grade A for physical imperfections and IR. IR should be 0.12 to 0.14 mR AC IR on 280’s and I believe the 304’s are very similar. Grade B are often 0.16 to 0.30 mR AC IR. In the EV application you need the low IR to stop cell voltage from plummeting when discharging in excess of 0.5C continuous and up to 2C burst. I would be VERY upset if I got cells measuring over 0.18mR AC IR on A Grade cells because that pretty much guarantees your ability to discharge above 0.75C is non existent. You’d see cell voltages drop below 3.20 from a fully charged state. SunFunKits (hate on them if you want to but the guy Austin? is right) has some great videos where he uses a big boy electronic load to discharge A vs B grade cells (some I suspect are used B grade). The difference is shocking. A grade cells will hold voltage well above 3.20V even with some pretty hefty discharge currents. The only downside to his testing is he does constant power testing and not constant current. In a way the CP testing is more representative of the real world because the power you are drawing with an inverter is in watts. If your battery voltage drops well that’s more amps required. And the battery voltage on B grade cells drops even further because you need to draw more amps to compensate for the lower voltage. It’s a good representation of the downfall of high IR cells. Now if your application is never discharging above 0.5C (the vast majority of all energy storage systems) then this issue isn’t terribly important. So long as your system can sustain the max power of your inverter you’ll never have major voltage drop issues. This is why I say for every 5kw of inverting power it’s best you have 15kwh of batteries to pull from so it’s at most a 0.33C. When you pair a 5kw inverter with a 5kwh server rack battery you are putting yourself in a situation to see voltage drop as low as 48V from full right off the bat if you are maxing out your inverter. What does that mean? Well instead of drawing 92A at say 54V for a 5kw power draw you will be drawing 104A at 48V. This is putting you on the edge of so many component limits for these 100A discharge batteries. If you just had A grade cells to start with you’d be fine but now you find yourself unable to max out your inverter even with a full battery, and a battery at 50% SOC….well shit 4kw might drive your cells to low voltage disconnect whereas the A Grade cells should still be well above 3V until you probably get down to about 10% or 20% SOC and then you might find a 5kw draw to be impossible. So as long as you have triple the battery storage vs inverting power this should be a non issue. At 2:1 you might see some issues at low SOC. At a 1:1 ratio you have essentially prevented yourself from using your inverters full power with B grade cells.

@@ericklein5097 interesting stuff. thanks for sharing. Considering my install is on a sailboat that will never see a load above 1500 watts, and its a 12 volt system that worked just fine on 12 volt lead acid batteries that would routinely draw down below 11.5 volts under load, it sounds like even B grade cells at their worst performance would be an substantial upgrade :P

The computer uses only 150W though under full load so that's not cutting it for excess power.

Neee, nix Französisch hier, nur Englisch!

Most power tool batteries are rated at 10C, some of the newer pouch packs do 15C reliably, and Li ion weighs significantly less than LFP / LFYP batteries for the same amount of power. There's no reason to carry around a 4kg battery pack just so that your drill can receive the 800W it's used to from a 0.8kg battery. For storage applications, anything above 0.5C is probably going to be considered wasteful, you're heating up the battery cells and cables instead of doing useful work, and you're only storing enough power for less than 2 hours' usage.

I have not tested these batteries. Is using LFP batteries for tools a thing? I have only ever seen Li-ion so far.

@@stefanvanzyl9090 I mean corded power tools connected to a inverter/battery system. You could of course use battery tools, but they are expensive and the tool options are limited when i comes to certain applications like saw tables and welders. 15C for li-ion is impressive but considering these li-ion batteries are in the 4-10ah range capacity wise the power output is not that high or practical. If you have a larger battery bank you could connect batteries in parallell and get sufficient power output, but im looking for a 200ah system atm. That's why im considering LFYP cells since they provide 3C instead of the normal 1C for LFP batteries and apparently can tolerate colder climate charging/discharing too.

⁠unless you're using industrial welders/saws, you're limited to circa 3500W for the appliance regardless, since 230V plugs are 14-16A rated. And since the preferred inverter DC battery voltage is 48V, even a 0.7C 100Ah LFP battery can achieve this, and last almost 1.5 hours doing it. If you use a 3C 40Ah battery (since you're min/maxing here to save costs and a 20Ah won't make 3500W), the battery will be flat in 30 minutes of work. As a corollary, the tesla powerwall 1 and 2 are 18650 cells in series/parallel to reach their 14kWh storage capacity - it's not unheard of. Lastly worth noting, most newer pure sine wave inverters have sufficient surge capability to handle striking an arc without drawing more from the batteries.

CCTV, dogs and an angry wife to protect everything. 100k... hahaha, that's a bit... haven't got these gold-batteries yet!

It's just a test when I have excess power... I'll do a separate video about that.

Yep, exactly. I didn't explain this correctly. I forgot to mention that in the first test, we still had the BMS connected while in the cell test, we only measured the bare cells. Smaller current, different temperatures, different test devices (smart shunt vs. ZKE). This all makes the difference.

Not quite, as we also had the BMS connected in the first test and the balancer caused a certain 'imbalance' in the battery (and your maths) as it discharges single cells.

Nope, tested all this in past videos with 200A, there is no difference in B-Grade and automotive cells. ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-6PyANZWtKnY.html All rumours and people trying to justify their more expensive spend on A-grade cells.

Well, there is no limit on how much fuel you can store in a shed apparently. I know 2 people which are bit nuts with that...

@@OffGridGarageAustralia Ah yes, the Blown-Off-The-Grid-Garage. Great channel. Explosive content.

Yes, exactly ⚡🐟

It's not a news channel, my friend.

Thanks for sharing. I think they will read here as well and hopefully send you the forgotten BMS quickly. Keep complaining.

Waiting for what?

I have one running for a while if I have surplus energy. I'll make a video about that setup... It's just for testing purposes at the moment.

@@OffGridGarageAustralia oh great... Which miner is it? Waiting for the video 👍

Have they not replied to you and offered a solution? Do your batteries have a B on the QR code as well as mine have?

@OffGridGarageAustralia it was an ordeal of arguing with the sales rep and their management. I provided photos, evidence of our conversation (when they promised flat grade A cells). All to no avail. Alibaba having all the evidence still sided with them. All are crooks.