Oh goodness, you trust those "evaluation certificates"?? I would never! Lishen grade A cells looked fantastic on paper and were "grade a certified", but the welds on the corner of the case were awful, and lead to leaks.
The company issues these grades at their own discretion, and they should not be trusted. Ever. Grade A can be very relative depending on the company. Typically, brand new Gangfeng, CATL, EVE cells can be trusted, but you should go by people's experience here on the forum and elsewhere before trusting their silly little "certificates".
Seriously Will! You had a bad experience with Lishen cells leaking and that condemns the whole process of testing batteries for 30 days plus X-Raying , weighing them and measuring there electrical characteristics for inconsistencies? It's all useless in your opinion?
These kinds of certificates are how we are able to sustain work flow in the industry. It gives us confidence that the products we are using conform to the standards we expect. And NOBODY in the industry just follows them blindly. You would be a fool to be in the EV industry and not have several employees that randomly pull out packs in each shipment and run them on $10,000 machines to measure all of their electrical parameters etc. I would suspect they also will X-Ray some sample batteries and they are all weighed etc. The information is matched against the certificates and if it is not reasonably matched then the whole order will be further examined and most likely returned.
We certainly tested sample batches of all the products we had made. Pulling them apart and checking every component for substitutes and using microscopes to view soldering joint and traces for defects. We would even desolder large electrolytic capacitors and check them to make sure they met the specifications listed on the manufactures charts. We suspected labels could be fabricated on parts and this was actually tried on a few occasions
As for your Lishen cells leaking, that is probably not an issue that could be tested for. The battery probably pasts every test, it's just that the material holding the cell together is faulty or the chemical composition of the electrolyte is to corrosive. That is no different than the Capacitor Plague of the late 1990's when the market got flooded with bad Capacitors for 6 years from all the major Taiwanese manufactures.
I worked with those Capacitors and they tested normally and they worked normally but after the product was out in the public's hands for 1-2 years the Capacitors would start leaking all over the circuit boards.
It was found out almost a decade later that it was a faulty formula in the electrolyte that had been given to a Chinese Spy working at Rubycon in Japan. His job was to figure out why their capacitors were superior and when he realized it was the Electrolyte the company covertly let him get access to a formula, but it was actually a formula that they had failed testing because it had major long term issues. They knew he would take the formula back to China and Taiwan and pass it on to their competitors and that's exactly what he did. They then used the bad formula in all their capacitors for around 5 years and anyone who bought them ended up with lots of warranty returned items.
Also, I would take most cycle life data with a grain of salt. Especially for solar, where calendar aging will probably kill the cells long before cycling will.
You keep saying that but is there any real world data that shows that to be true? I have seen a guy on here showing off an LFP battery bank that is 15 years old and it is still working fine.
Eflex fortress packs are fantastic, but their cells are not special. Same with valence and other more expensive brand packs. You will ALWAYS have a single cell that will deviate higher as the SOC rises. It is physically impossible to match lifepo4 cells 100%. Internal resistance or capacity. There is always a difference, cell to cell.
Yes there is always a difference but that is a vague statement. The tolerance values in a matched set of cells used by a Tier one company will be very tiny as compared to a company that just randomly throws Grade B cells into a pack. You should charge up that eFlex and monitor the cells and then compare it with an EG4LL and see the difference. You will find the eFlex has a very slight deviation as it charges while the EG4LL has a huge deviation.
Yes at the end of the day the EG4LL balances out the cells but you have to wonder about the longevity of those cells that are always going high.
Now of course, grade b or used cells will have a larger deviation, and I understand the difference there. But one cell will raise in voltage first, no matter what the grade of the cell is. There is natural cell drift with every cycle, and you will have one reach 100% before the others.
Yes one cell will always reach there first. What I don't expect is for three to get to 3.75V when the others are at 3.35V. I also don't expect it to be the same cells doing it all the time. That tells me that those cells are significantly different than the other ones in the pack.
EG4 has the same BMS as everyone else. Again, not special at all, and it is a fantastic BMS. It is all over the place? Can you post evidence of this difference in the firmware over time because I am not aware of it. What HVD threshold figures were programmed initially, and when did they change?
I don't remember mentioning a firmware difference or any complaints about the BMS. I will have to look back through the posts.
And the reason one cell is usually the culprit, but not always, in high quality and low quality packs, is because one cell will naturally have a slightly lower capacity than the others. And no matter how much you balance, it will show itself in due time. If you do not balance for a while and hang out around 50% soc, then do heavy shallow cycles in cold temps for example, that "trouble" cell will be first to have its voltage spike. That will always happen. Show me a battery where that does not occur, and I would be utterly amazed.
The eFlex battery does not have that issue unless you want to nit pick. As I said it's all about matching and the tolerances that are accepted.
All the cells will have slightly different ESR and capacity but if they are all producing good numbers and then batched together properly it will be very small differences.
What matters at the end of the day is the capacity over time. If it fades quickly, you will know. And it can be measured. And if it is terrible degradation, you will get a brand-new battery pack. This is not referring to signature solar either, all of these manufacturers offer a similar or better warranty. They plan for it. We all know that 1 out of 1000 packs will have a bad cell no matter what manufacturer you use. Same with tesla cars and any other lithium ion based product.
Agree but I would add that the warranty is only as good as the company that stands behind it. We have already seen threads hit 900 posts with one guy trying to get warranty service on a battery that simply did not work with his Inverter. I would hate to be trying to prove my capacity has dropped by 20% when a video of switching something on and it not working has been proven to be insufficient evidence to get warranty help.
Something else I wish to mention is the consistency of capacity test results of chins batteries and their clones. They are usually within .1Ah. These batteries are using the cheapest grade a cells around! Some companies told me they would never build with those cells. And they still pull fantastic figures. And people use them daily without issue. That is what matters most, do they pull full capacity and can you use them for decades
That's the big question! Lots of things work perfectly in the short term, but the long term is where the difference lies. I just finished working on my 14 Month old GE Washing Machine. Replaced a Burned out drain Pump and as soon as I put my ohm meter on it and got 9 ohms my heart sunk because the new pump measured 40 ohms.
Sure enough the motor control board is also dead. Why is it we can't make appliances that last anymore?
My old washing Machine lasted for 15 years with only minor issues.
. If people have a lot of issues with a pack, I will let the world know. The EG4 is using the same parts as everyone else. They are not special. It is cells in a steel box with a BMS that everyone else is using. --Snip-
Your swimming in batteries so I get your line of thinking, but for someone like me that wants a pack of batteries to last 10-12 years I am not going to turn a blind eye to so many leading people in the industry saying that cells in battery packs are not all the same. If it's anything like I suspect it is, I doubt that even one production run of packs is the same as another. I suspect it's all dependent on what batteries they can get for the price point they are trying to buy them at.
Now everyone here needs to learn how a shunt works and do a cycle test every year or so to see how well their system is holding up. Use a victron shunt or whatever you please. Do a single cycle and see what figure you get.
That's not as easy as you might think because these batteries are running the power in my house 24/7 and the Sol-Ark is constantly drawing power and putting back in power as the clouds pass over the PV.