Amp-Hour measurement and losses, or, why Voltage still doesn't matter.

[upnorthandpersonal]

Recently I got into discussion in the comment section of YouTube (I know, I should know better) of a video posted by @Off-Grid-Garage. Topic of the video was capacity testing and why voltage does not matter when doing this test specifically when measuring Amp-Hours. After some people still didn't get it, I decided to make an extreme video demonstrating this. I connected a 20Ah battery to a CBA IV and put 8 meters of cable in between, together with some bad alligator clip connections etc. I did an 8A discharge test (the maximum I could with the mess of wire: the voltage at the CBA dropped to 0.5/0.6V) - 80% losses in the wiring. Here are the results:

 

[upnorthandpersonal]

Thought I'd post the result at the end of the test in case one doesn't care for or wants to sit through the video:



You can see the losses here in that the CBA IV measured 13.386Wh at the end out of a total of 64Wh on the battery label. On the other hand, the 22.5Ah capacity reading is identical to the one I measured in a 'proper' way a cycle before:

 

[Steve_S]

I watched a couple of his video's and had to bite my tongue.
As you show, it's far too easy to skew tests, even by using craptastic wire....

My biggest "gripe" is folks are missing the entire charge saturation issue.
You can charge a cell to 3.65V and stop immediately and it will settle relative to the depth of its charge.
You can charge that same cell with Amps accepted drops to 1.0A or even 2.0A and it will NOT settle down like the other and will output a few extra AH.
Factories test with cells at 3.65V and fully saturated, not just surface charged.... (terms people love to argue over, I won't)
It's even outlined in some of the factory docs. Simply put, charge to 100% SOC (fully saturated) allow to rest for 60 Minutes & start test cycle.

I learned quite a while ago, not to use "clips" but actually have Ring Terminal on the wires (bench PS Wires & Cell Tester wires too.).
CC-CV Charging down to 1.0A taken is best on initial charge & for finish Top Charge & Balance.

The CBS Testers are great for several reasons but the Logging makes a big bonus.
Fortunately, now those 8 in 1 Battery Testers from China are also coming out with PC Logging and for much less cash.

 

[upnorthandpersonal]

Yes, I don't make any comments on the quality of the meter he uses. I don't mind them for relative testing, but I'd go with the CBA for anything real and objective. It's just frustrating to see so many misunderstandings and confusion on these topics, and people who continue to believe what they say and continue to spread this even though they obviously have no idea what they are talking about - comment sections on these kinds of videos are rife with this as @Will Prowse can probably attest. I'm an educator at heart (been a professor at Uni for over a decade) and these things get to me

 

[Steve_S]

My GF is a Medical University Proff.
I myself spent years teaching and tutoring.
I've also reached a similar "burn-out" like Will has where the constant repetition to counter FUD and mis/disinformation has become far too tedious to bother with anymore.

Having to apply Rule of Three on this too.
If the same question is asked more than 3X, document it and file the answers for later retrieval.
Do not answer more than 3X on same subject/topic.
Immediate 3-strikes if someone comes in heavy guns with FUD / BS or Bafflegab.
Immediate 3-strikes if anyone pulls a Political, Religious or other equally stupid pile of dwaddle into it.

BTW: Funny thing, I now have my GF using Rule of Three in ALL of her courses.... Ohhh Boy did some students have a kinipshin ! She's now finding herself less stressed because she just aims the students at the statement on Rule of Three. She is also an executive in the College and is now getting others to employ the Rule of Three which is limiting a LOT of Bull Hooey that had been going on. Funniest part about Rule of Three, people have used it for Millenia without realizing it consciously... think Baseball, 3rd strike, your out, third time is a charm, do it in triplicate, etc etc etc...

 

[Solarfun4jim]

My biggest "gripe" is folks are missing the entire charge saturation issue.
You can charge a cell to 3.65V and stop immediately and it will settle relative to the depth of its charge.
You can charge that same cell with Amps accepted drops to 1.0A or even 2.0A and it will NOT settle down like the other and will output a few extra AH.
Factories test with cells at 3.65V and fully saturated, not just surface charged.... (terms people love to argue over, I won't)
It's even outlined in some of the factory docs. Simply put, charge to 100% SOC (fully saturated) allow to rest for 60 Minutes & start test cycle.

I think for a lot of folks new to this DIY battery pack caper, (myself primarily) we are purchasing tools and equipment that we have never used before. For top balancing for example, many have purchased a 0-30v 10A BTPSupply. Now ideally on a 280AH cell your looking for 140CA with a clean 14A cut off point...job done, plenty of saturation at that point.
Unfortunately with the lower power PSU many are purchasing on the back of Will's video, you are lucky to be getting 5+Amps to start with...so no 14A cutoff threshold. Added to that, comments that charging at 3.65V and waiting till the amps drop to 0.1A etc, can lead to overcharging the cell....then wowser, caution applied and early termination. I understand the capacity issue with the potential 8AH etc not 'forced' into the cell through saturation, but i didnt expect it to have such an impact on bringing the cells to balance.
Would twin 10A psu's connected in parallel do a better job of achieving saturation and the 14A cut off(0.05C)?

 

[Steve_S]

I think when EVE says 0.05C as suitable for termination amps, I have my doubts that is correct. Translation being as lax as it is.
I know the beck power supply is set to 3.65V now when it's reaching 15A it is at 3.65 and stays there till amps drop to <2.0A.
People have bought 10A capable bench supplies not realizing that at 10A full bore you are talking a LOT of hours to charge a 280 from 2.50V.
I start at close to 40A which quickly dwindles down to 30A and stays there until the cells actually reach 3.330 then the amps slowly fall till we reach 3.500 by which time it's at 25A. The moment the cell voltage hits 3.65, the amps taken slowly decreases till it appears to stall at 2.0A but at that point it''s is trickling in and saturating. By 1.0A your quite topped & saturated.

A GOTCHA !
Bench supplies are tricky beasts.... You can set the voltage to 3.600 and start charging. The cell will still say 2.75V or whatever and climb as it gains charge. BUT setting the TOP must be done carefully with an Eagle Eye ! By the time the amperage has started to drop near the top, you have to keep an eye on the voltage limit, often that needs to be "tweaked" to stay within 3.600-3.650, if not, you could end up topping to 3.70 or even 4.00V.

Case in point, I am charging a cell right now... started at 2.60V after a capacity test.
Cell terminal says 3.360V. Charger terminal says 3.560 and pushing 25.4A
As Cell Volts go up, the Amps going out drop.
As the cell reaches 3.65, the amperage will drop while the Volts at the Charger terminals will also hit 3.650. At this stage we are roughly at 15A being taken, which then falls steadily to close to 2.0A at which point it really slows into a trickle for saturation.

A side note: I finished one cell off to 3.650V with 1.5A charge being accepted last night. It settled to 3.6303V in an hour. It has sat for 12 hours now and has only settled to 3.5584V. Had I done like my previous test of cutting off charge at 10A, it would have settled even lower, closer to 3.4500 ish. That's why I had to do my capacity tests twice, because I failed to saturate.

 

[Solarfun4jim]

I think when EVE says 0.05C as suitable for termination amps, I have my doubts that is correct. Translation being as lax as it is.
I know the beck power supply is set to 3.65V now when it's reaching 15A it is at 3.65 and stays there till amps drop to <2.0A.
People have bought 10A capable bench supplies not realizing that at 10A full bore you are talking a LOT of hours to charge a 280 from 2.50V.
I start at close to 40A which quickly dwindles down to 30A and stays there until the cells actually reach 3.330 then the amps slowly fall till we reach 3.500 by which time it's at 25A. The moment the cell voltage hits 3.65, the amps taken slowly decreases till it appears to stall at 2.0A but at that point it''s is trickling in and saturating. By 1.0A your quite topped & saturated.

A GOTCHA !
Bench supplies are tricky beasts.... You can set the voltage to 3.600 and start charging. The cell will still say 2.75V or whatever and climb as it gains charge. BUT setting the TOP must be done carefully with an Eagle Eye ! By the time the amperage has started to drop near the top, you have to keep an eye on the voltage limit, often that needs to be "tweaked" to stay within 3.600-3.650, if not, you could end up topping to 3.70 or even 4.00V.

Case in point, I am charging a cell right now... started at 2.60V after a capacity test.
Cell terminal says 3.360V. Charger terminal says 3.560 and pushing 25.4A
As Cell Volts go up, the Amps going out drop.
As the cell reaches 3.65, the amperage will drop while the Volts at the Charger terminals will also hit 3.650. At this stage we are roughly at 15A being taken, which then falls steadily to close to 2.0A at which point it really slows into a trickle for saturation.

A side note: I finished one cell off to 3.650V with 1.5A charge being accepted last night. It settled to 3.6303V in an hour. It has sat for 12 hours now and has only settled to 3.5584V. Had I done like my previous test of cutting off charge at 10A, it would have settled even lower, closer to 3.4500 ish. That's why I had to do my capacity tests twice, because I failed to saturate.

Great info

 

[upnorthandpersonal]

Personally I think people worry too much about the capacity of these cells (and trying to get every last bit out of it). They're an absolute bargain compared to the alternatives even if you assume you will only get 250Ah out of them when assembled in a pack. Yes, they might not perform as the seller says or advertises, but hey, that's part of dealing with Alibaba vendors in China. When you go in level-headed, there is no problem really. I'm still going to buy another 32 cells this year, probably Lishen cells and throw them with the other 32 EVEs I have been running for the past 6 months. It would be impossible where I am to even get close to this price point with lead acid.

 

[Solarfun4jim]

Personally I think people worry too much about the capacity of these cells (and trying to get every last bit out of it). They're an absolute bargain compared to the alternatives even if you assume you will only get 250Ah out of them when assembled in a pack. Yes, they might not perform as the seller says or advertises, but hey, that's part of dealing with Alibaba vendors in China. When you go in level-headed, there is no problem really. I'm still going to buy another 32 cells this year, probably Lishen cells and throw them with the other 32 EVEs I have been running for the past 6 months. It would be impossible where I am to even get close to this price point with lead acid.

Exactly right...if these were sold at the prices they are, with a spec of 250AH, ran between 3.2v - 3.4v with a delta less than 10mv.....folks would be creaming in their pants
Human nature i suppose!

 

[AussieSim]

Using this meter for a 16s EVE 280Ah capacity test

My Wh reading (16*3.2*280 denominator) is consistently higher than the Ah (280Ah denominator) reading throughout the test span.

I did not do the discharge test at a constant C rate - over 14kWh ! - and my BMS was balancing during the test so there would be some heat losses there.

 

[upnorthandpersonal]

I don't know that meter, but I would not trust that unless it's the only thing in the circuit (so no BMS that's doing balancing, or regular loads). Have you tried an isolated, single cell to check if those readings match at all? Or even just a power supply?

 

[AussieSim]

I don't know that meter, but I would not trust that unless it's the only thing in the circuit (so no BMS that's doing balancing, or regular loads). Have you tried an isolated, single cell to check if those readings match at all?

I haven't checked a single cell, I have 34 cells in 2 * 8s + 1 * 16s + 2 spares.

I've recorded just over 97% Wh useable capacity between 3.62v - 2.75v range so I'm happy with that.

 

[upnorthandpersonal]

I tested my cells way at the beginning when I bought them (over 6 months ago) and they all checked out at 280Ah or above. I'm planning on pulling one cell that I think is currently the weakest one, and do a capacity test on it to see what it is at now. I still have the readings I took back then so I can compare. Maybe that could be an interesting data point for someone.

 

[Dzl]

Personally I think people worry too much about the capacity of these cells (and trying to get every last bit out of it). They're an absolute bargain compared to the alternatives even if you assume you will only get 250Ah

Exactly right...if these were sold at the prices they are, with a spec of 250AH, ran between 3.2v - 3.4v with a delta less than 10mv.....folks would be creaming in their pants.
Human nature i suppose!

As one of the aforementioned worrywarts (I would characterize it as grumbling not worrying ?) I agree. But this (the fact that these cells would still be a good deal if honestly represented) is the source of my frustration.

These B grade cells, would be an equally good deal, if they were acknowledged to be B grade with the possibility of reduced capacity. The price already reflects this. In my eyes they would be a better deal if they were accurately represented, because trust the in seller adds value to the product.

Spoiler: Accidental Rant...

I think that most people that take issue with the misrepresentation (by sellers, and by many customers who want to believe they bought matched A grade cells, or don't know any better) are of a similar mindset as me. Its not that they are 'worried' or regret their purchase if there cells come up 3% or 5% short or are widely mismatched, its that they (and I) want to promote accurate information, and want to hold sellers to some basic standards. Anyone buying from these grey markets should be well aware of the widespread misrepresentation and lack of transparency and general 'buyer beware' culture, and should have realistic expectations, but there is a difference between recognizing something and accepting it. I recognize the reality, but don't accept or excuse it.

We have little power to change a whole sales culture, but we do have the power to educate other forum members and incrementally change some aspects of our little niche market, by pushing suppliers to be more honest/realistic, and by choosing to buy from suppliers who do better in this regard. At the same time I realize we have to understand its a different sales culture, and there is risk in being the first supplier to acknowledge something negative. I spend a lot of time in the beginners subforum, and at least weekly (and for many months almost daily) there is a post or comment about someones purchase of matched grade A 'xuba cells' or EVE cells. And more egregious examples of misrepresentation like the suppliers in the group buy explicitly claiming to individually capacity test each cell and that all tested above 280Ah, and only admit that was not true (they randomly capacity tested a small subset supposedly) after people started reporting their own capacity test results which fell short of spec. Bear in mind that many of the people upset about what seems like somewhat small discrepancies in capacity to you, bought in the group buy where they were explicitly promised individually capacity tested cells => 280Ah

This is a long winded way of saying, I agree, if accurately represented these cells would be a good deal (better deal really), and a few % isn't a huge deal in the realm of things. But its also not wrong to expect more of the sellers we choose to buy from, and to try to present information as accurately as possible, so others in the future can do a fuller and more accurate cost / benefit analysis.

 

[fafrd]

As one of the aforementioned worrywarts (I would characterize it as grumbling not worrying ?) I agree. But this (the fact that these cells would still be a good deal if honestly represented) is the source of my frustration.

These B grade cells, would be an equally good deal, if they were acknowledged to be B grade with the possibility of reduced capacity. The price already reflects this. In my eyes they would be a better deal if they were accurately represented, because trust the in seller adds value to the product.

Spoiler: Accidental Rant...

I think that most people that take issue with the misrepresentation (by sellers, and by many customers who want to believe they bought matched A grade cells, or don't know any better) are of a similar mindset as me. Its not that they are 'worried' or regret their purchase if there cells come up 3% or 5% short or are widely mismatched, its that they (and I) want to promote accurate information, and want to hold sellers to some basic standards. Anyone buying from these grey markets should be well aware of the widespread misrepresentation and lack of transparency and general 'buyer beware' culture, and should have realistic expectations, but there is a difference between recognizing something and accepting it. I recognize the reality, but don't accept or excuse it.

We have little power to change a whole sales culture, but we do have the power to educate other forum members and incrementally change some aspects of our little niche market, by pushing suppliers to be more honest/realistic, and by choosing to buy from suppliers who do better in this regard. At the same time I realize we have to understand its a different sales culture, and there is risk in being the first supplier to acknowledge something negative. I spend a lot of time in the beginners subforum, and at least weekly (and for many months almost daily) there is a post or comment about someones purchase of matched grade A 'xuba cells' or EVE cells. And more egregious examples of misrepresentation like the suppliers in the group buy explicitly claiming to individually capacity test each cell and that all tested above 280Ah, and only admit that was not true (they randomly capacity tested a small subset supposedly) after people started reporting their own capacity test results which fell short of spec. Bear in mind that many of the people upset about what seems like somewhat small discrepancies in capacity to you, bought in the group buy where they were explicitly promised individually capacity tested cells => 280Ah

This is a long winded way of saying, I agree, if accurately represented these cells would be a good deal (better deal really), and a few % isn't a huge deal in the realm of things. But its also not wrong to expect more of the sellers we choose to buy from, and to try to present information as accurately as possible, so others in the future can do a fuller and more accurate cost / benefit analysis.

I believe the issue is that the sellers / trading companies don’t know what they are selling...

They purchase these EVE reject/excess cells by the palette, probably through at least 1 if not several middlemen.

They check that the cell is near 50% charge, perhaps measure internal resistance to be sure it is not way off, and at best, they’ll do some matching of cells by voltage andIR before shipping.

They cannot afford to run a capacity test and have no idea what the actual delivered capacity of the cells they are shipping will be - only what capacity they were manufactured to deliver by EVE.

 

[Dzl]

I believe the issue is that the sellers / trading companies don’t know what they are selling...

They purchase these EVE reject/excess cells by the palette, probably through at least 1 if not several middlemen.

They check that the cell is near 50% charge, perhaps measure internal resistance to be sure it is not way off, and at best, they’ll do some matching of cells by voltage andIR before shipping.

They cannot afford to run a capacity test and have no idea what the actual delivered capacity of the cells they are shipping will be - only what capacity they were manufactured to deliver by EVE.

I think this is a very good assessment and I agree, at least insofar as I think it is one of the most probable possibilities.