Crude 2.5C Capacity Test Of Frey 36130190-60PF 60Ah Cell

[bestconcreteblock]

Just finished this crude test(see .pdf) for this Frey 36130190-60PF 60Ah Cell. These are power cells with 3C continuous discharge.

Looking at the data sheet discharge curve(also attached), I expected the terminal voltage to hold above 3.0V for more than half the test. That was not the case. Cell must have a little bit higher internal resistance than advertised. These were purchased directly from Frey via alibaba.

These cells have delivered around 62 Ah consistently with a 1/3C discharge rate.

Any thoughts?

 

[Marinepower]

Just finished this crude test(see .pdf) for this Frey 36130190-60PF 60Ah Cell. These are power cells with 3C continuous discharge.

Looking at the data sheet discharge curve(also attached), I expected the terminal voltage to hold above 3.0V for more than half the test. That was not the case. Cell must have a little bit higher internal resistance than advertised. These were purchased directly from Frey via alibaba.

These cells have delivered around 62 Ah consistently with a 1/3C discharge rate.

Any thoughts?

I don't have any informed thoughts and don't know much about high C rate cell behaviour. I suppose testing is normally done at 25 Celsius, so a bit warmer than your test. That would have impact on internal resistance and voltage drop? However, at 3C, i would think there would be a lot internal heat generated.

These cells list up to 1.5m ohm internal resistance, which seems high to me, even for a 60ah cell. In comparison, i have grade B, 280ah EVE cells with a measured IR of 0.018m ohm.

I have seen other user IR test data for their 100ah cells (measured at 0.6m ohm) - which significantly beats the spec sheet for that cell. Makes me wonder if Frey have a difficult time manufacturing consistent IR in their cells - and to compensate they give themselves a large range in the spec sheet?

This post is interesting to me - as I ordered 16 of the Frey 100ah cells from Maggie at Frey, via Alibaba at the beginning of January of this year. I was promised the cell's factory test data sheet for my shipment, but I never received it. Did you ask for or receive test data from Frey when you purchased?

MP

 

[bestconcreteblock]

I suppose testing is normally done at 25 Celsius, so a bit warmer than your test.

I think you may be right on this. These are much more sensitive to temperature than I had expected. When I first started testing these cells at 1/3C, I made the mistake of doing it in the basement during a cold spell. basement was around 50 deg F. I would only get around 58 Ah. After moving my discharge setup upstairs, I now get around 62 Ah.

This high current setup was done in the basement. I started with a warm battery and placed it in a 68F water bath, but by the time I got everything set up, the water had dropped to 64F.

My other theory is that my water bath idea backfired. I was thinking that the water bath would help stabilize the battery temp. I am sure it did, but the manufacturers testing was probably done in open air at 25C. In their test, the battery temp probably increased more and reduced internal resistance.

I ordered 16 of these from Emily at Frey. I chose the slow boat shipping option. Packaging looked good, but 6 of the plastic caps were cracked. I did not request test data.

 

[Marinepower]

I think you may be right on this. These are much more sensitive to temperature than I had expected. When I first started testing these cells at 1/3C, I made the mistake of doing it in the basement during a cold spell. basement was around 50 deg F. I would only get around 58 Ah. After moving my discharge setup upstairs, I now get around 62 Ah.

This high current setup was done in the basement. I started with a warm battery and placed it in a 68F water bath, but by the time I got everything set up, the water had dropped to 64F.

My other theory is that my water bath idea backfired. I was thinking that the water bath would help stabilize the battery temp. I am sure it did, but the manufacturers testing was probably done in open air at 25C. In their test, the battery temp probably increased more and reduced internal resistance.

I ordered 16 of these from Emily at Frey. I chose the slow boat shipping option. Packaging looked good, but 6 of the plastic caps were cracked. I did not request test data.

Thanks for your info on your shipment. Too bad about the broken plastic holders. The boxed cells look good however.

My order was from: https://frey.en.alibaba.com/?spm=a2...95a76e9fSX4ZV&tracelog=from_orderlist_company

There is a poster here "RCinFLA" who has expertise in voltage drop vs. temp and C rate. Hopefully he will chime in at some point.

MP

 

[bestconcreteblock]

There is a poster here "RCinFLA" who has expertise in voltage drop vs. temp and C rate. Hopefully he will chime in at some point.

Some good information here:
https://diysolarforum.com/threads/damaged-batteries-3-2v-to-1-2v.55024/post-706643

 

[Marinepower]

Some good information here:
https://diysolarforum.com/threads/damaged-batteries-3-2v-to-1-2v.55024/post-706643

Yes, that was exactly the chart i was thinking of - but could not locate yesterday.

My EVE cells are on my boat and in the winter they are often 5 - 7 degrees celsius (basically ocean temp).

When I charge at those temps (0.15 C rate) , i really notice the cell voltages go up fairly high (4.2v) even when they are at a 50% or less SOC.

 

[bestconcreteblock]

Some good information here:
https://diysolarforum.com/threads/damaged-batteries-3-2v-to-1-2v.55024/post-706643

My biggest question after looking over this information:

Do both R_ohmic and R_ionic generate heat inside the cell, or is it only R_ohmic?

My gut feeling is yes. I ask, because I am trying to figure out how much heat needs to be removed from cells to keep them in their happy place.

 

[Marinepower]

This post is interesting to me - as I ordered 16 of the Frey 100ah cells from Maggie at Frey, via Alibaba at the beginning of January of this year. I was promised the cell's factory test data sheet for my shipment, but I never received it. Did you ask for or receive test data from Frey when you purchased?

MP

Update on this: after a gentle reminder, cell data was sent to me. So that's a win and bodes well for the quality of the vendor....

 

[bestconcreteblock]

Update on this: after a gentle reminder, cell data was sent to me. So that's a win and bodes well for the quality of the vendor....

If you don't mind me asking, what kind of test data did they provide you with?

 

[Marinepower]

A capacity test (presumably at 0.5 C) and a 1K hz IR test.

No testing at higher C rates or voltage drop.

 

[bestconcreteblock]

.02 Ohm/150A Discharge Test - Google Drive

docs.google.com

Ran this test again with another cell. This time, battery was in open air(cold basement). Battery started at 56 deg F as measured by insulated(from air) probe on outside of battery case in the center.

Interesting to see how the terminal voltage increases as battery temp increases.

62 Ah approximate capacity.

I think this cell performed pretty well. My only concern is that it appears the internal resistance could be slightly higher than expected.

I am designing my system to remove about 50w of heat per cell under worst case(150A continuous discharge). Does this seem reasonable?

 

[bestconcreteblock]

A few thermal pics during the test. It looked like the hot spot was near the top of the cell first, but by the end it was centered:

 

[RCinFLA]

High discharge rate cells are made with thin electrodes. Since capacity is amount of LFP in cathode and and graphite material in anode electrodes, thinner electrodes needs more layers to achieve capacity. Extra layers cost more money in copper and aluminium foil and take up volume within cell, but extra current collector layers is lower current density per square cm of copper and aluminum which gives lower resistance.

First should measure 1 kHz impedance with YR1035+. The spec says less then 1.5 milliohms, so lacking accurate data on your cells let's assume it is 1.2 milliohms. This is primarily the conductive resistance of cell metal foil current collectors, call it R_ohmic resistance.

E_ionic is the overpotential voltage slump due to ion migration necessary to support the demanded cell current. It is the more important number. It's equivalent resistance is logarithm to current through cell. Call this piece R_ionic, R_ionic is non-linear with current. R_ohmic is fairly linear with cell current. Ionic voltage drop is also called polarization voltage of cell.

R_ionic also has an exponential time decay to equilibrium for given cell current. It takes roughly 2-3 minutes to reach equilibrium where voltage slump levels off.

Total cell terminal voltage slump is the R_ohmic plus the R_ionic resistance times cell current.

So, your 4 minute into 147 amp load data, starting from 3.4v open circuit rested no load voltage (OCV) and at 4 minutes cell voltage shows 2.964vdc. That is 3.400v - 2.964 = 0.436v total slump. 1.2 millohm R_ohmic contributes 147A x 1.2 milliohm = 176 mV of the total voltage slump.

The drop due to R_ionic contribution is 0.436v - 0.176v = 0.260v due to R_ionic voltage drop.

0.260v drop due to R_ionic is not too bad for 3C discharge rate. 147A x 0.436v cell voltage slump will produce about 60 watts of cell heating.

 

[bestconcreteblock]

R_ionic also has an exponential time decay to equilibrium for given cell current. It takes roughly 2-3 minutes to reach equilibrium where voltage slump levels off.

Thanks for taking the time to explain all of this. These visual aids are very helpful. Did you create these?

So a cell made for low C-rate storage applications is going to have a higher R_ionic at 3C because the electrode layers are thicker and the electrons cannot flow as quickly as they can in a cell with thinner electrode layers?

Obviously there are a lot of variables here, but does 60W of cell heating seem realistic?

Looking at it a slightly different way, when I capacity test these cells at 1/3C, I get around 202 Wh. In my second crude test, I would estimate 2.844V average x 62.4 Ah =177.5 Wh. This is a difference of 24.5 Wh. The length of the test was .436 hrs. 24.5 Wh / .436 hr = 56 W of cell heating. Obviously, this is ignoring cell heating that happening at 1/3C and a lot of other variables, but it seems to be in the ballpark.

 

[RCinFLA]

The thicker the electrode material the greater the AH capacity for given cell volume and weight but the greater the obstacles to ion migration causing layer ion starvation at high discharge current. This drives up the overpotential voltage slump to push the ion migration harder.

The difference in extractable capacity with greater load current is the result of additional R_ohmic and R_ionic losses. Losses rise quicker when cell gets into layer ionic starvation current range.

This is example of thick electrode EVE 280AH cells. They have electrode in 150-180 um thickness. On these cells ion starvation starts at about 0.5 C(A) current. A high C(A) rate cell will have electrode thickness in the 50um to 100 um range and ion starvation starts at a greater discharge rate.

Electrolyte has dissolved Lithium salt (LiPF6 typically). When dissolved it separates into Li+ ions and PF6- cat-ions. The electrolyte lithium ion transfer is like a relay race where the dissolved Li+ ions are the relay race batons. When cell is at cold temp or aged, the migration rate of lithium ions through the electrolyte to the electrodes is reduced in rate. It requires greater overpotential to push them along to meet the cell current demand.

At high cell current rate the Li+ ions bunch up at the electrode surfaces trying to find a parking spot in the electrode. During high charging rate the Li+ ions are vulnerable to attack by electrons escaping graphite causing lithium-ions to chemically converter to lithium metal. This is bad as it removes usable lithium from cell normal charging/discharging operation and grows lithium metal dendrites that can short out a cell across the separator layer.

High overpotential also drives other parasitic chemical reactions that are detrimental to cell. Electrolyte wants to stay charge neutral, with number of Li+ ions equaling the number of PF6+ cat-ions. When not at equilibrium the electrolyte is vulnerable to decomposition parasitic chemical reactions.

 

[Marinepower]

.02 Ohm/150A Discharge Test - Google Drive

docs.google.com

Ran this test again with another cell. This time, battery was in open air(cold basement). Battery started at 56 deg F as measured by insulated(from air) probe on outside of battery case in the center.

Interesting to see how the terminal voltage increases as battery temp increases.

62 Ah approximate capacity.

I think this cell performed pretty well. My only concern is that it appears the internal resistance could be slightly higher than expected.

I am designing my system to remove about 50w of heat per cell under worst case(150A continuous discharge). Does this seem reasonable?

Bestconcreteblock,

my 100ah Frey cells arrived late spring and I have been doing some test bench trials with them over the summer , before I install them on my boat.

I am noticing some of the same characteristics you report with these cells.

First they have quite high internal resistance when measured with a 1khz meter.

Second, under load they have a pretty large voltage drop in relation to, for eg, my EVE 280k.

I have only tested to about 0.4C ( and not 3C like you) but at that rate I am getting btw 100mv to 150mv voltage drop after 3 minutes of load ( which is also surprisingly inconsistent btw "matched" cells ) .

These are supposed to be high current output cells, but their IR and voltage drop seems characteristic of a storage cell? Very confusing.

However, they pulling full capacity and even in a 4P4S configuration (which is not ideal) , they are staying very well balanced ( around 15mv deviation at 3.45v)

Have you come to any further conclusions on these cells? I wonder if Frey is sending us their rejects?

Picture of a messy test set up attached....

MP

 

[bestconcreteblock]

Bestconcreteblock,

my 100ah Frey cells arrived late spring and I have been doing some test bench trials with them over the summer , before I install them on my boat.

I am noticing some of the same characteristics you report with these cells.

First they have quite high internal resistance when measured with a 1khz meter.

Second, under load they have a pretty large voltage drop in relation to, for eg, my EVE 280k.

I have only tested to about 0.4C ( and not 3C like you) but at that rate I am getting btw 100mv to 150mv voltage drop after 3 minutes of load ( which is also surprisingly inconsistent btw "matched" cells ) .

These are supposed to be high current output cells, but their IR and voltage drop seems characteristic of a storage cell? Very confusing.

However, they pulling full capacity and even in a 4P4S configuration (which is not ideal) , they are staying very well balanced ( around 15mv deviation at 3.45v)

Have you come to any further conclusions on these cells? I wonder if Frey is sending us their rejects?

Picture of a messy test set up attached....

MP

Are you sure you are isolating voltage drop across your wiring and measuring directly on cell terminal? Either way, 100-150mV slump for a .4C test on the Frey 100Ah cell doesn't seem bad to me. According to the EVE 280 charts posted by RCinFLA, a similar .4C test would slump around 70-80mV.

In my case, I have moved forward with my build as I am satisfied with my tests. In the end, I paid for a 60Ah cell, and I believe that is what I got. If you look at my second test above, you can see I was able to pull about 62Ah at this high discharge. Am I going to loose some energy to heat? Yes. Will the cells survive these high C-rates as long as I remove the heat? I think so. Would an EVE storage cell survive these high C-rates? I doubt it.

From what I have seen, this is a complex topic with not-so-simple answers. I will leave it to the those who have more experience. When RCinFLA said "0.260v drop due to R_ionic is not too bad for 3C discharge rate", that was enough for me to move forward.

 

[Marinepower]

Hey, I'm not dishing these cells and the capacity check out for me as well. Not only that, they seem to be in excellent balance after a Summer of neglect.

Your build looks professional and I note the cooling fins on each cell. Nice.

My 0.4C votage drop test was with each cell individually (at 80% SOC) and measured from no-load to 1amp, then 20amps , then 30amps and then 40amps. Volt drop was measured at the cell terminals with a Fluke meter. See attached spread sheet. I wasn't testing for high current performance, I was just trying to match the 16 batteries into groups of 4 (4P4S pack).

The EVE cells do drop about what you mention (80mv) , buts that's what is so weird. The 280k is a storage cell, not really made for a high C discharged. I expected the Freys to be less of a voltage drop and less IR. But they are not.

No big deal for me. I did not buy these for high a C application, although I do have the occasional inductive motor loads (windlass, 1000watt, 12v motor).

I think these cells are not a " jelly roll "construction, but folded ( like an accordion). Maybey that has an impact?

MP

 

[bestconcreteblock]

Thanks for the data. That is interesting.

I think these cells are not a " jelly roll "construction, but folded ( like an accordion).

I didn't realize this, do you you know which way the accordion folds go? I haven't been able to find any internal pics. This could explain the cans of the cells being concave. Are yours concave too?

 

[Marinepower]

I didn't realize this, do you you know which way the accordion folds go? I haven't been able to find any internal pics. This could explain the cans of the cells being concave. Are yours concave too?

This promo video gave me that impression...

My cells seem pretty flat-sided to me. The aluminum cell casing seems much thinker than an EVE cell. They do not seem to bloat or contract when high or low SOC.