Premature Battery Sulfation?

[walmsleyc]

So, my house bank seems like it may have suffered a premature death, perhaps due to sulfation? The bank is composed of three 1.5 year old Crown, group 31, deep cycle batteries. I've got a Promariner 1240P charger (40A). Total of about 375Ah battery capacity. Winter and summer, I monitor the electrolyte to keep it topped up. Last summer, I only went through a handful of ~50% discharge cycles. Always followed by shorepower charging with the 1240P. The batteries sat on the charger most of the winter. The charger typically goes into "sleep mode", and when I went to the boat over the winter, the charger would be in this state, and the Victron would read 12.75 or higher.

The boat is in the water now, and with the charger not turned on, I was testing various lights and lighting and I noticed the voltage drop surprisingly quickly. Here's what I see as my symptoms:

• With a 3-5A load, my Victron went from 12.75V to a reading of 12.3-12.4V within 15-20 minutes (confirmed with voltmeter on batteries)
• I checked the specific gravity of all cells after charging and before testing and got readings of 1.280-1.30 (seems good)
• After a test (as above) drawing the voltage down to about 12.4V, I checked the gravity on one test cell still read 1.28 (seems still fine?)
• All gravity readings were taken at 5-8C and they seem normal
• My TTG reading on my Victron plummetted as I watched even thought the Ah drawn from the battery about < 2Ah.
• I ran an equalization charge on the batteries yesterday: 240 minutes at >15V, and tested again, with similar results.

According to Mainesails article, a "under load" reading of 12.37V means about 75% and that seem unlikely in < 30 minutes at 4 Amp draw. Any suggestions on what to test, or how to confirm their state of death.

Chris

 

[Benny17441]

Disconnect and test them individually, one bad battery will bring the other two down. I don't enjoy messing with batteries so I take them down to the auto parts and they test them for free and give me a computer read out of their condition. Some batteries will refuse to die while others may fail within days; that is why the industry of batteries warranty is so developed. It is so that the price of the same battery may vary according to the warranty chosen.

 

[walmsleyc]

I was sort of hoping it might have been bad battery, but when separated, all 3 read identical voltages.

 

[Don S/V ILLusion]

When was the last time you removed and cleaned EVERY connection in the circuit?

A poor or high resistance connection is the most overlooked cause of what you describe.

 

[JimInPB]

I was sort of hoping it might have been bad battery, but when separated, all 3 read identical voltages.

Did you charge them after separating them & before testing them?

 

[Maine Sail]

• All gravity readings were taken at 5-8C and they seem normal

^^^^ 41F to 46F is not going to yield a good indication of battery health.. Also flooded batteries will initially sag and then rebound a bit once the chemical reactions stabilize under load.. See discharge graph below...

Here is a 130Ah rated Trojan SCS-225 under a 6.5A constant-current 20 hour discharge test. This battery was being tested at approx 75F. Note the initial voltage sag to 12.494V and the rebound to 12.601V once stabilized. The batter tested here, 99.4 Ah out of a 130Ah rating, passed the 50% of "actual capacity" threshold at 11.994V.

 

[walmsleyc]

The battery interconnects are new (and clean) and everything else in ~2 years old, and I'd say everything has be tightened. No corrosion or dirty connections. Direct reading of the battery produce the same readings. I did a quick test this morning before going to work, and got the following (similar) results:

The charger was on last night, hence the 13.26 starting voltage. The load was 5A and the temp is 7C. The cool temperature is perhaps drawing the voltage reading down by maybe 0.09v? As my best guess, it seems like I'm down to about 75% after only 30 minutes?

Data:
Time (mins), V
0, 13.28
3, 12.89
5, 12.67
7, 12.5
9, 12.47
11, 12.47
13, 12.47
15, 12.46
17, 12.45
19, 12.44
21, 12.42
23, 12.4
25, 12.37
27, 12.34
29, 12.31

Chris

 

[Maine Sail]

According to Mainesails article, a "under load" reading of 12.37V means about 75% and that seem unlikely in < 30 minutes at 4 Amp draw. Any suggestions on what to test, or how to confirm their state of death.

Chris

Chris,

You don't have Lifeline AGM batteries do you? Your situation is exactly why that article was written, to show how difficult it really is to infer SoH or SoC from voltage alone....

 

[walmsleyc]

They are Crown 31HDC-130 (x3). The tests are on the bank of 3 of these. What's a practical ay to determine if these are healthy or less than usable?

Chris,

You don't have Lifeline AGM batteries do you? Your situation is exactly why that article was written, to show how difficult it really is to infer SoH or SoC from voltage alone....

 

[Maine Sail]

They are Crown 31HDC-130 (x3). The tests are on the bank of 3 of these. What's a practical ay to determine if these are healthy or less than usable?

A controlled constant-current discharge test at Ah Capacity / 20 = Discharge Rate. So for a 100Ah battery this is:
100Ah / 20 = 5A

For a 130Ah battery this is:
130Ah / 20 = 6.5A

This graph is a 20 hour discharge test of a 130Ah rated Trojan G31 battery that tested at just 76.5% SoH:

Batteries should be at 75F to 80F, discharged with the discharge rate held steady, and the cut off voltage for the test is 10.5V. You can then calculate the SoH based on the percentage of time, of20 hours, that it ran before hitting 10.5V. So if your 130Ah battery ran for 15 hours before hitting 10.5V the battery is delivering just 75% of its rating..

Or invest in a Balmar SG200 ,let it learn the bank, and report its determination of both SoH and SoC...

 

[JimInPB]

A controlled constant-current discharge test at Ah Capacity / 20 = Discharge Rate. So for a 100Ah battery this is:
100Ah / 20 = 5A

For a 130Ah battery this is:
130Ah / 20 = 6.5A

This graph is a 20 hour discharge test of a 130Ah rated Trojan G31 battery that tested at just 76.5% SoH:

Batteries should be at 75F to 80F, discharged with the discharge rate held steady, and the cut off voltage for the test is 10.5V. You can then calculate the SoH based on the percentage of time, of20 hours, that it ran before hitting 10.5V. So if your 130Ah battery ran for 15 hours before hitting 10.5V the battery is delivering just 75% of its rating..

Or invest in a Balmar SG200 ,let it learn the bank, and report its determination of both SoH and SoC...

Thank you for yet another truly useful post.

 

[walmsleyc]

I think I will install one of these.
Chris

A controlled constant-current discharge test at Ah Capacity / 20 = Discharge Rate. So for a 100Ah battery this is:
100Ah / 20 = 5A

For a 130Ah battery this is:
130Ah / 20 = 6.5A

This graph is a 20 hour discharge test of a 130Ah rated Trojan G31 battery that tested at just 76.5% SoH:

Batteries should be at 75F to 80F, discharged with the discharge rate held steady, and the cut off voltage for the test is 10.5V. You can then calculate the SoH based on the percentage of time, of20 hours, that it ran before hitting 10.5V. So if your 130Ah battery ran for 15 hours before hitting 10.5V the battery is delivering just 75% of its rating..

Or invest in a Balmar SG200 ,let it learn the bank, and report its determination of both SoH and SoC...

 

[walmsleyc]

Installed a SG200 last night, and am running the "one pass" test to help it get the initial SoH reading. It's been running for about 22 hours so far and was reading still above/about 50% SoC. The voltage in in the 12.0 range and it is reporting something like ~16 hours to go. I suspect that until it has computed a SoH value, the "time to go" is based on the "design capacity" of the bank as the numbers seem to add up for 390Ah

Anyway ... I'm running it down to 37% as per the instructions, and then I wait >= 65 minutes, and then fully charge. At that point it says it should provide an initial SoH reading.

I will report what I get.

Cheers
Chris

 

[walmsleyc]

So, after running the "one cycle" setup procedure outlined in the back of the Balmar manual, the unit reports 100% SoH (if this is accurate, then it's fantastic). I am going to run a few more discharge/charge cycles to allow the unit to fine tune it's calculations.

So I guess my concern was caused by the "under load" voltage readings be lower than the resting voltage by more than I expected.

Chris

 

[Maine Sail]

So, after running the "one cycle" setup procedure outlined in the back of the Balmar manual, the unit reports 100% SoH (if this is accurate, then it's fantastic). I am going to run a few more discharge/charge cycles to allow the unit to fine tune it's calculations.

So I guess my concern was caused by the "under load" voltage readings be lower than the resting voltage by more than I expected.

Chris

It’s not accurate yet. Flooded lead acid can take approximately 10 to 15+ deep cycles to hone in on SoH.. LiFePO4 and AGM respond quicker than FLA.. in our opinion the current Balmar SG 200 manual need some tweaking in regards to laying out expectations and how fast it will hone in.

 

[JimInPB]

Silly question here:
What is SoH?

I assume that SoC is state of charge, but I don't know SoH.

 

[dlochner]

Silly question here:
What is SoH?

I assume that SoC is state of charge, but I don't know SoH.

State of Health, or how much capacity the battery still has.

 

[walmsleyc]

After a somewhat shallower DoD test, and recharge, it's now reading 90% SoH. Still acceptable, but I hope it doesn't creep too much further down. But, that's why I bought the SG200, to find out before we head out for a ~3000nm cruise doing the DECR. I will continue to test and report.

It’s not accurate yet. Flooded lead acid can take approximately 10 to 15+ deep cycles to hone in on SoH.. LiFePO4 and AGM respond quicker than FLA.. in our opinion the current Balmar SG 200 manual need some tweaking in regards to laying out expectations and how fast it will hone in.

 

[Maine Sail]

After a somewhat shallower DoD test, and recharge, it's now reading 90% SoH. Still acceptable, but I hope it doesn't creep too much further down. But, that's why I bought the SG200, to find out before we head out for a ~3000nm cruise doing the DECR. I will continue to test and report.

The SG200 typically reads three ---, then 100% SoH, then 90% SoH, on the fourth plus reading it will be much closer to actual. I suspect you'll wind up lower than 90% if it is there on the third display reading. Deeper cycles will get to a more accurate SoH faster than shallow cycles will.

 

[walmsleyc]

3 Cycles now, and a 4th one in progress. Results so far is

• fresh install, SoH = ---
• first cycle done according to manual, then SoH = 100%
• second cycle, not as deep and a little erratic load, then SoH = 90%
• third cycle, deep, and steady load, then SoH = 90%
• 4th cycle - in progress

will post result as they evolve.

I'm not practically able to control the temperature in the boat. So most of the tests are done at
around 10C. Does the SG200 account for temperature in computing the SoH value?

Chris

The SG200 typically reads three ---, then 100% SoH, then 90% SoH, on the fourth plus reading it will be much closer to actual. I suspect you'll wind up lower than 90% if it is there on the third display reading. Deeper cycles will get to a more accurate SoH faster than shallow cycles will.