Premature Battery Sulfation?

Feb 2, 2006
464
Hunter Legend 35 Kingston
4 cycles and the results:
  • 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%
  • fourth cycle down to !65%, then SoH = 90%
Some interesting notes:
  • The last two cycles, while in the discharge part of the test, the SG200 showed SoC readings that I'm guessing are driven by coulomb counting. When I turned off all loads, it seems the SG200 then switchs to a SoC reading that is driven by voltage. After I turned off all loads for the last two tests, the SoC dropped SIGNIFICANTLY. Third test was stopped at maybe 40%, and after the loads were removed, and after some handful of minutes, the SoC reading went down to ~10%. For the fourth test, I stopped at ~65%, and then when the loads were removed, the SoC reading dropped to 30%-40%. This makes me feel like the the batteries are perhaps in pretty bad shape, and the SoH reading has not yet converged on the real state of the batteries.
  • Even though it is reporting 90% SoH, during the discharge part of the tests, the calculations based on the current draw, and the reported "Time To Go" (TTG(h) * current draw ) / <design capacity of bank>, suggest that the SoH should be ~80%.
Test #5 is underway ...
 
Feb 2, 2006
464
Hunter Legend 35 Kingston
Test #5: Discharged to 65%, waited > 65 minutes, and then it read 29% SoC. Again, it seems that the SoC counting method while discharging and the SoC method while resting are different, and they're not in sync.
 
Feb 6, 1998
11,667
Canadian Sailcraft 36T Casco Bay, ME
Test #5: Discharged to 65%, waited > 65 minutes, and then it read 29% SoC. Again, it seems that the SoC counting method while discharging and the SoC method while resting are different, and they're not in sync.

See post 15....

How is it physically wired in regards to the shunt? Where is start batt neg? How close to house bank neg is the shunt? Can you post a photo of the house bank wiring? Again you’re looking at more cycles before it “learns” your bank and hones in better.
 
Feb 2, 2006
464
Hunter Legend 35 Kingston
See wiring diagram below:
  • Distance from shunt to house negative is about 1'-2' (the shunt is as close to the bank as possible)
  • Distance from shunt to negative bus is about 4"

I do have an Bluesea ACR, that will combine when charging (with the Bluesea "parallel on" master switch) if that would have any impact? I can post a picture later today, but it will be a little hard to see because of the batter bank and shunt location on my Hunter (requires some yoga to get at most things).

Chris

Ramble On - Battery, Charger & ACR - Wiring.png
 
Feb 6, 1998
11,667
Canadian Sailcraft 36T Casco Bay, ME
Voltage sense, for house, needs to directly sense the battery positive terminal not the battery switch..... Also with 75A fuses on the ACR, and what appears to be a DCP battery switch, you will want to be using the SI (start isolation feature) feature or you may eventually trip one of those fuses.... I will assume there are no other neg/ground wires on the battery side of the shunt other than batteries the house bank is in parallel with? No bilge pump negative or any other neg wire on the battery side of the shunt. Don't know why you are switching off the ACR with a multibank charger as the ACR will do that job just fine with the charger feeding house directly.
 
Feb 2, 2006
464
Hunter Legend 35 Kingston
Here are some pics. The low angle picture shows how close the shunt is to the bank. Battery wire are all short and direct to the batteries.
  • Start Isolation is installed, but not on diagram
  • There is nothing on the "battery" side of the shunt except the battery
  • Switching off the ACR when the 1240P is on seem'd like "an (good) idea" at the time, but I agree with your comment that is doesn't have much value.
I will move the house "Sense" wire directly to the battery. Are you suggesting this to avoid any voltage drop? or just to keep it "always connected" to the battery?

Again today, with a discharge down to 65%, once all load was removed, and waiting enough time (> hour) , the SoC went from 65% to 45%. The SoH still remains at 90%.

Does ambient temperature affect the SG200?


20190511_085204.jpg 20190511_085144.jpg
 
Feb 2, 2006
464
Hunter Legend 35 Kingston
Latest result summary:
  • fresh SG200 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%
  • fourth cycle down to ~65%, then SoH = 90%
  • fifth cycle down to ~65%, then SoH = 90%
Observations:
  • For the last few discharge cycles that I observed (perhaps all), the SG200 change its SoC reading jumped down by 20%-30% after all loads were removed and allowed to sit for 60+ minutes
  • For the last (fifth) charge cycle, when the 1240P charger stopped charging, the SG200 read 75%. This morning, (9 hours) later), the SG200 reading is now 100%. This was the first time I was present when the charger completed it's charge cycle, so I suspect that the pattern was the same for other charge cycles.
I "seems" that when the unit only relies on voltage it thinks things are about 20%-30% worse than when it watches Ah's.

Chris
 
Feb 2, 2006
464
Hunter Legend 35 Kingston
  • fresh SG200 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%
  • fourth cycle down to ~65%, then SoH = 90%
  • fifth cycle down to ~65%, then SoH = 90%
  • sixth cycle down to 79%, wait 4 hours, SoC read 46%, recharging ....
Again a big discrepancy between SoC reading under load and after resting.
 
Jul 23, 2009
857
Beneteau 31 Oceanis Grand Lake, Oklahoma
My suggestion is to take the batteries home and charge them one at a time with a known good charger. What is the peak charge voltage? Float voltage? Resting voltage after 24 hours? Run several 100 amp / 10 second load tests with a cheap battery tester. What are the results? What is the resting voltage several hours after the test? Do all this at room temperature, 77f or 25c if possible.

Is there any chance that the charger leads have a bad connection?
 
Dec 4, 2018
60
Balboa 27 Denver
Wonderful set of posts, and mini battery clinic all in one. Thanks MaineSail for your persistence in this.

Take home reminder to me, is pulling a deep cycle battery out and taking into a "battery shop" for load testing does almost nothing in the context of use on a boat. Just saying. Taking them home and individual subjecting them to a deep discharge test and cycle, is the only way to do this, or the Balmar thingee.
 
Feb 2, 2006
464
Hunter Legend 35 Kingston
Latest update:

Moving the house bank "voltage sense" wire from the SG200 from where I had it, to the battery caused the SG200 to lose it's "learning" about the batteries! Need to restart the process.

Chris
 
Jan 11, 2014
11,323
Sabre 362 113 Fair Haven, NY
Latest update:

Moving the house bank "voltage sense" wire from the SG200 from where I had it, to the battery caused the SG200 to lose it's "learning" about the batteries! Need to restart the process.

Chris
I may be incorrect here, however, it is my understanding that batteries have a finite number of charge/discharge cycles before the battery is no longer serviceable. If that is the case, then repeated cycling of the battery would accelerate the batteries demise. Perhaps it would be better to let the SG200 learn the batteries from the natural cycling that occurs from use, rather than forced cycling.
 
Feb 2, 2006
464
Hunter Legend 35 Kingston
I agree, but I'm trying to determine the SoH of these batteries before I depart on a long trip, so losing a bunch of cycles to testing is acceptable for my purposes. If I was just going to be summer cruising where I had easy access to ordering new batteries if necessary, then I would do as you suggest, and just let it learn naturally.

I may be incorrect here, however, it is my understanding that batteries have a finite number of charge/discharge cycles before the battery is no longer serviceable. If that is the case, then repeated cycling of the battery would accelerate the batteries demise. Perhaps it would be better to let the SG200 learn the batteries from the natural cycling that occurs from use, rather than forced cycling.
 
Feb 6, 1998
11,667
Canadian Sailcraft 36T Casco Bay, ME
Latest update:

Moving the house bank "voltage sense" wire from the SG200 from where I had it, to the battery caused the SG200 to lose it's "learning" about the batteries! Need to restart the process.

Chris
Yes this will happen.
 
Jul 23, 2009
857
Beneteau 31 Oceanis Grand Lake, Oklahoma
I do agree that load testing batteries with a cheap load tester isn't ideal but it's one more tool in your arsenal. I find that the cheap testers provide more repeatable data then the digital ones, unless your are talking about lab grade equipment.

Dave, your are absolutely correct in that batteries only have so many cycles in them. The OP seems to be OK with that.

Judging from the discharge curve provided earlier, these batteries don't look very healthy. I generally don't test batteries at that low a temperature though. I do work with batteries almost every day, group 34 AGM all the way up to 4,000Ah flooded. My house bank of 2 group 34 AGM start batteries (I'm aware that I shouldn't use start batteries for a house bank but the price was right) are over 5 years old and will hold more voltage at a 5 amp load for 30 minutes than the OP much larger bank of 3 group 31s. Something doesn't seem right.

Maybe the charger is going into float too soon.
Is the temp. probe connected and working? (I just looked at the pictures and can see that the probe is connected, I should check the pictures sooner.)
 
Last edited: