Batteries drain quickly

[CurtisM]

I've got an issue with my batteries dying overnight.

I just installed the Victron battery monitor to go with my newly installed Victron Multiplus Charger/Inverter. I did the install as per the instructions with the loads and such on the correct side of the shunt, etc. I installed these to replace a Xantrex system I thought was the problem. A couple thousand dollars to learn it was something else.

Last night at around 10pm with the inverter on and watching a movie on the laptop everything read like this:



I shut down the inverter and only had refrigeration on and a fan for sleeping. Not much amps at all being used at this point. However, this is what my batteries looked like at almost 7am. They actually dropped down like this by 1am just didn't seenshot it:



The way I read it I used about 42Ahrs between 10pm and 7am but my system got killed. It's also saying I could go for 10 days still on these batteries. While the Ahrs may be available, nothing works at 10.10 volts.

My system is two Duracell 8DAGM house batteries less than 1 yr old in parallel and one AGM start battery. The start battery is isolated with a ACR switch and is staying fully charged.

I have about 500 watts solar and a D400 windgen (hardly any wind last night). If there is wind, the batteries stay charged up. And, an 8kw Kohler genset, which I used previous to install to get the batteries to full chqrge.

QUESTIONS:

1. I charge the batteries and load test them and they test good. One thing here though is I read AGMs should sit for FOUR hours before being tested after receiving a charge to avoid false positives and negatives. Is this true?

2. Would an electrical short or drain somewhere in the boat show up in excessive Ahrs usage in addition to voltage drop?

This is driving me nuts to where I can't even sleep because the fan changes speed when there's a wind that picks up during the night. I wake up realizing the batteries are drained, again, so I go through everything in my head trying to figure this damned thing out

Curtia

[GordMay]

You indicate that you “did the install as per the instructions”; but did you set the Peukert exponent?
To what?

[Sailmonkey]

Are all negative leads passed through the shunt before they go to the battery?

[CurtisM]

Originally Posted by GordMay

You indicate that you “did the install as per the instructions”; but did you set the Peukert exponent?
To what?

It was set at the default of 1.2

[CurtisM]

Originally Posted by Sailmonkey

Are all negative leads passed through the shunt before they go to the battery?

Yes. Everything is on the load side of the shunt. From the ground bus, where everything is grounded, to the load side of the shunt thru to the battery bank

[wholybee]

How did you "load test" the battery. Yes, they need to sit before testing. The 2 8D batteries need to be tested separately, and disconnected from the boat. Did you do a capacity test? I would do a proper 20h rate capacity test, or take the batteries to an electrician that can.

If one cell is dead in the battery, you will be able to charge it to 12+ volts by overcharging the remaining cells, but it will very quickly drop to 10v as that overcharge is used. I suggest this, because that happened to me, with the same symptoms, on a 1 year old battery. In my case it was a FLA battery and easy to find the bad cell with a hydrometer.

[sailorboy1]

Your night time power use looks normal. The voltage at the 91% looks normal for that charge state and load, so it suggests that during the day you got pretty charged.

How do you decide you are charged 100%? Only acceptance amps at absorption can do this. You didnt say how often you get 100% charged.

What do you mean by you “Load tested them”?

I think your batteries have lost capacity.

[BillKny]

I don't know why everybody is fussing about the details of the SOC measurement. While those can be very important in many cases, in this situation the data is pretty clear, and I don't need the SOC data at all to know your batteries are BRICKED.

All I need to see it that you battery voltage is 10.1V in the AM. (I assume you have checked this simple measurement with a multimeter at the battery terminals...Yes?)

At that voltage your batteries are less that totally dead. Since the battery voltage reported in the evening is reasonable, I have to assume that you did not make a systemic error in connecting the battery voltage sense wires.

There is no way you could have enough power draw in a short circuit to pull the voltage down to 10V from full in a few hours. Smoke and flames would be the sign...

You say you have "load tested" the batteries. How did you do that--exactly.

Can you kill a new set of batteries in a year? You bet... Especially using a truck starting battery like the Duracell 8D AGM in deep cycle service. They are cheaper than real deep cycle batteries...

All the rest of this is just noise around the edges.

[ranger58sb]

First reading, 12.23V (relatively low) at 91% SOC... something wrong with that.

-Chris

[CurtisM]

I use a regular load tester to test the batteries. I disconnect them completely from everything else and test each battery at the naked post. It's just a normal load tester from NAPA.

[wholybee]

Originally Posted by CurtisM

I use a regular load tester to test the batteries. I disconnect them completely from everything else and test each battery at the naked post. It's just a normal load tester from NAPA.

You are not really testing the health of the battery with that. That only confirms it could theoretically start an engine. You need to do a 20h test to test the capacity the battery has. I think you will find the batteries are shot.

[more]

reading 12.23V is 50-60% not 91% SOC by
file:///C:/Users/davor/Downloads/Duracell%20Technischer%20Ratgeber%20E.pdf

[Be Free]

I'm still on the learning side of the curve here so I'm asking these questions so I can understand what is going on.


The first screen shot shows 12.23V and 91% SOC. I thought an AGM at 12.23V would be well below 50% SOC.


The second screen shot shows 10.1V and 84% SOC. That can't possibly be right. Is the Victron so set on counting Coulombs that it does not run a basic "sanity check" on the output values? I'd expect that on a cheaper monitor. Please tell me the Victron is better than that.



I would normally suspect a bad cell when I see a battery in the 10V range but this is a bank of 2 batteries. Would one bad cell pull them both down to 10V overnight?


Hypothetically, if the Victron was not programmed correctly and the OP was only looking at the SOC reading to decide when to cut back on usage and/or when to stop charging wouldn't this lead to the battery being fully charged and/or discharged deeper than it should have been?


If the previous hypothetical is true, could it account for an apparently damaged battery. I did not see where the OP indicated the age of his batteries.

[sailorboy1]

Originally Posted by CurtisM

I use a regular load tester to test the batteries. I disconnect them completely from everything else and test each battery at the naked post. It's just a normal load tester from NAPA.

Has posted that is meaningless really. You need to do a full charge till acceptance amps stop going down at absorption voltage. Probably somewhere in the 0.5-2C range. After that if the voltage still drops to less than 12 over night with just frig and maybe a fan on the batteries just dont have enough capacity. There isn't really any need to do a capacity test as they just dont have enough.

[Wotname]

To repeat...

Yep from the evidence posted, the house batteries are DEAD and unrecoverable. No need to do any further analysis to confirm that. However you need to check your charging parameters after you have replaced the house batteries to confirm all is in order.

FWIW - as posted above, the normal load tester from NAPA is not a suitable method to check the capacity of a house battery. Others have posted (briefly) on the correct method.