What drained the sealed batteries?

[DixieMtCR]

I finally got access to the Victron SmartShunt 500 via a new Cerbo GX and wifi AP. Connecting to the shunt via Bluetooth standing next to the electrical cabin in the snow wasn't cutting it. Nuff Said!

With the Victron Cerbo GX connected to the Smart Shunt all connected to an AP, connected to the VRM, I have been able to monitor battery levels. Last night was a real eye opener. An alarm went off notifying of low batteries.

Battery level was about 26.20 volts at 10:30p, and nothing was connected to be using power. Victron VRM data shows the batteries were at 97% and the current was "Idle." The only drain would have been the inverter which draws about 24w. The battery bank is made up of 8 100Ah Sealed Deep Cycle RV batteries. They are wired 2s4p to provide for a 24v system. There are 3 charge controllers feeding the batteries, each fed by their own 600w solar panel rack. They are 40a MPPT charge controllers.

This morning I checked the status on the VRM and noted that the battery voltage was in alarm and had reached the alarm min voltage of 20.00v. The data showed that the batteries were at 26v till approximately 1am. At that time the voltage dropped, within a couple of hours, to 20v. The current data during that time showed only about 500ma. Certainly not enough current to tank the batteries. The charge controllers have "load" outputs but nothing is connected to them. They are use for things like lights in an RV. It was DARK and Rainy, (Oregon's Best), all night so the panels never caught a single usable photon! The system should have just rested all night.

I have a 24v battery charger connected to the battery bank right now to bring it back up to full. I just don't get it. Oh, BTW, there still aren't any photons making it through the dense Oregon clouds and rain.

Any thoughts, ideas, or experience with a mess like this? Only current drains batteries.

 

[Rednecktek]

Cold also drains batteries, so add in the temperatures and the constant draw plus lack of charging and it really adds up quick.

 

[DixieMtCR]

Cold also drains batteries, so add in the temperatures and the constant draw plus lack of charging and it really adds up quick.

Yes, but last night the temp when the dive in voltage occurred was 45F. And, the voltage did not just drain linear or slowly like you would think with nothing connected to it. it literally took a dive. You can see in the photo below that the current was near totally flat and the voltage just dove into an alarm state.

BTW, the time is Eastern Time. I am Pacific time. I have never figured out how to change that.!

Strange

 

[HighTechLab]

Did you make absolutely certain nothing is connected to the batteries negative terminals other than the shunt? 100% of everything must be connected to the system side of the shunt and only interconnecting cables on the batteries otherwise the shunt can't register it.

Also, make certain that the shunt is installed in the correct direction (battery size to batteries - yes, I've seen that one many times before).

 

[DixieMtCR]

Did you make absolutely certain nothing is connected to the batteries negative terminals other than the shunt? 100% of everything must be connected to the system side of the shunt and only interconnecting cables on the batteries otherwise the shunt can't register it.

Also, make certain that the shunt is installed in the correct direction (battery size to batteries - yes, I've seen that one many times before).

Hi Dexter,

You have some good points here. I would not want to leave anything unnoticed. I have gone over things looking for clues as to what has happened to take out this system. It has been working just fine with the shunt, yielding good readings with VictronConnect, for the last year.

Looking at the VC Data, I could judge when and how long I could use this 3600w max off-grid system. A few weeks ago, I started noticing that the panels were not pushing as much current as normal. One rack of panels (600w in 6 panels) had a lower than normal voltage and was not producing much at all. It was the high producer at one time. I swapped MPPT CC with another rack and no difference. I inspected the panels in the rack and saw black jagged lines running through the cells below the front glass. No cracks were visible but the elements in the cells are discolored where the dark line cross those elements. I have worked overtime to try to see cracks in the glass and there are none. The open circuit voltage on each panel is within specs but the short circuit current is around 14 milliamps on all. Open circuit the voltage is there but they cannot push any current.

I'm convinced that the wiring on the shunt is correct and the shunt is in the right direction. I'm stumped.

Currently, the panel rack in question is disconneted and the Charge Controller is disconnected from the battery bank. Something is still amiss.

 

[mikefitz]

Best guess is a problem battery or more than one, in the pack. I suggest checking each battery out of circuit, at least for terminal voltage.

Mike

 

[Vigo]

Sounds to me like a cell decided to 'randomly' crap out and the rest of the pack discharged into it. You don't happen to have temp monitoring data for the battery space during this time, do you?

You will probably find the answer just by checking the individual battery voltages. If not, i would see if anything is visible (residue, melting) around the vents of only one battery, or if one battery has any distortion to the shape of the case. If one cell vented hundreds of amp-hours worth of heat you may even be able to detect a difference by weight. Of course you could easily tell by how each accepted charge if charged individually, but that's tedious vs the 'physical/mechanical' clues i just listed so i would try the easy stuff first. If you have a small load you could apply to each battery individually, you could also compare how the voltage sags with that given load on each battery as that may be a lot quicker than individually charging even if only for the same 10 seconds. But all of that is ONLY IF you can't immediately tell by the terminal voltages.

 

[DixieMtCR]

Thanks everyone. Looks I have some work cut out for me. Will have to wait till there is better weather since it is in the 30s, raining and snowing these days.

One of the racks of 6 panels has mystery black channels zig zaging back and forth across the cells of every panel. I am working with the Amazon Seller to replace them since they don't seem to be performing any more. I have to take them all off the rack to get into the junction box of each to check the diodes. I have sent the photos and other voltage and current measurements and they just keep asking more questions.

 

[Vigo]

and they just keep asking more questions.

That seems to be a common thread among companies who have no intention of warrantying anything. They just run out your patience, or the clock, with BS words that the customer service rep who is spewing them to you, probably doesn't even know what they mean anyway. You can play 'their game' for a while but at some point you will have to switch from compliance to threats so they understand that they are not going to talk you into going away.

 

[DixieMtCR]

I certainly will not buy any more items from that company. This is one of the problems with buying from amazon any more. Their paradigm shifted from having things and selling things, to totally hosting vendors from all over the world. As you say, companies who are not interested in standing behind their products. All they have to do is sell to a couple of thousand costomers ONCE, and they are rich. They are not counting on return customers. Also, I hate it when I buy something and expect it to be delivered within a reasonable amount of time only to find out that the company is in China and your purchase will not "SHIP" till the next container ship leaves China. Then, who knows after that. I have had items take more than 2 months with no word from the company during that time.