Brain fart on looking for bad battery in a bank of parallel.

[Dsg319]

Have a bank of 60 2volt battery cells all connected in parallel.

If searching for a single bad battery do they need to be disconnect or just check each point at every battery and it will show. I’m thinking you can leave them connected.

Also can you check while batteries are under charge. I think not but having different opinions among us.

 

[LarryFine]

Easiest would be a clamp-on DC ammeter, depending on the physical jumper arrangement.

 

[Hv&Lv]

When we check our 120VDC banks we leave them connected and check across each 2VDC cell. Turn the charger off.
Should be somewhere around 2.2VDC.

When's the last time they were equalized?

 

[winnie]

If the cells are in parallel, then all the cells will have essentially the same voltage, good or bad. (The only difference is voltage drop from current flow in the wires.)

Current flow will be different, so you could plausibly detect a bad cell with some sort of current sensor that can check the wires to each cell. This is what @LarryFine suggested.

In a series string, current is the same and cell voltage is different. Measure each cell as @Hv&Lv suggested.

Do you really have a 60 cell 2V parallel bank?

 

[Dsg319]

When we check our 120VDC banks we leave them connected and check across each 2VDC cell. Turn the charger off.
Should be somewhere around 2.2VDC.

When's the last time they were equalized?

That was my measurement and I’m not sure about the second question.

 

[Dsg319]

If the cells are in parallel, then all the cells will have essentially the same voltage, good or bad. (The only difference is voltage drop from current flow in the wires.)

Current flow will be different, so you could plausibly detect a bad cell with some sort of current sensor that can check the wires to each cell. This is what @LarryFine suggested.

In a series string, current is the same and cell voltage is different. Measure each cell as @Hv&Lv suggested.

Do you really have a 60 cell 2V parallel bank?

Yes that’s what I am currently checking.

 

[Hv&Lv]

If the cells are in parallel, then all the cells will have essentially the same voltage, good or bad. (The only difference is voltage drop from current flow in the wires.)

Current flow will be different, so you could plausibly detect a bad cell with some sort of current sensor that can check the wires to each cell. This is what @LarryFine suggested.

In a series string, current is the same and cell voltage is different. Measure each cell as @Hv&Lv suggested.

Do you really have a 60 cell 2V parallel bank?

I’m assuming a series bank like in a substation

 

[Hv&Lv]

That was my measurement and I’m not sure about the second question.

Around 2.2 with the charger off for a little bit?

 

[kwired]

If the cells are in parallel, then all the cells will have essentially the same voltage, good or bad. (The only difference is voltage drop from current flow in the wires.)

Current flow will be different, so you could plausibly detect a bad cell with some sort of current sensor that can check the wires to each cell. This is what @LarryFine suggested.

In a series string, current is the same and cell voltage is different. Measure each cell as @Hv&Lv suggested.

Do you really have a 60 cell 2V parallel bank?

60 of them in series would give you 120 volts.

All sixty in parallel with only 2 volt output seems unlikely but maybe something like groups of six in series to get 12 volts and then ten sets of those in parallel to get more capacity?

 

[Hv&Lv]

60 of them in series would give you 120 volts.

All sixty in parallel with only 2 volt output seems unlikely but maybe something like groups of six in series to get 12 volts and then ten sets of those in parallel to get more capacity?

We have ~23 of these type banks on our system

 

[Dsg319]

Around 2.2 with the charger off for a little bit?

That was 2.2v with charger on. We still need to test with charger off as I originally thought.

 

[wwhitney]

Have a bank of 60 2volt battery cells all connected in parallel.

Are they really in parallel? If each battery is 2V with a maximum output of 10A, then putting them in parallel gives you a bank at 2V with a maximum output of 600A. While if they are in series, you'd have a bank of 120V with a max output of 10A.

If they're in parallel, how do you even know that one battery is dead? As long as the dead battery can withstand 2V across its terminals, the only detectable symptom would be that the maximum current output is lower by 1.6%. OK, maybe you could notice that the source impedance went up by 1.6%.

If they're in series, then with one dead battery that produces 0V but can pass current OK, you'd notice that the 120V string voltage is low at 118V. I guess my point is that it seems easier to notice a 1.6% drop in voltage than a 1.6% rise in source impedance.

If they are in series, and the string voltage is low by 2V, you can find the cell that isn't producing voltage by divide and conquer. Measure the voltage from one end of the string to the midpoint of the string. If it's 58V, then the dead cell is between the two measuring points; if it's 60V, it's not. Now you've narrowed it down from 60 cells to 30 cells; repeat the process by measuring from one end of the 30 cell substring to its midpoint. Etc. Should take 6 measurements to narrow down the 0V cell.

Cheers, Wayne

 

[Dsg319]

Are they really in parallel? If each battery is 2V with a maximum output of 10A, then putting them in parallel gives you a bank at 2V with a maximum output of 600A. While if they are in series, you'd have a bank of 120V with a max output of 10A.

If they're in parallel, how do you even know that one battery is dead? As long as the dead battery can withstand 2V across its terminals, the only detectable symptom would be that the maximum current output is lower by 1.6%. OK, maybe you could notice that the source impedance went up by 1.6%.

If they're in series, then with one dead battery that produces 0V but can pass current OK, you'd notice that the 120V string voltage is low at 118V. I guess my point is that it seems easier to notice a 1.6% drop in voltage than a 1.6% rise in source impedance.

If they are in series, and the string voltage is low by 2V, you can find the cell that isn't producing voltage by divide and conquer. Measure the voltage from one end of the string to the midpoint of the string. If it's 58V, then the dead cell is between the two measuring points; if it's 60V, it's not. Now you've narrowed it down from 60 cells to 30 cells; repeat the process by measuring from one end of the 30 cell substring to its midpoint. Etc. Should take 6 measurements to narrow down the 0V cell.

Cheers, Wayne

Much apologies everyone yesterday was a long day. They are indeed series wired