Working on a large battery system

[sheldon]

You may think I am in the wrong area for this thread, but this is not about ghosts per say. I was working on a large battery system that is suppose to be floating in referance to ground. These batteries are gel cells and are haveing problems with leaks. These leaks are causing ground faults in the system. I was trying to troubleshoot the problem using a multimeter measuring the voltage from ground to the posts on the batteries. In theory the battery that has the least voltage or where the voltage flips from posative to negative is the location of the ground fault. I found several readings like this and thought I found my ground fault. I left and came back and the low or flipped point has moved without any interference from me. I was confused and thought there was more to this that meets the eye, so I conducted some research and found out that this has been labeled as ghost troubleshooting and no true procedure can accomplish pinpointing the problem and only experience can make it happen. Has anybody herd of this? I realize that ground is a general term as far as voltage potential zero. I would still think the ground fault would still be in the same location though? Any wizzards out there no the physics behind this?

 

[sheldon]

sorry for the wizzard reply, can anybody help with this

 

[wimms]

Well, if you have floating ground, and you have ground faults, then it seems natural that you have floating faults, isn't it?

There are faults that cannot be easily found by meters, good understanding about how it works and what can go wrong is needed. For eg. audio amp can loose all of its power and get hot. You start measuring it, and it recovers immediately. As soon as you touch it with a meter, it heals. Actual cause occurs to be feedback induced oscillation at very high (radio) frequencies. You not only hardly detect it due to low amplitude, but you simply don't even think about it, its audio amp afterall. Meter merely shuts down the oscillation. Knowing this is part of 'experience'.
There is hardly any true procedure to pinpoint the source of a problem, other than 1) understand how it works, 2) understand what can go wrong, 3) eliminate all that can go wrong

So, without understanding schematic its hard to tell anything. The more experience you have, the less 'eliminate all' you'd have to exercise, you'd go directly to source of problem 'by intuition'.

 

[Dx]

how many batteries are there?

R they in series or parallel? beisides gnd is gnd but the best way to troubleshoot this is by using a half split method. that is mewasuring in the middle to find the fault, if its in the middle 1/2 split that till you find where the voltage drops. I am thinking a leaking battery, you should be able to visually see this to locate your battery problem, sir.
Let me know if this helps?
Dx

 

[sheldon]

Originally posted by Dx
R they in series or parallel? beisides gnd is gnd but the best way to troubleshoot this is by using a half split method. that is mewasuring in the middle to find the fault, if its in the middle 1/2 split that till you find where the voltage drops. I am thinking a leaking battery, you should be able to visually see this to locate your battery problem, sir.
Let me know if this helps?
Dx

Yes I have already located the problems via careful inspection and cleaning. I am just trying to understand what caused the phantom readings. Your right about the how to troubleshoot the problem and it works to a point. Thanks for your help

 

[Tyger]

I've worked with amps too

and when things get really weird put a good o'scope on it, or even near it. If it has RF oscillations you can find them.

 

[NewWorldSamurai]

weird