This posting is in english because it is the language we used to discuss it.
A - PREPARATION
From the battery bank (BB) identify all the existing connections:
- Normally there should be only one going to the main control panel (CP) by the chart table.
- This his almost never the case as devices that were installed in a second or third phase are usually connected directly to the BB, like SSB, WaterMaker and so on ....
- The best way to identify those undocumented connections is to switch everything OFF on the CP and to start/switch ON all the other existing devices, one after the other, even those that are supposed to go through the CP like lamps or 12v plugs.
- The ones that will start/work are those connected directly and after their cables and connection have been identified it is a good idea to label them and take some pictures.
- Once this is done all those devices should be disconnected from the BB, up until nothing remains connected, except the CP with all its swiches OFF.
It is a good idea, at this point, to build a table of all devices and to use it as a checklist for the next steps
<!========================================>
Exemple 
| DEVICE | Direct | Via CP | Connected | Switched ON |
|---|---|---|---|---|
| Chart plotter | YES | no | no | |
| Mast lights | YES | YES | YES | |
| Water maker | YES | no | YES | |
| SSB Radio | YES | no | no |
Direct: means directly connected,
CP: means through the Control Panel
Connected: is (+) if the leaks exists after reconnection
Swiched ON: is (+) if the leaks appears when the device is switched ON
In our case:
- The chart plotter is OK
- The mast lights cables definitively have an issue
- The water maker generate a leak when it runs
- The SSB is OK
<!========================================> Additionally, if the cables are difficult to identify it migh be useful to have a digital multimeter, some wire, etc to do some testing.
B - TEST
B.1 - Direct connection
Now one can manually and carefully re-connect the directly connected devices, one after the other, and each time test for a leak in 2 steps:
1 - after re-connection whilst the device is still switched OFF and
2 - after swiching ON the device.
- Doing it this way, should help differenciating leaks due to the connection cable from those due to the device itself.
- Once tested the device may be kept connected or disconnect for more security but disconnection seems better as it eliminate all possible 'falsly positive'.
- Then the result should be logged on the check list and, even if positive, it is almost mandatory to move to the next device as there can be several devices leaking.
B.2 - Connections throuh the Control Panel (CP)
Wether a possible culprit has been found or not, at that stage the same process has to be reproduced with the devices connected to through the CP by switching them all, one after the other, and logging the result on the check list.
Here again it is the same 2 steps process:
- Connect through the switch panel, then
- Turn ON
C - RESULTS and LAST CONTROLS
- Normally at this point the leak's origin has been identified but there could be case where is still there and no culprit has been identified.
- If that's the case, it is worth investigating the connections between the different elements of the Battery Bank as the cables might have been suffering from chaffe if not properly protected or fixed.
- If this proves negative then it is time to check if it is not the leak detector that has a problem or if it wrongly wired.
All lights on the mast.
The cables from the wind turbine and the solar pannels because of chaffe.
The alternator.
The SSB radio and any of its components.
The watermaker pumps.
The boiler.
The windlass.
The bowthruster.
The electric winches and furlers.
This is definitely a painfull and time consuming exercise but it is worth doing it well by documenting the results by filling the checklist table and taking pictures of the connetion after they have been labelled.
