Blocking Diodes

ghostlymoron

ghostlymoron

Well-Known Member
Joined
9 Apr 2005
Messages
9,889
Location
Shropshire
Visit site
Last year I found myself in the unfortunate position of having all my batteries discharged and unable to start the engine. I have 3 batteries - 1 engine battery and 2 leisure batteries. These are connected to the alternator via a diode box. The reason the leisure batteries were flat was due to a problem with the bilge pump but this does not explain the flat engine battery. One possibility is that the blocking diodes are not blocking and allowing current to flow between the two battery banks. Is it possible to test the diode box using a simple multimeter?
 
[ QUOTE ]
Is it possible to test the diode box using a simple multimeter?

[/ QUOTE ]Most multimeters have a diode test function, but you'd need to look at the instructions for your meter. In my experience, blocking diodes usually fail as an open circuit, rather than a short. Could there be another reason, like a 1-2-Both switch left in the wrong position?

If you decide to replace the blocking diode, a simple VSR could be your best choice.
 
I suspect that there are 3 terminals on your diode box? - 1 in terminal and 2 out terminals.

Disconnect all terminals, and you should get a low reading with one lead on the in terminal and the other lead on each-in-turn of the other terminals. Now reverse the test leads and there should be no continuity between the in and 2 out terminals.
(You may get this test back to front, but it should still be the same i.e. first test with no continuity and second test with reversed leads with continuity.)

Now check between the two out terminals. There should be no continuity between these terminals whichever way around you connect the test leads.
If you get different readings from this, i.e. continuity, then there is a problem with your diode box. This will be why both your batteries were flat at the same time.
 
Testing blocking diodes with a digital multimeter can be confusing- make up a test light with a 12v bulb and two wires.
First make a sketch of all the connections on the diode block, then disconnect the battery and all the wires at the block.
Identify which wire came from the alternator positive to the diodes, ( possible labelled A or input) connect the diodes there to battery positive.
Connect one end of your test light to battery negative, and the other end to the output terminals of the diodes (possible labelled B1, B2), one at a time. The bulb should light up on each terminal. If not, your diode (s) are open circuit and bu**ered.
Now, reverse the two connections at the battery, and try again at each output terminal- the bulbs should not light now, if they do, your diodes have failed as a short, and are equally bug*ered.
You can try the same test substituting the bulb with your multimeter on a 12v scale, but expect to read 11.5v (bulb on) or about 1volt (bulb off)- or maybe not...depends on the make of diode, thats why the bulb test is more definitive.
Good luck.
 
Test the blocking diodes ? disconnect the service batteries, check engine battery is switched on / connected and turn the cabin lights on.

If the lights come on, the diode is faulty, if they do not come on, it's working.

Either way through it away and fit a relay, instantly more battery power.

Brian
 
people have made comments about throwing away the diodes. However if your system has been set up to use diodes then changing to VSR (voltage sensing relay) may introduce problems.
Diodes have a voltage drop (almost regardless of current) for the common silicon diodes this is .7 volt. If the diodes are simply wired in to an alternator producing 14v regulated then batteries will be undercharged as max voltage will be 13.3V. Two ways around this are to increase regulated voltage to 14.7V or to use what is called load sensing regulation. Here a wire is run from a battery that senses the regulated voltage from that point ie 14v at the battery. (downstream of the diode and its voltage drop)

So if you have diodes then either of these 2 methods or neither will be used. You need to know. It is quite conceivable that the diodes have been bypassed by wire because of poor charging. (hence no isolation) Diodes are usually very reliable.

So you need to test the diodes (try wired in situ) as well as disconnected.
As already said using a DVM on ohms can be very confusing although it should show very high resistance one way and probably hundreds of ohms the other way. (if good) good luck olewill
 
I had a similar situation. I found one battery very low when I was sure that the main switch was off. In the end I found that the VHF, although turned 'off' was still drawing some current because it was permanently wired to that battery. I re-connected it after the main switch. Problem solved.
Perhaps you have the same arrangement?
 
Charge up the engine battery and measure the current drawn from it with the service batteries removed.
It may be that your bilge pump problem prevented the engine battery ever getting charged.
Depends on the details of your diode split circuit. I have seen some quite poor ones!
 
Only ever seen splitter diodes fail when abused -underrated or when an anchor windlass was trying to run through one... and it failed shorted.
I would vote against the relay as it has moving parts and points to get splatted.
It is relatively simple to set up the charging system to produce the extra 0.6v required by diodes.
As for your problem, it depends on your meter. Most nowadays have a diode setting which will display the forward loss in volts one way, and opencircuit the other.
Changing the loads and switching batteries on and off shouldn't do much to diagnose, as the diodes should only be in the charge circuit from the alternator/mains charger- they should never have to carry load current.
 
You must log in or register to reply here.

Other threads that may be of interest

W
Charging problem
Replies
8
Views
410
lustyd
lustyd

Share this page

Top