Voltage Drop When Load Turned On

[workmaster2n]

Today, while cutting out a section of corroded wires, I've run into a situation my sleep deprived brain can't reconcile.

My internal lights are on one breaker, but splits off into 2 parallel circuits. The first circuit controls one light fixture, and the second circuit is itself a parallel circuit that controls 5 fixtures. All the fixtures are LED, so amperage should be fairly low (the one fixture I checked, is "max 5W", according to the label).

I was able to restore power to the 6 light fixtures this evening, and everything was running great. Until it wasn't.

I had the 6 lights on, to test that they would work, when, suddenly, they all turned off. I had installed a Blue Sea Dual Bus terminal strip to connect the one power source to the 2 light circuits. I threw a multimeter across the two buses and there was only 8 volts being registered. Turning off all the light fixtures would result in the voltage being read as 12.8 volts across the bus.

I checked the resistance of the incoming power line, and it read between 6 and 7 mili-ohms. When I tried to read the resistance of the 2 out bound circuits, I would get OL, even if there was a light fixture turned on in each circuit. This OL, combined with the general state of the outbound circuits, lead me to believe the problem was in the outbound circuits.

I tried removing one outbound circuit then the next from the bus bar, but always got the same result of flickering, very dim lights, and a voltage reading of 7-8 volts across the bus once I turned a light on.

Sometimes, however, the lights will turn on as if nothing bad has ever happened.

The fact that both outbound circuits turned off at the same time, and the voltage drop is the same regardless of which outbound circuit I have connected, now makes me think that the voltage drop is a result of a bad connection on the power input side of things. Does this make sense with what I've described? If the bad connection was on the outbound side of things, I'd expect to see one or the other outbound circuit start to fail, while the other is unaffected. And I guess the voltage drop would be at the light fixture itself, instead of at the bus bar?

I've got another bus bar for my MFDs and other marine electronics with spare outputs open, so for testing, I was thinking of running a new "power source" from the MFD bus bar to the newly installed light bus bar. I'm hoping this will at least help me narrow down if the voltage drop is coming from the power cable into the light bus or something else.

I'm also a bit confused why the lights worked just fine, and sometimes continue to work just fine, but then all of a sudden cut out and I see a 4v (pretty much 25%) voltage drop. Is this just a sign of a corroded wire/poor crimp? Or should I be looking elsewhere?

To add to the confusion, when debugging the system earlier today, I used crappy screw terminal connectors to try and find which sections of wire needed to be replaced. Both outbound light circuits worked just fine when plugged into the power source using this temporary junction. This evening, after diagnosing the failure for a few hours, I cut off all of my crimped on connectors and tried to replicate my earlier success with the screw terminal connectors. I get similar results as the bus bar, wherein as soon as I turn on a light, most of the time I drop 4v across the power side of things. But sometimes the lights work for a few minutes before failing again.

[smac999]

If you have 8v at the buss. The issue is between the buss and battery. Use voltage not ohms. A single strand of wire is still 0ohms but can’t carry current.

[donradcliffe]

I'd get a jumper wire with alligator clips. Clip one end to a good 12v source, then put the other end at various points in the circuits. If the lights come on bright, the circuits downstream of where you clipped are good, and there is something wrong upstream. One of the first things I would jump to is the breaker output.

I would also use my clamp on ammeter to make sure there isn't a short, but its not likely in your case, as both load circuits behave the same.

[ttex]

Bad connection feeding the bus bar. Bad crimp, corroded wire, loose nut, etc..

[glenn.225]

I would look for a partial/intermittent short in one of the circuits. Bulb, receptacle, switch or elsewhere.

Sent from my SM-G970F using Cruisers Sailing Forum mobile app

[Be Free]

Just throwing out a few things to consider....


Assuming that your lights are white LEDs and that they are all wired in parallel, the forward voltage is pretty close to the 4v drop you are seeing. The voltage drop may not be a problem.


You mention that you cut out a section of corroded wire. If there is still corrosion (or anything that significantly increases the resistance of the circuit) you may not be able to supply the current required to drive the LEDs. Without knowing how many actual LEDs are involved there is no way of knowing what current is required.


Remove the power from the circuit and check the resistance of the wire after isolating each section as much as possible working outward from the battery. I think you are going to find a section of wire with much higher than normal resistance.

[stormalong]

You say that you checked the resistance and got 6 - 7 mili-ohms.

Are you sure of that reading? Most ohm meters will not read mili-ohms (thousandths of an ohm). Could that be mega ohms (millions of ohms)?

Voltage drop is caused by the amount of current through a conductor. A high impedance multimeter will not show a voltage drop in a circuit with no load. As you turned on more load the voltage dropped.

As others have said you have a poor connection or a very corroded wire somewhere.

[workmaster2n]

Originally Posted by stormalong

Could that be mega ohms (millions of ohms)?

100% correct. I was misreading the multimeter. The fact that the resistance is 6-7 mega ohms makes a lot more sense as to why the cable from the battery to the bus is the issue. Great catch!

I'm off to the store to pick up pieces so I can wire the bus to another 12v source.

Thanks everyone for the suggestions, this is making a lot more sense this morning than it was last night.

[AKA-None]

Remember also that an led will read like an open circuit one way
Switch the leads and measure the same circuit