Troubleshooting DC Voltage loss under load

[smokey73]

After reading Nigel Caldors book on mechanical and electrical manual I have devised a plan to try to pinpoint the voltage loss I am experiencing when under a moderate load (11 amps or so).

When under load when I have 12.5 volts at the battery, I only have 12. 1 or so volts at the DC electrical panel at the input side of the DC power breaker. According to Calder, this would indicate that the loss is occuring somewhere between the battery and the DC power breaker rather than downstream of the breaker.

I intend to put an 11 amp load on the system and then start measuring Voltages at each point along the path. This includes the input and output of the battery protectioin fuse, the DC ON-Off switch input and output and finally at the DC breaker input. I will use the same ground point for all measurement to take that out of the equation. Wherever I have a voltage loss it would indicate a high resistance connection or an electrical cable too small to carry the load without voltage loss.

Since the loss occurs before I get to any load, it would seem that the loss occurs before it gets to the DC panel. Of course, I intend to clean and reland all connections regardless of the results. But if its in the switch itself cleaning the connections won't do any good so I'd like to eliminate that possiblity.

Does that seem like a reasonable plan?

 

[Rich Stidger]

Don't forget any ground connections between the battery and the DC panel.

 

[smokey73]

Don't forget any ground connections between the battery and the DC panel.

For this test I'm elimating the ground path losses by using a ground directly to the battery negative. If I don't get losses across the positive path in this test, then the loss has to be in the negative and then I can focus on that. Using Nigel Caldor's book though, from his troubleshooting procedure it looks like it is most probably on the power, rather than the gournd side.

PS: At no load, the voltage at the battery and at the input to the DC panel breaker is the same. so it is related to resisitve loss or possibly a DC leak of some sort.

 

[JohnShannon]

0.4 volts drop for 11A is a resistance of 0.036 ohms. How do you know it is 11 Amps? some of that drop will be in your current measurement system. How thick and how long is the wire from the battery to the panel and how thick and how long is the return from the panel? If we know these things we can calculate the resistance of the wiring (there are many online calculators) and that will give us a sense of how much excess resistance there is.

 

[srimes]

For this test I'm elimating the ground path losses by using a ground directly to the battery negative. If I don't get losses across the positive path in this test, then the loss has to be in the negative and then I can focus on that. Using Nigel Caldor's book though, from his troubleshooting procedure it looks like it is most probably on the power, rather than the gournd side.

PS: At no load, the voltage at the battery and at the input to the DC panel breaker is the same. so it is related to resisitve loss or possibly a DC leak of some sort.

You're on the right path. You're using voltage drop as a proxy for resistance. You can use an ohmmeter to measure resistance directly if the leads can reach.

I'm rusty here so correct me if I make a mistake. V=IR. For the leg you measure, v went from 12.5 to 12.1, so V=.4. If I = 11, R is 0.036.

edit: JohnShannon beat me to it

 

[dlochner]

If you get a clamp on ammeter you can measure the actual current being drawn through the system.

Assuming clean tight connections on both the positive and negative side, the 2 main culprits would be undersized cables or aging batteries.

First question, how old are the batteries, what type of batteries are they, and how are they being maintained/charged? A battery on its last legs can show an appropriate charge level of 12.5v, however the depth of that charge will be low and any load will quickly drop the voltage. Since alternators and battery chargers use voltage to determine the charge current, the batteries will not be fully charged. Voltage is not a reliable indicator of a battery's health. Consider this, 9 AAA batteries can be connected to yield 13.5 v, but they would not have nearly enough power to support an 11 amp draw for any length of time.

The size of the cables should be easy to determine, it should be printed on the cable every few inches. There are any number of charts available online that can guide you on cable size based on current draw and length. BlueSea.com has a nice one.

 

[Whatfiero1]

Home depot or harbor Freight sell a tone tracer great tool to trace path of wires just clamp on wires and follow beep to find bad spot in wire

 

[rgranger]

I had something similar happen to me years ago. I had 13 V at my bow light but when I flipped the switch it dropped to 9 V and the LED would not activate at that voltage. After ripping all of the wire out and replacing it I still had the same problem. Turned out to just be corrosion on the battery terminal.

 

[smokey73]

I should have explained more about what I have done so far and I apologize for that.
@dlochner - The batteries are about 3 years old FLA and I recently conducted a full 20 hour battery capacity test using @Maine Sail instructions with a constant current load cell. It tested at 93% Capacity when placed on a constant 11.5 Amp load for the test (18.75 hrs)

I have 2/0 cables from the batteries to the battery fuse (Blue Sea) and 2/0 cables to the Battery ON/OFF switch (just on/off, no both position) From there, it goes directly to the DC panel breaker and was wired from the factory by Hunter but I did not see a wire size printed on the cable but I'll look again today. It is a "brown" with some color stripe but I don't recall the stripe number. I know its not a 1/0 cable and if its not marked I can measure the wire diameter (metal only) with a micrometer and get the wire size.

The drop is on the order of 3% but I would not expect it to be this much for the main power connections to the panel itself. Any drop beyond the 3% going to the loads themselves like the instruments would put the voltage at a level that I would not want.

@JohnShannon - I am measuring the current using a my Victron 702 battery monitor which measures directly at a shunt located about 10 inches from the Battery negative pole and there are no negative leads located anywhere on the battery side of the shunt. I am also measureing Battery Voltage with the Victron but checked it with a digital Voltmeter directly at the battery and it was within .01 volts so I know my voltage and current values are pretty accurate. I am measuring the voltage supplied at the positive post of the DC breaker panel using the same digital voltmeter. When I am not under load, I get the same voltage at the Victron, Voltmeter across battery, and at the DC panel breaker positive post (before the breaker). When 11 amps is drawn I get a .4V difference.

AND YES, I know I can just remove and clean all connections, which I intend to do, but I'd like to know the cause of the increased resistance (sorry, Im an NGINEER). I never had this problem before so something has changed. I know its some increased resistance somewhere. Has anyone ever had a Battery ON/OFF switch build up corrsion internally to the switch contants (not the posts, but the switch itself)?

Quoting Caldor's DC circuit troubleshooting guide: "The source of the voltage drop can once again be narrowed down by connecting the meter from the equipment positive lead to a known good ground (J to L in Figure 3-15). Wit the load "on," if the meter shows less than system voltage, there is a voltage loss on the positive side."

The load here is everything downstream of the positive power lead to the DC breaker on the panel and the known good ground is the lead directly back to the battery negative terminal.

 

[kloudie1]

The "downloads section of the 40.5 info page has owner's manual. It has some wire size info.. looks like 12 ga from batt switch to panel?

https://sbo.sailboatowners.com/downloads/Hunter_40.5_72237343.pdf

 

[smokey73]

The "downloads section of the 40.5 info page has owner's manual. It has some wire size info.. looks like 12 ga from batt switch to panel?

https://sbo.sailboatowners.com/downloads/Hunter_40.5_72237343.pdf

I sure hope not 12 gauge from the panel to the battery. That would be way undersized for a 12 volt system as the main power wire I believe. Can't imagine why they would do 2/0 from the battery to the switch and then only 12 gauge from the switch to the panel? I think the page you are referring to is showing 12 gauge from the battery switch panel to the bilge pump, which is directly connected to the battery at the battery ON/OFF switch. The bilge pump wire looks like 12 gauge but the main power lead looks much bigger. I'll check when I get to the boat later today or tomorrow.

 

[dlochner]

It is a "brown" with some color stripe but I don't recall the stripe number.

Brown is used in bilge pump wiring. The cable to the panel should be red and pretty large.

 

[MitchM]

I heartily second DLoachner's recommmendation of a clamp- on ammeter. it helped me find a pinched wire in less than an hour that a 'professional' couldn't find in 1/2 a day. PS Don't buy a top of line Fluke $340 clamp on w/ Ac and DC features --mine quit doing 'clamp on' amp capability in less than 2 years of occasional use w a one- yr warranty of course.)

 

[smokey73]

Brown is used in bilge pump wiring. The cable to the panel should be red and pretty large.

Its a pretty good size wire but probably not 1/0. It's been awhile since I looked at it. There is a brown with striped wire going to the bilge circuit. The main power wire may be red but if it is red it has some sort of stripe on it as I recall. I'll check when I get to the boat tomorrow for the troubleshooting.

Thanks for the tip.

 

[JohnShannon]

2/0 cable is 0.077 ohms per 1000 feet. So that cable is thick enough. The "brown cable" if it was 12 ga would be 1.5 ohms per 1000 feet so 10 ft to the electrical panel and 10 feet on the return would give you 0.03 ohms. All connections, switches, fuses and wires have some resistance. So maybe it is just the wire from the switch to the panel.

Start with the black lead of the voltmeter on the battery negative and the red lead of the voltmeter on the positive terminal. then move the red lead along the circuit noting the voltage all the way. Eventually the red lead will be back on the negative terminal of the battery and the voltmeter will read 0.00.

 

[smokey73]

2/0 cable is 0.077 ohms per 1000 feet. So that cable is thick enough. The "brown cable" if it was 12 ga would be 1.5 ohms per 1000 feet so 10 ft to the electrical panel and 10 feet on the return would give you 0.03 ohms. All connections, switches, fuses and wires have some resistance. So maybe it is just the wire from the switch to the panel.

Start with the black lead of the voltmeter on the battery negative and the red lead of the voltmeter on the positive terminal. then move the red lead along the circuit noting the voltage all the way. Eventually the red lead will be back on the negative terminal of the battery and the voltmeter will read 0.00.

That's the plan. Hope I can pinpoint the larger voltage drops, if any and correct them as I go along. From the battery postive terminal to the point where I'm noting the voltage delta (at the main power lead to the DC panel) there are only 4 connections.

 

[smokey73]

That's the plan. Hope I can pinpoint the larger voltage drops, if any and correct them as I go along. From the battery postive terminal to the point where I'm noting the voltage delta (at the main power lead to the DC panel)

Well. I did some testing. I used a 10G wired for ground directly to the negative post of the battery hooked to the negative on my voltmeter long enough to reach all the testing points. That way I take out any ships ground wire effects on the testing. I then turned on equipment to have a 16A discharge rate measured by the Victron Battery monitor. All but one test showed a negligible voltage difference accross the connection or length of wire tested. I have 2/0 wire (solid Red) running from the Battery to the Battery Disconnect Switch. The wire from the switch to the Battery "pop up breaker" and from there to the electrical power is Orange/thin Red Stripe with no indication of size on the wire.

The one place where there was a significant voltage loss was the Orange/Red stripe power wire from the Battery pop up breaker (downstream of the popup breaker - on upper right connection on picture 2) so just measuring the wire loss to the input of the DC main power breaker on the electrical panel (lower left breaker on picture 3.) There was a .27V drop on this 16 or so foot length of wire or just over 10% loss. Way too much! Either the wire is undersized or there is some other problem on that wire. The wire is orange with a single red stripe on it (picture attached) There is no wire gauge written on the wire but it looks to be the same size as the ground wire which is 6G wire. I''ll have to check but it looks like the orange/res stripe wire may not be tinned? I'll take a closer look in a day or so. Too hot to do work on the boat during the heat of the day.

Looks like I need to do some wire chasing. I also need to see what those two Red/White stripe wires go to? They are live when the ON/Off switch is on. One is probably the AC powered Battery charger and the other might go to the ACR relay for the start battery?. (Its not a solar charger though). Seems like the battery charger leads should be on the power in side so the battery will charge if the Battery Selector switch is in the off position?

1. Has anyone had a similar problem with a Hunter of this vintage, particulary a Hunter 40.5?
2. Has anyone pulled a new larger wire and if so what gauge wire did/would you use. I would like to use something more like 2G but it may be limited by what I can pull. I didn't get a a chance to see the exact route or what troubles I might have.