Red hot studs and melted wires, could have been worse
The sad looking picture above is the back of Have Another Day’s 240v AC panel. Though it doesn’t look good here, it was scarier looking when I first opened the panel the dull grey stud with the red 6 gauge wire was glowing red along with some of the wires exposed copper conductor. As soon as I saw the state of affairs I disconnected shore power and began investigating. The culprit was pretty easy to identify and turns out to be a matter of basic good practices not being followed when this panel was constructed, but realizing we had a problem was a matter of relatively subtle hints.
My family and I set out for a summer cruise from our home port in Chicago up Lake Michigan and over to Lake Huron to cruise the Georgian Bay and North Channel. We had a three day weather window in which the forecast on Lake Michigan consisted of nearly all zeros for wave height. That doesn’t happen too often and when it does it’s best to take advantage, so we plotted a course of three long days running in order to travel from Chicago to Manistique, MI at the top of the lake in Michigan’s Upper Peninsula. Our first day took us to Port Washington, WI.
We arrived in Port Washington around 6:30 in the evening having been on the water for a little over 11 hours. Not exactly up to Ben Ellison’s 24 hour marathon, but a long day nonetheless, especially with two antsy kids. Have Another Day has two 50a 240v shore power circuits. As I glanced at the electrical panel just before transferring from generator to shore power I noticed that while on generator power we were only seeing about 200v on the 2nd (primarily air conditioning) circuit. With that piece of data I made a mental note to myself that I’d need to figure out at some point. Once load was transferred from generator to shore power the incoming voltage was now showing 190v and I knew my time to investigate was now.
Earlier in the season at Burnham Harbor we’d seen low voltage coming into the AC circuit, but that was on a 95+ degree day and I’d attributed that to voltage dip due to load at the marina. My boat is setup with isolating transformers on both shore power circuits but only shore power circuit 1 (which primarily feeds 120v domestic loads) has a boosting function in the transformer. This means that I can’t do an apples to apples comparison of the two voltage readings to identify a problem within the boat versus at the pedestal.
Now armed with the new information about low voltage readings on generator and on multiple shore power connections it was time to dig deeper. My investigation started, and as it turned out, ended at the main 240v AC panel. Upon dropping the panel I found the melted wire and cooked breaker.
I didn’t get a good picture of the problem when the breaker was still in place but here’s my recreation of the problem. Carver employs a matched pair of breakers for each circuit with one for the generator and one for shore power. There’s a sliding lockout that prevents both breakers from being energized at once. The fried breaker above is the shore power breaker for the second circuit. The wiring in this area appears to be original to when the boat was constructed 15 years ago. What’s not easy to see in any of the photographs is the smaller gauge ring terminal (12 I believe) below the 6 gauge ring terminal. Everything I know about wiring says this is a no-no.
The largest conductor should be on the bottom with successively smaller terminals stacked on top, wedding cake style. That wasn’t done here, the smaller ring terminal that feeds the volt-meter was stacked under the larger one. Over the years and trips inside this panel the nut worked it’s way slightly loose, and it was truly very slightly loose. This created resistance in the connection, which, in turn, created heat. The heat melted the insulation, the breaker case and scorched the conductors.
I feel fortunate nothing worse happened. The insulation melted off this wire exposed a hot leg in close proximity to ground, neutral and the other hot . The enclosure for this panel has always seemed a little under sized. When the panel is pushed into place its front bows out from the pressure of the wires in the box and the lack of adequate space for the wires. I think this slightly too small enclosure probably contributed to the connection working its way loose. Carver is a price point builder and the electrical does reflect it some. There’s none of the lovely, groomed connections with terminal strips for each circuit. Instead, most of the wires are direct run to the load side of the breakers. The crimps are carefully done and the work has survived 15+ years of service.
Upon figuring out what had happened I temporarily removed the melted breaker and jumper wire. I connected the line side supply to the distribution, which meant I didn’t have a breaker in the panel to interrupt load in the event of a problem. There is another 50 amp breaker for shore power where the shore power enters the boat, as well as the 50a breaker on the pedestal.
In removing the breaker, I’d also defeated the safety of the lockout preventing the generator and shore power from being energized simultaneously. Fortunately Have Another Day is equipped with both fore and aft shore power connections. To switch between the inputs there’s a rotary switch at the top of the panel. with aft-off-fore positions I was able to use that switch to turn off shore power altogether when using the generator and in so doing ensure shore power and generator power never meet. I was also very careful to only ever have one source of power available. So, if the generator was running the shore power cables were already disconnected and vice-versa for shore power.
Ward’s Electric in Fort Lauderdale had the breaker I needed in stock and had it to me in two days. Repairs were pretty simple to make; requiring a new breaker and crimping a new jumper cable. The greatest time was spent adjusting the way all ring terminals sat in order to make sure they were properly stacked and unlikely to become bent or pushed into contact with another conductor.
Lastly, I feel this is a good reminder to me and possibly others about the value of paying attention to the information available to you. Had I been more vigilant in noticing the declining voltage I likely could have found this problem before the electrical panel stood on the precipice of disaster. If I had scanned the AC panel with an IR thermometer periodically I could have found the increasing temperature of the face of the breaker. I’ll add the AC and DC panels to the list of things I try to periodically scan with an IR thermometer. When things break I’m frequently reminded how much easier it would be to diagnose and fix if I knew what the status should be when working properly.
On board Gizmo, the last frontier
February 25, 2013
Paneltronics AC load shedding, smarter power
July 7, 2010
OctoPlex, the three cable boat, and NMEA 2000
January 26, 2006
GFCI and ELCI breakers can be confounding, but heed their warning
August 6, 2019