Burnt Wiring On Airconditioning Board

[Doug4bass]

Hi guys.... After smelling a burning smell through the airconditioning vents, I found this ugly scene under the panels on a factory installed A/C. As you may see, the ac common has separated from its lug on the circuit board. I think this burn has gone on for a while. We smelled the burning ,thinking this was " dust" burning off while running the heat for the first time after the spring splash. Some of the guys at the marina seem to think this is the result of an overload over time witnessed by melted insulation on the surrounding wires further down from the terminal strip. Or maybe a bad crimp job?? This airconditioner is dedicated to its own supply from a second shore power inlet. I have read the many posts from Stu and Maine regarding bad connections. I have a new circuit board in hand, but would like some opinions. If this was the result of overload, why wouldnt the 30 amp breaker open? Thx in advance. Now for the nasty pics....

 

[Daveinet]

I do not believe it is a result of any kind of overload. it is a poor connection that generated heat over time. You have one wire that got seriously hot, as the insulation is melted off the wire. That would suggest a poor crimp and resulting corrosion.
Poor connections generate heat. Heat causes poor connections. That results in a condition we refer to as thermal runaway, as the worse it gets, the worse it gets. I would be interested to see the back side of that board, just to make sure it was a bad crimp, as opposed to a cold solder joint. A bad crimp will put some heat into the board, but typically a bad solder joint will get hot enough to cause all the solder to run out of the hole.

 

[Roadking Larry]

All the magic smoke leaked out of that one.
Like Dave said, possibly a poor connection causing an increased current draw. Possibly a failed component on the board itself. Hard to tell from the pic but there looks to be heat damage on some of the other components on the board, possible that one of those beefy transistors on heat sinks fried and caused your problem.
Either way it's horked

 

[kloudie1]

Typical bad connection.. I check my AC connections often since I leave it on all the time at the slip in its dehumidify mode.

 

[uncledom]

Doug, keep looking. Check the Neutral behind the A/C panel also. It's true that the worse it gets the worse it gets, but it's not necessarily the origin of the problem.

 

[capta]

Perhaps the power from the dock is the source of your problem? Is it possible it the voltage fluctuates or the connections to shore power are poor?
It always sends shivers down my spine when I walk by a boat in a marina with his A/C running and no one is aboard. Not only can this sort of thing happen, but it's pumping some serious water through the system and a leak in that could easily sink a boat in a few hours.
Good thing you were there to catch it.

 

[Doug4bass]

Thx, Guys....Ill do a little more snooping when I get home. I yep..I never run the AC unattended.

 

[Daveinet]

Voltage fluctuations will not cause wires to melt. Its only a driving a motor, and the motor can handle some voltage fluctuation. If the voltage is too low, then the motor will burn up due to lack of rotation or the run capacitor will swell and burn up. Low voltage will not hurt the wiring, it will only damage the motor and/or the run cap.

Generally speaking, it would be very rare for wires to melt, due to excessive current. To cause the amount of heat to melt the insulation, there would need to be a massive amount of current over a long time. Fuses will blow, long before the wiring would melt. (the primary function of a fuse is to prevent wiring from melting) To create heat, there needs to be a significant power across the device. (in this case, a connector)

 

[Bill Roosa]

Pull the board and check the circuit traces for overload. If everything looks OK then I'd agree with the OP that the connector is getting hot due to a bad crimp or loose connection/corrosion on the connection....
if the trace looks like it got hot over its entire length (overcurrent in it will fry the whole trace not just the part near the hot connector) then you are looking at a bad component. you might be able to replace and use the board as an emergency spair since you have a replacement in hand. I would want to know which component failed BEFORE I put the new board in as if it is off board it will fry the new board too!!!
and replace the fried wires and connectors of course.

 

[uncledom]

Voltage fluctuations will not cause wires to melt. Its only a driving a motor, and the motor can handle some voltage fluctuation. If the voltage is too low, then the motor will burn up due to lack of rotation or the run capacitor will swell and burn up. Low voltage will not hurt the wiring, it will only damage the motor and/or the run cap. Generally speaking, it would be very rare for wires to melt, due to excessive current. To cause the amount of heat to melt the insulation, there would need to be a massive amount of current over a long time. Fuses will blow, long before the wiring would melt. (the primary function of a fuse is to prevent wiring from melting) To create heat, there needs to be a significant power across the device. (in this case, a connector)

Dave, the problem is there is no fuse. There is a circuit breaker which is a very different beast. Circuit breakers do not simply trip due to over-voltage. It is over-voltage over time. There is a heat/voltage overtime curve that breakers are built to. The amount of heat generated can easily be hot enough to burn insulation. I found a burned Neutral connector at my Nav station Neutral bus. It was burning the wire right at the crimp connector. It doesn't take much to create heat due to loose connections of any kind.

 

[Daveinet]

Do some sample calculations for resistance in copper wire. You will find that even if you double the current going through the wire (which would surely pop a breaker or blow a fuse), the amount of heat created would not be enough to melt the wire. Fuses or breakers trip with current. While breakers do tend to be slower reacting, they still will trip when you exceed the current rating of the circuit they are protecting. That is why they are there. Basically to create heat, the resistance must go up substantially. The resistance of wire or a connector is very low, so there is no power dissipation in a working circuit. Even if the current rises, because the resistance is still low, you still do not get much heat. That is why the resistance must go up for it to melt a connector or wire. A poor connection means high resistance, which is where the heat is generated.

On Edit, I now have time to publish some sample calculations.
Power dissipation is what creates heat. So let’s say the wire gauge is 16ga. The resistance of copper for 16ga is .00473 ohms per foot. Typical AC unit will pull around 12 amps. I2R=watts. So 12X12X.00473=.68 watts. Dissipating .68 watts of heat over 1 foot of wire is pretty easy. Now let’s double the current (which would be impossible as the breaker would trip rather quickly) That comes to 2.7 watts. Still very easy to dissipate over the length of 1 foot of wire.
For perspective, think about your plug in portable heater that is 1200 watts. If you uncoil the internal wire that glows red hot, it is probably 3 or 4 feet total. So glowing wire takes 1200 watts over 4 feet 1200/4=300 watts. So compare 300 watts to 2.7 watts of copper wire, you can see you are not even in the ball park.
To get that kind of heat, the resistance must go up. A bad crimp can easily = 1 or 2 ohms resistance. Now multiply 12 amp squared X resistance of 2 ohms = 288 watts. Now that will melt some insulation, especially since the point of resistance is inside the small connector.



So the point is, an over voltage condition (marina problem) will not melt the wire, but rather the melted wire is from the heat created by the failed connector.

 

[ramblinrod]

Most likely bad connection.

Whenever I encounter a terminal strip, I check the screws for tightness. Usually, at least on is loose. That's all it takes, a loose screw and some current, and the heat starts building.

www.sheenmarine.com

 

[Doug4bass]

Im pulling the board off today. I will take some pics of the back and post. Thx again guys for the replies.

 

[Doc_holiday]

I agree, it's not the wire that failed but rather a loose or corroded crimp or connection. Voltage drops will cause higher current but the problem here is the regular or increased current was too much for the bad connection.

Let's say this is 10 AWG wire. It can safely handle 30 amps (just go with it here regardless of the accuracy). If the connection is poor, that connection may only be equivalent to having a section of 18 gauge wire in series? What happens when I run 30 amps through 18 gauge?

The 18 gauge, or in this case, the connection becomes the fuse. If it doesn't blow immediately the heat will begin to deteriorate the wire and insulation nearby until failure finally occurs.

Be sure to replace affected wires, you may have to cut back quite a way to get to good wire, don't just go by the darkening of the insulation. I see to many repeat failures where the repair is not properly addressed and soon you have a new similar but different problem.

 

[Doug4bass]

Well.......an update. Pompanette. ( who manufactures the A/C ) sent one of the designers up to the beautiful U.P to check the problem out. Way more than I can say for the dealer. Maybe a crimp, lightening or installation problem. A/C runs like new. New board, and even an extra. Boo to you Mr Dealer........not even a response. Shameful. Boat still under warranty. And once again....thanks guys for your expertise....I wish the yard could learn.

 

[Gunni]

So what was the problem? I've never heard of Pompanette air conditioning, or why Catalina would choose their gear. That wiring configuration looks completely different than any marine AC system I've ever worked on.