Trying to figure out cause of malfunction

[ygaudreau]

Hi everyone,

I'm currently banging my head trying to figure out exactly what caused a malfuction in a fairly simple design. I have a water heating system with multiple stages of heating done by a step controller and 1 stage failed after 3 weeks of operation. Each stage is protected via fuses before the contactor and to the heating elements. One stage overheated the wires between the line distribution block and the fuse holder. The wire melted and caused an arc flash which destroyed the connections to the fuse holder on the line side. I've checked all known possibilities of this failure and came up short. The arc happened before the protections, the contactor is still in working condition with no conductivity between lines and no resistance between line and load on each phase. there was no trace of leftover insulation under the terminal block either at the fuse holder or the distribution block. Heating element is still at right impedance and no conductivity to ground and passed megger test. No cracked isolating porcelain on the element and no apparent arc, weld or carbonisation of element bundle. Only one stage has faulted and all the others still working with no problem aside a couple with a little carbonisation left from the arc flash. Feeder wires are not bundled together with zip ties possibly inducing current in certain wires and there is no metal debris inside the electrical panel which could have caused a short circuit on the contactor or fuse block. If anyone has a clue on what could have happened, I`m all ears and very grateful because right now I'm at loss. Thank you for your help.

 

[TugboatEng]

When the contact initially closes and the heating element is cold, the electrical current is substantially higher than the advertised current. I don't know about heating elements specifically but in the case of tungsten filament light bulbs, this initial current can be up to 9X the advertised current. In the past switches often carried a tungsten rating. Perhaps this is a source of your high current.

 

[ygaudreau]

When the contact initially closes and the heating element is cold, the electrical current is substantially higher than the advertised current. I don't know about heating elements specifically but in the case of tungsten filament light bulbs, this initial current can be up to 9X the advertised current. In the past switches often carried a tungsten rating. Perhaps this is a source of your high current.

That is what I thought at first, but after inspection of the contacts inside the contactor, everything is still in perfect working order and no trace of heating or carbonisation due to arcing. We've been using these contactors with the same heating elements for a very long time without any incident. The only visible heat damage was on the wire about 6in from the contactor terminal block. Checking the elements, all impedance is still on point just like the other bundles in the unit.

 

[3DDave]

A surprising amount of damage can be done when the terminals aren't tightened properly. The installer can do 99 out of 100 correctly and that last one is going to burn up. If you can, get it all repaired and use a thermal camera to see if any wires or components are getting warm.

 

[waross]

Conductors undersized.
Fuses ovesized.

 

[Compositepro]

"The wire melted and caused an arc flash which destroyed the connections to the fuse holder on the line side."

That is very typical damage caused by a loose connection. The heater acts as a current limiting resistor for the arc that formed when the wire melted.

 

[LittleInch]

Everything seems to point to an issue with that wire.

Now its melted you'll probably.never know, but do you have any more you can look at?

Possibly.just damage/ Manufacturing fault.