Why did my junction box connectors melt after three years?

Nine times out of ten, this is caused by **thermal cycling** that eventually works a connection loose. Even if those lugs were tight three years ago, pulling 48 amps for hours every single night generates alot of heat. Metal expands when it gets hot and contracts when it cools down. Over hundreds of cycles, that movement can microscopically back off a screw or shift the wire strands, creating a tiny bit of resistance. Once you have high resistance at a connection point, it creates a feedback loop. The resistance generates heat, which causes more oxidation and more expansion, which leads to even more resistance. Eventually, the heat gets high enough to melt the insulation and the plastic housing long before the breaker ever thinks about tripping. Since the load was still under 60 amps, the breaker saw everything as normal while the junction box was literally cooking. When you rebuild this, you need to use a **torque wrench** or a torque screwdriver. Most people just crank down on those set screws til they feel tight, but every manufacturer has a specific inch-pound rating for a reason. If it's too loose, it arcs; if it's too tight, you can deform the wire and create a hot spot. Also, ditch standard wire nuts for a high-amperage continuous load like this. Use **Polaris connectors** or similar insulated tap blocks. They handle the heat much better and provide a far more secure mechanical connection for 6 AWG wire.

Double-check your wire type too. If that 6 AWG is NM-B (Romex), it's actually only rated for 55 amps at the 60°C column, which means a 48A continuous load is pushing it right to the limit. If it's THHN in conduit, you have more breathing room, but it's still vital to check the temprature ratings of your terminals. Make sure the new junction box is oversized so there is plenty of air space around the splices. Don't just stuff it all into a cramped 4-square box.

Have you ruled out a software glitch in the load-sharing logic? You mentioned they sync, but if that communication dropped, both chargers might have defaulted to a higher draw than the 6 AWG coud handle long-term. I've seen 'smart' setups fail into a high-draw state before. Also, take a look at the conduit entry points for any signs of moisture. Sometimes a slow drip of condensation onto a lug causes more heat than a loose screw ever could.

I would be curious to know if you utilized any antioxidant joint compound during the initial installation, particularly if you were working with aluminum conductors, though even with copper, the phenomenon of 'cold flow' or creep—where the metal permanently deforms under constant mechanical stress over several years—can result in a significant loss of contact pressure that traditional tightening methods fail to account for. You shoud also verify that your junction box is actually metallic to better dissipate the localized thermal energy, as a plastic enclosure acts as a thermal insulator and effectively traps the heat generated by those marginal resistance points until the internal temperature exceeds the thermoplastic's deformation threshold.

Quick update—the thermal cycling point makes alot of sense. I checked the other junction and those lugs were definitely looser than I remember. However, I’m still worried about a potential load-sharing glitch causing a spike. If I move to 4 AWG wire for the rebuild, woud that prevent this heat buildup, or is it overkill for a 60 amp breaker?