PSA: Check your connections!

[Caesium]

Well I didn't expect to find this on my preparations for reorganising my power cupboard today:





So I noticed the end of the positive battery lead coming out of the inverter looked a slightly different colour to the rest of the cable.. touched it.. yep, warm. Everything instantly switched off!

Took it apart and it's quite a mess. The screw holding the terminal in was still secure so I'm not quite sure where the loose part was; something has clearly not been making a very good connection. This burnt bit of cut-off cable still feels solid, I can't just pull the cable out of the crimp.

These are *not* DIY connections I must stress, these are commercially available Pylontech cables; the long ones. This particular connection has been in place for a year or so, its not something I've recently touched with my DIY adventures.

So morale of the story, check connections regularly. Just keeping an eye on the temperature of my re-crimped cable now, hopefully it fares better.

[Caesium]

And here's the remains of the cable sheath that used to be attached to the inverter, broke away in my hands as I was investigating:



Ouch.

[nowty]

Many times its been said on here that its loose connections that causes the vast majority of fires, not the batteries.

And your find is what many of us forget about, the commercial stuff which must be ok right ?

One of the issues with Pylontech (and other commercial batteries) is 25mm² cables, they can get warm or even hot. Thermal stress with constant expansion and contraction can loosen connections over time. Then they get even hotter. Getting hotter increases the electrical resistance so they get even hotter. Then they start to burn and oxidize, resistance increases again so they get even hotter.

My battery bank intermediate cables are 35mm² and 50mm² and the main cable back to the inverter is 70mm². They still get warm on the bolt down fuses, isolators and crimped connections.

My rule of thumb is as long as they are not uncomfortably hot to touch at full power, should be OK.

Once you've repaired the connection, I would still keep an eye on it in case the internal connection inside the inverter is loose and is the actual source of the problem. I would also clean up the terminal on the inverter end with a bit a fine sandpaper to remove any carbonisation.

[Oldgreybeard]

Good catch, could have been a lot more serious. Prompted me to go and check my cables, as the batteries have been charging at full whack for a few hours today. Thankfully they only seem to be a couple of degrees warmer than ambient.

I suspect this may have been a dodgy crimp, as the witness marks on the terminal seem to show that it was tight. I did wonder about the mismatch between the hole size in the terminals and the stud diameter when putting mine together, and made sure that I used the right mix of 8mm and 6mm eyes.

[Tinbum]

I had a similar experience recently with a connection on my DC fuse box. All was still working but I'm surprised it got to the stage it did as I go round with a thermal camera and check every so often. Obviously not often enough.

[Caesium]

Can I just ask, quick straw poll if I can.. those who have the means to measure it - when you're charging batteries up at full whack, so 70A+, how warm do your inverter terminals get, and the associated lengths of wire immediately connected to them?

[Oldgreybeard]

Can I just ask, quick straw poll if I can.. those who have the means to measure it - when you're charging batteries up at full whack, so 70A+, how warm do your inverter terminals get, and the associated lengths of wire immediately connected to them?

Measured mine with an IR thermometer earlier, as soon as I read your warning, after about an hour and a half of charging at 3kW. The cables at the inverter terminals were about 20 deg, the cables at the switch fuse terminals (all four) were slightly warmer, at about 22 deg. The temperature inside the outbuilding that houses the batteries and inverter was about 19 deg.

[Caesium]

Thanks OGB, so really negligble difference over ambient then. I think I may have a bigger problem on my hands, I closely monitored the battery terminals on the inverter as my Go Faster kicked in.

At first no problem, these terminals are passing 70A quite happily, but as the inverter itself warms up, I think it's conducting a lot of heat out through these cables and they're hitting 40c or more near the terminals. The body of the inverter is all pretty warm (30+), but especially the area around these terminals I'm measuring at 70+. So I'm wondering if its not the terminal connections at fault but rather excessive heat being transmitted out from inside the case.

I think it might be time to open it up and see if there's anything obviously wrong.

[Oldgreybeard]

My Sofar ME3000SP gets very hot when charging, less so when discharging. The reported temperature from it was regularly hitting 65 deg to 70 deg when charging, which concerned me enough to add an extractor fan hooked up to a length of 100mm diameter duct to reduce the heat sink temperature. I have the fan plugged into a thermostat with a remote sensor poked around the back by the heatsink, set to turn on at 40 deg and off at 35 deg, and that fan always starts to run after around 15 minutes or so of charging at full wack. Even with the fan running the inverter regularly reports a temperature of around 40 deg.

[Caesium]

Yep that's my current workaround, a bit of active ventilation pointed under the unit towards the rear heatsinks.

That's dropped the previous 70c area of the body down to 30c and the cables are all now cool to the touch. That at least assures me nothing is going to burn down tonight and I'll definitely make something more permanent tomorrow.