Immersion heater single internal wire overheats

Hello,

I am not an electrician but do have a reasonable understanding from working with electronics (soldering and working with discreet components)

I have an odd Immersion Heater issue:

Specifications:

Gledhil Accolade unvented cylinder (A210D – 210 Litres) Installed 1997 approx

- x2 Redring 3KW Immersion Heaters (GU11TC) side mounted

- Thermostats = Baker BT7

- Separate safety cut-out (screwed into the brass immersion base).

Series wiring under immersion cover:

redIn – [BT7] – whiteWireA – [CutOut] – whiteWireB – [Element] – BlueIn

The two whiteWires are stranded thick gauge wire. Conductor diameter 2.5mm and a consistent dull grey colour. The red/blue/earth wires are stranded copper and a bit thinner.

When manually testing the lower heater, within 30 seconds whiteWireA got too hot to touch.

Same result using thermostat from the booster, and two freshly purchased BT7 thermostats.

Every time I install, I make sure there is good contact from the thermostat cable screws which are firmly tightened without over tightening.

I trimmed whiteWireA to expose fresh wire on thermostat end hoping it was just oxide build up. No real change on re-test.

Next I pulled the other end off the Cut out (spade connector, very tight push fit). I cleaned the contacts up to a nice shine with a PCB rubber. Now the wire gets a little less hot, just about touchable.

Finally I take the whiteWireA from the booster (looks like new as almost never used) and install that. On re-test whiteWireA barely gets warm even after 5 mins.

Am I correct in assuming this is now normal operation?

Could the original whiteWireA have degraded somehow?

Perhaps the crimp on the base of the spade is loose?

Could oxidation have occurred further down wire than expected, so when I trimmed and exposed new wire, it was little better than before?

I am very interested to hear any thoughts!

Many thanks

Graham

You are looking at this deeper than most people would! , but logically if the original link gets hot, and swapping it for one from the backup does not when the same current is flowing then the original link is experience resistance heating that the replacement isn't, as to which end its likely hard to test since unless the resistance is quite a bit higher than it should be, for such a small link then this may be masked by any contact resistance of test probes (There do exist ways to eliminate this when for example joints in big busbars to carry multiple KAs have to be measured but thats beside the point - not relevant here). Of course a thermal camera would probably betray which end is generating the heat, but assuming you do not have access to one, then if you have restripped the end into the stat terminal and it does not look damaged (I would normally say bright copper, but as you say dull grey, I would assume tinned strands which would be expected with rubberised insulation) then its likely the other end. I think I'd just get a new bit of 2.5 silicon rubber or butyl and crimp a shrouded spade connector on it (decent crimp tool required, mind)

Great info, thank you Phoenix!

I was leaning toward the spade connector being the issue!

Locked to prevent being resurrected (again)