How does a terminal become loose?

[LewisM]

How does a terminal become completely loose?

My uncle had an old school outside socket installed ~40years ago he asked me to have a look as it had been intermittently working recently.

I had a look and the L+N terminals were completely loose and the cable came out as soon as pulled, however the cable had been folded over and quite a deep indentation in the copper so it was definitely connected nice and tight at one point, I know extreme heating and cooling can cause creep and vibrations can loosen terminals however this had only been used for lawnmowers/pressure washers.

 

[UNG]

You forgot to factor in the thermal effects of climate change

 

[littlespark]

To throw some shade on your detective work…. Maybe it was tight at one time… causing the tell tale dent in the copper, then disconnected for testing maybe?? And just not tightened up again as much.

 

[timhoward]

To throw some shade on your detective work…. Maybe it was tight at one time… causing the tell tale dent in the copper, then disconnected for testing maybe?? And just not tightened up again as much.

I nearly asked if he'd had an EICR.!

 

[littlespark]

I nearly asked if he'd had an EICR.!

After 40 years, I’d hope so!

 

[ipf]

After 40 years, I’d hope so!

but why disconnect an outside socket for testing.....if only 40 and no problem?

 

[Dustydazzler]

To throw some shade on your detective work…. Maybe it was tight at one time… causing the tell tale dent in the copper, then disconnected for testing maybe?? And just not tightened up again as much.

This is why I am not a fan of people who remove stuff when doing an Eicr , I have dropped switches and sockets in houses who have had a fairly recent Eicr done and a N or L or CPC just pop out the terminal or are snapped off in the terminal

A good initial connection should not become lose unless it has constant movement or some other factor like baking in the sun. Most bad connections in my experience happen when someone keeps undoing stuff and doing it back up badly.

 

[littlespark]

Last time I saw a 40 year old external socket….. it was metal clad in a poly bag.

 

[pc1966]

Biggest factor I guess is incorrect torque, both too tight (causing significant deformation of the wire) and too loose (the obvious one).

But vibration and thermal cycling just from room temperature will loosen screws that are not sufficiently under tension to stay put. For fasteners generally there are a whole load of methods to stop loosening under environmental factors (nylock nuts, spring washers, Nord-Lock washers, castellated nut and split pin, etc, etc) but few are applied to electrical connection (other than spring washers).

However, I have seen locking pain on screws before but most commonly for adjusters that are not under force but you want to stay put. At a push some nail varnish works fine as thread lock and usually easy to get at local shop, etc.

 

[R-fur]

In some areas when doing junction boxes in attics the sparks used to drip wax from their candles on to the screw heads of the porcelian connector blocks to partialy insulate the screws and to stop them working loose. I remember an old journyman showing me how to do it over 40 years ago.

 

[EricMark]

Had it with a WMDU (weather proof main distribution unit) first few times we blamed some one not torque them to correct torque, but had repeated cases, seems the bus bars were copper coated aluminium and heating and cooling caused the connections to become loose.

Some times it was current heating them up, but some times just the sun. Around 1990 and likely left over after Ian Smith and the copper shortage. But in a sun trap items can get quite hot, and really the 10 year EICR is too long with domestic.

The other is binding screws, had it with meter changes, they are so insistence on using torque wrenches that there is no feel, so a binding screw is tight on threads not the wire, once over heated however this removes the binding on the threads so it seems as though not torqued up. Had the same with car cylinder heads so they were always re-torqued at 500 miles.

Latter the idea was to torque up to lower figure then add x degrees.