Shock from switch due to dampness

[LastManOnline]

Does it happen?.Is it even possible.?.I ask the question because I have never experienced the issue professionally and have become increasingly convinced that with a standard plastic switch it is not even possible.The traditional reasoning behind mounting a light switch outside a bathroom is due to the perception that a credible risk of shock due to dampness/condense exists if it is mounted inside.However I have repeatedly tested the resistance between the live terminal of a switch and the outside of the switch (where a finger would be positioned) while submersing the switch in water and not got a resistance lower than 20,000 ohms.

 

[brianmoooore]

That's a Crabtree 2172, and the 2000 series were known to be susceptible to 'shocking' in a damp atmosphere. Crabtree virtually admitted to this when they emphasised the improved resistance to water entry in the publicity blurb, when the 4000 series was introduced.

 

[happysteve]

That's a Crabtree 2172, and the 2000 series were known to be susceptible to 'shocking' in a damp atmosphere. Crabtree virtually admitted to this when they emphasised the improved resistance to water entry in the publicity blurb, when the 4000 series was introduced.

Fantastic! Thanks for the information, that's superb!

It's all coming out now....

 

[LastManOnline]

That's a Crabtree 2172, and the 2000 series were known to be susceptible to 'shocking' in a damp atmosphere. Crabtree virtually admitted to this when they emphasised the improved resistance to water entry in the publicity blurb, when the 4000 series was introduced.

Very interesting.I would add that ,since plastic is technically a non-conductor of electricity, would it be more correct to describe these Crabtree makes as "non-plastic"? They must have contained a component or additive that allowed for conduction.

 

[DPG]

Very interesting.I would add that ,since plastic is technically a non-conductor of electricity, would it be more correct to describe these Crabtree makes as "non-plastic"? They must have contained a component or additive that allowed for conduction.

Could be poor quality or contaminated plastic maybe. I've seen that happen in a production environment where the plastic body of a batch of mains inlet sockets failed, although admittedly that was during 3kv flash tests.

 

[brianmoooore]

Water isn't technically a conductor either, but we all know that in most cases it is and this was the problem with 2000 series switches. There was a small and fairly short route between the plastic of the switch body and the plastic of the rocker, all the way into the live contacts. This drew water in by capillary action, causing the rocker to bite whoever touched it. Once dried out, they were fine again.
The promotional stuff when the 4000 series was introduced, referred to 'increased labyrinth design' or something like that, around the rocker.

 

[LastManOnline]

Water isn't technically a conductor either.This drew water in by capillary action, causing the rocker to bite whoever touched it. Once dried out, they were fine again.

That explains it then.Any insulator (wood) will be converted to a conductor when water gets in.So essentially it is,nt the "plastic part" of the switch doing the conducting.Its the water in the poorly designed switch thats responsible.

 

[Lucien Nunes]

Very interesting.I would add that ,since plastic is technically a non-conductor of electricity, would it be more correct to describe these Crabtree makes as "non-plastic"? They must have contained a component or additive that allowed for conduction.

There are many different plastics with very different characteristics, but electrical accessories like traditional UK light switches made of hard white 'plastic' aren't really plastic at all, but thermosetting resins with fillers (thermosets.) There are many different permutations of resins and fillers, for example one might use a phenolic resin with wood flour filler for a non-technical application, but prefer an inorganic filler for an electrical component. This makes a difference to the moisture absorption properties, e.g. some thermosets lose much of their insulation resistance when exposed to damp heat, because they are microscopically porous and absorb a measurable amount of water. Others are more resistant and as soon as the surface is dried off the insulation is good again. FWIW, 50 years ago in East Germany, there was a coding system where the type of resin and fillers used in the mix had to be identified on a product and were moulded into the approval mark. You would see these on plugs, sockets, switches etc.

Electrical components, accessories and importantly PCBs are (or should) be designed with minimum creepage (over the surface) and clearance (through air) spacing between parts that are supposed to be insulated from one another. The combination of material surface resistivity, creepage distance, environmental pollution and humidity determine the insulation resistance achieved in practice, and all should be considered in the design stage. Unfortunately sometimes things still get overlooked; a capillary such as Brian describes where a tiny amount of free liquid can be drawn across a long creepage path, or a material with an unsuitable surface that occludes moisture and dirt, carbonises and tracks across, even metal migration.

I have some connectors with paxolin insulators (same material as the brown part of the back of this switch) which is resin-bonded calendered paper, which have trees of metallic leakage paths across the surface where silver has migrated from one pin to another under the electric field. Remember when 'NON-TRACK' used to be moulded into some fittings like junction boxes? That's what it was referring to.

As for bulk resistivity, we've seen on this forum some thermoplastics with unsatisfactory performance. I think one was an IEC power cordset where people were getting shocks off the dry external surface of the connector. It was probably a substandard product with fake approvals, where the moulder had used an incorrectly graded recycled PVC feedstock containing carbon black.

 

[mart spark]

It could be due to phenolic breakdown. Sparky Ninja whom I believe is a forum member and his E5 group would be better placed than i to confirm it though.