Surge Protective Devices

[binky]

2 questions, prompted by the tick boxes on EICRs:-

1/You don't see many SPDs, as in the mains incoming type to protect say a house, but I was thinking how would you test a device is working? From what I've seen of the small single phase units, there is no way of testing a device.

2/ I've looked at SPDs a few times, and all the bumpf waffles on about lighting stirkes. What I can't work out is whether an SPD would protect a house from something like a pole transformer fail, or just the general voltage spikes you can get off the mains supply?

 

[Joseph Howarth]

Learning Lounge has a video on how it responds to various voltages and spikes, but then we just like the explosions

 

[Sidewinder]

As far as I know, they are self testing.

As they have a limited life, i.e. number of operations, then testing them would be self defeating.

Once the fault indicator comes on it's time to replace them.

They are a glorified VDR/MOV across the supply & ground, so they shunt fast rising transient voltages to earth.

Thus, they will protect against other transients.

As far as for examle a slowly rising voltage caused by a failing Tx N, causing the supply to skew I don't know, not my area, I don't work with the supply side really.

 

[Andy™]

megger it at 500v and itll go short circuit...

there should also be a visual indication on them if they are dead or alive

 

[binky]

but if you kill it and haven't got  replacement cartridge .....500v should be nothing to an SPD I suppose

 

[Andy™]

if meggering it is enough to kill it, is probably due for replacement anyway

 

[Tony S]

You need to look at the manufacturer’s data.

A 500V DC IR test won’t affect a 250V SPD when it is rated at 25000 amps. Threshold 250V*√2=354V.

They should be disconnected for IR tests as they will give a false reading.

 

[binky]

so in short we can't really test a SPD is really working?

 

[Lurch]

so in short we can't really test a SPD is really working?

No different to not being able to test a fuse or circuit breaker though, and RCD testing is only really of limited value IMO.

 

[binky]

I'm convinced RCDs are designed to trip when you don't want them to :^O

 

[Tony S]

so in short we can't really test a SPD is really working?

You could by having a variable voltage supply with a current sensor. The MOV/GDT should start to conduct once over its threshold value. In a way, a bit like ramp testing a RCD.

A 1000V Megger should in theory show a dead short.  ray