ECN Forum Forums General Topics of the Trade Non-US Electrical Systems & Trades Rcd trip times

Anybody have any extensive experience with
testing and inspection?
On a number of occasions over the years have tested Rcd's for compliance with trip times and found that the trip time registered at 30ma is faster than that at 150ma. Use a Chauvin Arnoux 6115 installation tester calibrated anually by the manufacturer. This has occured on all sorts of different makes of rcd's over time.
Anyone else ever experienced anything like it? Todays test was on a Federal GFI 30ma trip. Test at 30ma 12.4ms test at 150ma 32.8ms. Thoughts gentlemen please!

Gidday Alan,
I use an AVO Megger Installation tester to do my testing of RCD's.
I have seen these results pop up everytime I test.
Over here it has to be under 300mS trip time
at the stated (say 30mA) current.
I think the longest time I have had is 42mS, and that was at 300mA, with a Selective RCD.

It's interesting to tryand figure out why the breaker should take longer to trip at the higher current.

I'm thinking along the lines of hyteresis in the core maybe? It's quite a difference though -- After all, a full-cycle of 50Hz is 20mS.



[This message has been edited by pauluk (edited 09-17-2005).]

Guys,
I'm right into a thing like this.
Considering that my training teaches me that an RCD should trip quicker, with a higher amount of current.
If it was in fact Hysteresis I'd be worried by that.
If it took time to saturate the core of the RCD transformer, it could concievably kill someone.
An experiment to do at work in the weekend!.

Hi guys,

During power restoration (either the RCD being turned on again or the power being reapplied after an outage), there is usually a small amount of Earth inrush as power is seldom applied at the zero crossing of the waveform (not to mention that Live often connects first followed by the Neutral!). This inrush resulting from the natural (and unnatural) capacitance existing on the system.

RCDs need to ignore such inrush else one would simply never be able to restore power to a circuit (many who remember the very early RCDs will remember this problem especially as the "cure" was to trip all sub circuits, restore the main supply, and then, with fear, reenergise each circuit in turn praying the RCD would remain on!).

This anti-inrush-tripping is very simply achieved in the "non-electronic" versions by having two secondary windings on the RCD core. The first winding powers the operating solenoid (as shown in most explanations on RCDs). The second winding also feeds the solenoid, but is wired in reverse (i.e. the two windings cancel each other out). However, the second winding is fed via diodes and a capacitor meaning the solenoid can only operate once the capacitor has charged and thus removes the cancelling current.

At low currents the capacitor has very little charging to do as well as the diodes, because they have voltage drop, allow most of the main secondary current to flow in the solenoid with little cancelling current being allowed through - hence the faster operating times at lower currents.

Make sense?

M.

Ah...... It's beginning to make sense. I've never really looked at the detailed arrangements of RCDs before.

How exactly are the diodes arranged on that extra winding? Is it a simple bridge rectifier followed by a series capacitor?

Thanks for the explanation M. All sounds very plausible.
Why does this only happen on odd occasions then? Have asked the question of Chauvin Arnoux! no satisfactory answer as yet.