RCCD "sensitivity"

[Chris.J]

Does anyone know what appox. current the Trip button on a 30mA RCCD "applies"? Is it,say,around 35 mA, or much larger?
I have a 30mA RCCD "downstream" of a 100mA one, and when you press the test button on the 30mA one, both trip out (no appliances connected).
Is this right, I would have thought the 100mA one should have stayed in?

 

[ban-all-sheds]

Pressing the test button is a fairly crude test - don't know what fault current it generates but I'm not surprised to hear that it also trips the 100mA RCD.

 

[plugwash]

aren't test buttons normally wired between live and neutral accross the rcd... so they shuldn't trip a rcd further upstream even if it has the same sensitivity

is the 100ma a time delay type?

 

[kendor]

aren't test buttons normally wired between live and neutral accross the rcd... so they shuldn't trip a rcd further upstream even if it has the same sensitivity

is the 100ma a time delay type?

if it didn't then the supplying RCD would not be doing its job. The imbalance incorporates the entire circuit ie current imbalance is the same in the series-parallel circuit but apart from that your argument is flawed as a test button across LN of the rcd would produce no imbalance see http://www.memonline.com/rcd1.html for circuit diagram of actual RCD what in fact happens is that the neutral connection is before the toroidal and the live connection is after hence the imbalance of only one current carrying conductor passing through the toroidal during a push button test.

 

[plugwash]

yeah by accroiss the rcd i meant live on one side of the rcd (or at least its sense cuircuit, in that mem diagram the test button is all on one side of the isolating/disconnecting part) and neutral on the other so it creates an imbalance accross that rcd but should not create one accross rcds further upstream

 

[mapj1]

Sounds like the RCD has an earth connection (and so it may be a voltage ELCB rather than a current balance type)
As already mentioned a Z connection is more normal
The test resistor is perhaps 6k8 in a modern 30mA one, so no it shouldn't trip the 100mA, unless there are other leakages in the house adding in to the total..
How old are the 2 devices, and do they have live and neutral connections only or earth connections too?
Note that a '30mA' RCD may trip at a little over 15ma, and a 100mA one at a little over 50ma, albeit not particularly rapidly- see the curves here
http://www.memonline.com/f111.html
Note also that some RCD testers use a combination of resistor an diodes to either test on positive or negative half cycles only, as the trip time is different in this case, an for certain types of load this is important.

 

[kendor]

one thing that intrigues me is the newer generation of rcd tester that tests disconnection time without tripping! what baffles me is how then can they reliably know whether it would trip properly in practice?

 

[andy]

aren't test buttons normally wired between live and neutral accross the rcd... so they shuldn't trip a rcd further upstream even if it has the same sensitivity

is the 100ma a time delay type?

if it didn't then the supplying RCD would not be doing its job.

the 30mA should not trip the 100mA if the T button is pressed, simple because the 30mA shorts L-N thru a resistor tripping it, however no current has been leaked to earth so the 100mA should not trip

 

[Lectrician]

one thing that intrigues me is the newer generation of rcd tester that tests disconnection time without tripping! what baffles me is how then can they reliably know whether it would trip properly in practice?

New on me - All RCD Testers will HAVE to trip the the RCD to determine the time.

 

[Chris.J]

Thanks all for your replies.
Both RCCD's are about 8 years old, they are RCCD's, not ELCB with earth wire.
Think I might have to investigate how the trip sw/resistor is wired on the 30mA one. If I measure the resistor that will give me the test current.
Will also double check appliance leakage.

 

[has]

one thing that intrigues me is the newer generation of rcd tester that tests disconnection time without tripping! what baffles me is how then can they reliably know whether it would trip properly in practice?

I think you mean the ramp test. This just slowly increases the trip current until the rcd trips, but trip it it does.