Mains - upstream RCD tripped by test button on project's RCD

[cowana]

One of my recent projects I have been working on is a variac (variable mains autotransformer), in a nice project box with meters.

I've finished the mechanical side of the build, and wired everything up yesterday.

The setup:

• Filtered IEC inlet socket
• Standard fuse (F/4A) in live wire
• 30mA Residual Current Device (http://www.ced-cart.co.uk/downloads/RCB0_Data.pdf)
• 4A rated variac (0-270v)
• Analogue voltmeter
• Analogue ammeter
• Switched UK plug socket as outlet

This all works well, and everything seems to function as it should. However - when I press the 'test' button on the RCB on the project, two things happen.
 1) - The RCB on the project flicks off (as you would expect)
 2) - The RCB in the main fusebox that feeds my workshop also flicks off

This seems a bit weird - as from what I can tell, the test button just allows a small amount of current to bypass the toroid inside the RCD being tested. The fact that the whole workshop is tripping indicates it is actually allowing the test current to flow from live to neutral - this seems a bit weird.

I've found the main RCD covering the workshop to be very sensitive - once I cut through a mains cable (with live switched off at the breaker) - the act of the cutter shorting earth and neutral was enough to trip the RCD. This seems to indicate my neutral is slightly above earth potential?

Any thoughts would be very helpful!

Cheers,
Andy

[cowana]

This RCD has a white functional earth lead, which is connected to mains earth.

Looking into the workings of RCDs a little more, google seems to suggest the test button on RCDs with an earth lead do allow current to flow from live to earth - which would explain why both RCDs in the system trip simultaneously.

It does seem a little strange to me that it works like this rather than rather than just allowing current across the toroid - however all the signs point to it.

Oh well, I guess there's nothing that can be done to stop the main workshop breaker tripping, except for replacing the project RCD with a more standard one.

Cheers,
Andy

[Kremmen]

Why do you have two successive, identical circuit protectors in the same circuit? This is called a protection coordination failure and would be a code violation here (not a severe one but a design error anyway).
The RCD test button works this way by design - the test is supposed to test an actual ground fault and not to a simulated effect that may or may not reflect reality.

[ConKbot]

Why do you have two successive, identical circuit protectors in the same circuit? This is called a protection coordination failure and would be a code violation here (not a severe one but a design error anyway).
The RCD test button works this way by design - the test is supposed to test an actual ground fault and not to a simulated effect that may or may not reflect reality.

guessing he wants the one on the lab equipment to trip first so he doesnt have to go to to panel to reset the lab 

[Towger]

He would need a lower rated RCD than the main one on his equipment, and even then there is no guarantee it will activate first.