I found a paper (from South Africa) on the history of the RCD. I thought it could be of interest to anyone who wants to know how the RCD came to be.

http://www.cbi.co.za/papers/24/CBI2.pdf


-----------------------------------
By the way:

A European RCD/GFI look like this



This particular one is the cheapest you can buy here. The two pole $20, the four pole $30

Although it's not printed on the device, the manufacturer reports the trip time to be 100ms at 30mA and 40ms at 150mA, whereas more expensive models trip in 40 ms at 30 mA.

[This message has been edited by C-H (edited 11-08-2002).]

C-H,
How do you guys test your RCD's over there, after installation?.
We had a really hot debate, going on amongst us Electrical people over here, as to what constituted a true test of the effectiveness of the RCD, some fella's reckoned that pushing the test button, satisfied the test requirements.
I have a dedicated RCD tester, and use it in every installation that has RCD's installed,regardless of wether or not I installed them.
One trend over in Australia that I do not like, is the practice of using a 50A RCD(S) downstream of the Main Switch, this protects the whole installation, so if one power point trips, the whole house is left dead, not good!.:mad

Well, good question. I have an RCD-tester, but this is of the simplest type: It simply leads 30mA to earth. If the RCD trips, it's supposed to be OK. A professional will have a real meter, where you can set the test current (10mA, 30mA, 100mA, 300mA or similar) and perhaps also duration.
This sounds like a very good idea. They do fail. (I pushed the test button on one at a friends house, with no result. As it was running hot without load, I adviced him to get rid of it) Here users are advised to test them at least every six months. Reasons for this vary: Mechanical freeze, losened sensitivity or lightning damage. I don't dare to get involved in this debate...

The over-all RCD you describe is the norm here. The electric safety authority assume you use Diazed (Edison base) fuses, not breakers. These fuseboxes can only be split between two RCD:s (or no RCD vs. RCD) by sawing them apart physically. I'd love to see the look on the face of the house owner when the electrician starts sawing in the box (No, I don't think it is common to do this, although legal)

[This message has been edited by C-H (edited 11-11-2002).]

Notice the "press monthly" marking on the test button of the unit pictured above. 99% of users in this country won't even notice that instruction, nevermind carry it out.

Interesting document on the development of ELCB/RCD use in S.A. This passage caught my eye:
I've noticed that a large number of people on this side of the Atlantic seem to look upon 110V in this manner.

Question: Is a 110-0-110V system safer than 220V if there is a 30mA RCD? Or is the risk in fact greater with the former? (Yes, I'm serious)

C-H,
All that the "Test Monthly" advice,is to remind people to test their RCD's, failing this, the Mechanical part ot the RCD can freeze up, meaning that it will probably not Trip, when required to do so.
However, very few people even take any notice of their Switch-boards, unless something goes wrong.
I use a tester, that gives actual trip times and currents and you can use these on your Certificate Of Compliance, which shows you have actually tested the Installation.
An RCD is a life-saving device, why is it not treated accordingly!!.

Better make sure I'm reading this the right way.

Do you want to compare the systems with both protected by a 30mA RCD, or just one?

Both of them with RCD.

The advantage of supplying 240V to a big appliance under the American system is that no part of the wiring exceeds 120V to ground. The only way to get shocked at the full voltage is to touch both hots simultaneously.

Looking at a 220-240V British/European equivalent, there's only one side of the circuit which will deliver a shock if touched, but it is at the full voltage. Everything else being equal, the current through the body will therefore be much higher.

If the loop resistance to ground through the body were a little over 4000 ohms, then touching one side of the American supply would result in a current of just under 30mA, so one could be left holding onto it without the breaker tripping. At the same resistance on a 220V line the current would be over 55mA so the RCD should trip immediately -- Higher shock current, but only for a brief fraction of a second.

Is this what you were thinking?

Yes, this was what I thought. It can happen in theory, but will it also happen in real life?

Is this why US GFI's are 5 mA and European RCD:s 30 mA

Note that the resistance of the skin is drops with higher voltage.

In Austria it's advised to test GFIs twice a year, along with daylight savings (you have to reset your clocks anyway, so you can as well test your GFI). I don't know how many people actually do that.
NO shock risk at 110V???????????????????? Did some guy go nuts here??????? 110V is very well sufficient to kill a person, although it's definitely less dangerous than 220.
However, on my opinion stepping down from 220 to 110 just for safety reasons would be the worst nonsense I've ever heard. Just from an economic point of view!

I think it's more because the American 6mA GFI is intended to provide protection against shock whereas the European 30mA type grew up from the need to provide protection against ground faults on TT systems. As TT isn't used in the U.S., they had no such need.