I have an RCD socket outlet feeding my workbench - where I do electronics
type stuff. One of these:-

http://www.tlc-direct.co.uk/Products/CM4901M.html

Yesterday I managed to produce a fault while fooking around with an ATX
power supply which didn't trip it - but did the one in the split load CU.

Would a socket with a lower fault current limit sort this - and who makes
one? Not much point in having this one if the main one is going to trip
first. Although it does pre-date the split load CU.

--
*i souport publik edekashun.

Dave Plowman London SW
To e-mail, change noise into sound.

Dave Plowman (News) wrote:
Dave Plowman (News) wrote:
I have an RCD socket outlet feeding my workbench - where I do electronics
type stuff. One of these:-

http://www.tlc-direct.co.uk/Products/CM4901M.html

Yesterday I managed to produce a fault while fooking around with an ATX
power supply which didn't trip it - but did the one in the split load CU.

Would a socket with a lower fault current limit sort this - and who makes
one? Not much point in having this one if the main one is going to trip
first. Although it does pre-date the split load CU.


I have an RCD socket outlet feeding my workbench - where I do electronics
type stuff. One of these:-

http://www.tlc-direct.co.uk/Products/CM4901M.html

Yesterday I managed to produce a fault while fooking around with an ATX
power supply which didn't trip it - but did the one in the split load CU.

Would a socket with a lower fault current limit sort this - and who makes
one? Not much point in having this one if the main one is going to trip
first. Although it does pre-date the split load CU.

Dave,

Don't automatically blame your local bench RCD. It is possible that
there is enough leakage in the rest of your installation to be near the
threshold of your split CU RCD and a few mA extra from your ATX PSU
tripped that whilst being well below the threshold of your local one.

Maybe put a controlled leak (suitably rated variable resistor) on the
circuit before your bench and see how much current it will take before
the CU trips out?

The main problem I find with RCDs and experimental bench is the effect
of mains filters in test equipment can cause nuisance tripping. I'd
rather work carefully via an isolating transformer.

Bob

In article ,
"Dave Plowman (News)" writes:
I have an RCD socket outlet feeding my workbench - where I do electronics
type stuff. One of these:-

http://www.tlc-direct.co.uk/Products/CM4901M.html

Yesterday I managed to produce a fault while fooking around with an ATX
power supply which didn't trip it - but did the one in the split load CU.

Would a socket with a lower fault current limit sort this - and who makes

No.

one? Not much point in having this one if the main one is going to trip
first. Although it does pre-date the split load CU.

If you generated, say, a 1A leak, then it's a toss-up which one
trips (although for any given pair, it will often be the same one).
The tripping currents of the devices (providing less than the 1A
leak) won't make any difference. (Can even get both tripping sometimes.)

To get discrimination between RCD's, you have to use time-delay on
the up-stream one. If tripping everything is going to be a problem,
I would think about running a non-RCD protected supply to the RCD
socket.

--
Andrew Gabriel
[email address is not usable -- followup in the newsgroup]

In article ,
Bob Minchin wrote:
Would a socket with a lower fault current limit sort this - and who
makes one? Not much point in having this one if the main one is going
to trip first. Although it does pre-date the split load CU.

Dave,

Don't automatically blame your local bench RCD. It is possible that
there is enough leakage in the rest of your installation to be near the
threshold of your split CU RCD and a few mA extra from your ATX PSU
tripped that whilst being well below the threshold of your local one.

Yes - although I don't get nuisance tripping from the main one. Oh - it
wasn't a few extra mA from the ATX - try an earth to neutral short via the
case due to a trapped cable. ;-(.

Maybe put a controlled leak (suitably rated variable resistor) on the
circuit before your bench and see how much current it will take before
the CU trips out?

Worth a try.

The main problem I find with RCDs and experimental bench is the effect
of mains filters in test equipment can cause nuisance tripping. I'd
rather work carefully via an isolating transformer.

I do have an isolated output. But wasn't using it in this case.

--
*Do infants enjoy infancy as much as adults enjoy adultery? *

Dave Plowman London SW
To e-mail, change noise into sound.

On Aug 7, 9:48*am, (Andrew Gabriel) wrote:
In article ,
* * * * "Dave Plowman (News)" writes:

I have an RCD socket outlet feeding my workbench - where I do electronics
type stuff. One of these:-

http://www.tlc-direct.co.uk/Products/CM4901M.html

Yesterday I managed to produce a fault while fooking around with an ATX
power supply which didn't trip it - but did the one in the split load CU.

Would a socket with a lower fault current limit sort this - and who makes

No.

one? Not much point in having this one if the main one is going to trip
first. Although it does pre-date the split load CU.

If you generated, say, a 1A leak, then it's a toss-up which one
trips (although for any given pair, it will often be the same one).
The tripping currents of the devices (providing less than the 1A
leak) won't make any difference. (Can even get both tripping sometimes.)

To get discrimination between RCD's, you have to use time-delay on
the up-stream one. If tripping everything is going to be a problem,
I would think about running a non-RCD protected supply to the RCD
socket.

That's the best option, but there is an easier one if you're already
using an iso transformer. Simply feed your rcd socket from the iso
transformer. Now any earth leakage after the rcd socket isnt even seen
by the CU RCD.


NT

On Fri, 07 Aug 2009 09:53:07 +0100, Dave Plowman (News) wrote:

Don't automatically blame your local bench RCD. It is possible
that
there is enough leakage in the rest of your installation to be
near the
threshold of your split CU RCD and a few mA extra from your ATX
PSU
tripped that whilst being well below the threshold of your local
one.

Yes - although I don't get nuisance tripping from the main one. Oh - it
wasn't a few extra mA from the ATX - try an earth to neutral short via
the case due to a trapped cable. ;-(.

The additional leak doesn't have to be big. The RCD in the CU has the
leakage of *all* the circuits connected through it to contend with.
The one on your bench, down stream of the CU one, just the bench
circuits. If the house is sitting at say 20mA and the bench 5mA, 15mA
of extra leakage at the bench will trip most likely trip the CU RCD
(20+15 = 35mA) not the one on your bench (5+15=20mA)(*).

To give you the discrimnation the CU RCD needs to be a time delayed
one or feed the bench RCD on a radial taken before the CU RCD. I'd go
for the latter.

(*)Assuming a 30mA RCDs and they trip precisely at 30mA.

--
Cheers
Dave.

"NT" wrote in message
...
On Aug 7, 9:48 am, (Andrew Gabriel) wrote:
In article ,
"Dave Plowman (News)" writes:

I have an RCD socket outlet feeding my workbench - where I do
electronics
type stuff. One of these:-

http://www.tlc-direct.co.uk/Products/CM4901M.html

Yesterday I managed to produce a fault while fooking around with an ATX
power supply which didn't trip it - but did the one in the split load
CU.

Would a socket with a lower fault current limit sort this - and who
makes

No.

one? Not much point in having this one if the main one is going to trip
first. Although it does pre-date the split load CU.

If you generated, say, a 1A leak, then it's a toss-up which one
trips (although for any given pair, it will often be the same one).
The tripping currents of the devices (providing less than the 1A
leak) won't make any difference. (Can even get both tripping sometimes.)

To get discrimination between RCD's, you have to use time-delay on
the up-stream one. If tripping everything is going to be a problem,
I would think about running a non-RCD protected supply to the RCD
socket.

That's the best option, but there is an easier one if you're already
using an iso transformer. Simply feed your rcd socket from the iso
transformer. Now any earth leakage after the rcd socket isnt even seen
by the CU RCD.

I fail to see what advantage there is fitting an RCD after an
isolating transformer, as by definition the secondary is isolated
from earth, and you can earth ether side with impunity.
A boon when connecting test-equipment to live chassis
radios and TVs under restoration.

--
Graham.

%Profound_observation%

On Aug 7, 12:28*pm, "Graham." wrote:
"NT" wrote in message

...
On Aug 7, 9:48 am, (Andrew Gabriel) wrote:



In article ,
"Dave Plowman (News)" writes:

I have an RCD socket outlet feeding my workbench - where I do
electronics
type stuff. One of these:-

http://www.tlc-direct.co.uk/Products/CM4901M.html

Yesterday I managed to produce a fault while fooking around with an ATX
power supply which didn't trip it - but did the one in the split load
CU.

Would a socket with a lower fault current limit sort this - and who
makes

No.

one? Not much point in having this one if the main one is going to trip
first. Although it does pre-date the split load CU.

If you generated, say, a 1A leak, then it's a toss-up which one
trips (although for any given pair, it will often be the same one).
The tripping currents of the devices (providing less than the 1A
leak) won't make any difference. (Can even get both tripping sometimes.)

To get discrimination between RCD's, you have to use time-delay on
the up-stream one. If tripping everything is going to be a problem,
I would think about running a non-RCD protected supply to the RCD
socket.

That's the best option, but there is an easier one if you're already
using an iso transformer. Simply feed your rcd socket from the iso
transformer. Now any earth leakage after the rcd socket isnt even seen
by the CU RCD.

I fail to see what advantage there is fitting an RCD after an
isolating transformer, as by definition the secondary is isolated
from earth, and you can earth ether side with impunity.
A boon when connecting test-equipment to live chassis
radios and TVs under restoration.

it prevents the upstream rcd tripping. You can also have your supply
isolated or neutral earthed as you please, and the socket rcd will
still work in the latter case.


NT

Dave Plowman (News) wrote:
In article ,
Bob Minchin wrote:
Would a socket with a lower fault current limit sort this - and who
makes one? Not much point in having this one if the main one is going
to trip first. Although it does pre-date the split load CU.

Dave,

Don't automatically blame your local bench RCD. It is possible that
there is enough leakage in the rest of your installation to be near the
threshold of your split CU RCD and a few mA extra from your ATX PSU
tripped that whilst being well below the threshold of your local one.

Yes - although I don't get nuisance tripping from the main one. Oh - it
wasn't a few extra mA from the ATX - try an earth to neutral short via the
case due to a trapped cable. ;-(.

In which case it is a "race hazard" - either or both could trip. This is
why TT installs with cascaded RCDs use a time delay type at the head end
to ensure discrimination.



--
Cheers,

John.

/================================================== ===============\
| Internode Ltd - http://www.internode.co.uk |
|-----------------------------------------------------------------|
| John Rumm - john(at)internode(dot)co(dot)uk |
\================================================= ================/