> From [email protected] Sun Jul 19 09:39:35 PDT 1998
> Received: from hpsdlo.sdd.hp.com (hpsdlo.sdd.hp.com [15.255.160.64]) by
> hpsdlfsg.sdd.hp.com with ESMTP (8.7.6/8.7.3 TIS 5.0/sdd epg) id JAA16098 for
> <[email protected]>; Sun, 19 Jul 1998 09:39:34 -0700 (PDT)
> Received: from send1b.yahoomail.com (send1b.yahoomail.com [205.180.60.23])
> by hpsdlo.sdd.hp.com (8.8.6/8.8.5tis+epg) with SMTP id JAA18148
> for <[email protected]>; Sun, 19 Jul 1998 09:39:33 -0700 (PDT)
> Message-ID: <[email protected]>
> Received: from [202.54.30.131] by send1b; Sun, 19 Jul 1998 09:39:04 PDT
> Date: Sun, 19 Jul 1998 09:39:04 -0700 (PDT)
> From: Gautham Nalamada <[email protected]>
> Subject: GFCI/RCD
> To: [email protected]
> MIME-Version: 1.0
> Content-Type: text/plain; charset=us-ascii
>
>
> Hello Rich
>
> I chanced upon your article on GFCI in one of the newsgroups, I think
> sci.engr.electrical.compliance.
>
> I had a doubdt regarding RCD (Residual Current Devices) now coming out
> for Plusating DC currents, supposedly more suitable for Computers and
> Electronic equipment.
>
> If you do, could please carify the working and difference with
> ordinary ones.
>
> Sorry for this unsolicited mail.
>
> And also thanks ahead.
>
>
> Gautham Nalamada
>
>
>
>
> ==
>
> Gautham Nalamada
>
> Phone : 0091 - 40 - 7120100/7123310 (office)
> 0091 - 40 - 7123122 (fax)
> 0091 - 40 - 244839/216775 (home)
> Email: [email protected]
> _________________________________________________________
> DO YOU YAHOO!?
> Get your free @yahoo.com address at http://mail.yahoo.com
>
>
Hi Gautham:
The GFCI (USA/Canada) is an all-electronic device which
disconnects the supply if the difference current between
phase and neutral exceeds about 5 or 20 (depending on
application) milliamperes rms (or dc).
The RCD (Europe and other geographical areas) is an all-
magnetic device which disconnects the supply if the
difference current between phase and neutral exceeds
about 20 milliamperes rms. Since it is all-magnetic,
dc can cause the magnetics to be biased such that any
ac current has little or no effect.
Most computer equipment today employs switching-mode
power supplies. These operate by first rectifying the
mains to dc, and then switching the dc at frequencies
ranging from 30 kHz to 1 mHz, the typical frequency
being about 100 kHz. These high frequencies allow the
use of very much smaller and more efficient circuits
than power supplies operating at mains frequencies.
The fear arises when there is a mains-to-enclosure
insulation fault on the dc side of the rectifier in a
switching-mode power supply. If this should occur, the
fault current (which passes through the GFCI/RCD) is
comprised of both ac and dc, and could fit the
description "pulsating dc."
The GFCI will disconnect the supply. The RCD may not,
due to the dc component if the fault ("pulsating dc")
current.
This has been a major concern among European countries,
some of whom have required the dc portion of switching-
mode power supplies to be double-insulated with respect
to ground/earth and accessible conductive parts.
Unfortunately, I cannot describe how a RCD for pulsating
dc physically overcomes the dc-induced magnetic bias in
a RCD.
I have taken the liberty of distributing your question
and my answer to a mail list concerned with product
safety issues such as this one. Perhaps one of the
subscribers can describe the construction of a RCD
suitable for pulsating dc currents.
Best regards,
Rich
-------------------------------------------------------------
Richard Nute Quality Department
Hewlett-Packard Company Product Regulations Group
San Diego Division (SDD) Tel : 619 655 3329
16399 West Bernardo Drive FAX : 619 655 4979
San Diego, California 92127 e-mail: [email protected]
-------------------------------------------------------------
