Of course, Tony! In your example, the PCI bus clock is supplied by the host, but has a whole lot to of influence on a PCI expansion card (being that without the 33MHz PCI signal the card can't operate).
What I was alluding to in my statement was that clocks unrelated to the card (like that for a serial interface or for the processor itself) sometimes pollute the interface between the add-in card and the host computer. Taking the PCI bus as an example, a 33MHz signal mismanaged by the card is probably a problem with the card (but it would be a problem with the card in any host system, since any host system would supply the card with that signal), but a signal of, say, 50MHz *might* be a problem with the host computer polluting the card because there is coupling between the processor bus and the PCI bus (if your card doesn't use or generate such a signal). Of course, due care should have been taken to reduce any effects caused by the host computer, but it is doggone near impossible to harden an add-in card against all of the noises that can turn it into their avenue of escape. There are no absolutes here, but you want to make sure that you are testing *your equipment* and that you are not a victim of the host computer's manufacturer having taken short cuts in the interests of costs and time to market. Steve Chin StreamLogic Corp. Menlo Park, CA, USA The usual disclaimers apply! -------------------------------------- List-Post: [email protected] Date: 12/11/96 2:05 PM To: Steve Chin From: Tony Fredriksson Steve, I am assuming that "any of the clocks on your board" includes the host bus clock that the board uses as a part of the intended function. For example, a PCI bus clock generated and supplied by the host computer can be mismanaged by the expansion card. Thus, the board can create a coupling mechanism of the PCI clock to a poorly designed I/O port. However, this would tend to show up on any host computer, at least those with similar PCI clock edge rates at the PCI expansion connectors. Regards, [email protected] ---------- From: Steve Chin To: Charles Blackham DD-KB; emc-pstc; Martin Ginty Subject: Re: Computer board tests List-Post: [email protected] Date: Wednesday, December 11, 1996 11:50AM I fully agree with Charles. I also want to add that there are many host computers which will pass class B emissions with a full complement of external peripherals, but a few from this group which will fail emissions as soon as an add-in card is placed in them. In short, you should test your support equipment first, then test the support equipment with your add-in card installed. If the frequency of failure is not associated with any of the clocks in your board, you *might* have a problem with a host computer polluting its expansion bus. Once you have found your ideal test setup, *keep it under lock and key!* Steve Chin StreamLogic Corp. Menlo Park, CA [email protected] As always, the views expressed here are my own and do not represent those of anyone else! -------------------------------------- List-Post: [email protected] Date: 12/11/96 7:49 AM To: Steve Chin From: Charles Blackham DD-KB Martin Amendment A1, 1995, to EN55022:1994 states: "In the case of Printed Wiring Board Assemblies (PWBA), separately marketed for the enhancement of diverse host units, the PWBA (e.g. ISDN interface, CPU, adaptor cards, etc.) shall be tested in at least one appropriate representative host unit of the PWBA manufacturer's choice so as to ensure compliance of the PWBA with the entire population of hosts in which it is intended to be installed." So according to the standard, the required minimum number of hosts in one. I believe that one is sufficient, especially where you have no idea what computer the end user will have. I would always check the Emissions from support equipment before adding an adaptor card, as combining CE marked computers,monitors, keyboards etc. does not always produce a combination that is still meets class B limits ! Charlie Blackham x31592 Approvals Engineer Madge Networks Limited ------------------ RFC822 Header Follows ------------------ Received: by sledgehammer.com with SMTP;11 Dec 1996 07:49:28 -0800 Received: from ruebert.ieee.org by oz.sledgehammer.com (SMI-8.6/SMI-SVR4) id HAA04076; Wed, 11 Dec 1996 07:51:37 -0800 Received: (from daemon@localhost) by ruebert.ieee.org (8.7.5/8.7.3) id FAA04454 for emc-pstc-list; Wed, 11 Dec 1996 05:49:47 -0500 (EST) Message-ID: <[email protected]> In-Reply-To: <126FAE3201E03700> List-Post: [email protected] Date: Wed, 11 Dec 96 10:35:25 -0000 From: "Charles Blackham DD-KB" <[email protected]> Organization: Madge Networks To: [email protected], [email protected] (Martin Ginty) Subject: Re: Computer board tests X-mailer: Connect2-SMTP 4.00 MHS to SMTP Gateway Sender: [email protected] Precedence: bulk Reply-To: "Charles Blackham DD-KB" <[email protected]> X-Resent-To: Multiple Recipients <[email protected]> X-Listname: emc-pstc X-List-Description: Product Safety Tech. Committee, EMC Society X-Info: Help requests to [email protected] X-Info: [Un]Subscribe requests to [email protected] X-Moderator-Address: [email protected] ------------------ RFC822 Header Follows ------------------ Received: by sledgehammer.com with SMTP;11 Dec 1996 14:05:07 -0800 Received: from netpower-lan.netpower.com by oz.sledgehammer.com (SMI-8.6/SMI-SVR4) id OAA05482; Wed, 11 Dec 1996 14:07:08 -0800 Received: from ipcb.NeTpower.com (ipcb.netpower.com [198.102.245.2]) by netpower-lan.netpower.com (8.6.11/8.6.5) with SMTP id OAA10685; Wed, 11 Dec 1996 14:03:37 -0800 Received: from mailgate.netpower.com by ipcb.NeTpower.com (4.1/SMI-4.1) id AA14845; Wed, 11 Dec 96 14:03:35 PST Received: by mailgate.netpower.com with Microsoft Mail id <[email protected]>; Wed, 11 Dec 96 14:02:56 PST From: Tony Fredriksson <[email protected]> To: Charles Blackham DD-KB <[email protected]>, emc-pstc <[email protected]>, Martin Ginty <[email protected]>, Steve Chin <[email protected]> Subject: Re: Computer board tests List-Post: [email protected] Date: Wed, 11 Dec 96 13:58:00 PST Message-Id: <[email protected]> Encoding: 102 TEXT X-Mailer: Microsoft Mail V3.0
