The Cyber-Spy.Com Usenet Archive Feeds Directly
From The Open And Publicly Available Newsgroup
This Group And Thousands Of Others Are Available
On Most IS NNTP News Servers On Port 119.
Cyber-Spy.Com Is NOT Responsible For Any Topic,
Opinions Or Content Posted To This Or Any Other
Newsgroup. This Web Archive Of The Newsgroup And
Posts Are For Informational Purposes Only.
Date: Sat, 28 Sep 2002 10:41:14 -0400
From: Raymond Rogers
User-Agent: Mozilla/5.0 (X11; U; Linux i586; en-US; rv:1.0.1) Gecko/20020823 Netscape/7.0
X-Accept-Language: en-us, en
Subject: Re: PCB puzzle
X-Comments: This message was posted through Newsfeeds.com
X-Comments2: IMPORTANT: Newsfeeds.com does not condone, nor support, spam or any illegal or copyrighted postings.
X-Comments3: IMPORTANT: Under NO circumstances will postings containing illegal or copyrighted material through this service be tolerated!!
X-Report: Please report illegal or inappropriate use to
X-Abuse-Info: Please be sure to forward a copy of ALL headers, INCLUDING the body (DO NOT SEND ATTACHMENTS)
Organization: Newsfeeds.com http://www.newsfeeds.com 80,000+ UNCENSORED Newsgroups.
Kevin Kilzer wrote:
> About 8 months ago we made a batch of products and put them on the
> shelf. This month sales got around to selling them. Standard
> production testing involves a few temperature cycles, -40C to +80C.
> The process is hot first for 1 hour, then cold for 3 hours, then hot
> again for 1 hour, and so on.
> All units failed during the 2nd hot cycle, and most recovered at room
> The problem was traced to a single pin on a connector to a device. In
> our application, the pin is intentionally left open (per the device
> manufacturer), but was found to be a low-impedance to an adjacent
> signal trace within the PCB. Some units had higher impedance, some
> lower (range about 60 ohms to 10 Mohms).
> The PCB has 8 devices connected similarly. All 8 show the problem
> after enough temperature cycles. We have eliminated the device itself
> as the problem.
> Heating that area of the PCB with a heat-gun causes the problem. Two
> cycles in the oven cause the problem. Returning to room temperature
> usually clears the problem.
> The PCB manufacturer lapped down an un-assembled PCB and saw nothing
> unusual. The artwork was checked for scratches.
> The connector is a typical pin-header (0.050" spacing) from a major
> manufacturer. If we heat a connector with a heat-gun, we can get the
> pin-pin resistance below 10Mohms, but this is not very scientific
> (although it did not melt).
> I'm looking for suggestions as to the cause.
> Thanks in advance.
Looking at your cycling it occurs to me that outgasing of something
during the first hot cycle and then the subsequent condensation would
explain the results, more or less. I would guess that the subsequent
heating failure was due to a threshold shift in the IC making it
sensitive to the impedance. 10Meg is not to strange; almost any residual
can cause that; the heat sensitivity is a little unexpected. 600ohms is
radical. BTW: Is the area around the pin conformally coated?
In any case the testing of the hypothesis is easy; just clean around the
disconnected pin and see if the problem/impedance goes away. This is
not perfect because the contaminant might have soaked into the board.
How about cutting into the board around the pin. Presuming that it is
contamination you might be able to salvage the boards if you really bake
them and circulate air to prevent clear out the contaminant; although
the 600ohm is really worrisome. After clearing the contaminant, say 40
hours at the allowed upper storage temperature for the boards, then
cut/insulate that pin! Of course you should tell your customer that you
had this problem and how you attempted to fix it.
You must find the source of contamination!
Generically check with the IC manufacturer about terminating that pin.
If you can't, try to design in another IC. Leaving a sensitive pin
floating is nothing but a problem.
-----= Posted via Newsfeeds.Com, Uncensored Usenet News =-----
http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
-----== Over 80,000 Newsgroups - 16 Different Servers! =-----
Go Back To The Cyber-Spy.Com
Usenet Web Archive Index Of
The sci.electronics.design Newsgroup