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.
From: "Michael A. Terrell"
Subject: Re: Determining type of passive component on LCR meter
Date: Fri, 10 Jan 2003 01:05:26 -0500
Organization: Have you seen my bench? No, really! Where is it?
X-Mailer: Mozilla 4.61 [en]C-CCK-MCD (Win95; U)
References: <firstname.lastname@example.org> <_kjT9.30$BV6.email@example.com>
Jeff Verive wrote:
> "Jim Thompson" wrote in message
> > On Thu, 9 Jan 2003 12:38:50 -0600,
> > "Jeff Verive" ,
> We had an engineer who
> > |depended on the "leakiness" of free-wheeling diodes in a switching power
> > |supply in order to avoid having to add snubber components. When the
> > |manufacturer improved the diode's performance, the switching transistors
> > |started dropping like flies. The designer actually had the bright idea
> > |"screen" for the worst 5% or so of each lot of diodes in order to get the
> > |leakiest ones.
> > |
> > |
> > That's not "design", that's *hacking*. That's a worse stunt than
> > those who "design" to fit a specific transistor beta ;-)
> > ...Jim Thompson
> > --
> That's also the difference between an engineer who understands engineering
> and manufacturing economics and an engineer who understands only electronic
> device operation. This engineer knew his stuff from a technological point
> of view, but didn't seem to understand total costs to the company. The
> vendor would have been happy to screen for the leaky parts (at a price), but
> would not be able to guarantee that his process would yield enough leaky
> parts to support our production rate. Lead-time for the quantities we
> needed went from a couple of months to a couple of years!
> I was able to convince him to add a real snubber circuit (grudgingly on his
> part, of course), but we kept the manufacturing floor humming.
> Perhaps we should require engineers to spend a few months in the "component
> engineering" arena before we unleash them into R&D. Rotational assignments
> may not appeal to everyone, but they tend to give you insight that you
> probably could not get elsewhere.
Put them out on the plant floor for a little while to see the results
of bad designs. They will discover the real costs when they are buried
to their necks in bad boards while they try to tell you there is nothing
wrong with their design. You'll end up with a lot better engineer and
create good will between manufacturing and design at the same time.
Michael A. Terrell
Go Back To The Cyber-Spy.Com
Usenet Web Archive Index Of
The sci.electronics.design Newsgroup