Path: ...!eternal-september.org!feeder3.eternal-september.org!news.eternal-september.org!.POSTED!not-for-mail From: Phil Hobbs Newsgroups: sci.electronics.design Subject: Re: Operating temperature derating Date: Fri, 7 Jun 2024 22:28:21 -0000 (UTC) Organization: A noiseless patient Spider Lines: 60 Message-ID: References: MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Injection-Date: Sat, 08 Jun 2024 00:28:22 +0200 (CEST) Injection-Info: dont-email.me; posting-host="97b68f58497244e91e7f25f24f03cfd8"; logging-data="2386605"; mail-complaints-to="abuse@eternal-september.org"; posting-account="U2FsdGVkX1+Hg/AwmQkWnF/SZtKcek1h" User-Agent: NewsTap/5.5 (iPhone/iPod Touch) Cancel-Lock: sha1:AtWnzpkkqUNCjcKx1wwybixA7E8= sha1:tjU98LcrBP2JBTkTqIi/w5vAfig= Bytes: 3613 Don Y wrote: > On 6/7/2024 2:59 PM, legg wrote: >> On Fri, 7 Jun 2024 12:45:24 -0700, Don Y >> wrote: >> >>> On 6/7/2024 6:35 AM, legg wrote: >>>> Commercial considerations of free enterprise have always pushed >>>> human behaviour towards banditry, so what can I tell you? >>> >>> But banditry would suggest taking advantage of the customer. >>> This is the opposite; the customer gets *more* performance >>> than the stated capabilities. >>> >>> But, no way to know HOW MUCH more! >> >> What exactly IS your concern? > > I want too know how much MORE than the published/unpublished operating > limits one can reasonably expect from a piece of kit -- given that > these limits don't seem to TRULY represent "maxima". > > We have a general idea of the types of components used in these > things and realize that their operating limits usually exceed the > published limits for the composite device -- often by a lot! > > This suggests (to me) that the published limits aren't backed by > "real" engineering or stress testing. But, rather, likely arise > from marketing specs... someone making a SWAG as to how customers > will LIKELY use the devices and not what their actual design limits > happen to be. > > [Given that using ANYTHING beyond its operating limits leaves you > without a leg to stand on, it would be nice to have some idea as to > what a reasonable expectation for those limits might be, despite > the "fluff" on the spec sheet. E.g., I would be really pressing my > luck to use something at 80C in that most components would likely > not be specified at those extremes. But, 55C for a 50C-specified > device? 60C?] > > > We’ve had chipmaking folks here (including Miso, iirc) say that the abs max ratings are primarily marketing driven, like most of the rest of the datasheet. You find out what your major customers are going to do with the part, and derate it to match. That way you sell parts and still get to have a quiet life. Of course there are things like delamination, overvoltage and electromigration that cause lifetime issues rather than prompt failures. Cheers Phil Hobbs -- Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC / Hobbs ElectroOptics Optics, Electro-optics, Photonics, Analog Electronics