NetNews Usenet Archive 1992 #19/NN_1992_19.iso/spool/sci/electron/14985

Text File | 1992-08-29 | 1.9 KB | 39 lines

Newsgroups: sci.electronics
Path: sparky!uunet!zaphod.mps.ohio-state.edu!darwin.sura.net!sgiblab!daver!cypress.com!blp
From: [e-mail] (Bruce Prickett/PROM)
Subject: Re: How hot can the 486-50 cpu run ?
Message-ID: <[e-mail]>
Summary: Silicon Can Run > 125 C
Sender: [e-mail]
Nntp-Posting-Host: sunshine.cypress.com
Organization: Cypress Semiconductor, San Jose, CA
References: <[e-mail]> <[e-mail]> <[e-mail]>
Date: Sat, 29 Aug 1992 00:54:59 GMT
Expires: Mon, 28 Sep 1992 07:00:00 GMT
Lines: 24

>I'm sorry but you're talking nonsense.
>Silicon semiconductors have max. chip temperatures of about 75 to 125 deg C.
>I may be wrong, but I think the maximum for the 486-33 is 85 deg C. The case
>temperature depends on the case type (ceramics or plastics) but will always
>be lower, approximately 40 to 60 deg C.

I don't know who was talking nonsense, but: Silicon can run lots hotter
than 125 C at the die. Military chips are spec'ed to work over a range
of -55 to 125 degrees C AMBIENT temperature. Since the thermal resistance
("THETA J-A") from the die to the air can be > 50 degrees C per watt for ceramic
packages, then the die temperature can easily be more than 25 degrees higher than
the ambient if the chip burns just half a watt. Chip temperatures > 150 C
are not uncommon when chips are run at 125C . Plastic packages have a much
higher thermal resistance, and tend to degrade at internal 165 C, so they
are usually limited to the commercial temperature range of 0 - 70 C ambient.

>Anyway, in PC-486's sold today, the max. temperature will be exceeded in most
>cases unless you are using an additional cooling fan *near* the CPU and/or
>an ICECAP or s/t like that.

Sounds like good advice to me. Reliability degrades rapidly when things heat up!

    Bruce Prickett
    [e-mail]