Path: utzoo!utgpu!news-server.csri.toronto.edu!bonnie.concordia.ca!uunet!mcsun!ukc!strath-cs!ex-dcs!exua!CDHWilli From: CDHWilli@exua.exeter.ac.uk (Charles Williams) Newsgroups: sci.electronics Subject: Re: Tunnel diodes...who makes 'em these days? Message-ID: Date: 6 Jun 91 17:32:51 GMT References: <16303@life.ai.mit.edu> <1991Jun5.163259.1228@zoo.toronto.edu> Sender: CDHWilli@exua.exeter.ac.uk Organization: Computer Unit. - University of Exeter. UK Lines: 32 In-reply-to: henry@zoo.toronto.edu's message of 5 Jun 91 16:32:59 GMT In article <16303@life.ai.mit.edu> tk@wheat-chex.ai.mit.edu (Tom Knight) writes: >What makes you believe that tunnel diodes will work at 4K? I would >guess that carrier freezeout would start being a significant effect at >about 25K and would be near complete at 4K... You *are* guessing. See for example B. Lengeler, Cryogenics 14 (1974) pp439-447 In article <1991Jun5.163259.1228@zoo.toronto.edu> (Henry Spencer) says: >I would also predict some problems with differential thermal contraction >causing cracking of package, chip, and/or leads well before then. Back >when IBM was working on Josephson-junction computers, they put quite a >bit of effort into the problem of building circuit assemblies that could >go from room temperature to cryogenic temperatures and back repeatedly >and remain reliable. Thermal contraction can be a problem, but not always. A reasonably recent review with 82 exciting references is RK Kirshman, Cryogenics 25 (1985) pp115-122 BTW a Josephson version of the Am2901 4-bit Microprocessor has been protyped by Fujitsu using Nb-AlOx-Nb technology. Comparative performance: Material Clock (MHz) Power dissipation (W) -------- ----------- --------------------- Si 30 1.4 GaAs 72 2.2 Josephson 770 0.005 But that was over a year ago now..... Charles Williams.