Hi Bruce, I once had a similar problem with AD595 theromcouple IC's we had a whole bunch of them on a car when suddenly 2 at the rear kept failing. After the third failure I investigated further and it looked like on all 3 the feedback pin(8)had gone open circuit. I prised them open and on each, the bondwire had broken. We put this down to noise levels or possibly local supersonic buffeting exciting a resonance in that particular bond wire. A "blob" of high temperature RTV on the chip seemed to cure the problem. Hermetic packages have their own issues.
Robert G8RPI. --- On Fri, 19/6/09, Bruce Griffiths <[email protected]> wrote: > From: Bruce Griffiths <[email protected]> > Subject: [time-nuts] HP5370A self test fault characteristics and diagnosis > To: "Discussion of precise time and frequency measurement" > <[email protected]> > Date: Friday, 19 June, 2009, 5:53 AM > For the benefit of HP5370A and/or > HP5370B users I have added a webpage > detailing the characteristics and diagnosis of a fault that > my HP5370A > recently developed. > With the 10MHz rear panel output connected to both inputs > it developed > an overange error when set to measure the time delay > between the START > and STOP inputs. > Details are on the web page: > > http://www.ko4bb.com/~bruce/HP5370_Faults.html > <http://www.ko4bb.com/%7Ebruce/HP5370_Faults.html> > > At Didier's suggestion I've also included a link detailing > the imaging > setup and camera settings etc used to acquire the chip > image on the > above page. > Whilst it is possible to obtain good quality images with > less precise > multiaxis positioner than those used, at least it is very > easy using > such equipment > The chip image confirms the diagnosis inferred from > measurements of > various signals within the 5370A. > > Replacing A22U21 cured the problem, > > Bruce > > > _______________________________________________ > time-nuts mailing list -- [email protected] > To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. > _______________________________________________ time-nuts mailing list -- [email protected] To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
