[volt-nuts] Solartron 7081 randomly re-initialises
David C. Partridge
david.partridge at perdrix.co.uk
Fri Apr 15 08:09:22 EDT 2016
C2 is a hermetic tantalum, and doesn't appear leaky. +5V appeared on the mark and stable but C60 (220uF 10V in +5V supply) looked a bit crusty, so I replaced it on general principle with a 220uF 10V tant.
I think that C2 is used to hold off of the reset signal for a period of time after PWDN- is asserted to allow the processor to save stuff to RAM.
The ROMS appear all good and were all written pretty recently.
From: volt-nuts [mailto:volt-nuts-bounces at febo.com] On Behalf Of BIll Ezell
Sent: 14 April 2016 17:32
To: volt-nuts at febo.com
Subject: Re: [volt-nuts] Solartron 7081 randomly re-initialises
Repost from my response in the new Solartron group, :
>I assume you've eliminated obvious things like power-line spikes? One other thing comes to mind, the power-up reset logic for the CPU, the RESET >signal. My schematics are so bad it's hard to tell for sure, but it looks like that's generated by R6 and C2, along with IC4 on the earthy board. I'd check >C2, if it's very leaky, could be pulling IC4-1 down enough to give random triggering. IC6 is a possibility also, IC6-6 is used to also trigger a reset by >pulling C2 to ground.
I'd think if there was enough ripple on the 5V supply to trigger a reset, you'd be seeing other bad things.
Regading the nvram, the comments on that are dead-on, my cross post about that:
>The cal constants are in a non-volatile ram, it's possible you'll need to replace that. I've never seen a failure, but I've heard anecdotally that they can >fail. It's also possible that your meter has been sitting unused for so long that the nv-ness of the ram has leaked away.
>They work by injecting a stored charge into effectively a well-insulated gate. Over many years, the charge can leak away. That's why there's a rewrite >cycle the meters can do.
I forget what the nvram refresh command is, but it's in the manual.
Also, the nvram keeps two separate copies. I thought it would automatically try to read the second copy if the first was bad, and automatically rewrite the bad one. But, if enough charge has leaked, there could be multiple errors across the entire nvram.
volt-nuts mailing list -- volt-nuts at febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/volt-nuts
and follow the instructions there.
More information about the volt-nuts