RE: Re: VIOM locking up

["James Mitchelmore" <james@xxxxxxxxxxxxxxx>]

In my case, I don=92t think the unit freezing can have anything to do with
external sources as it is simply acting as an I/O interface to girder
(ie I=92m not using any of the internal routing/timing functions).  I=92m
also disinclined to simply power cycle it every now and again as that
could easily, if not definitely, result in it powering back in a
different state to that which it =91crashed=92 in.  From David=92s earlier
post, I suspect the issue may be due to the power supply as it has
become more apparent since I have been using more outputs (I have 2
relay boards attached) and I imagine that the current draw could be
quite high if a number of relays happened to be triggered at the same
time.  I think some calculations re. relay consumption/PSU are required.
I=92ll report my findings.
=20
Regards
=20
James
=20
-----Original Message-----
From: ukha_d@xxxxxxx [mailto:ukha_d@xxxxxxx] On Behalf
Of David Yeend
Sent: 15 November 2005 09:54
To: ukha_d@xxxxxxx
Subject: Re: [ukha_d] Re: VIOM locking up
=20
My experience is that these sort of spurious 'crashes' in relatively
simple
and otherwise stable equipment are generally due to external influences
rather than an exception in internal state.
That said, more complex systems or those which seem prone to regular
crashes can benefit from the self re-set approach.
For the case under consideration I would recommend an external
watchdog/reset system rather than relying on an in-built scheduled
reset.
Regards,
David

On 11/15/05, Jim Noble <yahoo-groups@xxxxxxx> wrote:
>
> rectorydp wrote:
>
> > I would agree that you don't want critical functions on
crash-prone
> > products, however, I do not think this is the case with the VIOM.
I
> > have now made sure that all my VIOMs are on good qulity psu's and
> > implemented a heartbeat over the serial interface in case
something
> > does go wrong. An automatic power cycle after a couple of missed
> > heartbeats has always worked.
>
>
> Why not use a spare output to reset a watchdog timer? If it crashes
and
> fails to reset the timer, the timer hits the reset line... (assuming
it
> has one there!) Do your bit to help deplete the european 555 timer
> mountain...
>
> And/or code in a hard reset every N minutes/hours/days?
>
> Jim
>
>
>
>
> Yahoo! Groups Links
>
>
>
>
>
>
>


[Non-text portions of this message have been removed]



"http://www.automatedhome.co.uk"http://www.automatedhome.co.uk




SPONSORED LINKS=20

HYPERLINK
"http://groups.yahoo.com/gads?t=3Dms&k=3DHome+repair+improvement&w1=3DHome+=
rep
air+improvement&w2=3DComputer+stuff&w3=3DHigh&w4=3DImprovement&c=3D4&s=3D76=
&.sig=3DT
cERUCDPQcJLbg9mtvQGJQ"Home repair improvement=20
HYPERLINK
"http://groups.yahoo.com/gads?t=3Dms&k=3DComputer+stuff&w1=3DHome+repair+im=
pro
vement&w2=3DComputer+stuff&w3=3DHigh&w4=3DImprovement&c=3D4&s=3D76&.sig=3Dp=
lS9-eUUfa
BXyLVFT87AJg"Computer stuff=20
HYPERLINK
"http://groups.yahoo.com/gads?t=3Dms&k=3DHigh&w1=3DHome+repair+improvement&=
w2=3D
Computer+stuff&w3=3DHigh&w4=3DImprovement&c=3D4&s=3D76&.sig=3Db5d547gPx7eFW=
NGqXQCl
7A"High=20

HYPERLINK
"http://groups.yahoo.com/gads?t=3Dms&k=3DImprovement&w1=3DHome+repair+impro=
vem
ent&w2=3DComputer+stuff&w3=3DHigh&w4=3DImprovement&c=3D4&s=3D76&.sig=3DO6pw=
_oratTOwe
p2G_91fIQ"Improvement=20
=20
=20
=20
_____=20=20

YAHOO! GROUPS LINKS=20
=20
*          Visit your group "HYPERLINK
"http://groups.yahoo.com/group/ukha_d"ukha_d";
on the web.
=20=20
*          To unsubscribe from this group, send an email to:
HYPERLINK
"mailto:ukha_d-unsubscribe@xxxxxxx?subject=3DUnsubscribe"ukha_d-un
subscribe@xxxxxxx
=20=20
*          Your use of Yahoo! Groups is subject to the HYPERLINK
"http://docs.yahoo.com/info/terms/"Yahoo!
Terms of Service.=20
=20
_____=20=20


--
No virus found in this incoming message.
Checked by AVG Free Edition.
Version: 7.1.362 / Virus Database: 267.13.0/168 - Release Date:
14/11/2005


--=20
No virus found in this outgoing message.
Checked by AVG Free Edition.
Version: 7.1.362 / Virus Database: 267.13.0/168 - Release Date:
14/11/2005
=20


[Non-text portions of this message have been removed]