Anyone have a sevice manual, or schematic, for this model?
Or ideas? Normally, my first thought would be that its PSU had failed,
but it has batteries as backup!
Thanks
Sylvia.
Thanks, but that appears to be from the 305 model, which seems to be
different.
In the mean, time I've discovered that the reason it's completely dead
even on battery is that the 40A fuse to the battery has blown.
Disconnecting the large transformer and shorting out the fuse allows the
display to light, go through its initialisation cycle, and then indicate
a fault.
The failure of the fuse wouldn't explain why the device won't run while
on mains power, so the fuse is presumably a symptom of something else.
Sylvia.
It could just be fatigue !
--
Best Regards:
Baron.
Uh, sorry ... not close enough ?
> In the mean, time I've discovered that the reason it's completely dead
> even on battery is that the 40A fuse to the battery has blown.
> Disconnecting the large transformer and shorting out the fuse allows the
> display to light, go through its initialisation cycle, and then indicate
> a fault.
I strongly suggest that you replace the fuse, because the
unknown fault might make things worse, and you may
blow up the whole unit ( which could solve the problem )
> The failure of the fuse wouldn't explain why the device won't run while
> on mains power, so the fuse is presumably a symptom of something else.
Sure it is. Sorry I didn't find the manual.
What is the error/fault message on the display ?
Seems to be different, though it's useful in showing the likely general
configuration.
>
>> In the mean, time I've discovered that the reason it's completely dead
>> even on battery is that the 40A fuse to the battery has blown.
>> Disconnecting the large transformer and shorting out the fuse allows the
>> display to light, go through its initialisation cycle, and then indicate
>> a fault.
>
> I strongly suggest that you replace the fuse, because the
> unknown fault might make things worse, and you may
> blow up the whole unit ( which could solve the problem )
>
>> The failure of the fuse wouldn't explain why the device won't run while
>> on mains power, so the fuse is presumably a symptom of something else.
>
> Sure it is. Sorry I didn't find the manual.
>
> What is the error/fault message on the display ?
"UPS fault" indicated by a continuous beep from the unit.
Given the number of transformer leads that I have unplugged, this is
hardly surprising.
Sylvia.
I've found one of the switching FETs shorted, and another with unlikely
resistances between the terminals. With both removed, it's possible to
get the UPS to power up on the mains.
What threw me for a significant while is that it appears that powering
up the UPS requires that the battery be present and providing power, but
once the thing's powered up, it no longer needs the battery.
So looks like the FET shorted, and took out the fuse. Seems reasonably
likely that replacing the FETs and the fuse will restore it to service.
Sylvia.
Of course ... sounds a bit like Monty Python to me.
> Given the number of transformer leads that I have unplugged, this is
> hardly surprising.
Since you don't know the cause of the original failure,
are you sure I won't happend again ?
Did a short-circuit occur on the output(s) of this UPS ?
I've commented elsewhere in this thread that I've identified a couple of
power switching FETs which are defective. One of them is shortcircuited,
across all three terminals, which would certainly have blown the fuse.
Of course, that just pushes the question down a level - am I sure that
whatever blew the FETs won't happen again. And of course, the answer has
to be no.
These transistors only switch when the UPS is in backup mode. At other
times, they're not switching, but their drains are connected to the
secondary of a transformer that has mains applied to the primary. The
failure could be attributed to a power spike, though I have to note that
the UPS is plugged into a surge arrestor, and nothing else in the house
seems to have suffered.
>
> Did a short-circuit occur on the output(s) of this UPS ?
>
No.
Have you asked SOLA?
Good Luck!
Rich
>On Apr 25, 2:28 am, Sylvia Else <syl...@not.at.this.address> wrote:
>> Jean-Christophe wrote:
>> > On 24 avr, 15:42, Sylvia Else <syl...@not.at.this.address> wrote:
>> >> Jean-Christophe wrote:
>> > What is the error/fault message on the display ?
>>
>> "UPS fault" indicated by a continuous beep from the unit.
>
>Of course ... sounds a bit like Monty Python to me.
The OP could interrogate the UPS via its RS232 port using
HyperTerminal:
http://www.networkupstools.org/protocols/sola.html
Unfortunately the status information probably won't tell her anything
useful.
BTW, I would suggest that sci.electronics.repair is probably a better
newsgroup for this type of question than sci.electronics.design.
- Franc Zabkar
--
Please remove one 'i' from my address when replying by email.
I've asked the company that took them over. I don't hold out much hope
though. They don't really have an interest in helping people repair old
gear.
Sylvia.
After I'd replaced the two FETs and the fuse, I turned it on. It went
through its initialisation cycle, and then promptly blew again, quite
dramatically, this time taking out all 4 FETs.
Guess this is one for the recycling bin :(
Sylvia.
I didn't quite get round to doing that. It bothered me that I couldn't
discover the failure mode.
The output circuit is similar to that here
http://web.telia.com/~u85920178/power/invert1_00.htm
in that an FET (or two in parallel in the case of the SOLA UPS) drives
each half of a centre-tapped transformer. However, the SOLA UPS is
creating a modified sineway, so there is a significant period between
one FET turning off, and the other turning on. The reverse diodes in the
FETs are required to prevent very high voltages appearing when an FET is
turned off. (Also note that it would seem to be a very *bad* idea to
insert a diode in the battery circuit to protect against reverse
polarity connection).
So it occurred to me that the two original FETs I left in the circuit
might not have been in as good a state as I'd supposed. When I sourced
replacement FETs, I bought 4, simply because of the mininum charge at
Farnell. This meant I still had two. The circuit only needs four when
it's driving a large load, so I thought I'd try one on each side. I also
installed a much lower rated fuse, in the hope of protecting the
transistors if things still went wrong.
But they didn't. It's actually working now. Perhaps the moral of the
story is "always replace all the switching transistors".
Sylvia.
Cheers.
-------------------------------------
Sylvia Else wrote:
-------------------------------------
Thanks
Sylvia.
##-----------------------------------------------##
Delivered via http://www.electrondepot.com/
Electronics Enthusiasts' Community of the Net
Web and RSS access to your favorite newsgroup -
sci.electronics.design,aus.electronics - messages and counting!
##-----------------------------------------------##