9200T going for landfill

On Sun, 18 Apr 2010 15:04:44 +0100, Alan
wrote:

In message , Albert Ross
wrote
Does any one have any recommendations for a replacement that will still
have manufacturers back up a couple of years down the line? Yes I know
it can never be guaranteed but surely some must be more caring of their
customers than others.

http://www.toppy.org.uk/index.php

worth a read, though I'm not sure what I'll do when my ever reliable
5800 turns up its toes

The Topfield is not supported by the manufacturer with regards
software/firmware. Their last effort at Freeview+ is over a year old,
it's as buggy as hell and not a single bug fix in site.

Damn, I haven't been keeping up. The original 5800 was quite usable
out of the box and I only updated the firmware a couple of times.
Played with some TAPs but by then was used to the original interface
anyway.

I knew the first edition 5810 was a bit unspecial but had assumed
they'd have mended it by now

It's just one of those things that sits there, does exactly what it
says on the tin and only needs a reboot a couple of times a year,
trust them to go and spoil it

"Johnny B Good" wrote in message

[1] I'm not totally excluding other causes such as chip faults or other
obscure components, it's just even less probable and a waste of time
considering them unless you're extremely familiar with the kit and its
documented 'stock fault' history.

Yep, fair comment. I bet most faults are caused by crap software. In the
hardware department it's possibly the hard disk (which has a high MTBF but a
limited life expectancy), and then the power supply components. And I quite
agree: high frequency/high ripple current is an area (in my experience as an
electronics tech) where electrolytics seem to struggle.

Really good electrolytics can be quite expensive, so it may well be that
costs are shaved in this area (also maybe a need to keep the physical size
down).

So, although PVR software seems to be the biggest cause of reported problems
by far, IF IT IS a hardware fault I completely accept that it is "probably"
the hard disk or the power supply caps, rather than anything else.

Glad to put the record straight! :-)

SteveT

In message , Albert Ross
wrote

The Topfield is not supported by the manufacturer with regards
software/firmware. Their last effort at Freeview+ is over a year old,
it's as buggy as hell and not a single bug fix in site.

Damn, I haven't been keeping up. The original 5800 was quite usable
out of the box and I only updated the firmware a couple of times.

And possibly one of those updates had very little to do with Topfield.
The ''Split NIT' bug was fixed by a customer and not Topfield.

--
Alan
news2009 {at} admac {dot} myzen {dot} co {dot} uk

In article , Steve Thackery
wrote:

"Johnny B Good" wrote in message

[1] I'm not totally excluding other causes such as chip faults or other
obscure components, it's just even less probable and a waste of time
considering them unless you're extremely familiar with the kit and its
documented 'stock fault' history.

Yep, fair comment. I bet most faults are caused by crap software. In
the hardware department it's possibly the hard disk (which has a high
MTBF but a limited life expectancy), and then the power supply
components. And I quite agree: high frequency/high ripple current is an
area (in my experience as an electronics tech) where electrolytics seem
to struggle.

I'm sure you remember the first thing to check - drummed into me as a mtce
engineer over 40 years ago. Check the power supply.

--
From KT24

Using a RISC OS computer running v5.16

"charles" wrote in message
...

I'm sure you remember the first thing to check - drummed into me as a mtce
engineer over 40 years ago. Check the power supply.

Indeed, although back then we didn't have switch-mode power supplies, so
subtle problems with high frequency noise creeping onto the power rails was
less of a problem.

Interestingly, I wonder how many people remember the "secondary SMPS" which
came out before the now ubiquitous "primary SMPS".

The secondary SMPS appeared on one of the teleprinter power supplies I
worked on back in the early 80s. It had a completely traditional mains
transformer, bridge rectifier and smoothing caps, but instead of the linear
5V or 12V regulator, it used a switch-mode regulator.

Quite a good compromise, actually - just as efficient, if not more so, than
the primary SMPS, and WAY more reliable! The only disadvantage was the need
for the usual large, heavy mains transformer and smoothing caps.

The first primary SMPSs I worked on (I mean where the switching is done on
the primary side, so you can use a teeny transformer) were for the "Cheetah"
telex machine and were fabulously unreliable, and exceptionally good for
firework displays.

When they failed they'd blow the plug-top 3A fuse. When we felt in need of
some excitement in the workshop, we would plug the failed power supply into
the kettle lead (always close to hand, of course) which had a 13A fuse. The
flashes, sparks, sizzles and bangs were a marvellous way to brighten a
humdrum day. And just occasionally, a shard off the top of one of the
tantalum capacitors would fly straight through one of the air vents and
whizz across the workshop.

It probably sounds irresponsible, but there was a smidgin of sense in it.
Fault location in an SMPS can be very complicated and difficult, especially
as almost all failures of the high voltage components (such as the switching
transistors) would cause ripple failures spreading widely through the
circuit. You could end up with "circular" faults, where a failed transistor
would cause a capacitor to go short-circuit. And then the short-circuit
capacitor would blow the new transistor.

By vapourising all the faulty components in one go, at least you could see
what needed replacing. :-)

SteveT

In article , Johnny B Good wrote:
It's probably bad caps in the power supply. 15 mins max to fix it.

Was this written as a joke? I only ask because - of course - it could be
loads of things (including noise on the 5V/12V lines), and it's probably a
bit "previous" to say what the fault "probably" is at this stage! :-)

"Bad Caps" is a pretty common failure in most electronic kit. It's been
the bane of PC motherboards for the past 15 years or so and it's, in
this case, "(including noise on the 5V/12V lines)" the most likely cause
(the alternative of inductors developing shorted turns is significantly
less probable[1]).

Although capacitor failure (electrolytic) is not good, it does have the
charm that it can _usually_ be spotted by close inspection with the MK1
eyeball (and a magnifying glass to be certain of those that, at first
glance, appear ok) and are also about the only component that can be
readily replaced by a reasonably competent amateur with moderate
soldering skills.

I'm astonished by the number of electronic faults that are diagnosed remotely
as requiring nothing more than replacement of all the capacitors, a procedure
that with most modern circuitry would probably do more harm than good. My
approach to electronic faultfinding, reinforced by what I was taught at the BBC
many years ago, has always been to use the appropriate measuring instruments
and service manuals to diagnose the component that is actually faulty, replace
that, and leave the others alone.

Rod.
--
Virtual Access V6.3 free usenet/email software from
http://sourceforge.net/projects/virtual-access/

In message en.co.uk,
Roderick Stewart wrote

I'm astonished by the number of electronic faults that are diagnosed remotely
as requiring nothing more than replacement of all the capacitors, a procedure
that with most modern circuitry would probably do more harm than good. My
approach to electronic faultfinding, reinforced by what I was taught at
the BBC
many years ago, has always been to use the appropriate measuring instruments
and service manuals to diagnose the component that is actually faulty, replace
that, and leave the others alone.

But in some consumer electronics the evidence is from many 100s of
reported failures and a blanket instruction to replace all capacitors of
certain type results in many faults being repaired.

For instance, the Topfield 5800 of a certain age is prone to power
supply failures where the fault is usually one of six electrolytic
capacitors that have failed. A measurement of the power supply voltages
will confirm the fault. The recommended action to replace all six
capacitors has resulted in many people restoring their machines to a
fully working condition at a cost of £5 to £20. Many people reporting a
successful repair have had no previous electronics experience or
training, they have just followed the comprehensive advice given on a
user forum.
http://en.wikibooks.org/wiki/Topfield_5800/PSU_Repair


--
Alan
news2009 {at} admac {dot} myzen {dot} co {dot} uk

In article en.co.uk,
Roderick Stewart scribeth
thus
In article , Johnny B Good wrote:
It's probably bad caps in the power supply. 15 mins max to fix it.

Was this written as a joke? I only ask because - of course - it could be
loads of things (including noise on the 5V/12V lines), and it's probably a
bit "previous" to say what the fault "probably" is at this stage! :-)

"Bad Caps" is a pretty common failure in most electronic kit. It's been
the bane of PC motherboards for the past 15 years or so and it's, in
this case, "(including noise on the 5V/12V lines)" the most likely cause
(the alternative of inductors developing shorted turns is significantly
less probable[1]).

Although capacitor failure (electrolytic) is not good, it does have the
charm that it can _usually_ be spotted by close inspection with the MK1
eyeball (and a magnifying glass to be certain of those that, at first
glance, appear ok) and are also about the only component that can be
readily replaced by a reasonably competent amateur with moderate
soldering skills.

I'm astonished by the number of electronic faults that are diagnosed remotely
as requiring nothing more than replacement of all the capacitors, a procedure
that with most modern circuitry would probably do more harm than good. My
approach to electronic faultfinding, reinforced by what I was taught at the BBC
many years ago, has always been to use the appropriate measuring instruments
and service manuals to diagnose the component that is actually faulty, replace
that, and leave the others alone.

Rod.

Assuming you can get service manuals .. rather difficult for most all
domestic TV AV equipment...
--
Tony Sayer

"Roderick Stewart" wrote in
message

I'm astonished by the number of electronic faults that are diagnosed
remotely
as requiring nothing more than replacement of all the capacitors, a
procedure
that with most modern circuitry would probably do more harm than good. My
approach to electronic faultfinding, reinforced by what I was taught at
the BBC
many years ago, has always been to use the appropriate measuring
instruments
and service manuals to diagnose the component that is actually faulty,
replace
that, and leave the others alone.

I completely agree with you, in principle. But let me argue to other case
with this illustration.

When I was an electronics bench techie we used to get circuit boards in from
all sorts of printers and telex machines. To begin with you would set up
all the gear, and work through what was often quite a lengthy and
sophisticated faulting procedure with storage scopes, logic analysers,
in-circuit emulators, and so on.

I'm sure you must have noticed in your time at the BBC the same thing as I
did: for each particular product, the fault was almost invariably in one of
more of the same three components ("three" is just an example).

I found I might spend 20-60 minutes diagnosing, and always end up at one of
those three components.

So after a while I did the logical thing: whenever a board came it, I'd
automatically replace all three components. Their total cost was a pound or
two, and I'd saved up to a man-hour. 98% of the time it fixed the board.
If it didn't, THEN I'd power up the analogue storage scope, the Huntron
Tracker, and all that stuff.

This was especially valuable for SPMSs, which as you know can be a nightmare
to fault locate on due to the cascade failures they suffer. Faulting one of
those might take literally hours, even a couple of days. Again, I
identified a list of about 15 components, some or all of which were blown up
whenever the switching transistors went short circuit (which they did all
the damn time).

Replacing all of those probably cost £20, but I could do two an hour, rather
than one or maybe two a day.

So yes, although I see your point, there is a lot to be said for a pragmatic
approach. And I have to say that electrolytic capacitors in switched mode
power supplies do seem to be rather more prone to failure than any other
electronic component in a typical consumer product. (In the early SMPSs it
was the high voltage switching transistors, but now they are pretty well
bullet-proof and it's the capacitors which are the next-weakest link).

So there MAY be a case for just swapping the lot, as I used to do with the
teleprinter power supply components.

I'm not sure why you say it may do more harm than good. If you use the
appropriate components and the appropriate tools, there's no reason why it
should do harm.

SteveT

In article , Steve Thackery wrote:
I'm sure you must have noticed in your time at the BBC the same thing as I
did: for each particular product, the fault was almost invariably in one of
more of the same three components ("three" is just an example).

I found I might spend 20-60 minutes diagnosing, and always end up at one of
those three components.

So after a while I did the logical thing: whenever a board came it, I'd
automatically replace all three components. Their total cost was a pound or
two, and I'd saved up to a man-hour. 98% of the time it fixed the board.
If it didn't, THEN I'd power up the analogue storage scope, the Huntron
Tracker, and all that stuff.

Yes, I've done the same sort of thing myself if there is a recognised "stock
fault" with a particular item of equipment.

However, I'm a little suspicious that the homespun advice one sees so freely
given on the internet about the repair of electronics is nearly *always* to
replace capacitors, sometimes to replace all of them, and hardly ever
mentioning any other type of component. Capacitors can fail of course, and
electrolytics can dry up, particularly the big ones in power supplies, but it
isn't in my experience that these are the only components that ever go wrong.
Without intending any direct criticism of any individual here, I can't help
wondering how much of this internet advice has a proper pedigree traceable
back to a source based on practical experience, and how much of it is self-
reinforcing through repetition.

Regarding my "more harm than good" comment, I was thinking of modern close-
packed circuit boards, which do not in general lend themselves to frequent
component replacement, if at all. If the advice to "change all the capacitors"
is followed as freely as it is given, particularly as in many cases it will
have been sought as a cheap solution by people who don't have the means or the
knowledge to fault-find properly, it can't overall be doing any good.

Rod.
--
Virtual Access V6.3 free usenet/email software from
http://sourceforge.net/projects/virtual-access/