I was toying with doing a wiki article on component level repair of
electronic stuff, since there seems to be so much otherwise decent kit
that dies these days for silly reasons like failing capacitors etc.

With modern lead free solder, and ever increasing layout densities, use
of multilayer boards etc, the job of component removal seems to be
getting ever harder. There was a time when a moderate iron with enough
bit mass to heat the joint quickly, was all you needed. To reflow the
joint and then either pull the device, or apply a solder sucker, all one
did was tin the bit, apply heat for a couple of secs, and job done.

Recently I have needed to recap a few motherboards (not usually an
economic exercise - but these were custom form factor jobbies where
replacements are not readily available), and they were a right PITA to
work on. Obviously multilayer, and since the caps are on the power
regulation sections, probably connected to fairly heavy power plane
traces in the board. Component removal was very difficult - often taking
excessive time for the joint to reflow. Tight pins in close holes with a
small annulus. However cleaning the holes for reuse proved impossible
with any of my normal irons. In the end I had to resort to using a hot
air paint stripper with small nozzle on the end to get enough heat into
the area to be able to suck the holes clear!

I was wondering what suggestions those of you who do this daily had to
offer?


--
Cheers,

John.

/================================================== ===============\
| Internode Ltd - http://www.internode.co.uk |
|-----------------------------------------------------------------|
| John Rumm - john(at)internode(dot)co(dot)uk |
\================================================= ================/

Recently I have needed to recap a few motherboards (not usually an
economic exercise - but these were custom form factor jobbies where
replacements are not readily available), and they were a right PITA to
work on. Obviously multilayer, and since the caps are on the power
regulation sections, probably connected to fairly heavy power plane
traces in the board. Component removal was very difficult - often taking
excessive time for the joint to reflow. Tight pins in close holes with a
small annulus. However cleaning the holes for reuse proved impossible
with any of my normal irons. In the end I had to resort to using a hot
air paint stripper with small nozzle on the end to get enough heat into
the area to be able to suck the holes clear!

I was wondering what suggestions those of you who do this daily had to
offer?



We had a similar problem on a Harris VHF/FM broadcast transmitter where
they recommend you break the cap off and solder to the wires that are
left behind;!..

This wasn't to do with cleaning out the holes just a lorra dismantling
to get to the other side of that board. Prolly not too practical on
Motherboards with very high component densities, but as you say a real
PITA to clear the holes..

Just bought a new Gigabyte board the other day boasts "solid" Electro
caps;!...
--
Tony Sayer

On 13/01/2012 18:55, tony sayer wrote:
Recently I have needed to recap a few motherboards (not usually an
economic exercise - but these were custom form factor jobbies where
replacements are not readily available), and they were a right PITA to
work on. Obviously multilayer, and since the caps are on the power
regulation sections, probably connected to fairly heavy power plane
traces in the board. Component removal was very difficult - often taking
excessive time for the joint to reflow. Tight pins in close holes with a
small annulus. However cleaning the holes for reuse proved impossible
with any of my normal irons. In the end I had to resort to using a hot
air paint stripper with small nozzle on the end to get enough heat into
the area to be able to suck the holes clear!

I was wondering what suggestions those of you who do this daily had to
offer?



We had a similar problem on a Harris VHF/FM broadcast transmitter where
they recommend you break the cap off and solder to the wires that are
left behind;!..

Hmm nice ;-)

Still in the past when I have needed to remove a QFP device off a board,
it was safer to run a knife down each side a few times to chop off all
the legs, and then flick each off individually after rather than try
doing the whole thing in one hit. Bit of a shame if the device was not
toast before you started though!

This wasn't to do with cleaning out the holes just a lorra dismantling
to get to the other side of that board. Prolly not too practical on
Motherboards with very high component densities, but as you say a real
PITA to clear the holes..

Actually a workable solution - although difficult getting cutters under
them sometimes. You can sometimes get a scalpel blade in there.


Just bought a new Gigabyte board the other day boasts "solid" Electro
caps;!...

Yup, some of the ASUS ones are now claiming mil-spec caps in full ali
enclosures rather than the normal plastic coated ones.

--
Cheers,

John.

/================================================== ===============\
| Internode Ltd - http://www.internode.co.uk |
|-----------------------------------------------------------------|
| John Rumm - john(at)internode(dot)co(dot)uk |
\================================================= ================/

On Jan 13, 6:55*pm, tony sayer wrote:
Recently I have needed to recap a few motherboards (not usually an
economic exercise - but these were custom form factor jobbies where
replacements are not readily available), and they were a right PITA to
work on. Obviously multilayer, and since the caps are on the power
regulation sections, probably connected to fairly heavy power plane
traces in the board. Component removal was very difficult - often taking
excessive time for the joint to reflow. Tight pins in close holes with a
small annulus. However cleaning the holes for reuse proved impossible
with any of my normal irons. In the end I had to resort to using a hot
air paint stripper with small nozzle on the end to get enough heat into
the area to be able to suck the holes clear!

I was wondering what suggestions those of you who do this daily had to
offer?

We had a similar problem on a Harris VHF/FM broadcast transmitter where
they recommend you break the cap off and solder to the wires that are
left behind;!..

That's it. The other problem with desoldering TH parts from multilayer
boards is its liable to damage connection between buried layers of the
pcb, which is then a nightmare to repair.


I see a couple of potential problems with tackling electronic repairs
in one page on a wiki - or anywhere else, aimed at the inexperienced.

First, no matter how long the page, a beginner simply isnt going to
end up competent after reading it. So to enable a newb to get a repair
done requires skipping key info, which will inevitably result in
problems and unsuitable advice.

The other issue is that popular myths have a habit of coming into
discussions about electronics in largely non-expert forums.


NT

In message , John
Rumm wrote


I was wondering what suggestions those of you who do this daily had to
offer?

Never use lead free solder. 60/40, with flux, is still readily
available.

Re-solder the joint before attempting to remove components or re-solder
the pad after component removal and before cleaning out the hole. This
will introduce some more flux to the joint and allow the solder to flow
better when using a solder sucker or de-soldering braid.

Most (all) of the capacitors that fail on PSUs tend to be the larger can
type electrolytic. On a single side PCB just heat one pad until the
solder melts and pull the lead through a bit by pulling on the capacitor
can. Repeat on the other leg - then go back to the original leg etc.
On a multi-layer board this method will cause damage to the board so cut
the component out and then remove the remaining soldered legs - if you
are going to replace a (cheap) component doesn't matter if you damage
the old one.

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

On 13/01/2012 18:55, tony sayer wrote:

Just bought a new Gigabyte board the other day boasts "solid" Electro
caps;!...

Nothing particularly new about those, see os-CON, etc.

--
Andy

On 13/01/2012 19:17, NT wrote:
On Jan 13, 6:55 pm, tony wrote:
Recently I have needed to recap a few motherboards (not usually an
economic exercise - but these were custom form factor jobbies where
replacements are not readily available), and they were a right PITA to
work on. Obviously multilayer, and since the caps are on the power
regulation sections, probably connected to fairly heavy power plane
traces in the board. Component removal was very difficult - often taking
excessive time for the joint to reflow. Tight pins in close holes with a
small annulus. However cleaning the holes for reuse proved impossible
with any of my normal irons. In the end I had to resort to using a hot
air paint stripper with small nozzle on the end to get enough heat into
the area to be able to suck the holes clear!

I was wondering what suggestions those of you who do this daily had to
offer?

We had a similar problem on a Harris VHF/FM broadcast transmitter where
they recommend you break the cap off and solder to the wires that are
left behind;!..

That's it. The other problem with desoldering TH parts from multilayer
boards is its liable to damage connection between buried layers of the
pcb, which is then a nightmare to repair.


I see a couple of potential problems with tackling electronic repairs
in one page on a wiki - or anywhere else, aimed at the inexperienced.

First, no matter how long the page, a beginner simply isnt going to
end up competent after reading it. So to enable a newb to get a repair
done requires skipping key info, which will inevitably result in
problems and unsuitable advice.

Can't see that no advice is going to be better than some. Also keep in
mind that in this day and age, there are not the army of professional
repairers around that there used to be. Hence the options often come
down to fix it yourself, or throw it away. Assuming you are starting
with something already broken, what is there to lose?

The other issue is that popular myths have a habit of coming into
discussions about electronics in largely non-expert forums.

Well possibly, but since we have a number of professional engineers here
with experience in a reasonably wide range of electronics and the
handling of same, that ought not be so much of a problem.


--
Cheers,

John.

/================================================== ===============\
| Internode Ltd - http://www.internode.co.uk |
|-----------------------------------------------------------------|
| John Rumm - john(at)internode(dot)co(dot)uk |
\================================================= ================/

John Rumm wrote:
I was wondering what suggestions those of you who do this daily had to
offer?

Hot air. It makes it hugely easier. There are relatively cheap hot air
stations around... go to certain backstreets of cities in the developing
world and you can see people doing component level service on recent TVs etc
there.

Not tried but, for example:
http://cpc.farnell.com/duratool/zd-9...ion/dp/SD01139
(I'm not sure I believe the listed temperature of '150' there)

You don't need dozens of tips for each and every package, but just some
narrow tips and some broad (square) ones will do. Having more heat can be
handy for the broader tips. Vacuum is nice, but really not essential. A
hotplate can be handy for preheat.

I think a selection of videos would be good for an FAQ: soldering is
something you learn from watching, not from reading.

Theo

In article ,
Alan wrote:
Re-solder the joint before attempting to remove components or re-solder
the pad after component removal and before cleaning out the hole. This
will introduce some more flux to the joint and allow the solder to flow
better when using a solder sucker or de-soldering braid.

A1 tip.

Most (all) of the capacitors that fail on PSUs tend to be the larger can
type electrolytic. On a single side PCB just heat one pad until the
solder melts and pull the lead through a bit by pulling on the capacitor
can. Repeat on the other leg - then go back to the original leg etc.
On a multi-layer board this method will cause damage to the board so cut
the component out and then remove the remaining soldered legs - if you
are going to replace a (cheap) component doesn't matter if you damage
the old one.

Yup. Don't attempt to save the suspect component - if possible cut it off
so you can attack each leg individually. The important bit you need to
avoid damage to is the PCB.

--
*If a parsley farmer is sued, can they garnish his wages?

Dave Plowman London SW
To e-mail, change noise into sound.

On 13/01/2012 18:37, John Rumm wrote:

I was toying with doing a wiki article on component level repair of
electronic stuff, since there seems to be so much otherwise decent kit
that dies these days for silly reasons like failing capacitors etc.

With modern lead free solder, and ever increasing layout densities, use
of multilayer boards etc, the job of component removal seems to be
getting ever harder. There was a time when a moderate iron with enough
bit mass to heat the joint quickly, was all you needed. To reflow the
joint and then either pull the device, or apply a solder sucker, all one
did was tin the bit, apply heat for a couple of secs, and job done.

Recently I have needed to recap a few motherboards (not usually an
economic exercise - but these were custom form factor jobbies where
replacements are not readily available), and they were a right PITA to
work on. Obviously multilayer, and since the caps are on the power
regulation sections, probably connected to fairly heavy power plane
traces in the board. Component removal was very difficult - often taking
excessive time for the joint to reflow. Tight pins in close holes with a
small annulus. However cleaning the holes for reuse proved impossible
with any of my normal irons. In the end I had to resort to using a hot
air paint stripper with small nozzle on the end to get enough heat into
the area to be able to suck the holes clear!

I was wondering what suggestions those of you who do this daily had to
offer?


I have a cheap hot-air SMT rework station (ebay) which is quite
effective at individual component removal on multi-layer boards. I
usually have the smallest tip fitted which has about a 4mm hole.
I use the same method for SMT ICs, run the air-flow around the pins
until the chip is free, then flip it over with a knife or similar.
Clean pads with decent solder wick and fit replacement, typically by
flow-soldering by hand all the pins on each side of the IC in turn.

Capacitor failure seems to be far worse than it used to be, when I
worked in electronic repair (~15 years ago) I rarely had any issues with
caps, but now it's one of the first things I look for, especially in
SMPSUs, monitors & TVs.

In message , John
Rumm writes

I was toying with doing a wiki article on component level repair of
electronic stuff, since there seems to be so much otherwise decent kit
that dies these days for silly reasons like failing capacitors etc.


A few random thoughts - more to follow


Repair would entail correct diagnosis of the fault

Some boards are so cheaply built that unless you are good with a
soldering iron, you can detach a pad from the pcb

Decent equipment and practice are essential if you are not going to do
more harm than good. Attacking a PCB by someone who is inexperienced is
liable to end in tears


plated through holes, once the component lead is removed, if you then
reapply solder to the joint and can clean the hole out better with a
solder sucker

Surface mount devices are often glued to the pcb


for surface mounted devices, you can now buy a budget hot air gun

http://cpc.farnell.com/duratool/zd-9...-rework-statio
n/dp/SD01139?in_merch=Featured%20Products&MER=e-bb45-00001001

You can also buy a budget desoldering station if you want a new toy.
We've been using one for a couple of months now and it hasn't fallen
apart yet

http://cpc.farnell.com/duratool/d006...dp/SD01384?in_
merch=Featured%20Products&MER=e-bb45-00001001



--
geoff

In article om,
Alan Deane wrote:
Capacitor failure seems to be far worse than it used to be, when I
worked in electronic repair (~15 years ago) I rarely had any issues with
caps, but now it's one of the first things I look for, especially in
SMPSUs, monitors & TVs.

A SMPS puts a far greater load on the caps than an analogue one.

--
*Indian Driver - Smoke signals only*

Dave Plowman London SW
To e-mail, change noise into sound.

In message , Theo Markettos
writes
I think a selection of videos would be good for an FAQ: soldering is
something you learn from watching, not from reading.

This is an excellent suggestion.

Brought up on valves, I'm still at the level of "cut or break off the
component and resolder to the old wires", but the resoldering is often
difficult with capacitors because they are over the wires.

I've got tons of stuff here that is just waiting for me to pluck up the
courage, so any info about updates to my technique would be very
welcome.
--
Bill

On Fri, 13 Jan 2012 23:47:55 +0000, Theo Markettos wrote:

John Rumm wrote:
I was wondering what suggestions those of you who do this daily had to
offer?

Hot air. It makes it hugely easier. There are relatively cheap hot air
stations around... go to certain backstreets of cities in the
developing world and you can see people doing component level service on
recent TVs etc there.

Not tried but, for example:
http://cpc.farnell.com/duratool/zd-9...ir-smd-rework-
station/dp/SD01139
(I'm not sure I believe the listed temperature of '150' there)

You don't need dozens of tips for each and every package, but just some
narrow tips and some broad (square) ones will do. Having more heat can
be handy for the broader tips. Vacuum is nice, but really not
essential. A hotplate can be handy for preheat.

Any recommendations for hotplates (DIY or OTS)?



--
Use the BIG mirror service in the UK:
http://www.mirrorservice.org

*lightning protection* - a w_tom conductor

On 14/01/2012 00:09, Alan Deane wrote:

Capacitor failure seems to be far worse than it used to be, when I
worked in electronic repair (~15 years ago) I rarely had any issues with
caps, but now it's one of the first things I look for, especially in
SMPSUs, monitors & TVs.

Initially this was supposed to be due to a supply of dodgy caps made
with a stolen (but incomplete) electrolyte formula. However the problems
seem to have endured far longer than that alone would explain!

It does seem to be any application in high ripple conditions they seem
to go tits up first. So SMPSUs and voltage regulation applications seem
particularly prone. I always try and replace them with low ESR caps
designed for this application, and with a 10K hours life rather than the
cheaper 2K ones.


--
Cheers,

John.

/================================================== ===============\
| Internode Ltd - http://www.internode.co.uk |
|-----------------------------------------------------------------|
| John Rumm - john(at)internode(dot)co(dot)uk |
\================================================= ================/

On 13/01/2012 23:47, Theo Markettos wrote:

John wrote:
I was wondering what suggestions those of you who do this daily had to
offer?

Hot air. It makes it hugely easier. There are relatively cheap hot air
stations around... go to certain backstreets of cities in the developing
world and you can see people doing component level service on recent TVs etc
there.


Not tried but, for example:
http://cpc.farnell.com/duratool/zd-9...ion/dp/SD01139
(I'm not sure I believe the listed temperature of '150' there)

No that does sound implausibly low... would not even melt leaded solder.
Not a bad price mind you given you can get a reasonable range of nozzles
for it.

You don't need dozens of tips for each and every package, but just some
narrow tips and some broad (square) ones will do. Having more heat can be
handy for the broader tips. Vacuum is nice, but really not essential. A
hotplate can be handy for preheat.

I think a selection of videos would be good for an FAQ: soldering is
something you learn from watching, not from reading.

We could probably find one of the many on youtube to link to. Having
said that I learnt from a written description many years ago.



--
Cheers,

John.

/================================================== ===============\
| Internode Ltd - http://www.internode.co.uk |
|-----------------------------------------------------------------|
| John Rumm - john(at)internode(dot)co(dot)uk |
\================================================= ================/

In message , John
Rumm writes
On 13/01/2012 23:47, Theo Markettos wrote:

John wrote:
I was wondering what suggestions those of you who do this daily had to
offer?

Hot air. It makes it hugely easier. There are relatively cheap hot air
stations around... go to certain backstreets of cities in the developing
world and you can see people doing component level service on recent TVs etc
there.


Not tried but, for example:

http://cpc.farnell.com/duratool/zd-9...md-rework-stat
ion/dp/SD01139
(I'm not sure I believe the listed temperature of '150' there)

No that does sound implausibly low... would not even melt leaded
solder. Not a bad price mind you given you can get a reasonable range
of nozzles for it.

I now have three of them, they are hot enough, they work. I have had a
couple of element burnouts, though. You only need the small (4mm?)
nozzle for normal work


You don't need dozens of tips for each and every package, but just some
narrow tips and some broad (square) ones will do. Having more heat can be
handy for the broader tips. Vacuum is nice, but really not essential. A
hotplate can be handy for preheat.

I think a selection of videos would be good for an FAQ: soldering is
something you learn from watching, not from reading.

We could probably find one of the many on youtube to link to. Having
said that I learnt from a written description many years ago.




--
geoff

On Jan 13, 9:14*pm, Owain wrote:
On Jan 13, 7:17*pm, NT wrote:

The other issue is that popular myths have a habit of coming into
discussions about electronics in largely non-expert forums.

Surely you're not suggesting this is a largely non-expert forums.

Owain

in electronic repair?


NT

On Jan 14, 12:21*am, geoff wrote:
In message , John
Rumm writes



I was toying with doing a wiki article on component level repair of
electronic stuff, since there seems to be so much otherwise decent kit
that dies these days for silly reasons like failing capacitors etc.

A few random thoughts - more to follow

Repair would entail correct diagnosis of the fault

exactly, and thats where ukdiy often falls down, and beginners cant
generally do it on their own.


plated through holes, once the component lead is removed, if you then
reapply solder to the joint and can clean the hole out better with a
solder sucker

you can, but I've never seen teh point. Just heat the hole and push
the component lead through.


NT

I have a puter here which is exhibiting all the signs of dodgy caps. It
occasionally clunks and clicks then lets through switching transients to the
analogue bits of the sound card. Its not the card. Its a real pity when such
an obviously simple thing means goodby.
Maybe they just cannot make capacitors that last any more.
Brian

--
Brian Gaff....Note, this account does not accept Bcc: email.
graphics are great, but the blind can't hear them
Email:
__________________________________________________ __________________________________________________ __________


"tony sayer" wrote in message
...
Recently I have needed to recap a few motherboards (not usually an
economic exercise - but these were custom form factor jobbies where
replacements are not readily available), and they were a right PITA to
work on. Obviously multilayer, and since the caps are on the power
regulation sections, probably connected to fairly heavy power plane
traces in the board. Component removal was very difficult - often taking
excessive time for the joint to reflow. Tight pins in close holes with a
small annulus. However cleaning the holes for reuse proved impossible
with any of my normal irons. In the end I had to resort to using a hot
air paint stripper with small nozzle on the end to get enough heat into
the area to be able to suck the holes clear!

I was wondering what suggestions those of you who do this daily had to
offer?



We had a similar problem on a Harris VHF/FM broadcast transmitter where
they recommend you break the cap off and solder to the wires that are
left behind;!..

This wasn't to do with cleaning out the holes just a lorra dismantling
to get to the other side of that board. Prolly not too practical on
Motherboards with very high component densities, but as you say a real
PITA to clear the holes..

Just bought a new Gigabyte board the other day boasts "solid" Electro
caps;!...
--
Tony Sayer

On 14/01/2012 03:32, NT wrote:
On Jan 14, 12:21 am, wrote:
In , John
writes



I was toying with doing a wiki article on component level repair of
electronic stuff, since there seems to be so much otherwise decent kit
that dies these days for silly reasons like failing capacitors etc.

A few random thoughts - more to follow

Repair would entail correct diagnosis of the fault

exactly, and thats where ukdiy often falls down, and beginners cant
generally do it on their own.

I don't think that anyone is suggesting detailed fault finding
techniques for digital electronics.

(which while not actually a massively difficult task with the right
equipment, its very easy to find yourself twiddling knobs on several
hundred grands worth of it in the process, which tends to put much of it
beyond most folks DIY budgets)

At the simplest level, a hell of allot of kit can be saved just by
spotting the caps that are bulging and spewing electrolyte all over the
place. If you want to get a bit deeper, then you can do plenty with a
multimeter, and even relatively specialist bits of kits like LCR
component analysers or ESR meters are not that expensive.

(I don't agree that ukdiy falls down on diagnosis either particularly -
we have some very good engineers here with plenty of experience in
relevant fields. That does not mean that all contributions or all advice
will be spot on, but its no reason to write it off either)

plated through holes, once the component lead is removed, if you then
reapply solder to the joint and can clean the hole out better with a
solder sucker

you can, but I've never seen teh point. Just heat the hole and push
the component lead through.

That works for many boards, but not all. In some cases a typical iron
delicate enough for the size of task can't supply the heat on a
multilayer board. That's where the hot air systems work well, either to
temper an area of board to give the iron a chance, or more typically on
their own just to reflow the obscured hole so it can be sucked clean.

--
Cheers,

John.

/================================================== ===============\
| Internode Ltd - http://www.internode.co.uk |
|-----------------------------------------------------------------|
| John Rumm - john(at)internode(dot)co(dot)uk |
\================================================= ================/

On Sat, 14 Jan 2012 01:52:16 +0000, geoff wrote:

In message , John
Rumm writes
On 13/01/2012 23:47, Theo Markettos wrote:

John wrote:
I was wondering what suggestions those of you who do this daily had
to offer?

Hot air. It makes it hugely easier. There are relatively cheap hot
air stations around... go to certain backstreets of cities in the
developing world and you can see people doing component level service
on recent TVs etc there.


Not tried but, for example:

http://cpc.farnell.com/duratool/zd-9...md-rework-stat
ion/dp/SD01139
(I'm not sure I believe the listed temperature of '150' there)

No that does sound implausibly low... would not even melt leaded solder.
Not a bad price mind you given you can get a reasonable range of nozzles
for it.

I now have three of them, they are hot enough, they work. I have had a
couple of element burnouts, though. You only need the small (4mm?)
nozzle for normal work

Good to have some actual feedback. May treat myself!



--
Use the BIG mirror service in the UK:
http://www.mirrorservice.org

*lightning protection* - a w_tom conductor

On Fri, 13 Jan 2012 19:32:39 -0800, NT wrote:

plated through holes, once the component lead is removed, if you then
reapply solder to the joint and can clean the hole out better with a
solder sucker

you can, but I've never seen teh point. Just heat the hole and push the
component lead through.

A danger of taking the through-hole plating off if it's a tight fit?
Especially in a multi-layer board.



--
Use the BIG mirror service in the UK:
http://www.mirrorservice.org

*lightning protection* - a w_tom conductor

Not tried but, for example:http://cpc.farnell.com/duratool/zd-9...-smd-rework-st...
(I'm not sure I believe the listed temperature of '150' there)


We've got exactly that model at work, and it's an excellent machine.

Very quiet in normal operation.

It has the disconcerting (but really very good) feature of - when you
switch it *off* - of turning the air feed rate to a noisy maximum for
a minute or so to cool itself down.

In article
,
NT wrote:
On Jan 14, 12:21 am, geoff wrote:
In message , John
Rumm writes



I was toying with doing a wiki article on component level repair of
electronic stuff, since there seems to be so much otherwise decent kit
that dies these days for silly reasons like failing capacitors etc.

A few random thoughts - more to follow

Repair would entail correct diagnosis of the fault

exactly, and thats where ukdiy often falls down, and beginners cant
generally do it on their own.


plated through holes, once the component lead is removed, if you then
reapply solder to the joint and can clean the hole out better with a
solder sucker

you can, but I've never seen teh point. Just heat the hole and push
the component lead through.

no use if the component you are trying to fit has multiple legs.


NT

--
From KT24

Using a RISC OS computer running v5.16

In message , John
Rumm wrote
On 14/01/2012 00:09, Alan Deane wrote:

Capacitor failure seems to be far worse than it used to be, when I
worked in electronic repair (~15 years ago) I rarely had any issues with
caps, but now it's one of the first things I look for, especially in
SMPSUs, monitors & TVs.

Initially this was supposed to be due to a supply of dodgy caps made
with a stolen (but incomplete) electrolyte formula. However the
problems seem to have endured far longer than that alone would explain!

A greater use of switching supplies or it could be taking
miniaturisation one step too far?

Replacement electrolytic capacitors from _reputable_ manufacturers are
often physically larger - usually the same diameter and pin spacing but
longer.


It does seem to be any application in high ripple conditions they seem
to go tits up first. So SMPSUs and voltage regulation applications seem
particularly prone. I always try and replace them with low ESR caps
designed for this application, and with a 10K hours life rather than
the cheaper 2K ones.

For the DIY the difference in price between the best and worst of
capacitors is often 10 pence. The low ESR figure is one of the
important parameters.



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

In message , geoff
wrote

Surface mount devices are often glued to the pcb

Not in my experience. Solder is the only thing holding the component on
the board on 99.99% of commercial gear.

There may be board coatings on boards for the military market which make
rework almost impossible for those with out the right tools.

The only time I've seen glue used is when a physically very large
component has been fitted and the manufacturer has used a "blob" of glue
to bond to component case to the board to prevent differential physical
movement - to help prevent vibration failures.

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

In article ,
Alan wrote:
Initially this was supposed to be due to a supply of dodgy caps made
with a stolen (but incomplete) electrolyte formula. However the
problems seem to have endured far longer than that alone would explain!

A greater use of switching supplies or it could be taking
miniaturisation one step too far?

The first SMPS I came across on a domestic product was in a VHS, and the
caps on that failed. It was made by Panasonic. so I'd guess they didn't
buy in dodgy caps.

--
*Hang in there, retirement is only thirty years away! *

Dave Plowman London SW
To e-mail, change noise into sound.

John Rumm brought next idea :
Initially this was supposed to be due to a supply of dodgy caps made with a
stolen (but incomplete) electrolyte formula. However the problems seem to
have endured far longer than that alone would explain!

It does seem to be any application in high ripple conditions they seem to go
tits up first. So SMPSUs and voltage regulation applications seem
particularly prone. I always try and replace them with low ESR caps designed
for this application, and with a 10K hours life rather than the cheaper 2K
ones.

I agree entirely...

In the past couple of years, I've had three failures which I've
actually been bothered to repair - all turned out to be failed
electrolytic caps. One the was on the two year old washing machine
micro controller board. Second was sat receiver 13 month old. Third was
the PWM speed control for my cars heater fan, a tiny bi-polar
electrolytic in the actual power switching module.

At one time, when repairing not very old equipment, the electrolytic
would be the last things I would be checking for faults. Now it is the
first thing I look to.

--
Regards,
Harry (M1BYT) (L)
http://www.ukradioamateur.co.uk

John Rumm formulated on Friday :
I was toying with doing a wiki article on component level repair of
electronic stuff, since there seems to be so much otherwise decent kit that
dies these days for silly reasons like failing capacitors etc.

I was wondering what suggestions those of you who do this daily had to offer?

One of my favourite techniques for de-soldering, where the PCB will
allow it, is to heat up the solder joint and then quickly give the PCB
a sharp rap on the edge of the bench. Inertia then causes the hot
solder to keep going and which then leaves the component lead free to
be removed.

--
Regards,
Harry (M1BYT) (L)
http://www.ukradioamateur.co.uk

In article ,
Alan wrote:
Surface mount devices are often glued to the pcb

Not in my experience. Solder is the only thing holding the component on
the board on 99.99% of commercial gear.

Not so. Most are glued in place as part of the manufacturing process then
flow soldered.

--
*I couldn't repair your brakes, so I made your horn louder *

Dave Plowman London SW
To e-mail, change noise into sound.

Harry Bloomfield wrote:
John Rumm brought next idea :
Initially this was supposed to be due to a supply of dodgy caps made
with a stolen (but incomplete) electrolyte formula. However the
problems seem to have endured far longer than that alone would explain!

It does seem to be any application in high ripple conditions they seem
to go tits up first. So SMPSUs and voltage regulation applications
seem particularly prone. I always try and replace them with low ESR
caps designed for this application, and with a 10K hours life rather
than the cheaper 2K ones.

I agree entirely...

In the past couple of years, I've had three failures which I've actually
been bothered to repair - all turned out to be failed electrolytic caps.
One the was on the two year old washing machine micro controller board.
Second was sat receiver 13 month old. Third was the PWM speed control
for my cars heater fan, a tiny bi-polar electrolytic in the actual power
switching module.

At one time, when repairing not very old equipment, the electrolytic
would be the last things I would be checking for faults. Now it is the
first thing I look to.

They always are the first thing on old valve equipment, after you have
checked all the valves.

It happens that The Natural Philosopher formulated :
Harry Bloomfield wrote:
John Rumm brought next idea :
Initially this was supposed to be due to a supply of dodgy caps made with
a stolen (but incomplete) electrolyte formula. However the problems seem
to have endured far longer than that alone would explain!

It does seem to be any application in high ripple conditions they seem to
go tits up first. So SMPSUs and voltage regulation applications seem
particularly prone. I always try and replace them with low ESR caps
designed for this application, and with a 10K hours life rather than the
cheaper 2K ones.

I agree entirely...

In the past couple of years, I've had three failures which I've actually
been bothered to repair - all turned out to be failed electrolytic caps.
One the was on the two year old washing machine micro controller board.
Second was sat receiver 13 month old. Third was the PWM speed control for
my cars heater fan, a tiny bi-polar electrolytic in the actual power
switching module.

At one time, when repairing not very old equipment, the electrolytic would
be the last things I would be checking for faults. Now it is the first
thing I look to.

They always are the first thing on old valve equipment, after you have
checked all the valves.

I agree, but this is fairly recent modern equipment.

--
Regards,
Harry (M1BYT) (L)
http://www.ukradioamateur.co.uk

On Jan 13, 6:37*pm, John Rumm wrote:
I was toying with doing a wiki article on component level repair of
electronic stuff, since there seems to be so much otherwise decent kit
that dies these days for silly reasons like failing capacitors etc.

With modern lead free solder, and ever increasing layout densities, use
of multilayer boards etc, the job of component removal seems to be
getting ever harder. There was a time when a moderate iron with enough
bit mass to heat the joint quickly, was all you needed. To reflow the
joint and then either pull the device, or apply a solder sucker, all one
did was tin the bit, apply heat for a couple of secs, and job done.

Recently I have needed to recap a few motherboards (not usually an
economic exercise - but these were custom form factor jobbies where
replacements are not readily available), and they were a right PITA to
work on. Obviously multilayer, and since the caps are on the power
regulation sections, probably connected to fairly heavy power plane
traces in the board. Component removal was very difficult - often taking
excessive time for the joint to reflow. Tight pins in close holes with a
small annulus. However cleaning the holes for reuse proved impossible
with any of my normal irons. In the end I had to resort to using a hot
air paint stripper with small nozzle on the end to get enough heat into
the area to be able to suck the holes clear!

I was wondering what suggestions those of you who do this daily had to
offer?

--
Cheers,

John.

/================================================== ===============\
| * * * * *Internode Ltd - *http://www.internode.co.uk* * * * * *|
|-----------------------------------------------------------------|
| * * * *John Rumm - john(at)internode(dot)co(dot)uk * * * * * * *|
\================================================= ================/

Desoldering braid. Like co-ax braid but loaded with rosin (although I
think there is also a plain braid).
Flux, either a bottle of wet for use with a cocktail stick or artists
brush or a pen type.
Mechanical pencil. Clears the wet hole.
Tip cleaner, wetting and fluxing. The greater the cleaned surface
area of the tip, the more solder it will draw.
Bulb type sucker, which can be reserved for blowing. They don't have
the draw capacity of the sprung syringe but they are easier and
quicker to place because the hand may be stabilised by resting upon
the board.
Scrapers and point tools, to remove dry blobs.
I also like to use a gas iron which seems to have a higher heating
rate on a small bit, and I can switch hands without bothering with a
wire or getting my snozzle over the work.

On Jan 14, 5:20*am, John Rumm wrote:
On 14/01/2012 03:32, NT wrote:



On Jan 14, 12:21 am, *wrote:
In , John
*writes

I was toying with doing a wiki article on component level repair of
electronic stuff, since there seems to be so much otherwise decent kit
that dies these days for silly reasons like failing capacitors etc.

A few random thoughts - more to follow

Repair would entail correct diagnosis of the fault

exactly, and thats where ukdiy often falls down, and beginners cant
generally do it on their own.

I don't think that anyone is suggesting detailed fault finding
techniques for digital electronics.

any fault finding for any electronics would be a move forward for the
beginner

(which while not actually a massively difficult task with the right
equipment, its very easy to find yourself twiddling knobs on several
hundred grands worth of it in the process, which tends to put much of it
beyond most folks DIY budgets)

At the simplest level, a hell of allot of kit can be saved just by
spotting the caps that are bulging and spewing electrolyte all over the
place. If you want to get a bit deeper, then you can do plenty with a
multimeter, and even relatively specialist bits of kits like LCR
component analysers or ESR meters are not that expensive.

(I don't agree that ukdiy falls down on diagnosis either particularly *-
we have some very good engineers here with plenty of experience in
relevant fields. That does not mean that all contributions or all advice
will be spot on, but its no reason to write it off either)

A problem I see here too often is a series of people simply taking
wild guesses. Its an approach that simply doesnt work often when
you're faced with an electronic system using 100s of parts.

For beginners to electronic repair I'd suggest
1. getting a multimeter
2. understanding that both power and signals flow through the circuit
stage by stage
3. Using the meter to spot where power or signal fails to be there,
thus narrowing the fault down to a specific small area of circuit.,
usually. I say usually as eg CRT TV PSUs can be extremely interlinked
with other parts of the tv.


NT

plated through holes, once the component lead is removed, if you then
reapply solder to the joint and can clean the hole out better with a
solder sucker

you can, but I've never seen teh point. Just heat the hole and push
the component lead through.

That works for many boards, but not all. In some cases a typical iron
delicate enough for the size of task can't supply the heat on a
multilayer board. That's where the hot air systems work well, either to
temper an area of board to give the iron a chance, or more typically on
their own just to reflow the obscured hole so it can be sucked clean.

On 14/01/2012 09:33, Alan wrote:
In message , geoff
wrote

Surface mount devices are often glued to the pcb

Not in my experience. Solder is the only thing holding the component on
the board on 99.99% of commercial gear.

Its slightly less common than it was, since solder paste is now almost
universally used, and that tends to hold components in place until
soldering. On the earlier surface mount stuff where it was flow soldered
(usually on the "chip wave" rather than in the main bath) they had to be
glued fairly well else they would simply fall off the board once it was
inverted and fed through the wave.

There may be board coatings on boards for the military market which make
rework almost impossible for those with out the right tools.

Conformal coatings don't always make it much harder to remove and
replace stuff (although some are a right pig!), but it can be a pain
getting the coating back in place properly.

The only time I've seen glue used is when a physically very large
component has been fitted and the manufacturer has used a "blob" of glue
to bond to component case to the board to prevent differential physical
movement - to help prevent vibration failures.

Wogging as GEC used to call it...

(real vibration situations would usually call for stuff being tied down,
or clamped in place. Often with holes designed into the boards to make
that possible - or sometimes by fixing to the heat ladder when there is one)

--
Cheers,

John.

/================================================== ===============\
| Internode Ltd - http://www.internode.co.uk |
|-----------------------------------------------------------------|
| John Rumm - john(at)internode(dot)co(dot)uk |
\================================================= ================/

On 14/01/2012 12:42, NT wrote:
On Jan 14, 5:20 am, John wrote:
On 14/01/2012 03:32, NT wrote:



On Jan 14, 12:21 am, wrote:
In , John
writes

I was toying with doing a wiki article on component level repair of
electronic stuff, since there seems to be so much otherwise decent kit
that dies these days for silly reasons like failing capacitors etc.

A few random thoughts - more to follow

Repair would entail correct diagnosis of the fault

exactly, and thats where ukdiy often falls down, and beginners cant
generally do it on their own.

I don't think that anyone is suggesting detailed fault finding
techniques for digital electronics.

any fault finding for any electronics would be a move forward for the
beginner

Perhaps there are two articles here - one on fault finding and one on
service technique?

(which while not actually a massively difficult task with the right
equipment, its very easy to find yourself twiddling knobs on several
hundred grands worth of it in the process, which tends to put much of it
beyond most folks DIY budgets)

At the simplest level, a hell of allot of kit can be saved just by
spotting the caps that are bulging and spewing electrolyte all over the
place. If you want to get a bit deeper, then you can do plenty with a
multimeter, and even relatively specialist bits of kits like LCR
component analysers or ESR meters are not that expensive.

(I don't agree that ukdiy falls down on diagnosis either particularly -
we have some very good engineers here with plenty of experience in
relevant fields. That does not mean that all contributions or all advice
will be spot on, but its no reason to write it off either)

A problem I see here too often is a series of people simply taking
wild guesses. Its an approach that simply doesnt work often when
you're faced with an electronic system using 100s of parts.

Perhaps not, but we are talking about a guide to method and technique
here, so that does not need to apply!

For beginners to electronic repair I'd suggest
1. getting a multimeter
2. understanding that both power and signals flow through the circuit
stage by stage
3. Using the meter to spot where power or signal fails to be there,
thus narrowing the fault down to a specific small area of circuit.,
usually. I say usually as eg CRT TV PSUs can be extremely interlinked
with other parts of the tv.

Indeed, and probably a good example of kit one ought not encourage the
inexperienced to go poking about in since they can bite...



--
Cheers,

John.

/================================================== ===============\
| Internode Ltd - http://www.internode.co.uk |
|-----------------------------------------------------------------|
| John Rumm - john(at)internode(dot)co(dot)uk |
\================================================= ================/

On 14/01/2012 12:17, thirty-six wrote:
On Jan 13, 6:37 pm, John wrote:

I was wondering what suggestions those of you who do this daily had to
offer?

Desoldering braid. Like co-ax braid but loaded with rosin (although I
think there is also a plain braid).

I never used to be a fan of braid in the exclusively through hole days,
but its certainly quite handy on SM boards...

Flux, either a bottle of wet for use with a cocktail stick or artists
brush or a pen type.
Mechanical pencil. Clears the wet hole.
Tip cleaner, wetting and fluxing. The greater the cleaned surface
area of the tip, the more solder it will draw.
Bulb type sucker, which can be reserved for blowing. They don't have
the draw capacity of the sprung syringe but they are easier and
quicker to place because the hand may be stabilised by resting upon
the board.

Not as common as they once were... The point about lower height is a
good one though. I tend to use an Antex Mini de-soldering pump for that
very reason - its easier to use with the hand resting on the board.

Scrapers and point tools, to remove dry blobs.
I also like to use a gas iron which seems to have a higher heating
rate on a small bit, and I can switch hands without bothering with a
wire or getting my snozzle over the work.

Yup, they can be good... (although the last really difficult board I
had, even that would not re-flow the joint!)


--
Cheers,

John.

/================================================== ===============\
| Internode Ltd - http://www.internode.co.uk |
|-----------------------------------------------------------------|
| John Rumm - john(at)internode(dot)co(dot)uk |
\================================================= ================/

On 14/01/2012 01:52, geoff wrote:
In message , John
Rumm writes
On 13/01/2012 23:47, Theo Markettos wrote:

John wrote:
I was wondering what suggestions those of you who do this daily had to
offer?

Hot air. It makes it hugely easier. There are relatively cheap hot air
stations around... go to certain backstreets of cities in the developing
world and you can see people doing component level service on recent
TVs etc
there.


Not tried but, for example:

http://cpc.farnell.com/duratool/zd-9...md-rework-stat
ion/dp/SD01139
(I'm not sure I believe the listed temperature of '150' there)

No that does sound implausibly low... would not even melt leaded
solder. Not a bad price mind you given you can get a reasonable range
of nozzles for it.

I now have three of them, they are hot enough, they work. I have had a
couple of element burnouts, though. You only need the small (4mm?)
nozzle for normal work

Element burnouts in months of "all day" use, or early failures?



--
Cheers,

John.

/================================================== ===============\
| Internode Ltd - http://www.internode.co.uk |
|-----------------------------------------------------------------|
| John Rumm - john(at)internode(dot)co(dot)uk |
\================================================= ================/

I recall that in the olden days (if films were to be believed) if you were
stranded on a desert island with a radio set then you could convert it to a
transmitter!

Seriously, we are getting less able to repair or make items that can
interface with modern systems.