Alright, so here we’re looking at an A600
motherboard. Now when it first came in, it looked like
this. So what has happened here is that C303 has
leaked a bit, causing some corrosion in this area. And also, the owner of this board has made
an attempt at repairing this by himself, only to end up making the problem even worse. We can see here that the pad for the negative
terminal of this capacitor has been ripped off the board,
and the pad for the positive terminal has completely deattached from the PCB substrate. So yet again, this will be a repair where
I have to rectify someone else’s mistake. Oh God! *sigh* The owner of the board has also put non-polarized
capacitors in here for the audio coupling. That is just wrong, I will fix that up later.
But the main focus in this video will be on cleaning the corrosion off the board, and
also repairing the pads for capacitor C303. Now actually I’m already done with the repair
at the time of recording this narration, but in this video, I will show you how this
motherboard went from this… to this… … and then finally … to this. So here is how this repair was done. I started to prepare the board by first putting
it on my preheater and then putting some aluminium tape on there.
After cleaning the area with isopropyl alcohol, I then fluxed the components within the corroded
area using some liquid rosin flux. After letting the board preheat for a little
while, I then used auxillary heating to make sure
the flux activates properly during the soak time. Note here that I’m using a temperature of the overhang which is much lower than reflow
temperature. We can here see that the flux is starting
to activate. I then increased the temperature of the overhang
to reflow temperature, and started to remove components one by one. Note here that this procedure should not be
done without a preheater. You can not just take a hot air rework station
and start spot-heating the board all the way up to reflow temperature without any preheating
at all. Doing such a thing can cause all sort of problems
because of the thermal expansion properties of the board.
We are talking about things like board warping, microcracks in the PCB substrate, increased
risk of board scorching etc etc etc. Since this preheater doesn’t have a cooling
function I usually like to use an external fan for that instead. Okay, now this might look a little bit messy
perhaps, but this is actually how a rosin-based liquid flux usually looks like after it has
activated in a corroded area. There is no way around it unfortunately. But
it can easily be cleaned off with isopropyl alcohol, so really nothing to worry about.
But hey, let’s stop the video here for a moment because I want to show you something. Okay, so let’s have a look at this area right
here. Do you see that the green PCB has a certain
orange tint to it? That is the flux. All the flux has activated, but this area is so far
from the capacitor that is wasn’t really affected
by corrosion that much. That is also why the flux has not taken on a darker color here
– It hasn’t mixed with the contaminants from the corrosion.
So when our little friend here, C303 (the bastard!) started to leak onto the PCB, it’s
natural that the most nearby components were the ones taking most of the crap.
So in other words, the blackened pattern here that we can find is exactly what is to be
expected considering the circumstances. Oh no! My stomach! [Buhwahuaaaa!] Wow, that guy sure likes to puke, doesn’t
he? So, if we compare these two images, we can
clearly see the correlation between them. Here for example, we can easily see that the
LF347 is heavily corroded. In fact, if we take a look at it through a microscope while
it’s out of circuit, it becomes even more obvious.
I mean, look at all that crap. It looks absolutely horrible! Of course, reusing this component
is completely out of the question. Now I have to say that I think it can be quite
interesting sometimes to take pictures before and after every step and then go back and
compare them on a computer because, I think it’s possible to learn a few things
from that actually, so that can be quite interesting sometimes.
Anyway, let’s continue with the repair. Okay, now at this point, it’s very tempting
to start cleaning away the flux residues, especially considering that this area is now
quite messy. And yes, of course, a good rule of thumb is
to clean the area between every step of the repair.
But before I clean the area, I will be sure to first remove the solder from the pads with
solder wick. Let me explain why. The solder that is left on the pads after
component removal can take on different shapes, and in some cases, the remaining solder on
a pad might look a little bit like a pyramid. This is due to the cohesive forces of the
solder alloy during component removal. Here I have drawn in an exaggerated example
of that, where the solder almost has the shape of a hook.
If a cotton bud is now used to clean the area, there is a possibility that the fibers of
the cotton bud will hook onto the solder that is left on the pad.
Since we are making sweeping motions with the cotton bud in order to clean the area,
there is a certain risk that the cotton bud will then rip the pad straight off the board.
In order to avoid this from happening, the remaining solder is removed before any cleaning
takes place. I would like to point out though that the
choice of removing solder before cleaning the board is purely based on personal experience.
It is not coming off an IPC training manual or something like that.
And of course, whether you choose to clean the board or not before removing solder from
the pads is completely up to you, but at least now you know about this little risk factor.
And this is how I prefer to do it, because after all, my job is not to rip pads off the
board – What I do is to repair them. Now it’s time to do some cleaning. It’s always a good idea to try to keep the
rework to a minimum, but here I felt that I needed to add some more flux and go over
the pads just one more time with some clean solder wick,
in order to get the last corrosion off of there. And of course now, there are flux residues
on the board again, so we need to get that off. And in case there is still any corrosion left
in this area, this should pretty much take care of that.
Here I’m using an ESD-safe brush together with some isopropyl alcohol. A final clean up with a lint free cleaning
foam swab will ensure that there are no cotton fibers left on the board. And here we have the board looking nice and
clean. Now you might also notice that the pad for
the positive terminal of C303 is not there anymore. That’s because I removed it. Let
me explain why. So here is how the pad looked like before
I removed it. We can clearly see that it has completely lifted from the PCB substrate.
Now at this stage, we have two options. We could either try reusing the pad by gluing
it back onto the PCB, or we could remove it and install a new one.
So which one of those two alternatives should we go for here? Well, let’s evaluate the situation.
Just by looking at this pad, we can easily see that it’s terribly corroded.
The pad is going to be attached to the PCB through under bonding, that is, a high temperature
resistant epoxy is going to be used to glue the pad into place.
On the upper side of the pad, a component is going to be soldered into place. These
steps can not be performed with reliable results when a pad has this much corrosion on it.
In fact, the IPC recommended procedure when dealing with damaged pads is to remove the
pad from the PCB and install a new one. An exacto knife is then used to cut the trace,
and the pad is removed from the board. However, in this case, the connecting trace
was so heavily corroded that it snapped straight off as soon as I touched the pad. Actually, the owner of the board brought the
pad for the negative terminal with him. So why don’t we take a look at these two pads
through the microscope to see how they look like?
Let’s start with the negative terminal. Okay, so this doesn’t look that good, now
does it? Alright, so let’s move on to the other one,
which is the one I removed. Wow, this one looks even worse than the other
one! I think the orange stuff on there might be flux, but look at all the oxidation. And
also, take a look at the trace where it snapped off.
See how corroded that is? No wonder it came off that easily!
Now, let’s go back to the other pad again. How does it look like on the bottom side?
So, okay, let’s flip it over and have a peek. Go figure… Kind of looks like it has been
thrown into a 5000 degree furnace! Anyway, let’s take a look at the bottom side
of the other one as well. Yeah… So if I had just glued this pad back
into place, that would have turned out to be a perfectly reliable repair, wouldn’t it?
Umm… no. Here I’m using a metal screwdriver to remove
any remaining adhesive and contaminants from the PCB substrate. The site is cleaned with isopropyl alcohol
to get rid of any remaining debris. A dental pick is used to scrape some of the
solder mask off of these traces. Although it’s not necessary, compressed air
can be used to make the alcohol dry up a little faster. The old pads are used as a template to create
new pads from a 1.4 mil copper foil. Copper can oxidize over time when exposed
to oxygen. To ensure reliable results, I am here using
a dental pick to scrape off any contaminants from the connecting trace on the new pad. The new pads are put on the board to verify
that they are of the correct size. According to proper procedures, the length
of the overlap should be at least twice the width of the trace. The traces going to the pads are cleaned with
isopropyl alcohol, and then fluxed in preparation for tinning. The traces are then being pretinned before
the new pads are installed. Flux residues are then removed from the site. At this point, I decided to fix up some of
the traces before continuing with the pad repair. And it looks pretty good. No problem here. Okay, so I think I will have to fix this trace
also. And there we are. No problem. Kapton tape is here used to protect the board
during bonding of the new pads. The epoxy I’m using for under bonding of the
pads is the CW2500 by CircuitWorks. After being fully cured, it can withstand
temperatures up to 315 degrees Celsius for 60 seconds, which makes it the highest temperature
resistant epoxy I have ever seen on the market. This makes it very suitable for this kind
of repair, since the capacitor can then later be soldered onto the new pads without losing
much of the bonding strength. This epoxy is ideal for solder mask repair
according to the datasheet and meets the requirements of IPC-7721 procedure 2.4.1, but I don’t see
any reason why it can’t be used for under bonding as well. A lint free foam swab is ideal for creating
a thin film of epoxy at the site of the pads. Since this epoxy has a tack free time of 30
minutes, there is plenty of time to also repair the solder mask of these traces before the
epoxy starts to cure. The pads are carefully put on the board and
aligned into place. Kapton tape is then put on top of the pads. I then cured the epoxy according to the manufacturer’s
recommendations. After this, I didn’t do anything more with
the board for the next 12 hours or so, because I wanted to make sure the epoxy was fully
cured before proceeding with the repair. The next day, I then started off cleaning
the traces for these pads. And I then lap soldered the traces of the
pads together with the traces on the board. As you can see, a piece of kapton tape is
put on the pads here. That is to prevent the solder from reflowing onto the pads themselves.
The pads are now electrically connected to the board. So I cleaned off the flux residues
and then started to remove the kapton tape, by carefully peeling it off sideways. I then cleaned again so that there wouldn’t
be any adhesive material from the kapton tape left on the pads. I’m here checking for continuity to verify
that the pads are connected to the traces on the board. The lap solder joint now has to be over bonded
in order to prevent it from reflowing when the new capacitor is installed.
A piece of kapton tape is again placed over the pads so that no epoxy accidently gets
on the actual pads. Here I am again using CW2500 but this time
for over bonding of the lap solder joints. Excess epoxy is removed. And I’m again being careful when removing
the kapton tape, making sure to peel towards the side. Once again, I cured the epoxy according to
the manufacturer’s recommendations, and then I let the motherboard sit over the night. The pads are now repaired and the board is
ready to accept new components again. New components were installed using standard
soldering procedures, including cleaning, fluxing and soldering.
You might note here that I’m using a very short dwell time when soldering at the site
of C303. And also, I have the iron set to a slightly
lower temperature than I normally would for this kind of surface mount soldering. By the way, the smaller electrolytic capacitor
to the left of the other two, C612, is a common cause of failure on these Amiga 600 motherboards.
That is because this capacitor is part of a timing circuit consisting of a 555 timer,
and the output of this circuit can pull the system reset line low.
If C612 has shorted out, the reset line will constantly be held low and the computer will
never start up. Since this is a common problem on Amiga 600
motherboards, and since I don’t know when this capacitor was last replaced, I decided
to install a new capacitor in there while I was at it.
Because after all, it’s much better that I do that now, rather than this capacitor failing
in a year or so, which then leads to that someone with inadequate soldering skills tries
to repair it, only to end up ripping the pads off the board.
So therefore, C612 has now also been replaced with a new one, without lifting or damaging
any of the pads, of course. Upon inspection, I here realized that I forgot
to repair the solder mask on this trace while I was working with the epoxy.
But that is no problem, because I can quite easily add some coating onto that trace using
the CW3300 overcoat pen. A final continuity check is a standard procedure
for this kind of repair. There is absolutely no need to have non-polarized
capacitors in there. That is just stupid. There are in fact some downsides to having
non-polarized capacitors in there like this, but that topic is obviously outside the scope
of this video. Okay, there we go. No burn marks, no lifted
pads etc etc etc. The repair of this motherboard is now complete.
So, without using any destuctive methods and without damaging the board during the repair
process, this board has now been repaired using controlled
temperatures and repair techniques. All that is left to do now is to test the
board to verify that it works. Well, I actually don’t really need to do that,
because I already know the board will work fine since I’m the one who repaired it, but
I will still test it quickly just as a standard procedure. I would like to conclude this video with a
little piece of advice. As you know, I had to go through the trouble
of doing a pad repair on these two pads in order to repair this board.
However, if the owner of the board hadn’t attempted to fix it up by himself, it’s very
much likely that the whole step of having to repair the pads could have been avoided. This would have meant less work time for me
and less expenses for various repair material. So of course, what it all comes down to is
that I will now have to ask for more compensation for this work than I probably
would have had if the board hadn’t undergone such a destructive repair attempt prior to
ending up on my bench. So don’t overestimate your abilities. If you
don’t have the experience and skill to perform a repair correctly, leave the board to someone
who can. Because if you don’t, not only do you risk
doing more harm than good, but you might also end up having to pay more than you normally
would when hiring someone else to do the job for you. One more thing… The owner of this board
told me that he tried to use the infamous technique of using a pair of pliers to forcefully
twist and pull the capacitor off the board. Sadly, I know that there are people out there
who are big fans of this “spectacular” technique and actually believe that this is a good method
for removing electrolytic capacitors from a PCB. And to those people, I only have one thing to say: Get a clue, will ya! Because if you
do get a clue, you won’t have to use your glue. Alright, so that’s gonna do it for now.
I hope you will like the video if you liked the video. See you later!