en-AU.vtt

WEBVTT
Kind: captions
Language: en

00:00:00.000 --> 00:00:01.000
English (AU) (Spoken) [Manually Transcribed Captions]
github.com/WizardTim/WizardTim-captions

00:00:01.000 --> 00:00:04.400
In the past I've tried to repair
low-cost consumer electronics that often

00:00:04.400 --> 00:00:07.759
use
phenolic resin-bonded PCBs as well as

00:00:07.759 --> 00:00:12.320
chip-on-board style ICs however the first two times that i

00:00:12.320 --> 00:00:15.920
attempted this
the devices unfortunately did not make

00:00:15.920 --> 00:00:20.000
it through their surgery
however in both of these situations all

00:00:20.000 --> 00:00:22.640
that i was doing was just using hot air
to

00:00:22.640 --> 00:00:25.840
replace some surface mount resistors and
capacitors

00:00:25.840 --> 00:00:28.960
in somewhat closer vicinity to the chip-on-board

00:00:28.960 --> 00:00:32.320
but despite this the chip-on-board
started to

00:00:32.320 --> 00:00:36.399
fizz and crackle and pop and then uh
parts of the epoxy would uh

00:00:36.399 --> 00:00:40.399
go missing and the device would stop
working so i have here

00:00:40.399 --> 00:00:44.399
a PCB that we shall be sacrificing for
science and

00:00:44.399 --> 00:00:51.520
this PCB is probably from about 2005
this is from a children's toy where you

00:00:51.520 --> 00:00:55.280
press the buttons on the keypad and it
will make a bunch of sounds

00:00:55.280 --> 00:00:59.520
but the thing that's important to us is
that it uses a chip-on-board IC

00:00:59.520 --> 00:01:02.719
which is under this black blob here and
on the back

00:01:02.720 --> 00:01:07.740
we can see that it's brown which is a
telltale sign that this is a phenolic PCB.

00:01:07.740 --> 00:01:10.479
So i've got the hot air gun set at 260 °C

00:01:10.479 --> 00:01:13.520
which is the sort of
temperature you want to use for

00:01:13.520 --> 00:01:16.880
lead-free soldering
and i've got it pointed at that blob so

00:01:16.880 --> 00:01:21.600
hopefully we can see some of that snap
crackle pop action

00:01:22.300 --> 00:01:23.600
[popping sound]

00:01:24.320 --> 00:01:28.320
yeah and that's the other thing not only
do these chip on boards not lack the

00:01:28.320 --> 00:01:34.000
heat but these are phenolic PCBs the
lamination is nowhere near as good as

00:01:34.000 --> 00:01:36.140
woven fiberglass PCBs

00:01:36.140 --> 00:01:40.080
[cracking & popping sounds]

00:01:42.140 --> 00:01:49.119
and as you can see here yeah it's developed a bit of a bump
where it's delaminated, so the lesson

00:01:49.119 --> 00:01:54.880
here is that if you have to rework a
resin bonded PCB or do anything near

00:01:54.880 --> 00:02:00.320
a chip on board device it's a good idea
to bake the PCB first to remove any

00:02:00.320 --> 00:02:03.200
moisture
as well as try to use a lower

00:02:03.200 --> 00:02:06.640
temperature
however the best course of action is to

00:02:06.640 --> 00:02:11.680
just use a soldering iron
and if you have to desolder a multi-pin

00:02:11.680 --> 00:02:15.360
SMT part instead of using hot air you
can just cut the leads

00:02:15.360 --> 00:02:19.200
and take the part off that way and then
re-solder it by hand with a soldering

00:02:19.200 --> 00:02:22.080
iron
now whilst i have this i want to

00:02:22.080 --> 00:02:26.720
actually take a dig
at this bubble here just to see

00:02:26.720 --> 00:02:30.400
what sort of damage it's done to the thing

00:02:39.040 --> 00:02:42.319
and it turns out this is actually a
pretty good way of getting the blob off

00:02:42.319 --> 00:02:45.360
you can just
get your tweezers under it and rip it

00:02:45.360 --> 00:02:49.599
off and then you can
put the epoxy blob in some solvent for a

00:02:49.600 --> 00:02:54.800
while
and then you can get some die shots like this

00:03:05.560 --> 00:03:07.920
but yeah it looks like uh the first

00:03:07.920 --> 00:03:12.239
layer of this phenolic board has
completely delaminated from all of the

00:03:12.239 --> 00:03:17.200
other layers and it's just made this uh
bubble here and you can really see that

00:03:17.200 --> 00:03:21.200
on the other side
of the pcb so let's do this again and

00:03:21.200 --> 00:03:24.640
see if we can delaminate all of the
layers in this pcb

00:03:24.640 --> 00:03:28.319
and of course i'll get that microphone
in nice and close for you so that you

00:03:28.320 --> 00:03:33.840
can hear that nice horrific snap crackle
pop ASMR

00:03:35.320 --> 00:04:36.980
[crackling & popping sounds]

00:04:39.440 --> 00:04:44.639
so this shot's at 10 times speed and the
hot air gun is set to 300 ° C

00:04:44.639 --> 00:04:49.040
and you can really see the
phenolic PCB discolor

00:04:49.040 --> 00:04:53.120
this is pretty much why you don't want
to use a phenolic PCB for anything apart

00:04:53.120 --> 00:04:56.560
from the
lowest cost consumer garbage because

00:04:56.560 --> 00:05:00.160
fiberglass FR-4
is pretty much so much better in every

00:05:00.160 --> 00:05:04.500
single way apart from cost
and even that is pretty marginal

00:05:04.500 --> 00:05:08.240
So here I'm just gonna
break the PCB in half so we can see all

00:05:08.240 --> 00:05:10.960
of those
layers that have hopefully delaminated

00:05:10.960 --> 00:05:13.840
so yeah it looks like we've delaminated
all of the layers

00:05:13.840 --> 00:05:17.280
it looks like it's made up of three
different distinct

00:05:17.280 --> 00:05:20.880
fiber layers that have been bonded
together

00:05:20.880 --> 00:05:24.800
interestingly the delamination here is
actually pretty extreme there's

00:05:24.800 --> 00:05:28.479
a surprising amount of space in between
these layers that have

00:05:28.479 --> 00:05:32.240
that's been made by the gas so i presume
either

00:05:32.240 --> 00:05:36.960
quite a lot of moisture has been in this
PCB that's vaporized and created quite

00:05:36.960 --> 00:05:40.639
high pressures
either that or maybe the adhesive that

00:05:40.640 --> 00:05:44.400
bonds these layers together
maybe it's decomposing or something

00:05:44.400 --> 00:05:47.460
which is adding to this

00:05:52.280 --> 00:05:57.840
here's part of the PCB that hasn't delaminated and as you can see
the layers are pretty much

00:05:57.840 --> 00:06:01.360
indistinguishable from each other
and then we've got this bit here where

00:06:01.360 --> 00:06:03.919
the layers have just completely splayed
apart

00:06:03.919 --> 00:06:08.000
and the three distinct layers are very
obvious you can also see the very

00:06:08.000 --> 00:06:13.680
thin piece of solder resist on the top
as well as the very thin copper trace

00:06:13.680 --> 00:06:16.720
so yeah i hope this video has taught you
something about

00:06:16.720 --> 00:06:20.960
what you shouldn't do with phenolic PCBs
and chip on board devices

00:06:20.960 --> 00:06:24.319
so hopefully you don't have to go
through the two first time failed

00:06:24.319 --> 00:06:28.479
attempts at this
now i was also going to talk about the

00:06:28.479 --> 00:06:33.280
moisture sensitivity level of
normal SMT components as well in this

00:06:33.280 --> 00:06:35.600
video but i think it's gotten a little
long

00:06:35.600 --> 00:06:38.800
so i'll leave that to a video in the
future

00:06:38.800 --> 00:06:46.800
but as always thanks for watching