en-AU.vtt

WEBVTT
Kind: captions
Language: en

00:00:00.001 --> 00:00:01.000
English (AU) (Spoken) [Manually Transcribed Captions]
github.com/WizardTim/WizardTim-captions

00:00:01.000 --> 00:00:08.640
So in my last video we had a look at phenolic PCBs 
and how bad they are at resisting moisture and  

00:00:08.640 --> 00:00:14.160
high temperatures when reflowing, if you haven't 
seen that video I'll leave it in a card here.

00:00:15.040 --> 00:00:21.280
But what about FR-4 PCBs that we're probably 
more familiar with if you've ordered them  

00:00:21.280 --> 00:00:26.400
from the various online manufacturers you 
might have received them packed in this  

00:00:27.280 --> 00:00:32.400
sealed packaging with a silica gel so this would 
lead you to believe that these are actually  

00:00:32.400 --> 00:00:38.720
quite moisture sensitive, but in that video I did 
say that FR-4 PCBs are pretty much better in every  

00:00:38.720 --> 00:00:46.867
single way both in temperature resistance and 
moisture sensitivity so I thought I'd make this video and put that to the test. 

00:00:46.867 --> 00:00:52.080
So here I have a 
FR-4 PCB from a online PCB manufacturer and we're  

00:00:52.080 --> 00:00:58.400
going to be putting it to the test at 300 °C 
and see how it holds up against the phenolic PCB.

00:01:07.840 --> 00:01:12.880
So having a look at the aftermath 
it's pretty difficult to see but  

00:01:12.880 --> 00:01:18.160
there is some slight discoloration, you 
can see some of the pads are oxidized.

00:01:20.880 --> 00:01:25.000
To make it easier to see I've 
upped the contrast in this frame ,

00:01:25.040 --> 00:01:30.400
so you can really see the dark area is the 
area that's been discolored by the high temperature,

00:01:30.400 --> 00:01:35.000
but it's it's only very minor in 
comparison to the phenolic PCB. 

00:01:35.000 --> 00:01:42.560
So seeing as 300 °C didn't do it let's try 500 °C and 
that's the highest my little heat gun can go.

00:01:45.360 --> 00:01:49.360
And as we see it pretty 
quickly starts to degrade

00:01:49.360 --> 00:01:54.080
and discolor, I'm not quite sure 
what this would be referred to as.

00:01:56.400 --> 00:02:01.520
So jabbing tweezers into it we can see it's 
pretty much lost all it's structural integrity and  

00:02:01.520 --> 00:02:08.160
we can see a lot of the fibers in the fiberglass 
are just pulling apart and these tracks aren't  

00:02:08.160 --> 00:02:13.840
held on anymore at all so pretty much if you're 
if your PCB gets to 500 °C you're screwed  

00:02:13.840 --> 00:02:19.120
but at that point you've got bigger problems. So 
looking at it we can actually see the actual  

00:02:19.120 --> 00:02:26.560
fiberglass weave under all of this burnt epoxy 
and here we have the individual fiberglass strands  

00:02:27.120 --> 00:02:33.440
interestingly the epoxy is completely gone you 
can see the individual strands there's no epoxy  

00:02:33.440 --> 00:02:37.138
holding it together anymore it's all burnt away.

00:02:37.138 --> 00:02:41.665
So we had a look at both 300 °C and 500 °C 

00:02:41.665 --> 00:02:48.800
and in both cases there wasn't any damage due 
to moisture like we saw on the phenolic PCB  

00:02:48.800 --> 00:02:55.840
so here I have soaked this FR-4 PCB in water so we 
should be able to get some pretty good action here  

00:02:56.560 --> 00:03:00.960
now looking at the side of the PCB you can 
see just how far the water has wicked up  

00:03:01.600 --> 00:03:08.240
now this PCB has been in here for about 
a week so it's pretty much soaked full of water.

00:03:09.440 --> 00:03:12.880
So here I've got it set to 
350 °C just so that  

00:03:12.880 --> 00:03:17.840
we can definitely see some action 
from this PCB and if we look at it.

00:03:30.080 --> 00:03:36.240
It's all going nice and then suddenly it oh yeah 
it just completely delaminates and forms this  

00:03:36.240 --> 00:03:42.905
massive bubble and we can see that the epoxy 
is starting to burn as well because we're at such a high temperature. 

00:03:43.764 --> 00:03:47.680
Now trying it again 
on the other side but this time with a normal  

00:03:47.680 --> 00:03:54.080
temperature of 260 °C we can see 
that it's resisting it a fair bit more

00:03:59.600 --> 00:04:04.560
but here we have exactly the same problem of 
the layers delaminating and it forming a nice...  

00:04:05.120 --> 00:04:09.520
well a terrible bubble in this case we 
definitely never want to see this but  

00:04:09.520 --> 00:04:14.080
this is exactly what I want to see because it 
proves the point that they are moisture sensitive  

00:04:14.080 --> 00:04:20.560
but nowhere near as bad as phenolic PCBs and if 
I take my tweezers just to show you that  

00:04:20.560 --> 00:04:27.840
this bubble is certainly pressurized with 
some sort of gas presumably water vapor.

00:04:32.400 --> 00:04:38.720
So I've let it cool down now and you can see the 
outline of where that bubble was and the where the  

00:04:38.720 --> 00:04:46.216
layers have delaminated so yeah all this
area here is where it stopped interestingly the vias,

00:04:46.216 --> 00:04:50.960
interestingly here the vias appear 
to have actually stopped the delamination  

00:04:51.760 --> 00:04:59.552
which is rather interesting it might be a good 
little trick to use vias to hold the layers together better 

00:04:59.552 --> 00:05:07.601
I know that you can often use 
vias to ensure pads adhere better to the PCB so they don't fall off.

00:05:07.601 --> 00:05:11.840
So I'm very interested to 
see just what's underneath this bubble so  

00:05:11.840 --> 00:05:16.800
we're going to do a destructive teardown and 
have a look at the individual layers so I'm  

00:05:16.800 --> 00:05:22.320
just going to cut it here with this X-ACTO knife, 
be careful with an X-ACTO knife if you do this  

00:05:23.600 --> 00:05:29.200
and now that we've cut the top copper layer 
we can just pry it up just like that and we  

00:05:29.200 --> 00:05:37.840
can see that we've just took the top copper 
layer off it's delaminated from the fiberglass.

00:05:48.960 --> 00:05:55.120
Interestingly we can see the pattern of the 
epoxy on the copper so it looks like when we  

00:05:55.120 --> 00:06:00.960
peeled up the copper actually took the epoxy with 
it although on the edge it didn't and this layer  

00:06:00.960 --> 00:06:07.600
of fiberglass is still quite loose so I'm going to 
cut away further at this to see if we can expose  

00:06:07.600 --> 00:06:13.360
the lower layers so cutting it with that X-ACTO 
knife again and see here I can actually put the  

00:06:13.360 --> 00:06:19.840
X-ACTO knife in underneath that fiberglass layer 
so again this layer here is completely delaminated  

00:06:20.480 --> 00:06:26.960
and yeah I can just pull this up with some 
tweezers it's rather rigid but yeah it's  

00:06:26.960 --> 00:06:32.480
done a fair bit of damage to this although 
interestingly this tab hasn't delaminated as  

00:06:32.480 --> 00:06:38.560
much as the other one so I'm not sure if that 
top layer delaminating was just a result of me

00:06:40.640 --> 00:06:45.267
so yeah just jamming the tweezers in and 
pulling it up works pretty good

00:06:45.267 --> 00:06:52.240
Now just to prove to you that this isn't just a result of 
cutting a normal FR-4 PCB that if you cut it it  

00:06:52.240 --> 00:06:56.720
falls apart like this so I'll just cut it over 
here where I didn't apply the high temperature  

00:06:57.280 --> 00:07:04.320
and as you can see if I cut it, it doesn't 
fall apart at all it's held together quite well  

00:07:05.040 --> 00:07:10.240
yeah I try to get the knife in here but it's just, 
it's not going to come up there's no delamination  

00:07:10.240 --> 00:07:13.641
it's all held together quite well without epoxy.

00:07:13.641 --> 00:07:23.920
So taking a closer look at the macro lens we can really see that the copper has pulled up the epoxy with it, so I'm not sure if this was part of the  

00:07:24.480 --> 00:07:33.520
delamination from the gases because the other 
flaps appeared to still be adhered quite well but  

00:07:33.520 --> 00:07:40.800
in the case of all four flaps the first fiberglass 
layer is definitely completely delaminated from  

00:07:40.800 --> 00:07:45.840
the second fiberglass layer, here you can see 
the difference between where the epoxy has  

00:07:45.840 --> 00:07:50.800
been pulled up by the copper and where it stops 
and the epoxy is still adhered to the fiberglass.

00:07:55.520 --> 00:08:02.880
So I've now pulled back the top copper as well 
as the first fiberglass layer a lot more and  

00:08:02.880 --> 00:08:09.600
we can see actually that the second fiberglass 
layer does actually have an air bubble behind it  

00:08:09.600 --> 00:08:15.840
we can see here the extremity of the bubble 
here and we can see that this area here is  

00:08:16.800 --> 00:08:23.440
is still adhered together and hasn't delaminated 
so i suspect that these lower layers were probably  

00:08:23.440 --> 00:08:28.720
delaminated by the high temperature on the 
other side. In hindsight i probably should  

00:08:28.720 --> 00:08:34.720
have just gotten another PCB and done a second 
test with that without doing it on the other side  

00:08:35.360 --> 00:08:41.152
but I couldn't be bothered to put another one 
of these in a glass of water for a week and let it do its thing.

00:08:41.152 --> 00:08:45.680
So here we can really see that 
this first layer is completely delaminated and  

00:08:45.680 --> 00:08:52.080
you can really see that there's quite a definite 
position of where it's delaminated and you can  

00:08:52.080 --> 00:08:58.320
see that this top layer is actually a smaller 
area than the second layer's delamination so  

00:08:59.280 --> 00:09:05.280
yeah I'm pretty sure that second layer is because 
of the damage from the first test and we can see  

00:09:05.280 --> 00:09:10.480
here that the surface is very smooth for 
where the copper has been pulled off the  

00:09:10.480 --> 00:09:19.120
epoxy that covers the fiberglass and here's just a 
close-up image of that and here's another close-up  

00:09:19.120 --> 00:09:25.200
you can see here to the left the epoxy still has 
the texture from the other layer of fiberglass  

00:09:25.200 --> 00:09:32.400
in it and this bit here still has the top copper 
and the first layer of fiberglass but where I've  

00:09:32.400 --> 00:09:39.040
ripped it you can see that every second strand 
is sticking out which is rather interesting way  

00:09:39.040 --> 00:09:46.745
how it tore and here's another close-up of 
where I've ripped off the top fiberglass layer.

00:09:50.233 --> 00:09:54.880
These are pieces of fiberglass unlike 
the first high temperature test where the  

00:09:54.880 --> 00:10:00.160
epoxy was actually burnt and was degraded 
these still have the epoxy in them so  

00:10:00.160 --> 00:10:07.040
if I poke and prod them with an implement you 
can see that they're still completely rigid  

00:10:07.040 --> 00:10:13.336
unlike before where you could see the individual 
fibers that just splayed apart.

00:10:13.336 --> 00:10:24.816
So that's it for this video so I hope you've learned 
something and as always thanks for watching