Homebrew PCBs

Has anyone had any success with 4 layer pcb's diy?

  If so, what about blind or buried vias?

  Thanks
  Ron

[Non-text portions of this message have been removed]

----- Original Message ----- 

From: "Ron Amundson" <ron_amundson@...>
To: <Homebrew_PCBs@yahoogroups.com>
Sent: Monday, November 24, 2003 6:26 AM
Subject: [Homebrew_PCBs] 4 layer diy


>
>   Has anyone had any success with 4 layer pcb's diy?
>
>   If so, what about blind or buried vias?

DIY PTH is difficult, let alone multilayer boards. Blind and buried vias
only make sense with six layers or more - completely out of the question at
home.

Leon
--
Leon Heller, G1HSM
Email: aqzf13@...
http://www.geocities.com/leon_heller

>  Has anyone had any success with 4 layer pcb's diy?
>
>  If so, what about blind or buried vias?
>
>  Thanks
>  Ron

Sorry for my ignorance, but what is a pcb diy?

As for multilayer - I do this, meanwhile on a regular basis. The main
thing you need is a through plating station. Once you figure out the
process and provided you get the FR4 material in thin enough sheets (I
got mine from Bungard) it's not that diffcult. To give you an idea, to
create a four layer PCB it takes ~5 hours. That's however (still)
without CNC drill (which would cut the time by ~40 minutes) and no
solder stop mask and thining aplied. The latter would add another 30
minutes. However I usually only need prototypes that don't have to
look goog but meet the physical dimensions (and must be working of
course) of the final product.

HTH

Markus

----- Original Message ----- 

From: "Markus Zingg" <m.zingg@...>
To: <Homebrew_PCBs@yahoogroups.com>
Sent: Monday, November 24, 2003 8:29 AM
Subject: Re: [Homebrew_PCBs] 4 layer diy


> >  Has anyone had any success with 4 layer pcb's diy?
> >
> >  If so, what about blind or buried vias?
> >
> >  Thanks
> >  Ron
> 
> Sorry for my ignorance, but what is a pcb diy?

DIY: 'Do It Yourself'. 8-)

Leon
--
Leon Heller, G1HSM
Email: aqzf13@...
http://www.geocities.com/leon_heller

>----- Original Message ----- 
>From: "Markus Zingg" <m.zingg@...>
>To: <Homebrew_PCBs@yahoogroups.com>
>Sent: Monday, November 24, 2003 8:29 AM
>Subject: Re: [Homebrew_PCBs] 4 layer diy
>
>
>> >  Has anyone had any success with 4 layer pcb's diy?
>> >
>> >  If so, what about blind or buried vias?
>> >
>> >  Thanks
>> >  Ron
>> 
>> Sorry for my ignorance, but what is a pcb diy?
>
>DIY: 'Do It Yourself'. 8-)

Now THAT's funny! :-)  I was even considering not replying to the
original poster thinking that it might be some sort of bonding
technology or such !

With regard to multilayer - it's really not that hard. It takes the
propper procedure and is not even so demanding. I will put a page
online showing the process in detail as soon as time permits (I know I
start to sound like a broken record...)

Markus

----- Original Message ----- 

From: "Markus Zingg" <m.zingg@...>
To: <Homebrew_PCBs@yahoogroups.com>
Sent: Monday, November 24, 2003 9:26 AM
Subject: Re: [Homebrew_PCBs] 4 layer diy


> >----- Original Message ----- 
> >From: "Markus Zingg" <m.zingg@...>
> >To: <Homebrew_PCBs@yahoogroups.com>
> >Sent: Monday, November 24, 2003 8:29 AM
> >Subject: Re: [Homebrew_PCBs] 4 layer diy
> >
> >
> >> >  Has anyone had any success with 4 layer pcb's diy?
> >> >
> >> >  If so, what about blind or buried vias?
> >> >
> >> >  Thanks
> >> >  Ron
> >>
> >> Sorry for my ignorance, but what is a pcb diy?
> >
> >DIY: 'Do It Yourself'. 8-)
>
> Now THAT's funny! :-)  I was even considering not replying to the
> original poster thinking that it might be some sort of bonding
> technology or such !

DIY in terms of home improvement is a popular pastime in the UK and we get
*lots* of very boring TV programmes about it (they are very cheap to make).
It seems to be getting popular in France, judging by the numbers of DIY
stores I've seen there recently.

Leon
--
Leon Heller, G1HSM
Email: aqzf13@...
http://www.geocities.com/leon_heller

> DIY PTH is difficult, let alone multilayer boards. Blind and buried vias
> only make sense with six layers or more - completely out of the question
at
> home.

The reason I would suggest 4 layer blind and buried vias is to avoid having
to drill 0.005" holes or smaller at home and then try to DIY PTH. I wonder
how the board houses get decent through hole activation when the holes get
that small. I would imagine the drills have a lot to do with it.

I've been following the Markus' posts for a while. It appears he is really
getting a process down for PTH on 2 layer.

What frustrates me is the time it takes to get a 4 layer proto. If I can run
them at home, even if it takes 30 hours, I'm still way ahead of the
commercial houses who take a minimum of 5 days plus another day for
shipping.

Thanks
Ron

Hi Ron

>I've been following the Markus' posts for a while. It appears he is really
>getting a process down for PTH on 2 layer.

I definately do have trouble with my english - not sure what you try
to tell, but I DO create 4 layer boards at home on a regular basis not
only 2 layer boards. Even 6 layer is not at all out of scope. A 6
layer board would just take 2 more hours to process.

>What frustrates me is the time it takes to get a 4 layer proto. If I can run
>them at home, even if it takes 30 hours, I'm still way ahead of the
>commercial houses who take a minimum of 5 days plus another day for
>shipping.

Welcome to the club. You definately have discovered my primary
motivation to having built my own through plating station. Another
point where I live (Switzerland) are the extreme high prices
boardhouses want if you say want a board within 3 working days. As an
example, my last quote for a 4 layer 200 hole 10x10cm board was
CHF3500 which is about $2K!

In short the process to create a multilayer at home is as follows:

1) Use photoresist laminate - that's a requirement in order to through
hole plate or else the just copper plated holes would loose the
copper.

2) First create the films for the inner and outer layers. Note, they
should be made so as the holes of the pads later on remain covered by
copper. Due to the nature of photoresist laminate they have to be
negataive - but that's not difficult either.

3) If you can't use a CNC drill machine, create one aditional film of
one of the outer most layers (does not matter which one) where the pad
holes are not covered. This will later serve as a helper mask to
manually drill the holes.

4) Put all layers on a stack. Create your desing so as you define two
aditional alignement holes of 3 or 4 millimeters diameter at diagonal
positions outside of the effective PCB area. Drill those two holes so
as you later can insert a shaft? (hope this is the right word for it)
into them. Drill one hole first, insert the shaft, then drill the
second one to make 100% sure all layers will fit together. Cut the
pieces of the layers with some oversize. This does not matter as you
can cut off thereby creating nice sides once you are done.

5) disasemble the stack, laminate the inner layer on both sides,
expose develop and etch it. But DON'T drill any holes yet. Use the two
diagonal alighement holes with the shaft to align the films (note the
films must get those holes also). I create the holes in the film with
a little stamp kind of tool I made with some stainless steel whire
where I drilled a small hole on one end with a center drill. They have
the right angle to create a very sharp ending that can later be used
as the stamp. It's otherwise quite difficult to "drill" holes into
film material precisely enough.

6) put a very thin layer of 2 minute epoxy (devcon or araldit is fine)
on both sides of the inner layer. Apply it to the INNER layer and do
it all in one step. The reason to apply it to the inner layer is to
make sure there is glue in the etched away parts. Then put the stack
together again using the two center shafts. Put as much pressure on
the stack from both endes as you can. I use two wooden blocks on both
endes and built a wooden construction that holds two vises which
actually press the stack togeter. I used two vises to have more
pressing area and power. Of course if you have a press available
that's better. Let it dry for 5 minutes.

7) Laminate one outer layer, expose and develop it with the special
film mentioned in step 3) Of course don't etch it! Use this as the
drilling mask and now drill all holes.

8) Remove the laminate clean the board and run it through the through
hole plating process. The just drilled holes will cross the copper
parts of the inner layer pads at those spots where they have to. As a
result the trhough hole plating process will create connections where
needed in the inner layer. To answer your question about blured /
burried vias - of course by selectively drill the holes before you
glue and eventually some cleaning after glueing you shold be able to
create such beast.

9) from this moment on, the remaining process is identical to
processing a two layer board. If you are about to create a 6 layer
board, you simply would continue with applying glue to the "semi"
inner layer, but would leave out the drilling and through hole plating
until you are finshed with glueing all layers etc. I figure you get
the picture.

10) laminate, expose develop and etch the outer layer - you are done
now.

I created picutres of the wooden construction to press the stack as
well as of the stamping tool mentioned. Again, I currently don't have
the time to put this onto a page. It's really not that complicated
once you have the aligment hole thingy working (which is also not so
complicated). In fact, this technology works so well that I sometimes
even use it to asure alignement of two layer boards.

I hope the above brabble makes some sense to you. Feel free to ask
whatever questions you might have.

Markus

----- Original Message ----- 

From: "Ron Amundson" <ron_amundson@...>
To: <Homebrew_PCBs@yahoogroups.com>
Sent: Monday, November 24, 2003 1:40 PM
Subject: Re: [Homebrew_PCBs] 4 layer diy


>
> > DIY PTH is difficult, let alone multilayer boards. Blind and buried vias
> > only make sense with six layers or more - completely out of the question
> at
> > home.
>
> The reason I would suggest 4 layer blind and buried vias is to avoid
having
> to drill 0.005" holes or smaller at home and then try to DIY PTH. I wonder
> how the board houses get decent through hole activation when the holes get
> that small. I would imagine the drills have a lot to do with it.
>
> I've been following the Markus' posts for a while. It appears he is really
> getting a process down for PTH on 2 layer.
>
> What frustrates me is the time it takes to get a 4 layer proto. If I can
run
> them at home, even if it takes 30 hours, I'm still way ahead of the
> commercial houses who take a minimum of 5 days plus another day for
> shipping.

I've just had some 4-layer proto boards made in 48 hours by PWT in Welwyn,
UK. The price was reasonable - about 500 GBP for three boards and the
quality was excellent. I even got the red solder resist my client wants for
their production boards.

Leon
--76  'p[;
Leon Heller, G1HSM
Email: aqzf13@...
http://www.geocities.com/leon_heller

>I've just had some 4-layer proto boards made in 48 hours by PWT in Welwyn,
>UK. The price was reasonable - about 500 GBP for three boards and the
>quality was excellent. I even got the red solder resist my client wants for
>their production boards.
>
>Leon

I already noticed that it greatly depends on where one lives. For
those interested, 500GBP's are currently equal to ~$850.

However, If you consider that I built my through hole plating station
for roughly $1k (including PCB material to make 50 four layer 10x10cm
boards and chemicals that last for the next 3 years) I still think
that doing them on my own is a better deal (for me) in the mid to
longer run. I also have solder stop mask (ok, green not red one) that
I can laminate and chemicals for thin plating. So, if the need arises,
I can create boards that are looking like made in a board house. 

On the other hand I do understand that it's not everyones thing to go
that far. I just try to point out that to me this decision payed off
already multiple times. I have the station now since this summer and
it already payed off. Aapart from the finacial gain I agree with the
original poster that the biggest win is having the boards quickly. I
out myself as being a person that is making mistakes. In other words I
never manage to design a board that is perfect right from begining. So
I really apreciate the option of doing corrections to the design and
finally have a homebrew PCB that looks and operates like the one I
finally need. Of course, the series production is then made in a board
house. It's just a dam good feeling to be 100% sure that all parts
will fit into the housing, that plugs will fit well, that everything
works like it should etc. etc. Before this I always got very nervous
when I had to order say 1000 PCB's and the timely delay with (over
here) afordable prototype PCB's made in a boardhouse just became a
critical factor. I can't wait three weeks (where I probably would pay
as much as you did for your five days) just to discover that another
change is needed leading to another three weeks etc. etc. I figure
though that this is also all a matter of experinece and well,
abilities. I'm just able to do what I can.... Luky are are those able
to do everything right for the first time!

Markus

Markus Zingg wrote:

Drill those two holes so
> as you later can insert a shaft? (hope this is the right word for it)
> into them. 

Actually the english word for what you describe is called a dowel.


> 6) put a very thin layer of 2 minute epoxy (devcon or araldit is fine)
> on both sides of the inner layer. Apply it to the INNER layer and do
> it all in one step. The reason to apply it to the inner layer is to
> make sure there is glue in the etched away parts. Then put the stack
> together again using the two center shafts. Put as much pressure on
> the stack from both endes as you can. I use two wooden blocks on both
> endes and built a wooden construction that holds two vises which
> actually press the stack togeter. I used two vises to have more
> pressing area and power. Of course if you have a press available
> that's better. Let it dry for 5 minutes.
> 

I have tried 5 minute epoxy on PCB material and it doesn't bond 
anywhere near as well as professional multilayer panels.

Professionals bond layers with a special purpose partly cured 
epoxy sheet called "prepreg". I think they insert the prepreg 
between layers. Then they press at something like 20kg/cm.sq 
pressure and heat to 175\ufffdC for several hours. The heat completes 
the cure of the prepreg layers and it develops a super strong 
bond.  Here is a data sheet of found on the web of some prepreg;
http://www.rogers-corp.com/mwu/pdf/RO4400dsandfab.pdf


Can you wedge the PCB layers apart by forcing a screwdriver in 
the edge ?

The only other step you are missing from professional multilayer 
process is the "epoxy etch back" or "hole desemaer" where after 
drilling there is epoxy smeared on the innerlayers and can 
interfere with the through hole plating, so they put the board in 
a hot solution of potassium permanginate/NaOH which removes a 
small amount of epoxy. I have no idea if this step is necessary 
for your application.

> I've just had some 4-layer proto boards made in 48 hours by PWT in Welwyn,
> UK. The price was reasonable - about 500 GBP for three boards and the
> quality was excellent. I even got the red solder resist my client wants
for
> their production boards.

Thats great service and the price is amazing. Six months ago, I could have
gotten 48 hour service for only a slightly higher fee from a number of
prototype board houses. Today, their workload has increased to the point
that one  week turns are even getting questionable. Every board house I
checked with is having the same problems. If it weren't for US customs
random delays, I'd be pretty tempted to send files to the UK.

 I bought some IC's a while back that were only in stock in Canada. Even
with overnight shipping it still took 6 days for then to arrive.

I guess on the positive side, the economy here in the US is starting up
again. A lot of projects sat dormant and now, everyone, my clients included
want everything right away.

Thanks for the UK tip,
Ron

Hi Adam

>Markus Zingg wrote:
>
>Drill those two holes so
>> as you later can insert a shaft? (hope this is the right word for it)
>> into them. 
>
>Actually the english word for what you describe is called a dowel.

Thanks for letting me know :-)

>> 6) put a very thin layer of 2 minute epoxy (devcon or araldit is fine)
>> on both sides of the inner layer. Apply it to the INNER layer and do
>> it all in one step. The reason to apply it to the inner layer is to
>> make sure there is glue in the etched away parts. Then put the stack
>> together again using the two center shafts. Put as much pressure on
>> the stack from both endes as you can. I use two wooden blocks on both
>> endes and built a wooden construction that holds two vises which
>> actually press the stack togeter. I used two vises to have more
>> pressing area and power. Of course if you have a press available
>> that's better. Let it dry for 5 minutes.
>> 
>
>I have tried 5 minute epoxy on PCB material and it doesn't bond 
>anywhere near as well as professional multilayer panels.

Well, I get reasonably well results. Let me put it this way, by doing
the boards with this method, so far all boards came out well, all of
them are still working today and perfectly served the purpose of
prototypes. So, why should I bother?

>Professionals bond layers with a special purpose partly cured 
>epoxy sheet called "prepreg". I think they insert the prepreg 
>between layers. Then they press at something like 20kg/cm.sq 
>pressure and heat to 175°C for several hours. The heat completes 
>the cure of the prepreg layers and it develops a super strong 
>bond.  Here is a data sheet of found on the web of some prepreg;
>http://www.rogers-corp.com/mwu/pdf/RO4400dsandfab.pdf

I do have prepregs here, but I did not had a press. I first thought
that I have to build/buy/organize such a press, but then thought to
give the glueing technology a try. Again, this works fairly well. The
key point probably is to really only apply a thin layer of glue, and
to do both sides of the inner layer in one step also applying high
pressure.

>Can you wedge the PCB layers apart by forcing a screwdriver in 
>the edge ?

I'm sure that the bonding you get from using  prepregs is better, but
what the heck. It works just fine. I NEVER had a single problem with
false contacts of the inner layer nor did I had a single non working
board as of now. The resulting boards are also stiff etc. I don't have
a reason to improve the process for the purpose I use the boards.
Since the boards were all useable and are in use I honestly must
admitt that I did not had the ocasion yet to destroy one of them. I
think with enough brute force applied a screwdriver could be hammered
between the layers but again, I don't have a reason as of now to do
this.

Btw, the tip to use glue came from Mr. Bungard. He told me that this
"technology" was sucessfully used in a german school so I thought that
if it works there, I should manage the same. :-)

>The only other step you are missing from professional multilayer 
>process is the "epoxy etch back" or "hole desemaer" where after 
>drilling there is epoxy smeared on the innerlayers and can 
>interfere with the through hole plating, so they put the board in 
>a hot solution of potassium permanginate/NaOH which removes a 
>small amount of epoxy. I have no idea if this step is necessary 
>for your application.

I heard of this. I think due to the manually monitored and carefull
handled process of mine - remember, I transport the boards myself from
one bath to the next etc. and due to the fact that I also take the
time needed to carefully drill the holes this is probably not such a
big issue.  Might be that the chemistry I use is so good that it still
works. I simply don't know. All I can say is that so far I never had a
single board that did not ended up just fine.

Markus

http://www.computronics.com.au/rota-spray/

another rotrary sprayer design, good photo how it is constructed.

st

Well, the proof is in the pudding as the saying goes. If you 
have had no problems with delamination or inner layer 
connectivity then it shows your method is ok.
I do recall somewhere that drilling at high speed is more 
likely to cause drill smear. It may be that relatively slow 
manual drilling inadvertedly avoids most of this problem. 
Something to do with heat generated at the drill tip. Higher 
RPM means hotter. I think in production 100k RPM is typical.

After reading all the necessary storage and handing of 
prepreg I knew this stuff was definitely not suitable for 
home use. So yes, liquid epoxy is the obvious answer.

The heating and pressure they use must cause some fusing of 
the epoxy layers. They also talk about oxidizing the copper 
to black oxide to help epoxy bond to the copper. well 
according  http://207.207.205.214/bmr/oxide.htm
They use NaOH/NaCl0 or bleach. Damn, now that's adding yet 
more tanks to the PCB station.

Thanks for sharing your methods. Multilayer sounds exciting, 
but I've got too many other things to finish first.

Adam


Markus Zingg wrote:

> Hi Adam
> 
> 
>>Markus Zingg wrote:
>>
>>Drill those two holes so
>>
>>>as you later can insert a shaft? (hope this is the right word for it)
>>>into them. 
>>
>>Actually the english word for what you describe is called a dowel.
> 
> 
> Thanks for letting me know :-)
> 
> 
>>>6) put a very thin layer of 2 minute epoxy (devcon or araldit is fine)
>>>on both sides of the inner layer. Apply it to the INNER layer and do
>>>it all in one step. The reason to apply it to the inner layer is to
>>>make sure there is glue in the etched away parts. Then put the stack
>>>together again using the two center shafts. Put as much pressure on
>>>the stack from both endes as you can. I use two wooden blocks on both
>>>endes and built a wooden construction that holds two vises which
>>>actually press the stack togeter. I used two vises to have more
>>>pressing area and power. Of course if you have a press available
>>>that's better. Let it dry for 5 minutes.
>>>
>>
>>I have tried 5 minute epoxy on PCB material and it doesn't bond 
>>anywhere near as well as professional multilayer panels.
> 
> 
> Well, I get reasonably well results. Let me put it this way, by doing
> the boards with this method, so far all boards came out well, all of
> them are still working today and perfectly served the purpose of
> prototypes. So, why should I bother?
> 
> 
>>Professionals bond layers with a special purpose partly cured 
>>epoxy sheet called "prepreg". I think they insert the prepreg 
>>between layers. Then they press at something like 20kg/cm.sq 
>>pressure and heat to 175\ufffdC for several hours. The heat completes 
>>the cure of the prepreg layers and it develops a super strong 
>>bond.  Here is a data sheet of found on the web of some prepreg;
>>http://www.rogers-corp.com/mwu/pdf/RO4400dsandfab.pdf
> 
> 
> I do have prepregs here, but I did not had a press. I first thought
> that I have to build/buy/organize such a press, but then thought to
> give the glueing technology a try. Again, this works fairly well. The
> key point probably is to really only apply a thin layer of glue, and
> to do both sides of the inner layer in one step also applying high
> pressure.
> 
> 
>>Can you wedge the PCB layers apart by forcing a screwdriver in 
>>the edge ?
> 
> 
> I'm sure that the bonding you get from using  prepregs is better, but
> what the heck. It works just fine. I NEVER had a single problem with
> false contacts of the inner layer nor did I had a single non working
> board as of now. The resulting boards are also stiff etc. I don't have
> a reason to improve the process for the purpose I use the boards.
> Since the boards were all useable and are in use I honestly must
> admitt that I did not had the ocasion yet to destroy one of them. I
> think with enough brute force applied a screwdriver could be hammered
> between the layers but again, I don't have a reason as of now to do
> this.
> 
> Btw, the tip to use glue came from Mr. Bungard. He told me that this
> "technology" was sucessfully used in a german school so I thought that
> if it works there, I should manage the same. :-)
> 
> 
>>The only other step you are missing from professional multilayer 
>>process is the "epoxy etch back" or "hole desemaer" where after 
>>drilling there is epoxy smeared on the innerlayers and can 
>>interfere with the through hole plating, so they put the board in 
>>a hot solution of potassium permanginate/NaOH which removes a 
>>small amount of epoxy. I have no idea if this step is necessary 
>>for your application.
> 
> 
> I heard of this. I think due to the manually monitored and carefull
> handled process of mine - remember, I transport the boards myself from
> one bath to the next etc. and due to the fact that I also take the
> time needed to carefully drill the holes this is probably not such a
> big issue.  Might be that the chemistry I use is so good that it still
> works. I simply don't know. All I can say is that so far I never had a
> single board that did not ended up just fine.
> 
> Markus
> 
> 
> Be sure to visit the group home and check for new Bookmarks and files:
> http://groups.yahoo.com/group/Homebrew_PCBs
> 
> To unsubscribe from this group, send an email to:
> Homebrew_PCBs-unsubscribe@yahoogroups.com
> 
>  
> 
> Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/ 
> 
>

In article <3FC2AC68.8030804@...>,
   Adam Seychell <adam_seychell@...> wrote:

> I have tried 5 minute epoxy on PCB material and it doesn't bond 
> anywhere near as well as professional multilayer panels.

I think you guys are amazing doing all this. I'm still on double sided
using pre-coated etch resist and no PTH.

The reason I've come in is that my experience is (from just ordinary
gluing jobs) that the 5min stuff isn't anywhere near as good as the
"normal" [OK "regular" for you guys across the pond :-)] Epoxy. I know
this means clamping up for maybe 24 hrs but I think you will find the bond
a lot stronger.

Stuart.

-- 
 __  __  __  __      __ ___   _____________________________________________
|__||__)/ __/  \|\ ||_   |   /
|  ||  \\__/\__/| \||__  |  /...Internet access for all Acorn RISC machines
___________________________/ stuartwinsor@...

101 uses for a Pentium: No2 - A greenhouse heater.

On Tue, 25 Nov 2003 23:07:27 +0100, you wrote:

>In article <3FC2AC68.8030804@...>,
>   Adam Seychell <adam_seychell@...> wrote:
>
>> I have tried 5 minute epoxy on PCB material and it doesn't bond 
>> anywhere near as well as professional multilayer panels.
>
>I think you guys are amazing doing all this. I'm still on double sided
>using pre-coated etch resist and no PTH.
>
>The reason I've come in is that my experience is (from just ordinary
>gluing jobs) that the 5min stuff isn't anywhere near as good as the
>"normal" [OK "regular" for you guys across the pond :-)] Epoxy. I know
>this means clamping up for maybe 24 hrs but I think you will find the bond
>a lot stronger.

Well, I cannot stress enough that the results I'm getting are more
than satisfying. The resulting PCB's are with regard to their
useability in everyday situations absolutely ok. 

Markus

> Well, I cannot stress enough that the results I'm getting are more
> than satisfying. The resulting PCB's are with regard to their
> useability in everyday situations absolutely ok.
>
> Markus
>
That's the point.

There is not much needed for what you want to glue.

I know that regular (for "them") or normal (for "us") ;-) epoxy is much 
stronger
than 5 min epoxy but look at your application.

You have a GIANT surface, and only very little separation forces.

I think roughing the copper is just overkill.
But if you peel off the copper of a precoated board you can see it is 
roughened at the adhesive side.
So there seems to be some point in that.

ST

One thing I wasn't clear about after reading the excellent descriptions
is exactly how the 'sandwich' is made up.   I had expected that for a
4-layer board one would use two double-sided boards with an insulation
layer between -- but it sounds as though in fact one uses a single
two-sided board (for the two inner layers) and two single-sided
boards for the outer layers?   Is there a particular advantage in
this?

M Cowlishaw

> >The reason I've come in is that my experience is (from just ordinary
> >gluing jobs) that the 5min stuff isn't anywhere near as good as the
> >"normal" [OK "regular" for you guys across the pond :-)] Epoxy. I know
> >this means clamping up for maybe 24 hrs but I think you will find
the bond
> >a lot stronger.

Markus said:
> Well, I cannot stress enough that the results I'm getting are more
> than satisfying. The resulting PCB's are with regard to their
> useability in everyday situations absolutely ok. 

Epoxy has definitely improved. I gained a strong distrust of 5 minute
epoxies from the 70's and 80's (dating myself) as they'd always end up
exuding a greasy residue and letting go, getting sort of crumbly.

Now, though, I've had great success with 5 and 15 minute epoxies. Some
epoxies that I use, like Devcon's Plastic Epoxy, only come in 5 minute
varieties and are -very- strong. It appears to have a plastic solvent
in it, so that in addition to just epoxy glue action, it also welds
itself to the plastic.

I glued a polystyrene squirt gun handle onto an ABS tube for a prop
blaster about 15 years ago with the stuff and it is strong as if it
were cast in one piece. I got to show it to a pro propmaker from
Hollywood after he said that you cannot glue anything to ABS reliably
except ABS and only with ABS cement.

Steve

>One thing I wasn't clear about after reading the excellent descriptions
>is exactly how the 'sandwich' is made up.   I had expected that for a
>4-layer board one would use two double-sided boards with an insulation
>layer between -- but it sounds as though in fact one uses a single
>two-sided board (for the two inner layers) and two single-sided
>boards for the outer layers?   Is there a particular advantage in
>this?

Yes, the main advantage is that it's easier to handle. No need to
isolate. I.e. if say at some spot there is no glue in the inner layer,
it would not necesairly be catastrophic. It really boils down of
applying glue to both inner layer sides, tap the outer ones on it and
press it together as good as you manage.

The inner "double sided" material is 0.5mm sized and the outer ones
0.3mm together with the glue you end up with something about the
thikness of a standard PCB.

Markus

Markus Zingg wrote:
> 
> Yes, the main advantage is that it's easier to handle. No need to
> isolate. I.e. if say at some spot there is no glue in the inner layer,
> it would not necesairly be catastrophic. It really boils down of
> applying glue to both inner layer sides, tap the outer ones on it and
> press it together as good as you manage.
> 
> The inner "double sided" material is 0.5mm sized and the outer ones
> 0.3mm together with the glue you end up with something about the
> thikness of a standard PCB.
> 

I was just going to say you can try making a 4 layer PCB using 
two 0.8mm double sided panels and laying a piece of glass cloth 
between the them. But now I've starting writing this email I 
realized the added difficulty in having to etch the inner layers 
separate from the outer layers. Your "three ply" technique (0.3mm 
+ 0.5mm + 0.3mm) still looks like the best way.

Did you have trouble finding a supplier for single sided 0.3mm 
and double sided 0.5mm board material ?  I assume these are of 
18um copper foil.

Adam

>Markus Zingg wrote:
>> 
>> Yes, the main advantage is that it's easier to handle. No need to
>> isolate. I.e. if say at some spot there is no glue in the inner layer,
>> it would not necesairly be catastrophic. It really boils down of
>> applying glue to both inner layer sides, tap the outer ones on it and
>> press it together as good as you manage.
>> 
>> The inner "double sided" material is 0.5mm sized and the outer ones
>> 0.3mm together with the glue you end up with something about the
>> thikness of a standard PCB.
>> 
>
>I was just going to say you can try making a 4 layer PCB using 
>two 0.8mm double sided panels and laying a piece of glass cloth 
>between the them. But now I've starting writing this email I 
>realized the added difficulty in having to etch the inner layers 
>separate from the outer layers. Your "three ply" technique (0.3mm 
>+ 0.5mm + 0.3mm) still looks like the best way.
>
>Did you have trouble finding a supplier for single sided 0.3mm 
>and double sided 0.5mm board material ?  I assume these are of 
>18um copper foil.
>
>Adam

You probably guessed it already - I have the material from Bungard. 

Markus