Homebrew PCBs

Hi all,  I am building a tube guitar amp PCB.  It requires heavy
copper on thick boards.  I would like to use 1/8" boards, but cannot
find any locally.  Anyways, here are my plans:

Toner transfer method using Phoenix TTS paper (all I can get here in
Canada).  Then I plan to etch using MG Chemicals Sodium Persulphate
and hand agitation.

Is this reasonable for this thick of copper, or should I try a
different method.  Any hints, tips, or related sites would be appreciated!

thanks
Don

--- In Homebrew_PCBs@yahoogroups.com, "d1camero" <a8723@...> wrote:
>
> Hi all,  I am building a tube guitar amp PCB.  It requires heavy
> copper on thick boards.  I would like to use 1/8" boards, but cannot
> find any locally.  Anyways, here are my plans:
> 
> Toner transfer method using Phoenix TTS paper (all I can get here in
> Canada).  Then I plan to etch using MG Chemicals Sodium Persulphate
> and hand agitation.
> 
> Is this reasonable for this thick of copper, or should I try a
> different method.  Any hints, tips, or related sites would be 
appreciated!
> 
> thanks
> Don
>

You could probably get this to work, but for thick through hole parts 
in tube amps, I believe the reason for 1/8" board is stability in the 
face of higher heat levels.  Single sided is not going to cut it 
though, as one thing that helps is having plated through holes to 
hold the part mechanically in place.  Supposedly, you should use 
double sided just to have a solid metal eyelet holding the part to 
the pcb.  This is even if traces only run on one side.  So for a high 
quality tube amp built on a pcb, you should farm out the pcb 
construction for the plated through holes.

Read this for more info:

http://www.aikenamps.com/PCBorPTP.htm

If you're going to homebrew you might want to try eyelets through a 
plain (no copper) pcb.  Here's more info:

http://www.eham.net/articles/16053

Although using eyelets seems rather low tech, it looks like it is 
considered a high quality (though labor intensive) construction 
technique.  Old Fender amps in the 50s and 60s (preCBS) used eyelets 
and apparently they are still reliable.

Cheers,
Don

Thanks for the info Don.  I am actually quite well versed with Aiken,
eyelet construction, PTP, turret construction versus PCB.  What I am
looking for is feedback on how best to expose and etch 2 oz copper on
thick boards.

Anyone else?

--- In Homebrew_PCBs@yahoogroups.com, "d1camero" <a8723@...> wrote:
>
> Hi all,  I am building a tube guitar amp PCB.  It requires heavy
> copper on thick boards.  I would like to use 1/8" boards, but cannot
> find any locally.  Anyways, here are my plans:
> 
> Toner transfer method using Phoenix TTS paper (all I can get here in
> Canada).  Then I plan to etch using MG Chemicals Sodium Persulphate
> and hand agitation.
> 
> Is this reasonable for this thick of copper, or should I try a
> different method.  Any hints, tips, or related sites would be
appreciated!
> 
> thanks
> Don
>

Hi Don,  I haven't tried TT on anything heavier than 1oz but with my
results I believe it would work well on 2oz boards, especially for
something like a guitar amp with wider traces.  I use Ferric Chloride
to etch mine and even left in for too long, the toner is still solidly
adhered to the board,. It takes some good scrubbing to remove it even
after all of the etching is done (I use fine steel wool with alcohol
to strip it off and polish up the copper - works great).

I haven't tried the Sodium Persulphate but I did try Hydrochloric Acid
twice and it does attack the toner after a while. The board needs to
come out as soon as etching is done or the toner starts to lift. I'm
not sure if it would hold up for a 2oz etch or not. And I'm not sure
where the Sodium Persulphate is compared to these others.

Here's one tip for 1/8" boards, do 2 1/16 inch single sided
boards and bond them together with epoxy (or maybe even contact cement
would work).

Phil
KA0HBG

Phil, thanks for the info.  Good idea about the sandwich - I think I
will try that!

"d1camero" <a8723@...> writes:
> Phil, thanks for the info.  Good idea about the sandwich - I think I
> will try that!

Beware that epoxy softens when heated.  If that's an issue, use
something like J.B.Weld, which is more heat tolerant.

> 4c. Re: Best method for 2oz copper on 1/16" boards?
>     Posted by: "d1camero" a8723@... d1camero
>     Date: Mon Apr 2, 2007 6:40 pm ((PDT))
>
> Thanks for the info Don.  I am actually quite well versed with Aiken,
> eyelet construction, PTP, turret construction versus PCB.  What I am
> looking for is feedback on how best to expose and etch 2 oz copper on
> thick boards.
>
> Anyone else?

For the etching, you might want to try using Muriatic Acid and Hydrogen 
Peroxide, instead of the Sodium Persulfate.  It would be much cheaper, much 
more readily available, and should work at least as well.

I have the details of the etchant "recipe" on a webpage, at 
http://www.fullnet.com/~tomg/gooteepc.htm .

Good luck.

- Tom Gootee

--- In Homebrew_PCBs@yahoogroups.com, DJ Delorie <dj@...> wrote:
>
> 
> "d1camero" <a8723@...> writes:
> > Phil, thanks for the info.  Good idea about the sandwich - I think I
> > will try that!
> 
> Beware that epoxy softens when heated.  If that's an issue, use
> something like J.B.Weld, which is more heat tolerant.
>

The MSDS (Material Safety Data Sheets) for J-B WELD shows that there is 
10-20% iron powder in the resin and hardner. Perhaps this would not be 
a good 
choice for laminating PCB material with the chance of a short via the 
adhesive layer. 

http://jbweld.net/techinfo/index.php

Actually, I believe that " J.B. Weld"
is simply a two-part epoxy system with iron
powder added.

Same basic characteristics as generic
epoxy material.

Since the iron powder does not "crosslink"
with the epoxy system, I'm not sure what
viable function it serves, save for making
the system appear to be metallic.

Once hardened, (crosslinked) it does not "soften" to any
great extent when heated, but rather, will
decompose and burn if the heat rise is
high enough.

Roland F. Harriston

************************

warrenbrayshaw wrote:
>
> --- In Homebrew_PCBs@yahoogroups.com 
> <mailto:Homebrew_PCBs%40yahoogroups.com>, DJ Delorie <dj@...> wrote:
> >
> >
> > "d1camero" <a8723@...> writes:
> > > Phil, thanks for the info. Good idea about the sandwich - I think I
> > > will try that!
> >
> > Beware that epoxy softens when heated. If that's an issue, use
> > something like J.B.Weld, which is more heat tolerant.
> >
>
> The MSDS (Material Safety Data Sheets) for J-B WELD shows that there is
> 10-20% iron powder in the resin and hardner. Perhaps this would not be
> a good
> choice for laminating PCB material with the chance of a short via the
> adhesive layer.
>
> http://jbweld.net/techinfo/index.php 
> <http://jbweld.net/techinfo/index.php>
>
>  


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

"warrenbrayshaw" <warrenbrayshaw@...> writes:
> The MSDS (Material Safety Data Sheets) for J-B WELD shows that there
> is 10-20% iron powder in the resin and hardner.

Doh!  Nevermind.  Still, epoxy and heat don't get along, something to
keep in mind.

(jbweld is known to me because it's used to hold model rocket engine
mounts in place, instead of epoxy)

On Tue, 03 Apr 2007 22:34:26 +0200, DJ Delorie <dj@...> wrote:

>> The MSDS (Material Safety Data Sheets) for J-B WELD shows that there
>> is 10-20% iron powder in the resin and hardner.
> Doh!  Nevermind.  Still, epoxy and heat don't get along, something to
> keep in mind.
> (jbweld is known to me because it's used to hold model rocket engine
> mounts in place, instead of epoxy)


Why do you get that impression?

FR4 _IS_ glass in an epoxy matrix.

In fact, JB weld itself is an epoxy if i'm not mistaken, only they take  
more $ for the brand and packaging.

There are different kinds of epoxy, for example the cheap 5 minute stuff  
is far less strong and resistant to chemicals than the longer curing  
versions. I expect it is also not very good with heat but i never used it  
in such an application. Shouldn't be too tricky to get hold of a high  
quality epoxy that is comparable to the PCB itself.
If stuff gets that hot in electronics there's a problem somewhere anyway.

ST

"Stefan Trethan" <stefan_trethan@...> writes:
> Why do you get that impression?

From all the stories of motor mounts that fail due to epoxy softening.
And from my dad telling me to use a torch to remove epoxy plugs from
car parts.  And from HowItWorks showing heat-softened epoxy sheets
used for vacuum forming.

Yes, if a pcb gets that hot, something else is usually wrong.  But if
you're suing 2oz copper and 1/8 boards, I'm guessing you're probably
concerned about heat anyway.

My only point is, if you're concerned about heat, be careful about the
adhesive you use.

There are accountable reasons why PC boards heat up. Usually it is 
because the designer
did not take into consideration how much current various traces on the board
need to carry.  There are formulas and calculations that the "pros" use 
to design
high current carrying PC boards, especially in laying out power supply 
boards.

With 2 oz copper, and proper design, one can make boards that can carry
a lot of current. I don't have a good feeling for what increasing the 
thickness
of the substrate would do to make a properly designed PC board dissipate
more heat. Epoxy-glass  is not a very good heat conductor.
Heat sinks and fans usually handle heating problems, along with
properly engineered circuit trace widths..

Epoxy systems generally "decompose" when exposed to high temperatures,
and applying heat is a quick and dirty way of breaking down an epoxy bond.
I suspect that there are some toxic fumes emitted when epoxy systems are
decomposed via the application on intense heat.

Roland F. Harriston

*****************************


DJ Delorie wrote:
>
>
> "Stefan Trethan" <stefan_trethan@... 
> <mailto:stefan_trethan%40gmx.at>> writes:
> > Why do you get that impression?
>
> >From all the stories of motor mounts that fail due to epoxy softening.
> And from my dad telling me to use a torch to remove epoxy plugs from
> car parts. And from HowItWorks showing heat-softened epoxy sheets
> used for vacuum forming.
>
> Yes, if a pcb gets that hot, something else is usually wrong. But if
> you're suing 2oz copper and 1/8 boards, I'm guessing you're probably
> concerned about heat anyway.
>
> My only point is, if you're concerned about heat, be careful about the
> adhesive you use.
>
>  



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

2oz is good,but I wanna know the PCB layers you layout? I mean you must consider the thermal derating in the preparation of your design.

----- Original Message ----
From: Roland Harriston <rolohar@...>
To: Homebrew_PCBs@yahoogroups.com
Sent: Wednesday, April 4, 2007 6:48:07 AM
Subject: Re: [Homebrew_PCBs] Re: Best method for 2oz copper on 1/16" boards?

There are accountable reasons why PC boards heat up. Usually it is 
because the designer
did not take into consideration how much current various traces on the board
need to carry. There are formulas and calculations that the "pros" use 
to design
high current carrying PC boards, especially in laying out power supply 
boards.

With 2 oz copper, and proper design, one can make boards that can carry
a lot of current. I don't have a good feeling for what increasing the 
thickness
of the substrate would do to make a properly designed PC board dissipate
more heat. Epoxy-glass is not a very good heat conductor.
Heat sinks and fans usually handle heating problems, along with
properly engineered circuit trace widths..

Epoxy systems generally "decompose" when exposed to high temperatures,
and applying heat is a quick and dirty way of breaking down an epoxy bond.
I suspect that there are some toxic fumes emitted when epoxy systems are
decomposed via the application on intense heat.

Roland F. Harriston

************ ********* ********

DJ Delorie wrote:
>
>
> "Stefan Trethan" <stefan_trethan@ gmx.at 
> <mailto:stefan_ trethan%40gmx. at>> writes:
> > Why do you get that impression?
>
> >From all the stories of motor mounts that fail due to epoxy softening.
> And from my dad telling me to use a torch to remove epoxy plugs from
> car parts. And from HowItWorks showing heat-softened epoxy sheets
> used for vacuum forming.
>
> Yes, if a pcb gets that hot, something else is usually wrong. But if
> you're suing 2oz copper and 1/8 boards, I'm guessing you're probably
> concerned about heat anyway.
>
> My only point is, if you're concerned about heat, be careful about the
> adhesive you use.
>
> 

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





 
____________________________________________________________________________________
It's here! Your new message!  
Get new email alerts with the free Yahoo! Toolbar.
http://tools.search.yahoo.com/toolbar/features/mail/

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

In article <op.tp77d0xemg0lsf@tu-x2pj5qeyp2u4>,
   Stefan Trethan <stefan_trethan@...> wrote:
> If stuff gets that hot in electronics there's a problem somewhere anyway.

Fact of life - valves get hot.

I have several times had to make PCBs to replace burnt out ones in old
commercial valve equipment. Admittedly the originals were Paxolin rather
than fibreglass but...

Usual scenario is the solder joints between the PCB and the valve-holder,
because of the heat and the cycling, start to go crystaline. They become
high resistance and "dry" which causes the generation of more heat which
passes back along the track softening the adhesive and causing the track
to lift off the board charring it at the the same time.

In article <eus4im+q40f@...>,
   donahn <donahn@...> wrote:
> You could probably get this to work, but for thick through hole parts 
> in tube amps, I believe the reason for 1/8" board is stability in the 
> face of higher heat levels. 

Possibly but there is also the weight and leverage caused by valves. Not
forgetting the forces involved in getting them in and out of the sockets
periodicaly to replace them.

At 2:52 pm ((PDT)) Thu Apr 12, 2007, stuart.winsor.lists wrote:

>Usual scenario is the solder joints between the PCB and the valve-holder,
>because of the heat and the cycling, start to go crystaline. They become
>high resistance and "dry" [snip]

At 3:24 pm ((PDT)) Thu Apr 12, 2007, stuart.winsor.lists wrote:

>[snip] but there is also the weight and leverage caused by valves. Not
>forgetting the forces involved in getting them in and out of the sockets
>periodicaly to replace them.

In your experience and opinion, do the joints usually fail
1. through the body of the solder or
2. alongside the solder tag?

1. suggests to me either
a) simple work hardening and stress fracture, possibly
b) combined with recrystallization near the melting point,
while 2. hints at
c) brittleness though formation of an intermetallic compound
with something dissolved from the tags.

A high-temperature solder might reduce (a) and (b) by
being stronger and more stable at the working temperature
while a solder doped with small amounts of other metals
might reduce (b) and (c).

There are a fair number of online articles on solder metallurgy,
some produced by metals suppliers such as Johnson Matthey
and some propelled by the recent RoHS drive to lead-free solder.

Whether it's worth the bother of seeking something different from
what's already on the workbench, I can't advise ;-)