Homebrew PCBs

Hey all - my first post after doing lots of searches here. I'm a long
time electronics tinkerer (20+ years) but I haven't made too many pc
boards myself in that time and of those, most were done by hand. I
gave the toner transfer system a try when it first came out but wasn't
impressed. Back then I was using an iron to do the transferring and
like many people today I got decent results only when the board was
smaller than the iron. Even still, I didn't trust it to reliably make
tracks thin enough to run between DIP pads, much less do 0.5mm pitch
SMT devices, etc. Anyway, the new TTS paper made by Pulsar, especially
when applied with their so-called TIA Laminator (aka - GBC HeatSeal
H210), can supposedly lay down tracks and pads tighter than 8/8 design
rule reliably. Does anyone here have any experience to confirm or
refute this?

"Jeffrey Jenkins" <tesseract@...> writes:
> H210), can supposedly lay down tracks and pads tighter than 8/8 design
> rule reliably. Does anyone here have any experience to confirm or
> refute this?

I have the whole Pulsar kit (GBC Creative), and I can do 6.7/6.7 (four
pixels on my laser printer) with a little care.  0.5mm pitch SMT parts
(9.8mil rules) are not a problem, 0.4mm pitch (7.8mil rules) works
too.

0.4mm pitch and 01005 parts on pulsar-type board:
http://www.delorie.com/pcb/smd-challenge/old/proto-boards.html

close-up of 01005 cap next to an 0.4mm pitch IC:
http://www.delorie.com/pcb/first.html

--- In Homebrew_PCBs@yahoogroups.com, DJ Delorie <dj@...> wrote:

> I have the whole Pulsar kit (GBC Creative), and I can do 6.7/6.7 (four
> pixels on my laser printer) with a little care.  0.5mm pitch SMT parts
> (9.8mil rules) are not a problem, 0.4mm pitch (7.8mil rules) works
> too.
>

I checked out your site and what you've done with the Pulsar stuff is
very encouraging! I'm working on my first mostly-SMT board but because
of my past experience I figured I would have to have all of the
prototypes made rather than whip 'em up myself. If what you've done
with the Pulsar system is representative of what it can do then the
toner transfer technique has definitely come a long way since I last
tried it!

BTW - do you have to run the boards through more than once? Preheat
them at all? Anything special? Pulsar's website is kind of sparse on
the details on using the laminator and I know from experience that the
devil is in the details when it comes to this stuff.

-jj

"Jeffrey Jenkins" <tesseract@...> writes:
> BTW - do you have to run the boards through more than once? Preheat
> them at all? Anything special? Pulsar's website is kind of sparse on
> the details on using the laminator and I know from experience that
> the devil is in the details when it comes to this stuff.

I normally run the board through a couple of times, at various angles.
Traces that run parallel to the rollers seem to get less well
transferred than those that aren't.  I also use different parts of the
rollers, if the board is small enough, although I have no proof that
that helps..  So far, I've yet to have a case where there was too much
heat applied (no smeared traces), although a couple of times I've had
traces not stick well enough (didn't wait for the TIA to heat up all
the way, was too impatient with the water bath).

The other trick is to include a thick trace surrounding your board(s),
which holds the paper down around the edges.  Otherwise, stress on the
paper hits the board's traces first.  If the boards are really small
(my smallest are often in the 0.3 x 0.5 inch range), I put more than
one on each "panel" so if one has a broken trace, I can just use
another one.

My etchant is warm FeCl, wiped on with a sponge (a Pulsar trick).
Very quick that way, but you have to use the green TRF.

I'm still trying to figure out the paper though.  I have two types, an
older white pack, and a newer blue pack.  At the moment, the white
gives better results, but I don't know why.  I've tried humidifying
the blue, but it just doesn't seem to take up as much toner out of the
printer as the white does.  Next time I'm going to try running it
through the printer twice (the first time, not printing anything) to
see if prepping the paper helps it take up more toner.

--- In Homebrew_PCBs@yahoogroups.com, DJ Delorie <dj@...> wrote:

> I'm still trying to figure out the paper though.  I have two types, an
> older white pack, and a newer blue pack.  At the moment, the white
> gives better results, but I don't know why.  I've tried humidifying
> the blue, but it just doesn't seem to take up as much toner out of the
> printer as the white does.  Next time I'm going to try running it
> through the printer twice (the first time, not printing anything) to
> see if prepping the paper helps it take up more toner.
>

It's like deja vu all over again!?! I am getting exactly the same
crappy results I got with the Dyn-art paper 10+ years ago...

I have the light blue paper as well and, like you, I'm having problems
with the toner density. I just put in a brand new OEM (not reman.)
toner cartridge into my Oki C5100 (not a "toy" laser printer) and
tried tweaking every setting available to bump up the density, all to
no avail. 

Then there's the laminator... I see why they keep recommending 0.032"
boards over 0.064" - the thicker board jams up the laminator almost
immediately.

Now, I imagine that with practice and tweaking I'll be able to get
everything right, but if it is this fickle to begin with I'm not
confident that any tricks or settings I determine now will work
consistently in the future.

I suppose I better read up on all those posts concerning *other*
papers to use for this process... Ugh.

-Jeff

It looks like I've nailed down a reasonably reliable process for
making boards with the Pulsar paper after not too much trial and
error. I have learned a few lessons that I'll pass along in case they
help someone else:

* totally forget about using an iron - you need a laminator

* the GBC H210 laminator CAN NOT handle 1/16" board (1/32" works)

* the Green TRF film makes a huge difference in trace quality

* 1:2 muriatic/hydrogen peroxide rocks as an etchant!

* you have to run whole sheets of Pulsar paper through - no "cutouts"

Let me expound a little on #2 and  #5... #2 - Maybe some people can
get 1/16" (0.062") PC board to run through their "TIA Laminator" (aka
GBC H210), but I certainly couldn't. 0.032" board (1/16") is a bit too
flexible to use for SMT designs but it is totally adequate for
prototyping designs before sending them off to a board house.

#5 - Earlier in this thread I complained about how the toner wouldn't
transfer evenly, leading to spotty/broken traces. Once I ran full
sheets of the Pulsar paper through, instead of cutting out pieces just
big enough to contain the board image, I got great results. Of course
this means that a lot of paper gets wasted. Still, at a $1.50 a page
it's a lot cheaper, and faster, than sending Gerber files to the board
house.

One thing that worked for me (I cut out little bits of paper too) is a
trick included with the old white sheets.  Put the paper in a closed
container, on a rack or something, glossy side up, with some water in
the container (either use a plastic container, or put a bowl in a
cardboard box).  Keep the paper there for three minutes, use right
away.  This increases the moisture in the paper, and it seems to grab
and hold on to toner much better than if it dries out.

Don't let the paper get wet at all!  It's the humidity, not the wet,
that does the trick.

> * the GBC H210 laminator CAN NOT handle 1/16" board (1/32" works)

> Let me expound a little on #2 and  #5... #2 - Maybe some people can
> get 1/16" (0.062") PC board to run through their "TIA Laminator" (aka
> GBC H210), but I certainly couldn't. 0.032" board (1/16") is a bit 
too
> flexible to use for SMT designs but it is totally adequate for
> prototyping designs before sending them off to a board house.

My exp is that the H210 can handle 1/16 independantly when the heat is 
low and also when the heat is full power (fully warmed) w/some 
moderate pushing and pulling. After the unit has been on full for 
awhile it starts getting really cranky and doesnt want to pass 
anything, this includes 1/32 boards.

Ive managed to get the double sided 1/16 to work w/some fiddling and 
patience. The 1/32 is unacceptable for larger boards, they are too 
flexible and warp easily during the process. It was suggested that the 
1/32 can be trued by hand and has no memory, but for lead free smd and 
pc mount xfer they just wont work. I contemplated sandwiching 2 1/32 
boards, but seems like a pita and no one make 1/32 lead free clad 
boards.

cheers
wayne

Hello Wayne and the group.

This web site might be helpful on modifying and using the GBC laminators:
http://www.qsl.net/ve2emm/pcb/pcb_mach_mod_e.html

Pulsar used to have another similar write up about converting the GBC
laminator and slowing it down to accept thicker pc board material but he has
apparently removed it from his site.

I modified mine and it takes the board material with no problem.  I can also
still use it as a laminator.

Good luck.

Jim Pruitt

----- Original Message ----- 
From: wayneosdias
To: Homebrew_PCB
Sent: Friday, June 08, 2007 10:22 PM
Subject: [Homebrew_PCBs] Re: Pulsar's "TIA" Laminator - any good?

> * the GBC H210 laminator CAN NOT handle 1/16" board (1/32" works)

> Let me expound a little on #2 and #5... #2 - Maybe some people can
> get 1/16" (0.062") PC board to run through their "TIA Laminator" (aka
> GBC H210), but I certainly couldn't. 0.032" board (1/16") is a bit
too
> flexible to use for SMT designs but it is totally adequate for
> prototyping designs before sending them off to a board house.

My exp is that the H210 can handle 1/16 independantly when the heat is
low and also when the heat is full power (fully warmed) w/some
moderate pushing and pulling. After the unit has been on full for
awhile it starts getting really cranky and doesnt want to pass
anything, this includes 1/32 boards.

Ive managed to get the double sided 1/16 to work w/some fiddling and
patience. The 1/32 is unacceptable for larger boards, they are too
flexible and warp easily during the process. It was suggested that the
1/32 can be trued by hand and has no memory, but for lead free smd and
pc mount xfer they just wont work. I contemplated sandwiching 2 1/32
boards, but seems like a pita and no one make 1/32 lead free clad
boards.

cheers
wayne

This web site might be helpful on using the GBC laminators:
http://www.qsl.net/ve2emm/pcb/pcb_mach_mod_e.html

Pulsar used to have another similar write up about converting the GBC
laminator and slowing it down to accept thicker pc board material but he has
apparently removed it from his site.

I modified mine and it takes the board material with no problem.

Good luck.

Jim Pruitt

----- Original Message ----- 
From: wayneosdias
To: Homebrew_PCBs@yahoogroups.com
Sent: Friday, June 08, 2007 10:22 PM
Subject: [Homebrew_PCBs] Re: Pulsar's "TIA" Laminator - any good?



> * the GBC H210 laminator CAN NOT handle 1/16" board (1/32" works)

> Let me expound a little on #2 and #5... #2 - Maybe some people can
> get 1/16" (0.062") PC board to run through their "TIA Laminator" (aka
> GBC H210), but I certainly couldn't. 0.032" board (1/16") is a bit
too
> flexible to use for SMT designs but it is totally adequate for
> prototyping designs before sending them off to a board house.

My exp is that the H210 can handle 1/16 independantly when the heat is
low and also when the heat is full power (fully warmed) w/some
moderate pushing and pulling. After the unit has been on full for
awhile it starts getting really cranky and doesnt want to pass
anything, this includes 1/32 boards.

Ive managed to get the double sided 1/16 to work w/some fiddling and
patience. The 1/32 is unacceptable for larger boards, they are too
flexible and warp easily during the process. It was suggested that the
1/32 can be trued by hand and has no memory, but for lead free smd and
pc mount xfer they just wont work. I contemplated sandwiching 2 1/32
boards, but seems like a pita and no one make 1/32 lead free clad
boards.

cheers
wayne

Hey Jim

Sorry I didnt see your post till now. Ya I saw that page and asked 
frank about it and he said it has no impact on the laminators ability 
to pass thicker media, he said it just slows the passing time. I 
would figure if your reducing the RPMs, your increasing the torque, 
so I dont see how it wouldnt help w/thicker media.

Anyway, ive noticed after using the HS210 for a bit now that itll 
pass double sided 1/16 no prob, even boards that are 6x6 w/photo 
paper both sides. I guess you just got to break her in a bit.

wayne

--- In Homebrew_PCBs@yahoogroups.com, "Jim Pruitt" <wa7duy@...> wrote:
>
> Hello Wayne and the group.
> 
> This web site might be helpful on modifying and using the GBC 
laminators:
> http://www.qsl.net/ve2emm/pcb/pcb_mach_mod_e.html
> 
> Pulsar used to have another similar write up about converting the 
GBC
> laminator and slowing it down to accept thicker pc board material 
but he has
> apparently removed it from his site.
> 
> I modified mine and it takes the board material with no problem.  I 
can also
> still use it as a laminator.
> 
> Good luck.
> 
> Jim Pruitt
> 
> ----- Original Message ----- 
> From: wayneosdias
> To: Homebrew_PCB
> Sent: Friday, June 08, 2007 10:22 PM
> Subject: [Homebrew_PCBs] Re: Pulsar's "TIA" Laminator - any good?
> 
> > * the GBC H210 laminator CAN NOT handle 1/16" board (1/32" works)
> 
> > Let me expound a little on #2 and #5... #2 - Maybe some people can
> > get 1/16" (0.062") PC board to run through their "TIA Laminator" 
(aka
> > GBC H210), but I certainly couldn't. 0.032" board (1/16") is a bit
> too
> > flexible to use for SMT designs but it is totally adequate for
> > prototyping designs before sending them off to a board house.
> 
> My exp is that the H210 can handle 1/16 independantly when the heat 
is
> low and also when the heat is full power (fully warmed) w/some
> moderate pushing and pulling. After the unit has been on full for
> awhile it starts getting really cranky and doesnt want to pass
> anything, this includes 1/32 boards.
> 
> Ive managed to get the double sided 1/16 to work w/some fiddling and
> patience. The 1/32 is unacceptable for larger boards, they are too
> flexible and warp easily during the process. It was suggested that 
the
> 1/32 can be trued by hand and has no memory, but for lead free smd 
and

> pc mount xfer they just wont work. I contemplated sandwiching 2 1/32
> boards, but seems like a pita and no one make 1/32 lead free clad
> boards.
> 
> cheers
> wayne
>