Homebrew PCBs

Okay, I have yellow durabite ink in a stock cartridge for a CX5400 
printer and no printer.

I will do some tests today (if time allows - it is spring time and the 
honey do list is LONG).

But, these will all have to be crude tests as all I have is PCB 
material, ferric chloride etchant, no printer, and the ink that is 
still in the cartridge.  I am not even sure how to get the ink out! 

But, if you all have some suggestions, I hope to have the time to try 
them to see what happens.

Chris

Is there something 'important' going on when the printer applies the ink to
the surface?  By that I mean has the ink been warmed up, vapourized,
electro-shocked or anything else?

Before you unseal your cartridge I'd make sure your tests will be
representative of what happens in an inkjet printer.  No sense in wasting
that ink until you know your test will be useful.

All I did notice about the Epson printers that use Durabrite ink is that
they use piezoelectric heads.   Is something special going on in there that
affects the ink?

If the printhead is just spraying ink, then drilling the top of the
cartridge and dipping a toothpick in there will get you ink.  Maybe get a
couple drops out and then use thin artist paintbrushes; I always pick up a
few packs at the local dollar store (the ones we used for model airplanes
and such).  You can then seal your cartridge with just about any quality
tape.

Robert
:)



-----Message d'origine-----
De : Homebrew_PCBs@yahoogroups.com [mailto:Homebrew_PCBs@yahoogroups.com] De
la part de lcdpublishing
Envoyé : avril 15 2006 07:24
À : Homebrew_PCBs@yahoogroups.com
Objet : [Homebrew_PCBs] Epson durabrite ink tests - I have ink -no
printer...


Okay, I have yellow durabite ink in a stock cartridge for a CX5400 
printer and no printer.

I will do some tests today (if time allows - it is spring time and the 
honey do list is LONG).

But, these will all have to be crude tests as all I have is PCB 
material, ferric chloride etchant, no printer, and the ink that is 
still in the cartridge.  I am not even sure how to get the ink out! 

But, if you all have some suggestions, I hope to have the time to try 
them to see what happens.

Chris






Be sure to visit the group home and check for new Links, Files, and Photos:
http://groups.yahoo.com/group/Homebrew_PCBs

If Files or Photos are running short of space, post them here:
http://groups.yahoo.com/group/Homebrew_PCBs_Archives/ 
Yahoo! Groups Links

On Sat, 15 Apr 2006 13:23:57 +0200, lcdpublishing  
<lcdpublishing@...> wrote:

> Okay, I have yellow durabite ink in a stock cartridge for a CX5400
>
> printer and no printer.
>
>
> I will do some tests today (if time allows - it is spring time and the
>
> honey do list is LONG).
>
>
> But, these will all have to be crude tests as all I have is PCB
>
> material, ferric chloride etchant, no printer, and the ink that is
>
> still in the cartridge.  I am not even sure how to get the ink out!
>
>
> But, if you all have some suggestions, I hope to have the time to try
>
> them to see what happens.
>
>
> Chris


I think if you can get some ink out there it would be interesting to see  
if it forms a sheet on the PCB or if it beads up. You'll also be able to  
see how it holds up in the etchant.

I would also be highly interested in a CuCl test with this (and the other)  
ink, because CuCl can be more agressive than FeCl, especially if run with  
a high HCl molarity (but that can be left down if needed), but i don't  
think you use CuCl.

Without a printer those tests will of course be limited, but i would still  
be curious enough to try.

ST

Well, in about 10~15 minutes we should know something, what, I don't 
know :-)

I got some ink out of that cartridge.  I used a piece of wire to 
smear some around on the PCB board. I also shook it to get a few 
drops out and used a coarse brush to "paint it" on the board.  Using 
two sample boards....

1) I hit with the heat gun for a few seconds to make sure the ink 
was dry.

2) didn't do anything special - just let the ink air dry for a few 
minutes.  

Do not expect positive results though.  This is yellow ink, and 
being yellow it won't have much pigment in it like the black stuff 
has.  Which is why I think certain color sharies work bettter than 
others.

To answer stefans main question and the one I too suspected would be 
a problem - beading.  Nope, I could not see any evidence that this 
ink would bead up.  

I also did not do anything the to PCB material.  It was a cut off 
piece that I know is pretty decent quality as I just etched a few 
other boards from it a few days ago.

I am etching in ferric chloride so that is the etchant for my test.

Hmmm, this is interesting!

I let the board etch until I really felt it was "done".  With that 
yellow ink, it is very tough to tell the "masked" areas once the 
etchant gets on the board.  But, it looks as though I can see the 
bristle lines from the coarse brush I used.  The solid area I 
covered appears to be solid.  

I am running another test right now.  This time I pre-etched the 
board just for 10 seconds or so - to remove the gloss.  I then got 
some more ink out of the cartridge and wrote my name with a short 
piece of solder that I was dipping into the puddle.  Using a 
sharpie, I drew a square box around that so I would know where to 
look.  I also did another "Flood fill" area to see how it works 
again in that.

I really think this is going to work guys - more in a few minutes!

Chris




--- In Homebrew_PCBs@yahoogroups.com, "lcdpublishing" 
<lcdpublishing@...> wrote:
>
> Well, in about 10~15 minutes we should know something, what, I 
don't 
> know :-)
> 
> I got some ink out of that cartridge.  I used a piece of wire to 
> smear some around on the PCB board. I also shook it to get a few 
> drops out and used a coarse brush to "paint it" on the board.  
Using 
> two sample boards....
> 
> 1) I hit with the heat gun for a few seconds to make sure the ink 
> was dry.
> 
> 2) didn't do anything special - just let the ink air dry for a few 
> minutes.  
> 
> Do not expect positive results though.  This is yellow ink, and 
> being yellow it won't have much pigment in it like the black stuff 
> has.  Which is why I think certain color sharies work bettter than 
> others.
> 
> To answer stefans main question and the one I too suspected would 
be 

> a problem - beading.  Nope, I could not see any evidence that this 
> ink would bead up.  
> 
> I also did not do anything the to PCB material.  It was a cut off 
> piece that I know is pretty decent quality as I just etched a few 
> other boards from it a few days ago.
> 
> I am etching in ferric chloride so that is the etchant for my test.
>

Well, it looks to me like someone with an epson printer that has 
durabrite ink has to give this a go.  

The ink is resisting the ferric chloride for a while.  Then, it 
starts to breakdown and the etchant starts to eat away at the 
copper.  So, it does appear to work for a while.

I certainly can read my name from the previous etch test.  However, 
I also drew a box and some lines around my name with a sharpie.  
When I run my finger over the board, I can easily tell the 
difference in thickness of the copper.  The areas that were masked 
with the sharpie are much thicker copper than with the ink.

The problem with my tests is that I have no way to consistantly 
control the application thickness (Film thickness) of the ink.  So, 
in areas where the ink is thick, it protects good - I suspect.  This 
also appears to be the case where I tried another "Flood fill" 
area.  Some areas resisted well, others etched through like there 
was nothing there.

So, I really think the next step is for someone to try it on an ink 
jet printer.  that will provide consistant thickness of ink and of 
course, something accurate to see after the etching.

that's all I got folks.

Chris

lcdpublishing wrote:
> Well, it looks to me like someone with an epson printer that has 
> durabrite ink has to give this a go.  
> 
> The ink is resisting the ferric chloride for a while.  Then, it 
> starts to breakdown and the etchant starts to eat away at the 
> copper.  So, it does appear to work for a while.
> 
> I certainly can read my name from the previous etch test.  However, 
> I also drew a box and some lines around my name with a sharpie.  
> When I run my finger over the board, I can easily tell the 
> difference in thickness of the copper.  The areas that were masked 
> with the sharpie are much thicker copper than with the ink.
> 
> The problem with my tests is that I have no way to consistantly 
> control the application thickness (Film thickness) of the ink.  So, 
> in areas where the ink is thick, it protects good - I suspect.  This 
> also appears to be the case where I tried another "Flood fill" 
> area.  Some areas resisted well, others etched through like there 
> was nothing there.
> 
> So, I really think the next step is for someone to try it on an ink 
> jet printer.  that will provide consistant thickness of ink and of 
> course, something accurate to see after the etching.
> 
> that's all I got folks.

Might you try baking the Durabrite in a warm oven or under a lamp for 
a while?  It might take a while for the ink to cure into a tough solid.

Hi john,

That would be a good test, but it won't show me much with the crude 
tests I am doing.  It really needs to be tested on an actual 
printer.  The thickness of the ink is a critcal part in this and I 
can't apply an even film with what I have to work with. So, with an 
uneven film of ink, the results at best will be "uneven" ;-)

Chris



--- In Homebrew_PCBs@yahoogroups.com, John Popelish <jpopelish@...> 
wrote:
>
> lcdpublishing wrote:
> > Well, it looks to me like someone with an epson printer that has 
> > durabrite ink has to give this a go.  
> > 
> > The ink is resisting the ferric chloride for a while.  Then, it 
> > starts to breakdown and the etchant starts to eat away at the 
> > copper.  So, it does appear to work for a while.
> > 
> > I certainly can read my name from the previous etch test.  
However, 
> > I also drew a box and some lines around my name with a sharpie.  
> > When I run my finger over the board, I can easily tell the 
> > difference in thickness of the copper.  The areas that were 
masked 
> > with the sharpie are much thicker copper than with the ink.
> > 
> > The problem with my tests is that I have no way to consistantly 
> > control the application thickness (Film thickness) of the ink.  
So, 
> > in areas where the ink is thick, it protects good - I suspect.  
This 
> > also appears to be the case where I tried another "Flood fill" 
> > area.  Some areas resisted well, others etched through like 
there 
> > was nothing there.
> > 
> > So, I really think the next step is for someone to try it on an 
ink 
> > jet printer.  that will provide consistant thickness of ink and 
of 
> > course, something accurate to see after the etching.
> > 
> > that's all I got folks.
> 
> Might you try baking the Durabrite in a warm oven or under a lamp 
for 
> a while?  It might take a while for the ink to cure into a tough 
solid.
>

Chris

What about this? I see it as being like water-based paint. If you wash a
semi-gloss paint the day after you put it on, you will get some paint coming
off ... but wait a week or two, and the paint holds up to repeated water
exposures. Now I know the ink is probably not "curing" in the same way the
paint is, but if there was some residual water content in the ink when you
put it in the FeCL3, you could be redissolving some of the ink. That's why I
think John's suggestion has some merit.

Randy

> -----Original Message-----
> From: Homebrew_PCBs@yahoogroups.com
> [mailto:Homebrew_PCBs@yahoogroups.com]On Behalf Of lcdpublishing
> Sent: Saturday, April 15, 2006 4:38 PM
> To: Homebrew_PCBs@yahoogroups.com
> Subject: [Homebrew_PCBs] Re: RE : Epson durabrite ink tests - I have ink
> -no printer...
>
>
> Hi john,
>
> That would be a good test, but it won't show me much with the crude
> tests I am doing.  It really needs to be tested on an actual
> printer.  The thickness of the ink is a critcal part in this and I
> can't apply an even film with what I have to work with. So, with an
> uneven film of ink, the results at best will be "uneven" ;-)
>
> Chris
>
>
>
> --- In Homebrew_PCBs@yahoogroups.com, John Popelish <jpopelish@...>
> wrote:
> >
> > lcdpublishing wrote:
> > > Well, it looks to me like someone with an epson printer that has
> > > durabrite ink has to give this a go.
> > >
> > Might you try baking the Durabrite in a warm oven or under a lamp
> for
> > a while?  It might take a while for the ink to cure into a tough
> solid.
> >
>
>
>

I agree.  After that message, I went back out to the shop and put 
some ink on a PCB to let air dry.  After about an hour or so (don't 
know the exact amount of time), I rubbed my finger over it and some 
of the ink came off on my finger.  

I am going to leave it overnight and see if any more rubs off 
tomorrow. 

I did hit the test sample that I etched with the hot air gun.  I 
would have expected that to "Cure" the ink.  I did get the board hot 
enough where it was very uncomfortable to hold by it's edges.  So, I 
would have thought the ink was cured by that point.

I won't be around much tomorrow so I can't do too much more.  Over 
the week though I am going to try some other things if the ink 
doesn't dry up in the cartridge.  

At this point though, I know the ink will act as a resist. I also 
know it won't bead up on the copper.  Like you guys, I suspect that 
proper drying is key.  So, that is what I am going to focus on as I 
have time - getting the ink to dry, perferably quickly.

Chris



--- In Homebrew_PCBs@...m, "Randy Ledyard" 
<rll_groups@...> wrote:
>
> Chris
> 
> What about this? I see it as being like water-based paint. If you 
wash a
> semi-gloss paint the day after you put it on, you will get some 
paint coming
> off ... but wait a week or two, and the paint holds up to repeated 
water
> exposures. Now I know the ink is probably not "curing" in the same 
way the
> paint is, but if there was some residual water content in the ink 
when you
> put it in the FeCL3, you could be redissolving some of the ink. 
That's why I
> think John's suggestion has some merit.
> 
> Randy
> 
> > -----Original Message-----
> > From: Homebrew_PCBs@yahoogroups.com
> > [mailto:Homebrew_PCBs@yahoogroups.com]On Behalf Of lcdpublishing
> > Sent: Saturday, April 15, 2006 4:38 PM
> > To: Homebrew_PCBs@yahoogroups.com
> > Subject: [Homebrew_PCBs] Re: RE : Epson durabrite ink tests - I 
have ink
> > -no printer...
> >
> >
> > Hi john,
> >
> > That would be a good test, but it won't show me much with the 
crude
> > tests I am doing.  It really needs to be tested on an actual
> > printer.  The thickness of the ink is a critcal part in this and 
I
> > can't apply an even film with what I have to work with. So, with 
an
> > uneven film of ink, the results at best will be "uneven" ;-)
> >
> > Chris
> >
> >
> >
> > --- In Homebrew_PCBs@yahoogroups.com, John Popelish <jpopelish@>
> > wrote:
> > >
> > > lcdpublishing wrote:
> > > > Well, it looks to me like someone with an epson printer that 
has
> > > > durabrite ink has to give this a go.
> > > >
> > > Might you try baking the Durabrite in a warm oven or under a 
lamp
> > for
> > > a while?  It might take a while for the ink to cure into a 
tough

> > solid.
> > >
> >
> >
> >
>

Also, don't forget that Volkan had excellent results with a 'substitute'
ink, not the original Durabrite.

Robert
:)



-----Message d'origine-----
De : Homebrew_PCBs@yahoogroups.com [mailto:Homebrew_PCBs@yahoogroups.com] De
la part de lcdpublishing
Envoyé : avril 15 2006 17:43
À : Homebrew_PCBs@yahoogroups.com
Objet : [Homebrew_PCBs] Re: RE : Epson durabrite ink tests - I have ink -no
printer...


I agree.  After that message, I went back out to the shop and put 
some ink on a PCB to let air dry.  After about an hour or so (don't 
know the exact amount of time), I rubbed my finger over it and some 
of the ink came off on my finger.  

I am going to leave it overnight and see if any more rubs off 
tomorrow. 

I did hit the test sample that I etched with the hot air gun.  I 
would have expected that to "Cure" the ink.  I did get the board hot 
enough where it was very uncomfortable to hold by it's edges.  So, I 
would have thought the ink was cured by that point.

I won't be around much tomorrow so I can't do too much more.  Over 
the week though I am going to try some other things if the ink 
doesn't dry up in the cartridge.  

At this point though, I know the ink will act as a resist. I also 
know it won't bead up on the copper.  Like you guys, I suspect that 
proper drying is key.  So, that is what I am going to focus on as I 
have time - getting the ink to dry, perferably quickly.

Chris

lcdpublishing wrote:
(snip)
> At this point though, I know the ink will act as a resist. I also 
> know it won't bead up on the copper.  Like you guys, I suspect that 
> proper drying is key.  So, that is what I am going to focus on as I 
> have time - getting the ink to dry, perferably quickly.

My point is that it may not be just drying, in the sense of some 
solvent evaporating.  It may actually cure to a different chemical 
structure over some period of time, after the solvent evaporates. 
Lots of paints do this sort of thing.  Let it sit in the sunshine for 
a few days.  It might react with UV, too.

Yup, I put fingerprints into the engine block paint I tested on the PCBs.
But today I can lightly scratch at it with my nail and not damage the
surface.  If I had cured the paint AS RECOMMENDED (but too darn impatient),
I could probably take a nail to it.

Robert
:)


-----Message d'origine-----
De : Homebrew_PCBs@yahoogroups.com [mailto:Homebrew_PCBs@yahoogroups.com] De
la part de John Popelish
Envoyé : avril 15 2006 17:54
À : Homebrew_PCBs@yahoogroups.com
Objet : Re: [Homebrew_PCBs] Re: RE : Epson durabrite ink tests - I have ink
-no printer...


My point is that it may not be just drying, in the sense of some 
solvent evaporating.  It may actually cure to a different chemical 
structure over some period of time, after the solvent evaporates. 
Lots of paints do this sort of thing.  Let it sit in the sunshine for 
a few days.  It might react with UV, too.

Yup, well aware of that.  As I had the durabrite ink on hand and it 
would have ended up in the garbage, it was worth it to give it a try.  
So far so good, I really think we will all be ink-jet printing our 
etch masks within a few months  - at most.

It would also figure that it would work now.  The last 10 ~ 12 PCBs I 
did with toner transfer using the HP glossy paper were 100% perfect 
transfers - zero touch up.  While my stock of that paper is about to 
run out, I did atleast achieve VERY reliable results using toner 
transfer.

On the other hand, I won't miss it all if I can get the ink jet 
process to work as good or better!



--- In Homebrew_PCBs@yahoogroups.com, Robert Hedan <robert.hedan@...> 
wrote:
>
> Also, don't forget that Volkan had excellent results with 
a 'substitute'

> ink, not the original Durabrite.
> 
> Robert
> :)
>

What excites me most about this process is the possibility of
achieving dimensional accuracy, it would make CNC drilling a whole lot
easier !

I guess if the tests go ok, the next step is to find the cheapest
available durabrite compatable printer and strip it with the goal of
producing a straight paper path and clearance for the head over the
board. It is not always possible to use thin board material. From my
press-n-peel experience on a 600 x 600 laser, the lowest dpi durabrite
inkjets would be fine.

I'm looking on eBay already !


Chris  (-=Spiyda=-)


--- In Homebrew_PCBs@yahoogroups.com, "lcdpublishing"
<lcdpublishing@...> wrote:

>
> Yup, well aware of that.  As I had the durabrite ink on hand and it 
> would have ended up in the garbage, it was worth it to give it a try.  
> So far so good, I really think we will all be ink-jet printing our 
> etch masks within a few months  - at most.
>

Hi Chris,
  You are right the most important advantage is the accuracy. Indeed that's why I have started. My goal is to use my cnc to drill and cut the pcb. 
  If you are looking for a cheap Epson printer C68 is a good candidate another option can be C88. Today, I did shopping and bought it. I disassembled it and did some minor modifications which resulted a nice printed out on 1.6 mm PCB. I need to think about a little bit more how to fool the printer to get aligned double side printing. It seems not so easy.
  
 


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

--- In Homebrew_PCBs@yahoogroups.com, "Chris Horne" <chris@...> wrote:
>
> What excites me most about this process is the possibility of
> achieving dimensional accuracy, it would make CNC drilling a whole 
> lot
> easier !
> 
> I guess if the tests go ok, the next step is to find the cheapest
> available durabrite compatable printer and strip it with the goal of
> producing a straight paper path and clearance for the head over the
> board. It is not always possible to use thin board material. From my
> press-n-peel experience on a 600 x 600 laser, the lowest dpi 
> durabrite
> inkjets would be fine.
> 
> I'm looking on eBay already !
> 
> 
> Chris  (-=Spiyda=-)

I have no intention of disassembling my new R340 but I'll look over 
it to see if I can get any ideas. I think the CD printers would be 
more easily modified to have a straight-through paper path, since 
they've already got the straight-through slot for the CD tray, than 
the C-series which have no design for that at all. I'm sure you'd be 
able to use the Durabrite ink in the R-printers with a chip-swap 
trick. Keep in mind that taking apart an Epson printer can ruin it, as 
the entire printing mechanism needs to be kept 'aligned' by the frame. 
This is why most people can't empty or clean the waste ink tank on 
their own (thus the need to modify the printer to drain waste ink into 
a separate tank) because it's difficult to re-assemble the printer 
'properly'.

--- Volkan Sahin <vsahin@...> wrote:

>   ... My goal is to use my cnc to drill and cut the pcb. 

what CNC do you have?
anybody made 'his own' CNC drill/router?


__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com

Volkan,

I wouldn't bother trying to print double-sided, so many details.

I would think it easier to print on thin PCB, drill 2-3 guide holes, glue
back-to-back and then drill the rest.  But you're the expert here, what does
your experience tell us about this idea?  I haven't even touched thin PCB
yet.

Robert
:)


-----Message d'origine-----
De : Homebrew_PCBs@yahoogroups.com [mailto:Homebrew_PCBs@yahoogroups.com] De
la part de Volkan Sahin
Envoyé : avril 15 2006 21:38
À : Homebrew_PCBs@yahoogroups.com
Objet : Re: [Homebrew_PCBs] Re: RE : Epson durabrite ink tests - I have ink
-no printer...


Hi Chris,
  You are right the most important advantage is the accuracy. Indeed that's
why I have started. My goal is to use my cnc to drill and cut the pcb. 
  If you are looking for a cheap Epson printer C68 is a good candidate
another option can be C88. Today, I did shopping and bought it. I
disassembled it and did some minor modifications which resulted a nice
printed out on 1.6 mm PCB. I need to think about a little bit more how to
fool the printer to get aligned double side printing. It seems not so easy.

Robert,
  Definitely it is much more easier. You don't need to modify printer.  In some of my applications I am also planning to use it like that. Only issue is it is highly application dependent if you have bulky components then you need mechanically stable board, in such cases you can glue them on a thick  non copper clad board. During my experiments as I said, board was like paper to thin for most applications it is not acceptable.   
   
  Volkan,
   
   
   


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

It is a home made 3-axis step motor controlled one. 
  Volkan
   
  
"Herbert E. Plett" <cachureos@...> wrote:
  
--- Volkan Sahin <vsahin@...> wrote:

>   ... My goal is to use my cnc to drill and cut the pcb. 

what CNC do you have?
anybody made 'his own' CNC drill/router?


__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 


Be sure to visit the group home and check for new Links, Files, and Photos:
http://groups.yahoo.com/group/Homebrew_PCBs

If Files or Photos are running short of space, post them here:
http://groups.yahoo.com/group/Homebrew_PCBs_Archives/ 


    
---------------------------------
  YAHOO! GROUPS LINKS 

    
    Visit your group "Homebrew_PCBs" on the web.
    
    To unsubscribe from this group, send an email to:
 Homebrew_PCBs-unsubscribe@yahoogroups.com
    
    Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service. 

    
---------------------------------
  




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

--- Volkan Sahin <vsahin@...> wrote:

> It is a home made 3-axis step motor controlled one. 
>   Volkan

how did you build the mechanics for it? did you buy/make them?
the drivers for the steppers via LMD18245 is the easy part...  (I guess)

If the mechanical part is stable enough, then using a china-ink pen will plot a
board better than a jet printer! May take longer, but you can use whatever ink
you desire and the registration can be made 100% perfect. May be the solution
for the 'other' (less complicated) side.


__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com

Hi Herbert,
  Bingo.... I used LM18245 (motor driver)  and MSP430F123 (micro controller)  for step control. I bought ballscrews and linear slides from ebay. Mechanics seem to be stable as far as for light load but I don't thing that  it is possible to reach 3 mils (~ 75 micro meters) with plotter pen. May be you can say that who needs 3 mils. Mostly I am using 5-8 mils as a  minimum trace width.  I think easiest one is to adapt inkjet head to cnc. But neither  to find cheap piezo inkjet set head nor datasheet of Epson inkjet head is possible. Because of that I have decided to modify printer. I completely agree this is the difficult path. It seems it will be very difficult to find a solution for alignment problem and up to now I could not eliminate paper detection sensors of Epson C68  most probably they are embedded to head plastic or they are using capacitive detection. 
   


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

I agree that plotting is not a good solution, i think using a printer is  
the right way to do it.
Even if you attach a printhead to the cnc, which is a HUGE effort to  
control surely, it would be VERY slow because inkjet head speeds are  
really fast in a printer, that is if you succeeded at all.

About alignment, i think i may have worked that out some time ago.

What about converting the printer to a moving flatbed version, with some  
rigid, but light, plate moving in the paper direction? It would be driven  
directly by the printers rubber rollers, but also ride on some support  
rail to keep it horizontal. This table would activate the usually optical  
paper detector sensor. On this table one would stick some strip of PCB, at  
the leading edge, against which the PCB in print is pushed. This strip  
would also have a mark in the exact center of the page width (or wherever  
the center line ends up in a mirrored print). This mark is important,  
because it allows you to align the other side. So the PCB you want to  
print would need a mark also that is visible on both sides, for example a  
notch filed into one side.

Now if you align the PCB against the edge guide, and align the mark with  
the center mark on the table, you can print the bottom side. Then you turn  
the PCB over, and again align it against the edge guide, and the center  
mark as well, and you have it positioned in the exactly same place only  
turned over. Then you simply mirror the top layer artwork and print it out.


I hope the instructions were somewhat understandable, if not i'll try  
again or make sketches.


by the way the same edge guide and center mark could be used on your CNC  
table and you would have no trouble to align the PCB 100% accurate between  
printing and driling. with almost no effort at all.

Sure, it will require more modification to the printer than just making it  
pass pieces of PCB, but i think it will give you a much better system. I  
did such a modification to a HP deskjet back then when i still thought  
those would work on copper. I was stupid and modified it that way before  
actually trying to print on copper.

ST


On Sun, 16 Apr 2006 11:31:42 +0200, Volkan Sahin <vsahin@...>  
wrote:

> Hi Herbert,
>
>   Bingo.... I used LM18245 (motor driver)  and MSP430F123 (micro  
> controller)  for step control. I bought ballscrews and linear
> slides from ebay. Mechanics seem to be stable as far as for light load  
> but I don't thing that  it is possible to reach 3 mils (~ 75
>  micro meters) with plotter pen. May be you can say that who needs 3  
> mils. Mostly I am using 5-8 mils as a  minimum trace width.
> I think easiest one is to adapt inkjet head to cnc. But neither  to find  
> cheap piezo inkjet set head nor datasheet of Epson inkjet
> head is possible. Because of that I have decided to modify printer. I  
> completely agree this is the difficult path. It seems it
> will be very difficult to find a solution for alignment problem and up  
> to now I could not eliminate paper detection sensors of
> Epson C68  most probably they are embedded to head plastic or they are  
> using capacitive detection.

I CNC'd a Micro mill so that I can use it for other stuff also.

However, it is a little slow when drilling pcb's, (around 5 seconds
per hole) so I have the makings of a cnc drill machine just for pcbs,
just haven't got round to assembling it into a finished machine yet.. 
The new machine uses bushes on ground round guides for the ways and a
pneumatic 'Z'axis...  just need time to finish it..

I'm too busy with injection moulding, vacuum forming and aluminium
casting...  (too many hobbies !)

Chris  (-=Spiyda=-)

--- In Homebrew_PCBs@yahoogroups.com, "Herbert E. Plett"
<cachureos@...> wrote:

>
> 
> --- Volkan Sahin <vsahin@...> wrote:
> 
> >   ... My goal is to use my cnc to drill and cut the pcb. 
> 
> what CNC do you have?
> anybody made 'his own' CNC drill/router?
> 
> 
> __________________________________________________
> Do You Yahoo!?
> Tired of spam?  Yahoo! Mail has the best spam protection around 
> http://mail.yahoo.com
>

Can I put this ink, or sharpie ink, into cartidges for the photo 900p ?

Its originals are t007 and t009

Its has a lovely a4 flatpath upto about 4mm thick.



On 16/04/06, Volkan Sahin <vsahin@...> wrote:
> Robert,
>   Definitely it is much more easier. You don't need to modify printer.  In some of my applications I am also planning to use it like that. Only issue is it is highly application dependent if you have bulky components then you need mechanically stable board, in such cases you can glue them on a thick  non copper clad board. During my experiments as I said, board was like paper to thin for most applications it is not acceptable.
>
>   Volkan,
>
>
>
>
>
> [Non-text portions of this message have been removed]
>
>
>
> Be sure to visit the group home and check for new Links, Files, and Photos:
> http://groups.yahoo.com/group/Homebrew_PCBs
>
> If Files or Photos are running short of space, post them here:
> http://groups.yahoo.com/group/Homebrew_PCBs_Archives/
> Yahoo! Groups Links
>
>
>
>
>
>
>


--



Lez

- looking for a car, silver or white, 4 doors -
- diesel with a bosch pump, abs and pas  -
- aircon would be nice, and not a ford       -

FYI: 
PARALLAX is offering a 96 DPI HP Inkjet Cartridge and Cartridge Holder. 
If the DURABRIGHT refill ink could be loaded into it, it may be a 
starting point. At 96 DPI the best I would expect would be JUST over
10 mil

Link 
<http://www.parallax.com/html_pages/products/hpinkjetcontest/inkjetconte
st.asp>

--- In Homebrew_PCBs@yahoogroups.com, Volkan Sahin <vsahin@...> wrote:
SNIP
>
>  I think easiest one is to adapt inkjet head to cnc. But neither  to 
> find cheap piezo inkjet set head nor datasheet of Epson inkjet head 
> is possible. Because of that I have decided to modify printer. 
> I completely agree this is the difficult path. 
>
 SNIP END

>I would think it easier to print on thin PCB, drill 2-3 guide 
>holes, glue back-to-back and then drill the rest.

I've seen 1/32" single sided copperclad sold at PCB supply stores so 
it's probably a standard size along with 1/16". But quite frankly, 
photo etching is neither that expensive nor that difficult to warrant 
all these hassles IMO.

--- In Homebrew_PCBs@yahoogroups.com, Robert Hedan <robert.hedan@...> 
wrote:
>

It is kind of funny how some of us refuse to go down the path of 
photosensitive boards (myself being in this category).  I don't know 
why, I used have a darkroom and develop photos and such.  However, 
using it for PCBs just isn't my "cup of tea" I guess.

I believe there will be a variety of options for doing double sided 
with ink jets.  It's just a matter of getting things going and seeing 
how everything works out.





--- In Homebrew_PCBs@yahoogroups.com, "derekhawkins" <eldata@...> 
wrote:
>
> >I would think it easier to print on thin PCB, drill 2-3 guide 
> >holes, glue back-to-back and then drill the rest.
> 
> I've seen 1/32" single sided copperclad sold at PCB supply stores so 
> it's probably a standard size along with 1/16". But quite frankly, 
> photo etching is neither that expensive nor that difficult to 
warrant 

> all these hassles IMO.
> 
> --- In Homebrew_PCBs@yahoogroups.com, Robert Hedan <robert.hedan@> 
> wrote:
> >
>

>anybody made 'his own' CNC drill/router?

Here's a homebrew CNC drill setup (not only for PCBs), does 18 holes 
per minute;

http://www.pbase.com/eldata/cnc

--- In Homebrew_PCBs@yahoogroups.com, "Herbert E. Plett" 
<cachureos@...> wrote:
>

>I don't know
>why, I used have a darkroom and develop photos and such. 

Possibly because you think developing PCBs is as involved as 
developing photos. The following, in reference to photographic 
method of doing PCBs, is from the Pulsar website;

================================
This has been the most common method for the serious "midnight 
engineer" as well as R&D divisions of larger corporations who do 
their own designs.

This process requires a lot of support equipment. Darkroom, hot roll 
laminator to apply dry film (or buy expensive pre-sensitized PCB 
blanks), use one of several methods to make your film negative, UV 
exposure lamp & counter, contact print frame, trays and processing 
chemicals! There is a lot to it on the "setup" side. 

As you can see, this can be a pretty involved procedure and the 
system takes up quite a bit of dedicated space. It's not the fastest 
and it's not the cheapest method, however, results can be extremely 
good and it does allow for limited mass production capabilities.
================================

Trust me, the above is a long outdated approach. However, convenient 
to use if one is in the business of selling TT supplies.

--- In Homebrew_PCBs@yahoogroups.com, "lcdpublishing" 
<lcdpublishing@...> wrote:
>

Being able to flip a PCB over and print the other side properly requires a
perfectly 'square' PCB.  A lot of us cut pieces from larger pieces of PCB,
we'd have to stop that practice.

Robert
:)



-----Message d'origine-----
De : Homebrew_PCBs@yahoogroups.com [mailto:Homebrew_PCBs@yahoogroups.com] De
la part de Stefan Trethan
Envoyé : avril 16 2006 05:55
À : Homebrew_PCBs@yahoogroups.com
Objet : Re: [Homebrew_PCBs] inkjet second side alignment (was: homebrew CNC)


...

Now if you align the PCB against the edge guide, and align the mark with  
the center mark on the table, you can print the bottom side. Then you turn  
the PCB over, and again align it against the edge guide, and the center  
mark as well, and you have it positioned in the exactly same place only  
turned over. Then you simply mirror the top layer artwork and print it out.


...

ST

>Being able to flip a PCB over and print the other side properly 
>requires a perfectly 'square' PCB.

I would be surprised if one could even achieve less than 20 mils 
misalignment with that approach. 


--- In Homebrew_PCBs@yahoogroups.com, Robert Hedan <robert.hedan@...> 
wrote:
>

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

From: "lcdpublishing" <lcdpublishing@...>
To: <Homebrew_PCBs@yahoogroups.com>
Sent: Sunday, April 16, 2006 3:23 PM
Subject: [Homebrew_PCBs] Re: RE : Epson durabrite ink tests - I have ink -no 
printer...


> It is kind of funny how some of us refuse to go down the path of
> photosensitive boards (myself being in this category).  I don't know
> why, I used have a darkroom and develop photos and such.  However,
> using it for PCBs just isn't my "cup of tea" I guess.

I don't know why, either. I've dabbled with TT but it just seems too much 
like hard work, whereas photo-etch works every time with the minimum of 
fuss.

>
> I believe there will be a variety of options for doing double sided
> with ink jets.  It's just a matter of getting things going and seeing
> how everything works out.

I think that the main problem is still going to be the ink. Professional 
systems using ink jet technology use special ink with very expensive 
printheads, with UV curing directly after the ink is deposited. Also, 
double-sided boards will always be problematical with ordinary printers.

I heard of a new inkjet technology for PCBs on the radio recently that uses 
conductive ink which is being developed here in the UK. It's still very 
experimental. It's probably similar to this Epson system:

http://www.epson.co.jp/e/newsroom/news_2004_11_01.htm

Leon


Leon

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

From: "Robert Hedan" <robert.hedan@...>
To: <Homebrew_PCBs@yahoogroups.com>
Sent: Sunday, April 16, 2006 3:56 PM
Subject: [Homebrew_PCBs] inkjet second side alignment (was: homebrew CNC)


Being able to flip a PCB over and print the other side properly requires a
perfectly 'square' PCB.  A lot of us cut pieces from larger pieces of PCB,
we'd have to stop that practice.

A couple of holes for registration would get round that problem, but four 
pins would be required.

Leon
--
Leon Heller, G1HSM
leon.heller@...
http://www.geocities.com/leon_heller

I started with photo-resist PCB, I found it to be very expensive, that's why
I switched to toner transfer.  Yeah, it's more tedious trying to pass paper
through a printer, but I save money in the long run and right now that is a
concern.

Now if you would point me to a source of photo-resist liquid that I could
just spray onto copper-clad PCB,  I'd reconsider the whole thing.  But
you're still faced with the same alignment problems with double-sided PCBs.
My solution would be to use transparencies, tape them into a sandwich and
use a double-sided exposure setup.  I used transparencies on my laser
printers and was very satisfied with the results.  But you also have to deal
with one extra chemical.

I just like to explore all avenues first, and then make a decision after
I've heard reports on 'all' techniques.  Right now toner transfer is my
favourite, but if I can print directly onto thin substrate, I might switch
to that technique.  But I just won a Epson C84 on EBay for $10.50 this
morning and I'm going to test this process.

Robert
:)


-----Message d'origine-----
De : Homebrew_PCBs@yahoogroups.com [mailto:Homebrew_PCBs@yahoogroups.com] De
la part de lcdpublishing
Envoyé : avril 16 2006 10:24
À : Homebrew_PCBs@yahoogroups.com
Objet : [Homebrew_PCBs] Re: RE : Epson durabrite ink tests - I have ink -no
printer...


It is kind of funny how some of us refuse to go down the path of 
photosensitive boards (myself being in this category).  I don't know 
why, I used have a darkroom and develop photos and such.  However, 
using it for PCBs just isn't my "cup of tea" I guess.

I believe there will be a variety of options for doing double sided 
with ink jets.  It's just a matter of getting things going and seeing 
how everything works out.

You only need a minimum of 2 holes along the top or bottom edge really.  The
PCB will be perfectly aligned once you flip it, of course this depends on
how good your drilling abilities are.

4 holes = better alignment correction for the drilling-impaired  <-- that be
me!

Robert
:)


-----Message d'origine-----
De : Homebrew_PCBs@yahoogroups.com [mailto:Homebrew_PCBs@yahoogroups.com] De
la part de Leon Heller
Envoyé : avril 16 2006 11:09
À : Homebrew_PCBs@yahoogroups.com
Objet : Re: [Homebrew_PCBs] inkjet second side alignment (was: homebrew CNC)

A couple of holes for registration would get round that problem, but four 
pins would be required.

Leon

Oh gosh no!  I realize the photo process is MUCH simpler than the 
photographic developing process.  It's just that I have done that 
sort of thing before and it no longer interests me.  However, 
hacking an ink jet printer into doing something like what we are 
discussing - that sounds like fun!


--- In Homebrew_PCBs@yahoogroups.com, "derekhawkins" <eldata@...> 
wrote:
>
> >I don't know
> >why, I used have a darkroom and develop photos and such. 
> 
> Possibly because you think developing PCBs is as involved as 
> developing photos. The following, in reference to photographic 
> method of doing PCBs, is from the Pulsar website;
> 
> ================================
> This has been the most common method for the serious "midnight 
> engineer" as well as R&D divisions of larger corporations who do 
> their own designs.
> 
> This process requires a lot of support equipment. Darkroom, hot 
roll 
> laminator to apply dry film (or buy expensive pre-sensitized PCB 
> blanks), use one of several methods to make your film negative, UV 
> exposure lamp & counter, contact print frame, trays and processing 
> chemicals! There is a lot to it on the "setup" side. 
> 
> As you can see, this can be a pretty involved procedure and the 
> system takes up quite a bit of dedicated space. It's not the 
fastest 
> and it's not the cheapest method, however, results can be 
extremely 
> good and it does allow for limited mass production capabilities.
> ================================
> 
> Trust me, the above is a long outdated approach. However, 
convenient 

> to use if one is in the business of selling TT supplies.
> 
> --- In Homebrew_PCBs@yahoogroups.com, "lcdpublishing" 
> <lcdpublishing@> wrote:
> >
>

Well, not really. Right now you cut the PCB material to size then 
transfer & etch.  

Rather, you have your "leading" or "reference" edge cut straight and 
you mark the center point on that edge.  You then print one side, 
flip, print opposite side.  Now the artwork is square to that same 
edge and aligned from face to face.

Etch the board, then trim the rest to size.

I believe that will work out okay.






--- In Homebrew_PCBs@...m, Robert Hedan 
<robert.hedan@...> wrote:
>
> Being able to flip a PCB over and print the other side properly 
requires a
> perfectly 'square' PCB.  A lot of us cut pieces from larger pieces 
of PCB,
> we'd have to stop that practice.
> 
> Robert
> :)
> 
> 
> 
> -----Message d'origine-----
> De : Homebrew_PCBs@yahoogroups.com 
[mailto:Homebrew_PCBs@yahoogroups.com] De
> la part de Stefan Trethan
> Envoyé : avril 16 2006 05:55
> À : Homebrew_PCBs@yahoogroups.com
> Objet : Re: [Homebrew_PCBs] inkjet second side alignment (was: 
homebrew CNC)
> 
> 
> ...
> 
> Now if you align the PCB against the edge guide, and align the 
mark with  
> the center mark on the table, you can print the bottom side. Then 
you turn  
> the PCB over, and again align it against the edge guide, and the 
center  
> mark as well, and you have it positioned in the exactly same place 
only  
> turned over. Then you simply mirror the top layer artwork and 
print it out.

> 
> 
> ...
> 
> ST
>

Hmmm, that's a good one; just one straight edge and a center alignment pin,
so simple.

Maybe that's what Stefan was explaining back there, but I'm a visual guy, I
got lost after the 1st sentence, I need pictures.  Yup, I just glanceed back
at Stefan'S post and that's exactly what he said.  It went totally over my
head.  :D

"On this table one would stick some strip of PCB, at  
the leading edge, against which the PCB in print is pushed. This strip  
would also have a mark in the exact center of the page width (or wherever  
the center line ends up in a mirrored print). This mark is important,  
because it allows you to align the other side."

I'd just use a very small hole placed alongside the bottom border, just big
enough for a short pin.

Robert
:)


-----Message d'origine-----
De : Homebrew_PCBs@yahoogroups.com [mailto:Homebrew_PCBs@yahoogroups.com] De
la part de lcdpublishing
Envoyé : avril 16 2006 11:21
À : Homebrew_PCBs@yahoogroups.com
Objet : [Homebrew_PCBs] Re: inkjet second side alignment (was: homebrew CNC)


Well, not really. Right now you cut the PCB material to size then 
transfer & etch.  

Rather, you have your "leading" or "reference" edge cut straight and 
you mark the center point on that edge.  You then print one side, 
flip, print opposite side.  Now the artwork is square to that same 
edge and aligned from face to face.

Etch the board, then trim the rest to size.

I believe that will work out okay.

On Sun, 16 Apr 2006 16:56:28 +0200, Robert Hedan  
<robert.hedan@...> wrote:

> Being able to flip a PCB over and print the other side properly requires  
> a
>
> perfectly 'square' PCB.  A lot of us cut pieces from larger pieces of  
> PCB,
>
> we'd have to stop that practice.
>
>
> Robert
>
>
>


No, that's not the case.
To ensure a match all you need is to have one point match up and the angle  
of rotation be zero (or alternatively a second point match up)

The edge guide on one side with a centermark is sufficient, how/where the  
other edges are is of no relevance.

I prefer this over the two holes because you don't need to drill anything  
(you also don't need to drill two holes at exactly the right distance).  
making the center mark is simple enough, for example a nick with a nife to  
the edge is easily seen on both sides and very easy to align within 0.1mm  
or so.

ST

On Sun, 16 Apr 2006 17:20:39 +0200, lcdpublishing  
<lcdpublishing@...> wrote:

> Well, not really. Right now you cut the PCB material to size then
>
> transfer & etch.
>
>
> Rather, you have your "leading" or "reference" edge cut straight and
>
> you mark the center point on that edge.  You then print one side,
>
> flip, print opposite side.  Now the artwork is square to that same
>
> edge and aligned from face to face.
>
>
> Etch the board, then trim the rest to size.
>
>
> I believe that will work out okay.


At least one got it! fantastic...

ST

derek

could you describe your photo approach?

tnx
jim ab3cv

[...]

Trust me, the above is a long outdated approach. However, convenient 
to use if one is in the business of selling TT supplies.

On Sun, 16 Apr 2006 17:03:14 +0200, Leon Heller  
<leon.heller@...> wrote:

>> It is kind of funny how some of us refuse to go down the path of
>
>> photosensitive boards (myself being in this category).  I don't know
>
>> why, I used have a darkroom and develop photos and such.  However,
>
>> using it for PCBs just isn't my "cup of tea" I guess.
>
>
> I don't know why, either. I've dabbled with TT but it just seems too much
> like hard work, whereas photo-etch works every time with the minimum of
> fuss.


Well, i have tried both methods too, and photo etch seems too much like  
hard work, whereas TT works every time with the minimum of fuss. ;-)


>> I believe there will be a variety of options for doing double sided
>> with ink jets.  It's just a matter of getting things going and seeing
>> how everything works out.
>
>
> I think that the main problem is still going to be the ink. Professional
> systems using ink jet technology use special ink with very expensive
> printheads, with UV curing directly after the ink is deposited. Also,
> double-sided boards will always be problematical with ordinary printers.

Since i have seen just great results from Volkan i'm not that worried that  
there is a problem. I did not believe it is possible any time soon before  
seeing that, but those photos are clearly superb results and one can see  
they are inkjet printed as he says, so that's all the convincing i need.

>
>
> I heard of a new inkjet technology for PCBs on the radio recently that  
> uses
> conductive ink which is being developed here in the UK. It's still very
>
> experimental. It's probably similar to this Epson system:
>

That's the next step. Direkt resist printing is halfway there.
I have no doubt it will be a long time before we can print traces, but  
just printing and etching is as good as it gets until then.

ST

In metal working and in woodworking that concept is used quit 
frequently, so for me to get it was easy.  It is a tough one to 
grasp because it is hard to visualize that the printing will be 
square to the one edge - regardless of what the other three edges 
look like.  In reality, this could be accurate to within .003" 
depending on a couple of things...

1) The locating or reference surfaces are clean - dust has thickness!

2) The center mark is precise - don't try to line up a magic marker 
line with a pencil line.  Rather, use a razor blade to make both 
refernce marks (on the fixed stop and on the PCB).  With a fine line 
like that, accuracy with the naked eye is very easily done (even 
with my horrible eye sight).

3) The PCB does not move while printing (relative to the carrier 
jig).

4) The printer can sense the leading edge of the board (or jig) 
within .003"

It really should be difficult.  In reality, I think it will be more 
accurate than we can do with toner transfer or photo-imaging.  With 
the ink jet process, you eliminate two problem areas:

1) End to end miss-alignment as you will be bumped against a fixed 
point.

2) Rotational miss-alignment as you can have the board rotated if it 
is tight against the stop.

Even the third variable - side-to-side alignment can be very 
accurate too depending on the method used there.




--- In Homebrew_PCBs@yahoogroups.com, "Stefan Trethan" 
<stefan_trethan@...> wrote:
>
> On Sun, 16 Apr 2006 17:20:39 +0200, lcdpublishing  
> <lcdpublishing@...> wrote:
> 
> > Well, not really. Right now you cut the PCB material to size then
> >
> > transfer & etch.
> >
> >
> > Rather, you have your "leading" or "reference" edge cut straight 
and
> >
> > you mark the center point on that edge.  You then print one side,
> >
> > flip, print opposite side.  Now the artwork is square to that 
same

> >
> > edge and aligned from face to face.
> >
> >
> > Etch the board, then trim the rest to size.
> >
> >
> > I believe that will work out okay.
> 
> 
> At least one got it! fantastic...
> 
> ST
>

There's another problem with toner transfer or any other system which has an
intermediate printing step, and that is heat expansion.

On ordinary components it is insignificant, an 8-pin DIP will be good even
with moderate distortion, but on 'some' long components expansion becomes an
issue.  Yes, I have a ratio adjustment feature on QCAD, but the toner
transfer paper will not expand at identical rates.  Humidity in the air (and
paper) will give varying results, I'd have to test 'each' piece of paper
'cause my house is not a controlled environment.

I have one PCB that was affected by this, it had a 30-pin connector and it
came out 'short'.  The problem is that I only noticed that once etched.  I
had done a test print on ordinary paper, but that does not expand the same
way as Pulsar paper, everything was ok.

Printing directly onto the copper will reduce the risk of errors creeping in
the results.  Ideally, we'd have that direct etch system, but that will have
to wait.

Robert
:)


-----Message d'origine-----
De : Homebrew_PCBs@yahoogroups.com [mailto:Homebrew_PCBs@yahoogroups.com] De
la part de lcdpublishing
Envoyé : avril 16 2006 13:23
À : Homebrew_PCBs@yahoogroups.com
Objet : [Homebrew_PCBs] Re: inkjet second side alignment (was: homebrew CNC)


In metal working and in woodworking that concept is used quit 
frequently, so for me to get it was easy.  It is a tough one to 
grasp because it is hard to visualize that the printing will be 
square to the one edge - regardless of what the other three edges 
look like.  In reality, this could be accurate to within .003" 
depending on a couple of things...

...

http://www.fatwallet.com/t/24/602066/

  If we can find something with the durabrite inks on the list, this 
could make it a very nice deal.  $45 credit, or $25 check and $20 or $25 
credit.  Have until May 8th or so to purchase too.  Haven't even read 
the thread fully, but I think this isn't 100% finalized but looks like 
it's going to happen..

Alan

Now that's funny.  I'm picking up a C84 for $10.50 later, and I might be
able to claim $45 back from Epson.

Robert
LMAO!!!!!

-----Message d'origine-----
De : Homebrew_PCBs@yahoogroups.com [mailto:Homebrew_PCBs@yahoogroups.com] De
la part de Alan King
Envoyé : avril 16 2006 14:18
À : Homebrew_PCBs@yahoogroups.com
Objet : [Homebrew_PCBs] Epson printer class action rebates..


http://www.fatwallet.com/t/24/602066/

  If we can find something with the durabrite inks on the list, this 
could make it a very nice deal.  $45 credit, or $25 check and $20 or $25 
credit.  Have until May 8th or so to purchase too.  Haven't even read 
the thread fully, but I think this isn't 100% finalized but looks like 
it's going to happen..

Alan



Be sure to visit the group home and check for new Links, Files, and Photos:
http://groups.yahoo.com/group/Homebrew_PCBs

If Files or Photos are running short of space, post them here:
http://groups.yahoo.com/group/Homebrew_PCBs_Archives/ 
Yahoo! Groups Links

In a message dated 4/16/2006 9:28:50 A.M. Central Daylight Time,  
eldata@... writes:

anybody  made 'his own' CNC drill/router?



See mine here:  Photos;  Jan Rowland (near end of  all).  The PET you see on 
the left once controlled it.  The last PET  finally died several years back, 
and I am now converting the drill-control  circuitry to be controlled off the 
LPT-1 of an ancient PC running  MSDOS/GWBASIC.  Does about 50 holes per minute.
 
This was a very educational project, built in 1984.  Based upon some  of the 
details I taught myself doing that "PCB Drill," I subsequently built some  
serious PC-controlled X-Y router gear for pipeorganbuilding shops, even one in  
the UK.  Too much work for lazy people, so I retired while still  living!       
 JRR


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

Hi Chris, Volkan, Herbert, all,

I'm in the process of designing and building a small PCB router.  I'm 
using parts from two linear ways I collected a few years ago, and 
stepper motors I had on hand.  I'll probably time-share one of the 
controllers I built for my Sherline CNC lathe and mill.

I'm using the linear ball screw sleeves and ACME threaded nuts from two 
ways, and a similar but slightly smaller unit for the Z.  I'll need to 
make up translation blocks" (drilled for the bushings and ACME nuts) for 
the Y and Z axis of my gantry style mill.  Simple rectangular aluminum 
stock makes up the frame and the two uprights.

If anyone is interested, I can post pix of some of the stages of 
construction.  I might be posting it on The SherlineCNC list as well.

Alan  KM6VV


Chris Horne wrote:

> I CNC'd a Micro mill so that I can use it for other stuff also.
> 
> However, it is a little slow when drilling pcb's, (around 5 seconds
> per hole) so I have the makings of a cnc drill machine just for pcbs,
> just haven't got round to assembling it into a finished machine yet.. 
> The new machine uses bushes on ground round guides for the ways and a
> pneumatic 'Z'axis...  just need time to finish it..
> 
> I'm too busy with injection moulding, vacuum forming and aluminium
> casting...  (too many hobbies !)
> 
> Chris  (-=Spiyda=-)
> 
> --- In Homebrew_PCBs@yahoogroups.com, "Herbert E. Plett"
> <cachureos@...> wrote:
> 
>>
>>--- Volkan Sahin <vsahin@...> wrote:
>>
>>
>>>  ... My goal is to use my cnc to drill and cut the pcb. 
>>
>>what CNC do you have?
>>anybody made 'his own' CNC drill/router?

Hi LCD,

Use a center point for the edge.  I wonder if that might be a good 
solution for mechanical etching (isolation milling).

I normally use the LL corner for the origin; and then when the stock is 
turned over, the origin is now over on the lower right corner.  Not the 
best way to to work a second origin.  One either has to work from two 
origins, or subtract the exact width of the board.

An alignment pin or pins would simplify some of that.  I'm thinking of 
making a fixture, but I don't want to make a special fixture for each 
size of board.

Alan  KM6VV

lcdpublishing wrote:

> Well, not really. Right now you cut the PCB material to size then 
> transfer & etch.  
> 
> Rather, you have your "leading" or "reference" edge cut straight and 
> you mark the center point on that edge.  You then print one side, 
> flip, print opposite side.  Now the artwork is square to that same 
> edge and aligned from face to face.
> 
> Etch the board, then trim the rest to size.
> 
> I believe that will work out okay.
> 
>

On Sun, 16 Apr 2006 23:21:31 +0200, Alan Marconett <KM6VV@...> wrote:

> Hi LCD,
>
>
> Use a center point for the edge.  I wonder if that might be a good
>
> solution for mechanical etching (isolation milling).
>
>
> I normally use the LL corner for the origin; and then when the stock is
>
> turned over, the origin is now over on the lower right corner.  Not the
>
> best way to to work a second origin.  One either has to work from two
>
> origins, or subtract the exact width of the board.
>
>
> An alignment pin or pins would simplify some of that.  I'm thinking of
>
> making a fixture, but I don't want to make a special fixture for each
>
> size of board.
>
>
> Alan  KM6VV


re-read what you replied to and you'll see the problem is solved easily,  
precisely, with no pins or special fixtures.

ST

--- Robert Hedan <robert.hedan@...> wrote:

> Being able to flip a PCB over and print the other side properly requires a
> perfectly 'square' PCB.  A lot of us cut pieces from larger pieces of PCB,
> we'd have to stop that practice.
> ...

> 
> Now if you align the PCB against the edge guide, and align the mark with  
> the center mark on the table, you can print the bottom side. Then you turn  
> the PCB over, and again align it against the edge guide, and the center  
> mark as well, and you have it positioned in the exactly same place only  
> turned over. Then you simply mirror the top layer artwork and print it out.
> 

I believe that no 'square' is needed, only one perfectly straight header side
where to make the reference notch into, and not necessarily at the middle. The
other three can be ragged.
This one side will be fixed 'flat' against the reference border, can be
anything  (leading, trailing, side) as long it is 'fixed' relative to the
printer (motion). This 'reference' has to insure the registration during the
print process, not the board. This reference is part of some sort of tray that
has to move through the printer with +-5 mil tolerance... here is the quid!

__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com

--- Robert Hedan <robert.hedan@...> wrote:

> You only need a minimum of 2 holes along the top or bottom edge really.  The
> PCB will be perfectly aligned once you flip it, of course this depends on
> how good your drilling abilities are.
> 
> 4 holes = better alignment correction for the drilling-impaired  <-- that be
> me!

nope, no holes at all. the 'tray' upon which you fix the board is the trick!
A tray that runs through the printer with +-5 mil repeatability...


__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com

Hi Alan,
  I am always interested.  
  Thanks

Alan Marconett <KM6VV@...> wrote:
  Hi Chris, Volkan, Herbert, all,

I'm in the process of designing and building a small PCB router.  I'm 
using parts from two linear ways I collected a few years ago, and 
stepper motors I had on hand.  I'll probably time-share one of the 
controllers I built for my Sherline CNC lathe and mill.

I'm using the linear ball screw sleeves and ACME threaded nuts from two 
ways, and a similar but slightly smaller unit for the Z.  I'll need to 
make up translation blocks" (drilled for the bushings and ACME nuts) for 
the Y and Z axis of my gantry style mill.  Simple rectangular aluminum 
stock makes up the frame and the two uprights.

If anyone is interested, I can post pix of some of the stages of 
construction.  I might be posting it on The SherlineCNC list as well.

Alan  KM6VV


Chris Horne wrote:
> I CNC'd a Micro mill so that I can use it for other stuff also.
> 
> However, it is a little slow when drilling pcb's, (around 5 seconds
> per hole) so I have the makings of a cnc drill machine just for pcbs,
> just haven't got round to assembling it into a finished machine yet.. 
> The new machine uses bushes on ground round guides for the ways and a
> pneumatic 'Z'axis...  just need time to finish it..
> 
> I'm too busy with injection moulding, vacuum forming and aluminium
> casting...  (too many hobbies !)
> 
> Chris  (-=Spiyda=-)
> 
> --- In Homebrew_PCBs@yahoogroups.com, "Herbert E. Plett"
> <cachureos@...> wrote:
> 
>>
>>--- Volkan Sahin <vsahin@...> wrote:
>>
>>
>>>  ... My goal is to use my cnc to drill and cut the pcb. 
>>
>>what CNC do you have?
>>anybody made 'his own' CNC drill/router?




Be sure to visit the group home and check for new Links, Files, and Photos:
http://groups.yahoo.com/group/Homebrew_PCBs

If Files or Photos are running short of space, post them here:
http://groups.yahoo.com/group/Homebrew_PCBs_Archives/ 



  SPONSORED LINKS 
        Electrical engineering degree online   Electrical engineering degree   Printed circuit board     Electrical engineering   Electrical engineering course   Electrical engineering graduate school 
    
---------------------------------
  YAHOO! GROUPS LINKS 

    
    Visit your group "Homebrew_PCBs" on the web.
    
    To unsubscribe from this group, send an email to:
 Homebrew_PCBs-unsubscribe@yahoogroups.com
    
    Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service. 

    
---------------------------------
  




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

On Mon, 17 Apr 2006 01:44:24 +0200, Herbert E. Plett <cachureos@...>  
wrote:

> I believe that no 'square' is needed, only one perfectly straight header  
> side
>
> where to make the reference notch into, and not necessarily at the  
> middle. The
>
> other three can be ragged.
>
> This one side will be fixed 'flat' against the reference border, can be
>
> anything  (leading, trailing, side) as long it is 'fixed' relative to the
>
> printer (motion). This 'reference' has to insure the registration during  
> the
>
> print process, not the board. This reference is part of some sort of  
> tray that
>
> has to move through the printer with +-5 mil tolerance... here is the  
> quid!


Strictly speaking this edge does not even have to be straight.
It only needs the same two points to sit against the edge, so any concave  
shape would do, only a covex edge is forbidden.


About the tray, my plan is that this tray is a moving carriage,  
permanently situated on some guiderail and directly driven by the rubber  
rollers. This would ensure good positioning. It would need to be returned  
to the rear position after each print, maybe as simple as pusing agaist  
the motors with power off or maybe requiring a switch to take over control  
of the motor and run it backward with a simple driver.


This would need much more of a modification to the printer, but it it not  
hard to do with printers that have a metal frame and where you can remove  
the plastic case to have a free path.


This would basically create a flatbed printer, gantry style.
It would of course require a dedicated printer to modify.


ST

>could you describe your photo approach?

When the dust settles perhaps.

--- In Homebrew_PCBs@yahoogroups.com, "Jim Miller" <jtmiller2@...> 
wrote:
>