Homebrew PCBs

>
> Although I have been making successful single-sided PCBs for a while now, I just recently decided to try making a double sided board with SMD (.5mm TQFP) parts. So, I went out and bought a new inexpensive Samsung ML-2165 printer. The problem I am having is getting the scaling right. It's inconsistent and varies from print to print. Plain paper works pretty good, but the Pulsar paper comes out too short. I tried adjusting the scaling in the Kicad print dialog, but it seems to come out either too wide or too short. Very frustrating, and I have wasted four sheets of TTS paper so far.
> I have a couple of boxes of HP glossy photo paper, so I may try that.
> 
> Any advice?
> 
> Thanks,
> 
> Pete
>

>
> Although I have been making successful single-sided PCBs for a while now, I just recently decided to try making a double sided board with SMD (.5mm TQFP) parts. So, I went out and bought a new inexpensive Samsung ML-2165 printer. The problem I am having is getting the scaling right. It's inconsistent and varies from print to print. Plain paper works pretty good, but the Pulsar paper comes out too short. I tried adjusting the scaling in the Kicad print dialog, but it seems to come out either too wide or too short. Very frustrating, and I have wasted four sheets of TTS paper so far.
> I have a couple of boxes of HP glossy photo paper, so I may try that.
> 
> Any advice?
> 
> Thanks,
> 
> Pete
>

>
> 
> I use multiple printers and suddenly all of them seemed to develop a similar problem to yours at the same time.
> Turned out it was my son's A/C cycling and pulling the power down on that circuit. On my fastest printer you can actually hear the printer slow in sync with the power drop. It shouldn't effect the scaling but it did.
> My oldest printer doesn't do this at all, so it tells me that manufacturers are probably making the power supplies as cheap as they can.
> I've now got a power stabilizer thing for big home theaters and the problem is gone.
> You may also want to make sure there are no power saving or toner saving settings turned on, no telling how the power or toner is "saved"!
> 
> --- In Homebrew_PCBs@yahoogroups.com, "godzilera" <yah.5.peteben@> wrote:
> >
> > Although I have been making successful single-sided PCBs for a while now, I just recently decided to try making a double sided board with SMD (.5mm TQFP) parts. So, I went out and bought a new inexpensive Samsung ML-2165 printer. The problem I am having is getting the scaling right. It's inconsistent and varies from print to print. Plain paper works pretty good, but the Pulsar paper comes out too short. I tried adjusting the scaling in the Kicad print dialog, but it seems to come out either too wide or too short. Very frustrating, and I have wasted four sheets of TTS paper so far.
> > I have a couple of boxes of HP glossy photo paper, so I may try that.
> > 
> > Any advice?
> > 
> > Thanks,
> > 
> > Pete
> >
>

> From: "godzilera" <yah.5.peteben@...>
> To: "Homebrew PCBs" <Homebrew_PCBs@yahoogroups.com>
> Sent: Monday, June 25, 2012 12:59:27 PM
> Subject: [Homebrew_PCBs] Re: Inconsistent scaling with Samsung ML-2165
> Hello everyone,
> thanks for the responses.
> 
> I forgot to mention that I was only getting scaling problems in the
> 'Y' direction. Speed of the paper through the printer is what affects
> this scaling.
> After a few tests, I have gotten much better results, at least with
> the HP photo paper. I figured that the inexpensive printer was
> struggling to drag the paper out of the bin and into the unit,
> especially since the paper path of this printer is not straight at all.
> So, instead of printing my PCB design at the top of the page, I asked
> the driver to flip the page upside down, so that the design now comes
> out at the bottom.
> I think the printer is struggling to pull out the page and slows down
> at the beginning. Once the paper hits the fuser, I think this
> stabilizes, and the bottom of the page comes out a the right speed. I
> think this is more pronounced with the thicker paper, as there is a
> one inch difference between the position of the PCB layout on a page
> of regular paper and photo paper. This seems to confirm that the
> thicker photo paper moves more slowly through the printer, at least
> initially.
> 
> Thanks,
> 
> 
> --- In Homebrew_PCBs@yahoogroups.com, "Dennis" <nojoeco@...> wrote:
> >
> >
> > I use multiple printers and suddenly all of them seemed to develop a
> > similar problem to yours at the same time.
> > Turned out it was my son's A/C cycling and pulling the power down on
> > that circuit. On my fastest printer you can actually hear the
> > printer slow in sync with the power drop. It shouldn't effect the
> > scaling but it did.
> > My oldest printer doesn't do this at all, so it tells me that
> > manufacturers are probably making the power supplies as cheap as
> > they can.
> > I've now got a power stabilizer thing for big home theaters and the
> > problem is gone.
> > You may also want to make sure there are no power saving or toner
> > saving settings turned on, no telling how the power or toner is
> > "saved"!
> >
> > --- In Homebrew_PCBs@yahoogroups.com, "godzilera" <yah.5.peteben@>
> > wrote:
> > >
> > > Although I have been making successful single-sided PCBs for a
> > > while now, I just recently decided to try making a double sided
> > > board with SMD (.5mm TQFP) parts. So, I went out and bought a new
> > > inexpensive Samsung ML-2165 printer. The problem I am having is
> > > getting the scaling right. It's inconsistent and varies from print
> > > to print. Plain paper works pretty good, but the Pulsar paper
> > > comes out too short. I tried adjusting the scaling in the Kicad
> > > print dialog, but it seems to come out either too wide or too
> > > short. Very frustrating, and I have wasted four sheets of TTS
> > > paper so far.
> > > I have a couple of boxes of HP glossy photo paper, so I may try
> > > that.
> > >
> > > Any advice?
> > >
> > > Thanks,
> > >
> > > Pete
> > >
> >

From: godzilera <yah.5.peteben@neverbox.com>
Subject: [Homebrew_PCBs] Re: Inconsistent scaling with Samsung ML-2165
To: Homebrew_PCBs@yahoogroups.com
Date: Monday, June 25, 2012, 12:59 PM
















 



  


    
      
      
      Hello everyone, 

thanks for the responses.



I forgot to mention that I was only getting scaling problems in the 'Y' direction. Speed of the paper through the printer is what affects this scaling.

After a few tests, I have gotten much better results, at least with the HP photo paper. I figured that the inexpensive printer was struggling to drag the paper out of the bin and into the unit, especially since the paper path of this printer is not straight at all.

So, instead of printing my PCB design at the top of the page, I asked the driver to flip the page upside down, so that the design now comes out at the bottom. 

I think the printer is struggling to pull out the page and slows down at the beginning. Once the paper hits the fuser, I think this stabilizes, and the bottom of the page comes out a the right speed. I think this is more pronounced with the thicker paper, as there is a one inch difference between the position of the PCB layout on a page of regular paper and photo paper. This seems to confirm that the thicker photo paper moves more slowly through the printer, at least initially.



Thanks,









    
     

    
    






  










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

>
> Although I have been making successful single-sided PCBs for a while now, I just recently decided to try making a double sided board with SMD (.5mm TQFP) parts. So, I went out and bought a new inexpensive Samsung ML-2165 printer. The problem I am having is getting the scaling right. It's inconsistent and varies from print to print. Plain paper works pretty good, but the Pulsar paper comes out too short. I tried adjusting the scaling in the Kicad print dialog, but it seems to come out either too wide or too short. Very frustrating, and I have wasted four sheets of TTS paper so far.
> I have a couple of boxes of HP glossy photo paper, so I may try that.
> 
> Any advice?
> 
> Thanks,
> 
> Pete
>

>
> Hello everyone, 
> thanks for the responses.
> 
> I forgot to mention that I was only getting scaling problems in the 'Y' direction. Speed of the paper through the printer is what affects this scaling.
> After a few tests, I have gotten much better results, at least with the HP photo paper. I figured that the inexpensive printer was struggling to drag the paper out of the bin and into the unit, especially since the paper path of this printer is not straight at all.
> So, instead of printing my PCB design at the top of the page, I asked the driver to flip the page upside down, so that the design now comes out at the bottom. 
> I think the printer is struggling to pull out the page and slows down at the beginning. Once the paper hits the fuser, I think this stabilizes, and the bottom of the page comes out a the right speed. I think this is more pronounced with the thicker paper, as there is a one inch difference between the position of the PCB layout on a page of regular paper and photo paper. This seems to confirm that the thicker photo paper moves more slowly through the printer, at least initially.
> 
> Thanks,
> 
> 
> --- In Homebrew_PCBs@yahoogroups.com, "Dennis" <nojoeco@> wrote:
> >
> > 
> > I use multiple printers and suddenly all of them seemed to develop a similar problem to yours at the same time.
> > Turned out it was my son's A/C cycling and pulling the power down on that circuit. On my fastest printer you can actually hear the printer slow in sync with the power drop. It shouldn't effect the scaling but it did.
> > My oldest printer doesn't do this at all, so it tells me that manufacturers are probably making the power supplies as cheap as they can.
> > I've now got a power stabilizer thing for big home theaters and the problem is gone.
> > You may also want to make sure there are no power saving or toner saving settings turned on, no telling how the power or toner is "saved"!
> > 
> > --- In Homebrew_PCBs@yahoogroups.com, "godzilera" <yah.5.peteben@> wrote:
> > >
> > > Although I have been making successful single-sided PCBs for a while now, I just recently decided to try making a double sided board with SMD (.5mm TQFP) parts. So, I went out and bought a new inexpensive Samsung ML-2165 printer. The problem I am having is getting the scaling right. It's inconsistent and varies from print to print. Plain paper works pretty good, but the Pulsar paper comes out too short. I tried adjusting the scaling in the Kicad print dialog, but it seems to come out either too wide or too short. Very frustrating, and I have wasted four sheets of TTS paper so far.
> > > I have a couple of boxes of HP glossy photo paper, so I may try that.
> > > 
> > > Any advice?
> > > 
> > > Thanks,
> > > 
> > > Pete
> > >
> >
>

>
> 
> When I did TT, my process was to print a "calibration board" and go
> through all the steps until just before etching, and measure the pcb
> itself to find out what the X and Y scale factors are.  Each step along
> the way might alter the dimensions somewhat, so it's important to just
> check the results you care about.
> 
> gEDA/PCB, btw, has separate X and Y calibration values you can fill in
> to adjust printouts, just for toner transfer.
>

On 06/28/2012 02:46 AM, godzilera wrote:
> It's not a question of just adjusting the scaling; the problem is that it's inconsistent. Sometimes it would come out too small, sometimes it would come out too big...
>
> However I have now managed to print at least three perfect copies using the HP photo paper by printing my layout at the bottom of the page. This seems to confirm that the speed of the paper is only inconsistent at the top of the page. I think that is because the roller that pulls the paper out of the tray is also responsible for pushing it over the drum, and some slippage occurs with the thicker and glossier paper. Once the paper hits the fuser though, then there is enough pressure to pull the paper with a constant speed.
>
> Anyways, my board has been printed and etched, and it looks perfect. In the future, I think I will probably cut strips of TTS or photo paper and tape it to another sheet to make sure I always print on the bottom of the page without wasting a whole sheet.
>
> In general, I think the straighter the paper path, the better. With the old  HP LJ 4 at the office I used to use the manual feed and open up the back, so that the paper would not curve up back to the front, but that is not possible on this ultra-small Samsung.
>
> For now, the little Samsung at $90 works fine as long as I can work around its limitations.
>
> Thanks,
>
> Pete
>
>
> --- In Homebrew_PCBs@yahoogroups.com, DJ Delorie<dj@...>  wrote:
>>
>> When I did TT, my process was to print a "calibration board" and go
>> through all the steps until just before etching, and measure the pcb
>> itself to find out what the X and Y scale factors are.  Each step along
>> the way might alter the dimensions somewhat, so it's important to just
>> check the results you care about.
>>
>> gEDA/PCB, btw, has separate X and Y calibration values you can fill in
>> to adjust printouts, just for toner transfer.
>>
>
>
>
> ------------------------------------
>
> Be sure to visit the group home and check for new Links, Files, and Photos:
> http://groups.yahoo.com/group/Homebrew_PCBsYahoo! Groups Links
>
>
>

-----Original Message-----
From: Homebrew_PCBs@yahoogroups.com [mailto:Homebrew_PCBs@yahoogroups.com]
On Behalf Of eugenio
Sent: Thursday, June 28, 2012 4:39 AM
To: Homebrew_PCBs@yahoogroups.com
Subject: Re: [Homebrew_PCBs] Re: Inconsistent scaling with Samsung ML-2165

Hi guys, in the last two years i've implemented every imaginable way of
making pcb's at home.
One of these ways was Mark Lerman's direct laserjet printing, when i
finished the printer-machine i realized that the printer was inconsistent in
scaling.
After studying the inners of the machine i saw that the printer was perfect
in the execution, the brushless motor was synced perfectly with laser and
with a perfect constant speed and the clutches were working perfect. The
guilty is the micro-slips of the pcb with the rollers. It can be seen
testing that transparencies are more inconsistent than normal paper for
example.
The easy solution is to harden the springs of the rollers, and the perfect
but difficult solution is to make a carrier with the same speed as the
photosensible roller.
I was able to direct print multiple times in the same pcb and you wouldn't
notice, and i had perfect registration in two sides. I used a fpga to
substitute the sensors stuff of the machine.
For the curious, Mark Lermans's way didn't fit my needs so i discarded it,
but for other problems different that consistency, the most important one
was that toner is not a good resist for plating with 4-5 mils precission.




On 06/28/2012 02:46 AM, godzilera wrote:
> It's not a question of just adjusting the scaling; the problem is that
it's inconsistent. Sometimes it would come out too small, sometimes it would
come out too big...
>
> However I have now managed to print at least three perfect copies using
the HP photo paper by printing my layout at the bottom of the page. This
seems to confirm that the speed of the paper is only inconsistent at the top
of the page. I think that is because the roller that pulls the paper out of
the tray is also responsible for pushing it over the drum, and some slippage
occurs with the thicker and glossier paper. Once the paper hits the fuser
though, then there is enough pressure to pull the paper with a constant
speed.
>
> Anyways, my board has been printed and etched, and it looks perfect. In
the future, I think I will probably cut strips of TTS or photo paper and
tape it to another sheet to make sure I always print on the bottom of the
page without wasting a whole sheet.
>
> In general, I think the straighter the paper path, the better. With the
old  HP LJ 4 at the office I used to use the manual feed and open up the
back, so that the paper would not curve up back to the front, but that is
not possible on this ultra-small Samsung.
>
> For now, the little Samsung at $90 works fine as long as I can work around
its limitations.
>
> Thanks,
>
> Pete
>
>
> --- In Homebrew_PCBs@yahoogroups.com, DJ Delorie<dj@...>  wrote:
>>
>> When I did TT, my process was to print a "calibration board" and go 
>> through all the steps until just before etching, and measure the pcb 
>> itself to find out what the X and Y scale factors are.  Each step 
>> along the way might alter the dimensions somewhat, so it's important 
>> to just check the results you care about.
>>
>> gEDA/PCB, btw, has separate X and Y calibration values you can fill 
>> in to adjust printouts, just for toner transfer.
>>
>
>
>
> ------------------------------------
>
> Be sure to visit the group home and check for new Links, Files, and
Photos:
> http://groups.yahoo.com/group/Homebrew_PCBsYahoo! Groups Links
>
>
>



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

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

>
> Possibly a printer like the old Espon Stylus 3000 which has a tractor feed
> mechanism could produce more accurate direct prints? I would not be able to
> modify and re-engineer the printer to work for this application, but it
> seems you would be able to do it. 
> 
> -----Original Message-----
> From: Homebrew_PCBs@yahoogroups.com [mailto:Homebrew_PCBs@yahoogroups.com]
> On Behalf Of eugenio
> Sent: Thursday, June 28, 2012 4:39 AM
> To: Homebrew_PCBs@yahoogroups.com
> Subject: Re: [Homebrew_PCBs] Re: Inconsistent scaling with Samsung ML-2165
> 
> Hi guys, in the last two years i've implemented every imaginable way of
> making pcb's at home.
> One of these ways was Mark Lerman's direct laserjet printing, when i
> finished the printer-machine i realized that the printer was inconsistent in
> scaling.
> After studying the inners of the machine i saw that the printer was perfect
> in the execution, the brushless motor was synced perfectly with laser and
> with a perfect constant speed and the clutches were working perfect. The
> guilty is the micro-slips of the pcb with the rollers. It can be seen
> testing that transparencies are more inconsistent than normal paper for
> example.
> The easy solution is to harden the springs of the rollers, and the perfect
> but difficult solution is to make a carrier with the same speed as the
> photosensible roller.
> I was able to direct print multiple times in the same pcb and you wouldn't
> notice, and i had perfect registration in two sides. I used a fpga to
> substitute the sensors stuff of the machine.
> For the curious, Mark Lermans's way didn't fit my needs so i discarded it,
> but for other problems different that consistency, the most important one
> was that toner is not a good resist for plating with 4-5 mils precission.
> 
> 
> 
> 
> On 06/28/2012 02:46 AM, godzilera wrote:
> > It's not a question of just adjusting the scaling; the problem is that
> it's inconsistent. Sometimes it would come out too small, sometimes it would
> come out too big...
> >
> > However I have now managed to print at least three perfect copies using
> the HP photo paper by printing my layout at the bottom of the page. This
> seems to confirm that the speed of the paper is only inconsistent at the top
> of the page. I think that is because the roller that pulls the paper out of
> the tray is also responsible for pushing it over the drum, and some slippage
> occurs with the thicker and glossier paper. Once the paper hits the fuser
> though, then there is enough pressure to pull the paper with a constant
> speed.
> >
> > Anyways, my board has been printed and etched, and it looks perfect. In
> the future, I think I will probably cut strips of TTS or photo paper and
> tape it to another sheet to make sure I always print on the bottom of the
> page without wasting a whole sheet.
> >
> > In general, I think the straighter the paper path, the better. With the
> old  HP LJ 4 at the office I used to use the manual feed and open up the
> back, so that the paper would not curve up back to the front, but that is
> not possible on this ultra-small Samsung.
> >
> > For now, the little Samsung at $90 works fine as long as I can work around
> its limitations.
> >
> > Thanks,
> >
> > Pete
> >
> >
> > --- In Homebrew_PCBs@yahoogroups.com, DJ Delorie<dj@>  wrote:
> >>
> >> When I did TT, my process was to print a "calibration board" and go 
> >> through all the steps until just before etching, and measure the pcb 
> >> itself to find out what the X and Y scale factors are.  Each step 
> >> along the way might alter the dimensions somewhat, so it's important 
> >> to just check the results you care about.
> >>
> >> gEDA/PCB, btw, has separate X and Y calibration values you can fill 
> >> in to adjust printouts, just for toner transfer.
> >>
> >
> >
> >
> > ------------------------------------
> >
> > Be sure to visit the group home and check for new Links, Files, and
> Photos:
> > http://groups.yahoo.com/group/Homebrew_PCBsYahoo! Groups Links
> >
> >
> >
> 
> 
> 
> ------------------------------------
> 
> Be sure to visit the group home and check for new Links, Files, and Photos:
> http://groups.yahoo.com/group/Homebrew_PCBsYahoo! Groups Links
>

Subject: [Homebrew_PCBs] pcb-specific laminator/exposure/etc printer...
To: Homebrew_PCBs@yahoogroups.com
Date: Thursday, June 28, 2012, 1:36 PM
















 



  


    
      
      
      

Is it time for someone to purpose-design a direct-to-pcb printer?

Probably way too expensive to consider, but still...



Maybe something that laminates UV film on, and has a linear array of

high power UV LEDs to expose the film as it's being laminated, on both

sides...  Include an infeed tray with registration pins for pre-drilled

PCBs...  pre-drill, drop on pins, press button, when pcb comes out the

other side you wait 15 minutes, peel off the outer protective film, and

develop.  Maybe it could laminate (or just use a sticky-backed film) on

a UV-opaque film over the UV film after exposure, so you don't even need

a darkroom :-) Add a tiny inkjet head on the way out to print timestamp

and job information, and you're done!



ok, blue-skying over :-)



    
     







  










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

On 06/28/2012 08:36 PM, DJ Delorie wrote:
> Is it time for someone to purpose-design a direct-to-pcb printer?
> Probably way too expensive to consider, but still...
>
> Maybe something that laminates UV film on, and has a linear array of
> high power UV LEDs to expose the film as it's being laminated, on both
> sides...  Include an infeed tray with registration pins for pre-drilled
> PCBs...  pre-drill, drop on pins, press button, when pcb comes out the
> other side you wait 15 minutes, peel off the outer protective film, and
> develop.  Maybe it could laminate (or just use a sticky-backed film) on
> a UV-opaque film over the UV film after exposure, so you don't even need
> a darkroom :-) Add a tiny inkjet head on the way out to print timestamp
> and job information, and you're done!
>
> ok, blue-skying over :-)
>
>
> ------------------------------------
>
> Be sure to visit the group home and check for new Links, Files, and Photos:
> http://groups.yahoo.com/group/Homebrew_PCBsYahoo! Groups Links
>
>
>

>
> 
> 
> --- In Homebrew_PCBs@yahoogroups.com, DJ Delorie <dj@> wrote:
> >
> > 
> > Is it time for someone to purpose-design a direct-to-pcb printer?
> > Probably way too expensive to consider, but still...
> > 
> > Maybe something that laminates UV film on, and has a linear array of
> > high power UV LEDs to expose the film as it's being laminated, on both
> > sides...  Include an infeed tray with registration pins for pre-drilled
> > PCBs...  pre-drill, drop on pins, press button, when pcb comes out the
> > other side you wait 15 minutes, peel off the outer protective film, and
> > develop.  Maybe it could laminate (or just use a sticky-backed film) on
> > a UV-opaque film over the UV film after exposure, so you don't even need
> > a darkroom :-) Add a tiny inkjet head on the way out to print timestamp
> > and job information, and you're done!
> > 
> > ok, blue-skying over :-)
> >
> 
> I suspect it will be difficult to come up with something that will be affordable yet offer decent performance.
> 
> I'm intrigued at the thought of a laser photoplotter that would cure UV resist. Either an X-Y plotter table with a 405nm laser in place of a pen, or a scanner assembly from a laser printer with a feeder to run the board through.
> 
> Really though I've got the laser printer toner transfer process pretty well dialed in, and for any time I want more than one board and don't mind waiting, Seeed Studio is ridiculously cheap so it's easier to just send off for a batch after I've etched a prototype at home.
>

>
> I like the idea of an area of LEDs.  I wonder what kind of resolution /
> LEDs/inch one could get.
> 
> Sent from my iPad
> 
> On 2012-06-29, at 8:30 AM, designer_craig <cs6061@...> wrote:
> 
> 
> 
> Janes,
> The UV laser has been done -- check back in the archives Zoltan built a
> system basically replacing the print head with a UV laser diode and raster
> scanning the board. It was a little slow but had stunning resolution and
> chrispness in the image. It's on my project list, just not near the top.
> 
> Craig
> 
> --- In Homebrew_PCBs@...m, "James" <jamesrsweet@> wrote:
> >
> >
> >
> > --- In Homebrew_PCBs@yahoogroups.com, DJ Delorie <dj@> wrote:
> > >
> > >
> > > Is it time for someone to purpose-design a direct-to-pcb printer?
> > > Probably way too expensive to consider, but still...
> > >
> > > Maybe something that laminates UV film on, and has a linear array of
> > > high power UV LEDs to expose the film as it's being laminated, on both
> > > sides... Include an infeed tray with registration pins for pre-drilled
> > > PCBs... pre-drill, drop on pins, press button, when pcb comes out the
> > > other side you wait 15 minutes, peel off the outer protective film, and
> > > develop. Maybe it could laminate (or just use a sticky-backed film) on
> > > a UV-opaque film over the UV film after exposure, so you don't even need
> > > a darkroom :-) Add a tiny inkjet head on the way out to print timestamp
> > > and job information, and you're done!
> > >
> > > ok, blue-skying over :-)
> > >
> >
> > I suspect it will be difficult to come up with something that will be
> affordable yet offer decent performance.
> >
> > I'm intrigued at the thought of a laser photoplotter that would cure UV
> resist. Either an X-Y plotter table with a 405nm laser in place of a pen,
> or a scanner assembly from a laser printer with a feeder to run the board
> through.
> >
> > Really though I've got the laser printer toner transfer process pretty
> well dialed in, and for any time I want more than one board and don't mind
> waiting, Seeed Studio is ridiculously cheap so it's easier to just send off
> for a batch after I've etched a prototype at home.
> >
> 
>  
> 
> 
> [Non-text portions of this message have been removed]
>

 From: Benjamin Blumer <b3nzilla@...>
To: "Homebrew_PCBs@yahoogroups.com" <Homebrew_PCBs@yahoogroups.com> 
Sent: Saturday, June 30, 2012 11:41 AM
Subject: Re: [Homebrew_PCBs] Re: pcb-specific laminator/exposure/etc printer...
 
I was thinking a single row of LEDs with the board being carried through by
push wheels.  Not quite as snazzy as one LED/pixel, though more feasible
perhaps.  Though, I don't really know how feasible it is to have a row of
2000+ Individually addressable LEDs anyways.

Although,  this does remind me of the 3D printers that harden layer by
layer of plastic using a projector.  Though, it would mean we'd need a
photoresist sensitive to visible light.

Sent from my iPad

On 2012-06-30, at 9:25 AM, AlienRelics <alienrelics@yahoo.com> wrote:



A few people here did some tests with toner transfer to determine if 300dpi
or 600dpi Laser printer resolution was sufficient.

The end result was that 600dpi was required.

So to make a board 2x3 inches, you'll need 2x3x600 = 30,000 LEDs. And
they'll need to be about a thousandth of an inch square in size.

Steve Greenfield AE7HD

--- In Homebrew_PCBs@yahoogroups.com, Benjamin Blumer <b3nzilla@...> wrote:
>
> I like the idea of an area of LEDs. I wonder what kind of resolution /
> LEDs/inch one could get.
>
> Sent from my iPad
>
> On 2012-06-29, at 8:30 AM, designer_craig <cs6061@...> wrote:
>
>
>
> Janes,
> The UV laser has been done -- check back in the archives Zoltan built a
> system basically replacing the print head with a UV laser diode and raster
> scanning the board. It was a little slow but had stunning resolution and
> chrispness in the image. It's on my project list, just not near the top.
>
> Craig
>
> --- In Homebrew_PCBs@yahoogroups.com, "James" <jamesrsweet@> wrote:
> >
> >
> >
> > --- In Homebrew_PCBs@yahoogroups.com, DJ Delorie <dj@> wrote:
> > >
> > >
> > > Is it time for someone to purpose-design a direct-to-pcb printer?
> > > Probably way too expensive to consider, but still...
> > >
> > > Maybe something that laminates UV film on, and has a linear array of
> > > high power UV LEDs to expose the film as it's being laminated, on both
> > > sides... Include an infeed tray with registration pins for pre-drilled
> > > PCBs... pre-drill, drop on pins, press button, when pcb comes out the
> > > other side you wait 15 minutes, peel off the outer protective film,
and
> > > develop. Maybe it could laminate (or just use a sticky-backed film) on
> > > a UV-opaque film over the UV film after exposure, so you don't even
need
> > > a darkroom :-) Add a tiny inkjet head on the way out to print
timestamp
> > > and job information, and you're done!
> > >
> > > ok, blue-skying over :-)
> > >
> >
> > I suspect it will be difficult to come up with something that will be
> affordable yet offer decent performance.
> >
> > I'm intrigued at the thought of a laser photoplotter that would cure UV
> resist. Either an X-Y plotter table with a 405nm laser in place of a pen,
> or a scanner assembly from a laser printer with a feeder to run the board
> through.
> >
> > Really though I've got the laser printer toner transfer process pretty
> well dialed in, and for any time I want more than one board and don't mind
> waiting, Seeed Studio is ridiculously cheap so it's easier to just send
off
> for a batch after I've etched a prototype at home.
> >
>
>
>
>
> [Non-text portions of this message have been removed]
>




[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_PCBsYahoo! Groups Links



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

>
> On Sat, 30 Jun 2012 12:41:50 -0600, you wrote:
> 
> >I was thinking a single row of LEDs with the board being carried through by
> >push wheels.  Not quite as snazzy as one LED/pixel, though more feasible
> >perhaps.  Though, I don't really know how feasible it is to have a row of
> >2000+ Individually addressable LEDs anyways.
> >
> >Although,  this does remind me of the 3D printers that harden layer by
> >layer of plastic using a projector.  Though, it would mean we'd need a
> >photoresist sensitive to visible light.
> >
> >Sent from my iPad
> >
> >
> 
> What says you can't space the LEDs farther apart across the narrow dimension and
> place the board in a carrier? Then make multiple passes with the carrier along 
> the long dimension stepping the LED assembly sideways a small increment each
>  pass.  This would require good tolerance/repeatability on the long dimension (which 
> wouldn't be difficult as it would only move back and forth). You get the same 
> result as a high density line of LEDs, just slower.
> 
> Writing the driver software would be interesting too 8=)
> 
> If you are into machining think of the way an automatic surface grinder works.
> Keith Bowers WB4LSJ- Thomasville, NC
>

From: Homebrew_PCBs@yahoogroups.com [mailto:Homebrew_PCBs@yahoogroups.com]
On Behalf Of Tony Smith
Sent: Sunday, July 01, 2012 1:14 AM
To: Homebrew_PCBs@yahoogroups.com
Subject: RE: [Homebrew_PCBs] Re: pcb-specific laminator/exposure/etc
printer...

 

  

> I had a Brother Printer that was an LED printer about 10 years ago, it
worked
> great. I used it for years. It had a bar that went across the paper with 4
rows of
> LEDs. Very rugged machine, my guess it would expose UV film also. The LEDs
> looked red but I think they put out white light to fuse the toner to the
paper.
> Check around you may be able to find a used one to hack.

Surprisingly they still make those, they don't seem to be very common
though. They've been around for a while, I remember Brother & OkiData
models from over 20 years ago.

They work the same way as laser printers, except without the need to scan
the laser back & forth (so no moving parts). In theory that makes them more
reliable and faster, but they've always been a bit of an also-ran. I know
they'll do 1200 DPI, so the quality is ok. I don't know why they never
really took off, licencing issues maybe?

I've never bother to have a look at the innards of one. I assume the
technology is similar to the LED scanners you can get (Canon LIDE etc); LED
chips are pretty small so you can cram a few in there, but 1200 of them per
inch is rather impressive, especially across 8-11 inches or so. Staggered
maybe? Driving one of those would be an 'interesting' challenge, at 8"
that's 10,000 LEDs (triple for colour?)

There are also LCD printers, those just use a 'shutter' to control the
light. I haven't seen one of those for ages.

Tony





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

>
> In "Re: pcb-specific laminator/exposure/etc printer...", on 2012-07-01,
> tda7000 wrote:
> 
> > Much easier to do this, I would think:
> > 
> > http://www.youtube.com/watch?v=fi4P-Bwc6g8
> > 
> > http://www.das-labor.org/wiki/LaserExposer
> 
> This looks difficult to do, by most standards, since there is customised
> hardware and two microcontrollers.  There would be significant problems
> calibrating it, correcting for the out-of-focus laser beam and the
> non-linear sweep across the board.
> 
> Also, I think it would be difficult to align exposures for double-sided
> compared to the alternative of aligning two phototools (or perhaps
> transfer paper with toner) facing each other, and then slipping the PCB
> material in between them.
> 
> I think it is a really great project, but I don't feel inclined to do
> all this work myself.
> 
>   - Robin
>

>
> Robin, thanks for providing the links for an outstanding Laser Direct Imaging project. Does anyone know of any similar projects?
> 
> Wayne
> 
> --- In Homebrew_PCBs@yahoogroups.com, Robin Whittle <rw@> wrote:
> >
> > In "Re: pcb-specific laminator/exposure/etc printer...", on 2012-07-01,
> > tda7000 wrote:
> > 
> > > Much easier to do this, I would think:
> > > 
> > > http://www.youtube.com/watch?v=fi4P-Bwc6g8
> > > 
> > > http://www.das-labor.org/wiki/LaserExposer
> > 
> > This looks difficult to do, by most standards, since there is customised
> > hardware and two microcontrollers.  There would be significant problems
> > calibrating it, correcting for the out-of-focus laser beam and the
> > non-linear sweep across the board.
> > 
> > Also, I think it would be difficult to align exposures for double-sided
> > compared to the alternative of aligning two phototools (or perhaps
> > transfer paper with toner) facing each other, and then slipping the PCB
> > material in between them.
> > 
> > I think it is a really great project, but I don't feel inclined to do
> > all this work myself.
> > 
> >   - Robin
> >
>

> Robin, thanks for providing the links for an outstanding Laser
> Direct Imaging project. Does anyone know of any similar projects?
> 
> Wayne

>
> I believe I posted the links about the German project? ...Anyway...
> 
> Another idea was here: http://4hv.org/e107_plugins/forum/forum_viewtopic.php?66471
> 
> The first guy seems to have had an idea of using a grid of LEDs, but I don't think you can get them small enough.
> 
> However, in the 2nd half of that thread, there was another guy with a UV laser diode on a plotter table, the photos show that it looks like the board came out pretty well.
> 
> This is something I thought of trying out, but it looks expensive. Also double-sided would be a pain.
> 
> Laser-printed transparencies and toner-transfer still works well for me so far...
> 
> 
> --- In Homebrew_PCBs@yahoogroups.com, "wayne.causey" <wayne.causey@> wrote:
> >
> > Robin, thanks for providing the links for an outstanding Laser Direct Imaging project. Does anyone know of any similar projects?
> > 
> > Wayne
> > 
> > --- In Homebrew_PCBs@yahoogroups.com, Robin Whittle <rw@> wrote:
> > >
> > > In "Re: pcb-specific laminator/exposure/etc printer...", on 2012-07-01,
> > > tda7000 wrote:
> > > 
> > > > Much easier to do this, I would think:
> > > > 
> > > > http://www.youtube.com/watch?v=fi4P-Bwc6g8
> > > > 
> > > > http://www.das-labor.org/wiki/LaserExposer
> > > 
> > > This looks difficult to do, by most standards, since there is customised
> > > hardware and two microcontrollers.  There would be significant problems
> > > calibrating it, correcting for the out-of-focus laser beam and the
> > > non-linear sweep across the board.
> > > 
> > > Also, I think it would be difficult to align exposures for double-sided
> > > compared to the alternative of aligning two phototools (or perhaps
> > > transfer paper with toner) facing each other, and then slipping the PCB
> > > material in between them.
> > > 
> > > I think it is a really great project, but I don't feel inclined to do
> > > all this work myself.
> > > 
> > >   - Robin
> > >
> >
>

>
> tda7000,
> 
> Thanks for all three links concerning Laser Direct Imaging for PCBs. Any further links/sources for info on this topic will also be appreciated.
> 
> Thanks.
> 
> Wayne
> 
> --- In Homebrew_PCBs@yahoogroups.com, "tda7000" <Tda7000@> wrote:
> >
> > I believe I posted the links about the German project? ...Anyway...
> > 
> > Another idea was here: http://4hv.org/e107_plugins/forum/forum_viewtopic.php?66471
> > 
> > The first guy seems to have had an idea of using a grid of LEDs, but I don't think you can get them small enough.
> > 
> > However, in the 2nd half of that thread, there was another guy with a UV laser diode on a plotter table, the photos show that it looks like the board came out pretty well.
> > 
> > This is something I thought of trying out, but it looks expensive. Also double-sided would be a pain.
> > 
> > Laser-printed transparencies and toner-transfer still works well for me so far...
> > 
> > 
> > --- In Homebrew_PCBs@yahoogroups.com, "wayne.causey" <wayne.causey@> wrote:
> > >
> > > Robin, thanks for providing the links for an outstanding Laser Direct Imaging project. Does anyone know of any similar projects?
> > > 
> > > Wayne
> > > 
> > > --- In Homebrew_PCBs@yahoogroups.com, Robin Whittle <rw@> wrote:
> > > >
> > > > In "Re: pcb-specific laminator/exposure/etc printer...", on 2012-07-01,
> > > > tda7000 wrote:
> > > > 
> > > > > Much easier to do this, I would think:
> > > > > 
> > > > > http://www.youtube.com/watch?v=fi4P-Bwc6g8
> > > > > 
> > > > > http://www.das-labor.org/wiki/LaserExposer
> > > > 
> > > > This looks difficult to do, by most standards, since there is customised
> > > > hardware and two microcontrollers.  There would be significant problems
> > > > calibrating it, correcting for the out-of-focus laser beam and the
> > > > non-linear sweep across the board.
> > > > 
> > > > Also, I think it would be difficult to align exposures for double-sided
> > > > compared to the alternative of aligning two phototools (or perhaps
> > > > transfer paper with toner) facing each other, and then slipping the PCB
> > > > material in between them.
> > > > 
> > > > I think it is a really great project, but I don't feel inclined to do
> > > > all this work myself.
> > > > 
> > > >   - Robin
> > > >
> > >
> >
>

>
> Where do you get UV SMD LEDs.
> They seem very expensive?
> What wavelength should I look for?
> 
> Duane
> 
> -- 
>      Home of the $35 Solar Tracker      Receiver
>     http://www.redrok.com/led3xassm.htm      [*]
>    Powered by             \  \     \        //|
>   Thermonuclear   Solar Energy from the Sun / |
> Energy (the SUN)           \  \     \  /  /  |
> Red Rock Energy             \  \     /   /   |
> Duane C. Johnson   Designer  \  \  /  \ /    |
> 1825 Florence St  Heliostat,Control,& Mounts |
> White Bear Lake, Minnesota    === \   / \    |
> USA      55110-3364                ===   \   |
> (651)426-4766        use Courier New Font \  |
> redrok@...     (my email: address)  \ |
> http://www.redrok.com  (Web site)          ===
>