Homebrew PCBs

I have been away for awhile and noticed several recent posts investigating various printers and inks for making really dark transparencies for photo exposure masks.

I have been using the positive photoresist method for more than 10 years.  Initially, I used an HP inkjet on a high quality (expensive) transparency medium produced for commerical artists.  That worked well, but infrequent use led to dried ink cartridges and all the problems associated with that.

More recently, I simply make my transparencies with a laserjet (HP 4101).  It doesn't print particularly dark.   I use 3M transparency CG3300.  That probably doesn't matter.  After printing, I let the transparency cool and then go over the entire image with a black, dry erase marker.  I use Expo brand.  Try to do it with single, overlapping strokes. The solvents in the marker will soften the toner image slightly. Don't scrub the marker back and forth over the transparency.  Let it dry thoroughly (10 minutes), then I use folded facial tissue (Kleenex brand) to remove the dry erase. I usually make a first couple of swipes to remove the bulk of the excess, then open a clean surface and remove the rest.  The dry erase sticks in the crevices and areas between toner particles, but removes cleanly from the clear areas. You can rub pretty hard to remove last little bits that might be stuck around thermals or other very fine details.  In my experience, colored dry erase will not work, as the dyes used in them are not as easily removed.

I will try to upload a picture of a recent board I did for a QFN package.  Traces on the QFN sub-board are 16 mil.  On the main board they are 24 mil, and you can see that the letters in my name and date are considerable smaller.

John

Aside from obvious concerns like dimensional stability for really high
resolution work, why not use a fresh print on the lightest weight paper
your printer will feed? 

Sure, the paper will act as a UV attenuator but it's not going to block
a lot of it.  It should let you get a sufficiently uniform exposure for
most work, particularly if you dither the UV source position a bit to
compensate for the slight density variations in the exposed media. Most
resist is a high contrast media, so slight density variations probably
won't matter.

I expect dimensional stability won't be as big an issue if the tool is
used soon after printer.

-- 
mailto:lists@...     http://www.n1oz.net       ARS: N1OZ
Rich Osman;  POB 93167; Southlake, TX 76092 (Near DFW Airport)

It can be done, they even sell a spray that makes the paper more
transparent (I think orange peel oil, according to the smell).

ST

On Sun, Jan 8, 2012 at 4:20 PM, Rich Osman <lists@...> wrote:
> Aside from obvious concerns like dimensional stability for really high
> resolution work, why not use a fresh print on the lightest weight paper
> your printer will feed?
>
> Sure, the paper will act as a UV attenuator but it's not going to block
> a lot of it.  It should let you get a sufficiently uniform exposure for
> most work, particularly if you dither the UV source position a bit to
> compensate for the slight density variations in the exposed media. Most
> resist is a high contrast media, so slight density variations probably
> won't matter.
>
> I expect dimensional stability won't be as big an issue if the tool is
> used soon after printer.
>
> --
> mailto:lists@...     http://www.n1oz.net       ARS: N1OZ
> Rich Osman;  POB 93167; Southlake, TX 76092 (Near DFW Airport)
>
>
>
>
>
>
> ------------------------------------
>
> Be sure to visit the group home and check for new Links, Files, and Photos:
> http://groups.yahoo.com/group/Homebrew_PCBsYahoo! Groups Links
>
>
>

On 08/01/2012 15:20, Rich Osman wrote:
> Aside from obvious concerns like dimensional stability for really high
> resolution work, why not use a fresh print on the lightest weight paper
> your printer will feed?
>
> Sure, the paper will act as a UV attenuator but it's not going to block
> a lot of it. It should let you get a sufficiently uniform exposure for
> most work, particularly if you dither the UV source position a bit to
> compensate for the slight density variations in the exposed media. Most
> resist is a high contrast media, so slight density variations probably
> won't matter.
>
> I expect dimensional stability won't be as big an issue if the tool is
> used soon after printer.


There is an old technique that I tried once, which involved making the 
paper translucent with vegetable oil after printing. It worked fairly 
well with wide tracks, but was a bit messy.

Leon
-- 
Leon Heller
G1HSM

I have watched the discussions on printers and their limitations on this
forum.
I have seen the use of DLP projectors to harden resin in 3D printers.
I was wondering if a similar adaption will be effective in exposing
photoresist material at very high resolutions.
Maybe use high resolution CRT or LCD monitor as a contrasting source of PCB
image. A PCB could be prepared and taped on the monitor for exposed.
...raj

On Sun, Jan 8, 2012 at 9:53 AM, jpanhalt <janhalt@...> wrote:

> **
>
>
> I have been away for awhile and noticed several recent posts investigating
> various printers and inks for making really dark transparencies for photo
> exposure masks.
>
> I have been using the positive photoresist method for more than 10 years.
> Initially, I used an HP inkjet on a high quality (expensive) transparency
> medium produced for commerical artists. That worked well, but infrequent
> use led to dried ink cartridges and all the problems associated with that.
>
> More recently, I simply make my transparencies with a laserjet (HP 4101).
> It doesn't print particularly dark. I use 3M transparency CG3300. That
> probably doesn't matter. After printing, I let the transparency cool and
> then go over the entire image with a black, dry erase marker. I use Expo
> brand. Try to do it with single, overlapping strokes. The solvents in the
> marker will soften the toner image slightly. Don't scrub the marker back
> and forth over the transparency. Let it dry thoroughly (10 minutes), then I
> use folded facial tissue (Kleenex brand) to remove the dry erase. I usually
> make a first couple of swipes to remove the bulk of the excess, then open a
> clean surface and remove the rest. The dry erase sticks in the crevices and
> areas between toner particles, but removes cleanly from the clear areas.
> You can rub pretty hard to remove last little bits that might be stuck
> around thermals or other very fine details. In my experience, colored dry
> erase will not work, as the dyes used in them are not as easily removed.
>
> I will try to upload a picture of a recent board I did for a QFN package.
> Traces on the QFN sub-board are 16 mil. On the main board they are 24 mil,
> and you can see that the letters in my name and date are considerable
> smaller.
>
> John
>
>  
>



-- 
Imagine no possessions
I wonder if you can
No need for greed or hunger
A brotherhood of man
Imagine all the people
Sharing all the world ... John Lennon


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

A very high resolution projector or monitor would be 1280x1024, do you agree?

Some here have tested Laser printers and found 600dpi models to be wanting, 1200dpi being de rigueur.

So that is only about 2x1.5 at 600dpi, 1x0.8 at 1200dpi.

A monitor is normally around 72dpi to 96dpi.

A clever thought, however.

Steve Greenfield AE7HD


--- In Homebrew_PCBs@yahoogroups.com, Raj Jain <r.k.jain.or@...> wrote:

>
> I have watched the discussions on printers and their limitations on this
> forum.
> I have seen the use of DLP projectors to harden resin in 3D printers.
> I was wondering if a similar adaption will be effective in exposing
> photoresist material at very high resolutions.
> Maybe use high resolution CRT or LCD monitor as a contrasting source of PCB
> image. A PCB could be prepared and taped on the monitor for exposed.
> ...raj
>

I Think its definitely doable but complex, the main obstacle will be finding the optics and necessary calculatons  to proyect the 96dpi image onto the board with this image scaled down to get 600 or more dpi, from there the board could be moved in x and y in a raster way. 
I think it would be a little bit easier to adapt a cnc to work as a laser raster scanner, see bungard, they are doing exactly this idea.




--- In Homebrew_PCBs@...m, "AlienRelics" <alienrelics@...> wrote:

>
> A very high resolution projector or monitor would be 1280x1024, do you agree?
> 
> Some here have tested Laser printers and found 600dpi models to be wanting, 1200dpi being de rigueur.
> 
> So that is only about 2x1.5 at 600dpi, 1x0.8 at 1200dpi.
> 
> A monitor is normally around 72dpi to 96dpi.
> 
> A clever thought, however.
> 
> Steve Greenfield AE7HD
> 
> 
> --- In Homebrew_PCBs@yahoogroups.com, Raj Jain <r.k.jain.or@> wrote:
> >
> > I have watched the discussions on printers and their limitations on this
> > forum.
> > I have seen the use of DLP projectors to harden resin in 3D printers.
> > I was wondering if a similar adaption will be effective in exposing
> > photoresist material at very high resolutions.
> > Maybe use high resolution CRT or LCD monitor as a contrasting source of PCB
> > image. A PCB could be prepared and taped on the monitor for exposed.
> > ...raj
> >
>

I really like this idea myself. Could even try the route of a diy lcd 
projector:
http://users.telenet.be/cool_things_with_fresnel_lenses/LCD_gettingstarted.htm
for instance, just replace the projection lens with a perspex table and 
use a fresnel that images clearly onto it. The biggest problems would be 
the contrast of the blacks and the sharpness of the edges, but I can't 
help but think that if the image is clear and crisp on screen it could 
be clear on the pcb once it's reduced down from full screen to small 
enough pcb. But yeah, it's never going to be as accurate as a really 
good high res transparency, but if you just need a fairly simple pcb 
it'd be very quick to use.

Certainly, a laser based pico projector could be even better, depending 
on minimum dot size / resolution. It would also depend on whether the 
blue laser is bright enough / high enough wavelength to expose the 
resist. just put a ruler in your image beside the pcb, and sit a real 
ruler beside your real pcb, and use that to get the right distance / 
projection size.

With either of these techniques, you could show the pcb image dimly in 
red initially to line up the pcb, then switch it to blue/white full 
brightness to hopefully expose the resist.

Sure, a cnc with a blue / uv laser could/should work even better, but it 
would probably be slower, possibly harder to align, and requires a cnc 
bed. As much as I would like one of them, I don't have one.

Andrew

On 9/01/2012 11:18 PM, psykhon@... wrote:
> I Think its definitely doable but complex, the main obstacle will be finding the optics and necessary calculatons  to proyect the 96dpi image onto the board with this image scaled down to get 600 or more dpi, from there the board could be moved in x and y in a raster way.
> I think it would be a little bit easier to adapt a cnc to work as a laser raster scanner, see bungard, they are doing exactly this idea.
>
>
>
>
> --- In Homebrew_PCBs@yahoogroups.com, "AlienRelics"<alienrelics@...>  wrote:
>> A very high resolution projector or monitor would be 1280x1024, do you agree?
>>
>> Some here have tested Laser printers and found 600dpi models to be wanting, 1200dpi being de rigueur.
>>
>> So that is only about 2x1.5 at 600dpi, 1x0.8 at 1200dpi.
>>
>> A monitor is normally around 72dpi to 96dpi.
>>
>> A clever thought, however.
>>
>> Steve Greenfield AE7HD
>>
>>
>> --- In Homebrew_PCBs@yahoogroups.com, Raj Jain<r.k.jain.or@>  wrote:
>>> I have watched the discussions on printers and their limitations on this
>>> forum.
>>> I have seen the use of DLP projectors to harden resin in 3D printers.
>>> I was wondering if a similar adaption will be effective in exposing
>>> photoresist material at very high resolutions.
>>> Maybe use high resolution CRT or LCD monitor as a contrasting source of PCB
>>> image. A PCB could be prepared and taped on the monitor for exposed.
>>> ...raj
>>>
>>

Ligth intensity in my experience is only  a problem on white solder mask, for evrething else, if in this case the proyector is too dim, you just expose a little bit longer:-)  the test metodology is very simple
When i talk about a cnc , its because I think that the downscaling of the image should be enough to achive a dpi resolution that can match a photoploter resolution wich are about 2400 dpi for the lower range oneswicth enough resolution for pcb procesing. So you proyect a small squarean then the cnc performs a raster scann on the pcb.

Why do you bother with pcb registration? A few tooling pins should suffice
--- In Homebrew_PCBs@yahoogroups.com, Andrew Leech <coronasensei@...> wrote:

>
> I really like this idea myself. Could even try the route of a diy lcd 
> projector:
> http://users.telenet.be/cool_things_with_fresnel_lenses/LCD_gettingstarted.htm
> for instance, just replace the projection lens with a perspex table and 
> use a fresnel that images clearly onto it. The biggest problems would be 
> the contrast of the blacks and the sharpness of the edges, but I can't 
> help but think that if the image is clear and crisp on screen it could 
> be clear on the pcb once it's reduced down from full screen to small 
> enough pcb. But yeah, it's never going to be as accurate as a really 
> good high res transparency, but if you just need a fairly simple pcb 
> it'd be very quick to use.
> 
> Certainly, a laser based pico projector could be even better, depending 
> on minimum dot size / resolution. It would also depend on whether the 
> blue laser is bright enough / high enough wavelength to expose the 
> resist. just put a ruler in your image beside the pcb, and sit a real 
> ruler beside your real pcb, and use that to get the right distance / 
> projection size.
> 
> With either of these techniques, you could show the pcb image dimly in 
> red initially to line up the pcb, then switch it to blue/white full 
> brightness to hopefully expose the resist.
> 
> Sure, a cnc with a blue / uv laser could/should work even better, but it 
> would probably be slower, possibly harder to align, and requires a cnc 
> bed. As much as I would like one of them, I don't have one.
> 
> Andrew
> 
> On 9/01/2012 11:18 PM, psykhon@... wrote:
> > I Think its definitely doable but complex, the main obstacle will be finding the optics and necessary calculatons  to proyect the 96dpi image onto the board with this image scaled down to get 600 or more dpi, from there the board could be moved in x and y in a raster way.
> > I think it would be a little bit easier to adapt a cnc to work as a laser raster scanner, see bungard, they are doing exactly this idea.
> >
> >
> >
> >
> > --- In Homebrew_PCBs@yahoogroups.com, "AlienRelics"<alienrelics@>  wrote:
> >> A very high resolution projector or monitor would be 1280x1024, do you agree?
> >>
> >> Some here have tested Laser printers and found 600dpi models to be wanting, 1200dpi being de rigueur.
> >>
> >> So that is only about 2x1.5 at 600dpi, 1x0.8 at 1200dpi.
> >>
> >> A monitor is normally around 72dpi to 96dpi.
> >>
> >> A clever thought, however.
> >>
> >> Steve Greenfield AE7HD
> >>
> >>
> >> --- In Homebrew_PCBs@yahoogroups.com, Raj Jain<r.k.jain.or@>  wrote:
> >>> I have watched the discussions on printers and their limitations on this
> >>> forum.
> >>> I have seen the use of DLP projectors to harden resin in 3D printers.
> >>> I was wondering if a similar adaption will be effective in exposing
> >>> photoresist material at very high resolutions.
> >>> Maybe use high resolution CRT or LCD monitor as a contrasting source of PCB
> >>> image. A PCB could be prepared and taped on the monitor for exposed.
> >>> ...raj
> >>>
> >>
>