Homebrew_PCBs

Hello,

I just got a box of Kodak Polychrome recording film PRD on eBay for a
steal. probably $100+ of silver
in it, for $9.99 plus postage. I built a laser photoplotter years ago,
and have had to recalibrate it for film a
couple times as films became unavailable. I have almost used up my box
of Agfa HTR3 film, which worked
great, but Agfa is out of the film business, so I have to retool again.
I get an image with this Kodak PRD film,
but when I put it in the fixer, the clear areas turn quite pink! I
called Kodak, they don't sell this stuff anymore,
but somebody there said to dilute the fixer. I tried that, didn't seem
to make any difference.
I am using Kodak rapid fixer, not really designed for graphic arts
films, but worked fine on the Agfa film.
Anybody who works with litho film ever seen anything like this?

My photoplotter uses a single 670 nm 5mW laser diode and some lenses to
produce 1000x1000 DPI
raster images at 0.6 inches per minute (raster lines can be up to 18"
long wrapped around a drum rotating
at 600 RPM.)

If I can get this film to work, it looks like it will be fine for PCB
master artwork, which I also use for solder
stencils made from .003" brass shim stock. I'm just afraid that reddish
background will interfere with the
UV passing through the film.

Jon

John,
I wouldn't think the fixer would cause the emulsion to turn pink. Could it be just the color of the base stock, I have seen tinted base stock before but don't remember seeing any in pink, its usually just a little brown like sepia.

In the past I have used Liford Ilfoline IN-5 for PC flim work but a few years back I attempted get more and couldn't find any. I did find that Freestyl Photo still has a fairly good selection of sheet films but the prices are bloody expensive these days. 25 8x10 Sheets of Ilford FP4+ is $102 about $4 per sheet. www.freestylephoto.biz

Tell me more aobut you photo plotter. I always wanted to build one but never had the time. What did you use for the drum? How are you doing the pixel timing as the drum rotates? Are you using a rotary encoder or just timing from an index pulse. What RPM are you running?

After I finish up my CNC mill project I want to build UV laser diode raster based direct PCB writer. Volkan Sahin has already done some testing with great results. This eliminate the film and photo plotting entirely.

Craig

--- In Homebrew_PCBs@yahoogroups.com, Jon Elson <elson@...> wrote:
>
> Hello,
>
> I just got a box of Kodak Polychrome recording film PRD on eBay for a
> steal. probably $100+ of silver
> in it, for $9.99 plus postage. I built a laser photoplotter years ago,
> and have had to recalibrate it for film a
> couple times as films became unavailable. I have almost used up my box
> of Agfa HTR3 film, which worked
> great, but Agfa is out of the film business, so I have to retool again.
> I get an image with this Kodak PRD film,
> but when I put it in the fixer, the clear areas turn quite pink! I
> called Kodak, they don't sell this stuff anymore,
> but somebody there said to dilute the fixer. I tried that, didn't seem
> to make any difference.
> I am using Kodak rapid fixer, not really designed for graphic arts
> films, but worked fine on the Agfa film.
> Anybody who works with litho film ever seen anything like this?
>
> My photoplotter uses a single 670 nm 5mW laser diode and some lenses to
> produce 1000x1000 DPI
> raster images at 0.6 inches per minute (raster lines can be up to 18"
> long wrapped around a drum rotating
> at 600 RPM.)
>
> If I can get this film to work, it looks like it will be fine for PCB
> master artwork, which I also use for solder
> stencils made from .003" brass shim stock. I'm just afraid that reddish
> background will interfere with the
> UV passing through the film.
>
> Jon
>

--- In Homebrew_PCBs@yahoogroups.com, "designer_craig" <cs6061@...> wrote:

>
> John,
> I wouldn't think the fixer would cause the emulsion to turn pink. Could it be just the color of the base stock, I have seen tinted base stock before but don't remember seeing any in pink, its usually just a little brown like sepia.
>

Well, I wouldn't call it "sepia", but somebody else might call it that. But, there is more redness than sepia.

I am REALLY sure it happens during the fixing step, as I can see the developed film is quite white, and then after just SECONDS in the fixer, I can see it turning darker.

I am not using the right chemistry at all, just using the same stuff I use with my Agfa HTR3 film. The developer appears to work perfectly, but could be setting up some kind of chemical reaction when the film goes into the fixer. These red-sensitive films have a dye in the front to make it red-sensitive, and an anti-haliation backing dye on the back to stop any light that makes it through the film. These are dissolved in the developer.

> In the past I have used Liford Ilfoline IN-5 for PC flim work but a few years back I attempted get more and couldn't find any. I did find that Freestyl Photo still has a fairly good selection of sheet films but the prices are bloody expensive these days. 25 8x10 Sheets of Ilford FP4+ is $102 about $4 per sheet. www.freestylephoto.biz
>

There's this outfit called Ultrafine that is apparently making films that other makers have discontinued. They have something that is supposed to be the same as the Kodak PRD for $83 for 12" x 100 feet!
I was afraid if I had to buy new film it would cost $1000 a box. I paid ~$350 for a box of 100 sheets of the Agfa film over a decade ago.

> Tell me more aobut you photo plotter. I always wanted to build one but never had the time. What did you use for the drum? How are you doing the pixel timing as the drum rotates? Are you using a rotary encoder or just timing from an index pulse. What RPM are you running?
>

OK, it has an ~ 6.5" diameter drum, spun by a DC motor with a PLL control at 10 RPS. An encoder picks up a clock and it is multiplied by 20 by a digital PLL to produce 20480 pixel clock pulses per rev. The emulsion of the film is at a diameter of 6.519" x Pi = 20.480"
So, that makes it 1000 pixels per inch. I have a stepper motor that moves the recording carriage at 1000 steps/inch for the other axis. So, it lays down 0.6" of raster per minute. A 6" wide plot takes 10 minutes, regardless of complexity.

I have been using this system since 1998. The software is now extremely obsolete. I originally made it plot directly from Gerber files to raster slice buffers and to the machine, as the raster files are REALLY large. (I was going to make it an 8-beam plotter, but never made that mod. So, it stores only one bit per byte. Each square inch is one megabyte.) A problem is that anything past Windows 95 needs a driver to deal with DMA transfers to an interface. In Win 95, you can do it all from user-mode. But, the memory environment in Win 95 is very cramped. What I do now is compose the raster files on a program in Win 2K, and then send the file over the net to the plotter computer, still running Win 95.

Eventually, the old computer will croak and I'll have to re-do it all under Linux.

> After I finish up my CNC mill project I want to build UV laser diode raster based direct PCB writer. Volkan Sahin has already done some testing with great results. This eliminate the film and photo plotting entirely.
>

I also use this setup to make master artwork for panel labels, and make many copies of those. So, at least for a while, I think I will stick with silver film.

Jon

--- In Homebrew_PCBs@yahoogroups.com, "designer_craig" <cs6061@...> wrote:

>

> Tell me more aobut you photo plotter. I always wanted to build one but never had the time. What did you use for the drum? How are you doing the pixel timing as the drum rotates? Are you using a rotary encoder or just timing from an index pulse. What RPM are you running?

Just to give a little more specific answers to your questions:
I made the drum out of Plexiglas as I had planned to make it a transparency scanner as well as a plotter. I no longer need to make a scanner myself, so I never built that option.

I made a mandrel out of an old tree trunk, and wrapped the Plexi around it after heating in the home oven. I made two end plates, with screws drilled and tapped into the ends of the Plexi. I then turned it to diameter, making a finish pass at work where they have more accurate lathes.

Yes, a shaft encoder with 1024 lines/rev plus index. No way would a single index pulse be accurate enough.

600 RPM, but actually with the PRD film, I very likely could go to 1200 RPM if I can get the film to stay on the drum. This PRD is significantly more sensitive than the HTR3. I tape it on with black electrical tape to the drum.

Jon

John,
Thanks for the info, that is exactly what I was thinking but will most likely use some aluminum tubing for the drum. With the size and cost of memory these days I was planning on using a little Atmel uP possibly assisted by a FPGA to buffer up a full sheet worth of data prior to writing the drum.

My initial thoughts were something like 600 dpi do you find you need the 1000 dpi?

Craig

--- In Homebrew_PCBs@yahoogroups.com, "jmelson2" <elson@...> wrote:
>
>
>
> --- In Homebrew_PCBs@yahoogroups.com, "designer_craig" <cs6061@> wrote:
> >
>
> > Tell me more aobut you photo plotter. I always wanted to build one but never had the time. What did you use for the drum? How are you doing the pixel timing as the drum rotates? Are you using a rotary encoder or just timing from an index pulse. What RPM are you running?
>
> Just to give a little more specific answers to your questions:
> I made the drum out of Plexiglas as I had planned to make it a transparency scanner as well as a plotter. I no longer need to make a scanner myself, so I never built that option.
>
> I made a mandrel out of an old tree trunk, and wrapped the Plexi around it after heating in the home oven. I made two end plates, with screws drilled and tapped into the ends of the Plexi. I then turned it to diameter, making a finish pass at work where they have more accurate lathes.
>
> Yes, a shaft encoder with 1024 lines/rev plus index. No way would a single index pulse be accurate enough.
>
> 600 RPM, but actually with the PRD film, I very likely could go to 1200 RPM if I can get the film to stay on the drum. This PRD is significantly more sensitive than the HTR3. I tape it on with black electrical tape to the drum.
>
> Jon
>

--- In Homebrew_PCBs@yahoogroups.com, "designer_craig" <cs6061@...> wrote:

>
> John,
> Thanks for the info, that is exactly what I was thinking but will most likely use some aluminum tubing for the drum. With the size and cost of memory these days I was planning on using a little Atmel uP possibly assisted by a FPGA to buffer up a full sheet worth of data prior to writing the drum.
>
> My initial thoughts were something like 600 dpi do you find you need the 1000 dpi?

Yes, if I could have found the right size aluminum tube, I might have used that to start. Something slightly over 6.5" just didn't seem available. I guess I could have rolled some thick sheet, but even the roller at work probably could not have handled 1/4" thick aluminum.

A 12 x 16" sheet of film would take 192 MB of memory. What I would probably do now is either use the parallel port in DMA mode, or use something like the Cypress FX2 (EZ-USB) chip (a fast USB chip that is marginally programmable, and can stream byte-wide data at incredible speeds. I got one at work, I wasn't able to program it to do what I wanted, but I think it could run a photoplotter pretty well.

Note that my photoplotter as it is cranks out a pixel every 5 us, so even as 8-bit bytes, that is 625,000 bytes a second.

Yes, the 1000 DPI is really marginal for PCB work, I routinely make .008" lines and spaces. Even the cheap photoplotters do 2500 DPI and the more expensive ones are 10,000+.

Jon

A picture of my photoplotter is at
http://pico-systems.com/photoplot.html

Jon

What kind of resolution can you obtain with your plotter? What dot size do
you get with the 670nm laser and 13mm FL lens?



_____

From: Homebrew_PCBs@yahoogroups.com [mailto:Homebrew_PCBs@yahoogroups.com]
On Behalf Of jmelson2
Sent: Thursday, March 03, 2011 10:15 PM
To: Homebrew_PCBs@yahoogroups.com
Subject: [Homebrew_PCBs] Laser Photoplotter





A picture of my photoplotter is at
http://pico-systems.com/photoplot.html

Jon





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

If you want large acrylic plastic tubes, check out eBay. Look for "Clear
acrylic Plastic Plexiglas Pipe tube". This vendor sells 4 and 6" diameter
pipe in foot long segments.

Andy

_____

From: Homebrew_PCBs@yahoogroups.com [mailto:Homebrew_PCBs@yahoogroups.com]
On Behalf Of jmelson2
Sent: Thursday, March 03, 2011 7:11 PM
To: Homebrew_PCBs@yahoogroups.com
Subject: [Homebrew_PCBs] Re: Kodak PRD film ?






--- In Homebrew_PCBs@yahoogroups.com
<mailto:Homebrew_PCBs%40yahoogroups.com> , "designer_craig" <cs6061@...>
wrote:

>
> John,
> Thanks for the info, that is exactly what I was thinking but will most

likely use some aluminum tubing for the drum. With the size and cost of
memory these days I was planning on using a little Atmel uP possibly
assisted by a FPGA to buffer up a full sheet worth of data prior to writing
the drum.

>
> My initial thoughts were something like 600 dpi do you find you need the

1000 dpi?

Yes, if I could have found the right size aluminum tube, I might have used
that to start. Something slightly over 6.5" just didn't seem available. I
guess I could have rolled some thick sheet, but even the roller at work
probably could not have handled 1/4" thick aluminum.

A 12 x 16" sheet of film would take 192 MB of memory. What I would probably
do now is either use the parallel port in DMA mode, or use something like
the Cypress FX2 (EZ-USB) chip (a fast USB chip that is marginally
programmable, and can stream byte-wide data at incredible speeds. I got one
at work, I wasn't able to program it to do what I wanted, but I think it
could run a photoplotter pretty well.

Note that my photoplotter as it is cranks out a pixel every 5 us, so even as
8-bit bytes, that is 625,000 bytes a second.

Yes, the 1000 DPI is really marginal for PCB work, I routinely make .008"
lines and spaces. Even the cheap photoplotters do 2500 DPI and the more
expensive ones are 10,000+.

Jon






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

Thanks Andy,

Good Info.

Also PVC pipe can be had in even much bigger sizes at a plumbing store.

Bertho



From: Andrew Volk Sent: Friday, March 04, 2011 00:43



If you want large acrylic plastic tubes, check out eBay. Look for "Clear
acrylic Plastic Plexiglas Pipe tube". This vendor sells 4 and 6" diameter
pipe in foot long segments.

Andy





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

--- In Homebrew_PCBs@yahoogroups.com, "Jeff Heiss" <jeff.heiss@...> wrote:

>
> What kind of resolution can you obtain with your plotter? What dot size do
> you get with the 670nm laser and 13mm FL lens?

The machine runs at 1000 x 1000 DPI. I defocus slightly so the raster lines blend in to a solid black.

Jon

On 04/03/11 18:40, jmelson2 wrote:

>> What kind of resolution can you obtain with your plotter? What dot size do
>> you get with the 670nm laser and 13mm FL lens?
> The machine runs at 1000 x 1000 DPI. I defocus slightly so the raster lines blend in to a solid black.

Is there any chance of you posting some more info about the optics block?

I've been playing around with an RGB laser set (red diode, green DPSS,
BluRay/HD-DVD playback diode) and I'm curious how you got the spot size
down. The best I've managed is about 0.5mm, which is nowhere near small
enough to get any reasonable amount of resolution (about 50DPI by my
reckoning)...

Thanks,
--
Phil.
ygroups@...
http://www.philpem.me.uk/

I'm interested in the optics, too. If we could produce a spot of .001 inch or so, we could use it to remove paint resist from a pcb, just like a cnc router. Then etch and done. The plotter can be a simple xy - used ones with a resolution of .001 inch seem to be readily available.

Mark


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

>From: Philip Pemberton <ygroups@...>
>Sent: Mar 4, 2011 6:58 PM
>To: Homebrew_PCBs@yahoogroups.com
>Subject: Re: [Homebrew_PCBs] Re: Laser Photoplotter
>
>On 04/03/11 18:40, jmelson2 wrote:
>>> What kind of resolution can you obtain with your plotter? What dot size do
>>> you get with the 670nm laser and 13mm FL lens?
>> The machine runs at 1000 x 1000 DPI. I defocus slightly so the raster lines blend in to a solid black.
>
>Is there any chance of you posting some more info about the optics block?
>
>I've been playing around with an RGB laser set (red diode, green DPSS,
>BluRay/HD-DVD playback diode) and I'm curious how you got the spot size
>down. The best I've managed is about 0.5mm, which is nowhere near small
>enough to get any reasonable amount of resolution (about 50DPI by my
>reckoning)...
>
>Thanks,
>--
>Phil.
>ygroups@...
>http://www.philpem.me.uk/
>
>
>------------------------------------
>
>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, Philip Pemberton <ygroups@...> wrote:

>
> On 04/03/11 18:40, jmelson2 wrote:
> >> What kind of resolution can you obtain with your plotter? What dot size do
> >> you get with the 670nm laser and 13mm FL lens?
> > The machine runs at 1000 x 1000 DPI. I defocus slightly so the raster lines blend in to a solid black.
>
> Is there any chance of you posting some more info about the optics block?
>

I'm not an optics guy, I just fooled around with it until it worked.
First, I got a 670 nm 5mW laser diode, but it sprays the light in a big elliptical pattern. So, I put a 3mm sphere lens in front of it, almost touching the glass window on the front of the laser. I just glued the lens into a hole in a plate. I have it set up so I can slide the laser around so the more collimated beam from the sphere lens is pointing straight ahead.

I then put a piece of aluminum foil with a large pinhole in it slightly in front of the sphere lens. I blacked the foil facing toward the film with marker pen to kill reflections.

I have a 13mm microfiche objective lens about 6" in front of the laser, and about 1/4" from the film. The whole assembly can be slid in and out for focus adjustment. This made the spot too large, so I put a double-meniscus lens in the middle of the tube. I don't have any specs on this lens, it is out of something I scrapped. I would guess this lens reduced the spot size about 2 or 3:1.

Jon

--- In Homebrew_PCBs@yahoogroups.com, mlerman@... wrote:

>
> I'm interested in the optics, too. If we could produce a spot of .001 inch or so, we could use it to remove paint resist from a pcb, just like a cnc router. Then etch and done. The plotter can be a simple xy - used ones with a resolution of .001 inch seem to be readily available.

I doubt a 5 mW laser could remove paint off copper sheet. Maybe a 50 mW IR laser. This would be really tough to focus and set up safely.

Also, even a 50 mW laser would probably be really slow.

Jon

> Phil P wrote:
> Is there any chance of you posting some more
> info about the optics block?
> I've been playing around with an RGB laser set
> (red diode, green DPSS, BluRay/HD-DVD playback
> diode) and I'm curious how you got the spot
> size down. The best I've managed is about 0.5mm,
> which is nowhere near small enough to get any
> reasonable amount of resolution (about 50DPI
> by my reckoning)...

If you have a bad shaped laser to start with
(Most Laser Diodes do) then you will need a
couple of lenses to get to a small spot.

Either one lens to do something about the
astigmatism and one to shrink it OR one lens to
spread, go through an aperture (pinhole) and
another lens to get the small spot

Blue Sky Research have some laser diodes with
a micro lens built on the front.

http://www.blueskyresearch.com/c/b/posts/view/38

You will note they do down to 0.025mm spot size
(perfect for 1000dpi)

The other place to look at it

http://www.optima-optics.com/ld_kit.htm


When you build your laser plotter, you actually
want you spot size LARGER than the DPI you are
aiming for.

Also the elliptical beam is not so much a
problem either. You just align it so the short
axis is with the drum.

I got as far as electronics, print driver and
a crude drum before I put my project on hold
a while back. All this talk has got me back
in the mood :D

> Joh Elson wrote:
>
> Note that my photoplotter as it is cranks
> out a pixel every 5 us, so even as 8-bit
> bytes, that is 625,000 bytes a second.

16.384" drum
1000 dpi
600 RPM (10 r/s)
1 bit per pixel
8 bits per byte

I get 20 kilobyte per second

Am I missing something

> Jon wrote:
> <SNIP>
> I get an image with this Kodak PRD film,
> but when I put it in the fixer, the clear
> areas turn quite pink! I

Jon,

Is it rapid-fixer or hypo ?

I'd say the math is wrong on both sides. :)

IIRC, his drum is around 6.5" diameter, to achieve 20.48 inches circumference. He is generating 20480 bits per revolution, giving the 1000 bits per inch.

The drum is rotating 6 times per second, so 6 * 20480 = 122880 bits per second. Somewhere in the earlier discussion, I think I saw that he is using 1 byte per bit (??? presumably to save the bit-shifting circuitry?), so that would work out to 122880 bytes/second. If he is using bit-shifting circuitry, so that each byte is 8 bits, then he can reduce that to 15360 bytes per second.

Either way, it is a long way from 625000 bytes per second!!

--- In Homebrew_PCBs@yahoogroups.com, "cunningfellow" <andrewm1973@...> wrote:
>
> > Joh Elson wrote:
> >
> > Note that my photoplotter as it is cranks
> > out a pixel every 5 us, so even as 8-bit
> > bytes, that is 625,000 bytes a second.
>
> 16.384" drum
> 1000 dpi
> 600 RPM (10 r/s)
> 1 bit per pixel
> 8 bits per byte
>
> I get 20 kilobyte per second
>
> Am I missing something
>

--- In Homebrew_PCBs@yahoogroups.com, "cunningfellow" <andrewm1973@...> wrote:

>
> > Joh Elson wrote:
> >
> > Note that my photoplotter as it is cranks
> > out a pixel every 5 us, so even as 8-bit
> > bytes, that is 625,000 bytes a second.
>
> 16.384" drum
> 1000 dpi
> 600 RPM (10 r/s)
> 1 bit per pixel
> 8 bits per byte
>
> I get 20 kilobyte per second
>
> Am I missing something
>

Yes, probably that I wrote that in my sleep. OK, my machine has a 6.519" diameter drum, or 20.480" circumference, and a 1024 pulse encoder multiplied by 20 to give 20480 pixels per rev, at 10 RPS.
So, that gives 204800 pixels/second. Since this was designed to eventually allow 8 lasers to lay down 8 raster lines at a time, it now only uses one bit per byte, so the DMA port has to supply 204.8 K bytes/second, which is no problem for even an ancient PC.

Jon

--- In Homebrew_PCBs@yahoogroups.com, "cunningfellow" <andrewm1973@...> wrote:

>
> > Jon wrote:
> > <SNIP>
> > I get an image with this Kodak PRD film,
> > but when I put it in the fixer, the clear
> > areas turn quite pink! I
>
> Jon,
>
> Is it rapid-fixer or hypo ?
>

It is Kodak Rapid Fixer, really sold for print paper, I suppose.
Kodak suggested diluting the fixer, but that didn't help. The film looks great after development, completely white with black marks where intended. About 5 seconds after putting it in the fixer, I can see it starting to darken.

I am quite sure the film is not SUPPOSED to look like this, as the pink-orange color is blotchy. I would expect it to be very consistent if it was the intentional color of the backing.

I am working on getting the right Rapid access developer and fixer, but few shops use silver film anymore, so it is getting harder to get supplies.

Jon

I'm not familiar with Kodak PRD film, however, this is from my photography days.
Kodachrome colour slide film has a very distinct colour cast as it ages. The processed slides turn pink! Your film will probably be old stock, albeit unprocessed which could have something to do with it.

Exposing photographic slide film in a camera to capture an image is usually for only a fraction of a second - 1000th of a second sometimes. It follows, any handling of film must be done in total darkness otherwise it will fog.

The same applies to printing paper although it's not so light sensitive. Small amounts of stray light on the paper before fixing will turn it pink. If that happens, the pink colour cast is 'fixed' and will not clear.

Maybe there's a clue in there somewhere will throw some light on your problem. (pardon the pun!)

Peter
MM5PSL

----- Original Message -----
From: jmelson2
To: Homebrew_PCBs@yahoogroups.com
Sent: Sunday, March 06, 2011 4:41 PM
Subject: [Homebrew_PCBs] Re: Kodak PRD film ?





--- In Homebrew_PCBs@yahoogroups.com, "cunningfellow" <andrewm1973@...> wrote:
>
> > Jon wrote:
> > <SNIP>
> > I get an image with this Kodak PRD film,
> > but when I put it in the fixer, the clear
> > areas turn quite pink! I
>
> Jon,
>
> Is it rapid-fixer or hypo ?
>
It is Kodak Rapid Fixer, really sold for print paper, I suppose.
Kodak suggested diluting the fixer, but that didn't help. The film looks great after development, completely white with black marks where intended. About 5 seconds after putting it in the fixer, I can see it starting to darken.

I am quite sure the film is not SUPPOSED to look like this, as the pink-orange color is blotchy. I would expect it to be very consistent if it was the intentional color of the backing.

I am working on getting the right Rapid access developer and fixer, but few shops use silver film anymore, so it is getting harder to get supplies.

Jon




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

No virus found in this message.
Checked by AVG - www.avg.com
Version: 10.0.1204 / Virus Database: 1435/3485 - Release Date: 03/06/11


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

On Sun, 6 Mar 2011 18:24:10 -0000, you wrote:

>I'm not familiar with Kodak PRD film, however, this is from my photography days.
>Kodachrome colour slide film has a very distinct colour cast as it ages. The processed slides turn pink! Your film will probably be old stock, albeit unprocessed which could have something to do with it.
>
>Exposing photographic slide film in a camera to capture an image is usually for only a fraction of a second - 1000th of a second sometimes. It follows, any handling of film must be done in total darkness otherwise it will fog.
>
>The same applies to printing paper although it's not so light sensitive. Small amounts of stray light on the paper before fixing will turn it pink. If that happens, the pink colour cast is 'fixed' and will not clear.
>
>Maybe there's a clue in there somewhere will throw some light on your problem. (pardon the pun!)
>
>Peter
>MM5PSL
>
>

You might be better to try these questions in a photography group
if there are any that cover processing these days.

A call to Freestyle sales might turn up some old employee that could help.

You might also find some medical facility that still uses silver halide x-ray film.


From around 1960 through the late 1980's I was heavily into photography
and mixed my own chemicals from "scratch". I got the formulas from a
newsletter published by Patrick Dignan of Dignan Photographic in California.
He told all the things Kodak didn't want the customers to know :D

I don't remember much mention of line copy films, but he had formulas for
E-6,Anscochrome slides plus all sorts of B&W formulas. If you were really
a glutton for punishment (and had too much money) he even published
formulas and instructions for E-4. He took some side trips into DuPont and
other lesser known B&W films.

I did a bit of Internet searching and it appears a few university libraries
may have the collection in their archives. Mine is somewhere in the attic.

Keith Bowers WB4LSJ- Thomasville, NC

> Jon wrote:
>
> It is Kodak Rapid Fixer, really sold for
> print paper <SNIP>
> The film looks great after development,
> completely white with black marks where
> intended. About 5 seconds after putting
> it in the fixer, I can see it starting
> to darken.
>
> I am quite sure the film is not SUPPOSED
> to look like this, as the pink-orange
> color is blotchy. I would expect it to
> be very consistent if it was the
> intentional color of the backing.

Long ago I saw some high contrast paper
that went light yellow in the acidic
rapid fixer. It had to be used with
non-rapid (hypo) fixer.

Or it could be something that happens to
the film when old.

>> Me wrote:
>>
>> 16.384" drum
>> 1000 dpi
>> 600 RPM (10 r/s)
>> 1 bit per pixel
>> 8 bits per byte
>>
>> I get 20 kilobyte per second
>>
>> Am I missing something
>>

> Jon wrote:
> Yes, probably that I wrote that in my
> sleep. OK, my machine has a 6.519"
> diameter drum, or 20.480" circumference

OK - Me bad that time. I have a 16.384"
circumference. Now I have drum envy.

> and a 1024 pulse encoder multiplied by
> 20 to give 20480 pixels per rev, at 10
> RPS. So, that gives 204800 pixels/
> second. Since this was designed to
> eventually allow 8 lasers to lay down
> 8 raster lines at a time it now only
> uses one bit per byte, so the DMA port
> has to supply 204.8 K bytes/second,
> which is no problem for even an ancient
> PC.

OK - thats he big difference to mine.
I am doing 8 pixels per byte.

I had never even considered the posibilty
of trying to line up 8 lasers at 0.025mm
spacing.

I cant even fathom how it would be done
at home.

We dont have the option at home of
giant lasers and DMD (Digital Micro
Mirrors) devices.

And I guess at home we also don't care
about waiting 16 minutes for a page to
print.

In article <4D6F23AB.6050501@...>,
Jon Elson <elson@...> wrote:

> but somebody there said to dilute the fixer. I tried that, didn't seem
> to make any difference. I am using Kodak rapid fixer,

It's quite an interesting phenomena.

You could try changing the fixer completly of course. Most rapid fixers
are based on Ammonium Thiosuphate, often with additional (proprietry)
additives, instead of the traditional Sodium Thiosulphate but I would not
have thought there would have been such a result.

It can do no harm to experiment with a simple fixer.

This is about as simple as you can get:

240g Sodium Thiosuphate.
10g Sodium Sulphite (anhydrous)
25g Sodium Metabisulphite

Made up to 1l water. (pure if possible)

This is non-hardening and contains no acid to arrest development.

Are you using a stop bath and are you rinsing well between processes?

--
Stuart Winsor

Midland RISC OS show - Sat July 9th 2011

--- In Homebrew_PCBs@yahoogroups.com, "Peter Leybourne" <peter.ley08@...> wrote:

>
> I'm not familiar with Kodak PRD film, however, this is from my photography days.
> Kodachrome colour slide film has a very distinct colour cast as it ages. The processed slides turn pink! Your film will probably be old stock, albeit unprocessed which could have something to do with it.
>
> Exposing photographic slide film in a camera to capture an image is usually for only a fraction of a second - 1000th of a second sometimes. It follows, any handling of film must be done in total darkness otherwise it will fog.
>
> The same applies to printing paper although it's not so light sensitive. Small amounts of stray light on the paper before fixing will turn it pink. If that happens, the pink colour cast is 'fixed' and will not clear.
>

This is a special litho film with a sensitizing dye to make it red sensitive. It is designed specifically for use in a laser photoplotter. I have been using similar films, first from Kodak 15 years ago, and then two different Agfa films. This new experimentation is because I am down to the last 5 sheets of the Agfa film, and they are out of the business. I have the recommended safelight (a dark greenish-blue tube over a 48" fluorescent tube), they spec the same one as agfa did. This film is a good bit more sensitive than the Agfa film, but can still be handled for minutes under this safelight without fogging. Yes, this is pretty old film, but the Agfa film I have been using is over 12 years old, and still working fine. The PRD is about 9 years old, so age could be a factor.

Are you saying that exposing the plastic film backing to light sensitizes it, and the fixer "develops" that exposure into the pink color? Gee, I hope not. It is true that this film is mostly intended for automated processing, so most users could probably avoid this trouble. They also usually use a machine developer with hot developer and fixer.

My exposure is 5 microseconds (per pixel).

Jon

> >
> You might be better to try these questions in a photography group
> if there are any that cover processing these days.
>
>

Well, there is an outfit (Ultrafine) that claims they have a drop-in replacement for the discontinued Kodak PRD film. Also, Kodak still makes some photoplotter films that use essentially the same chemistry. So, I think I ought to be able to get the correct developer and fixer. One place I was talking to had a $400 minimum order, but I think somebody ought to be able to deal with smaller customers.

Jon

--- In Homebrew_PCBs@yahoogroups.com, "cunningfellow"

> I had never even considered the posibilty
> of trying to line up 8 lasers at 0.025mm
> spacing.
>
> I cant even fathom how it would be done
> at home.

Well, I actually built the laser pattern component. What I did was get a roll of communication optical fiber. I stripped one end of 8 fibers and carefully scraped off the protective plastic cladding, just leaving the bare glass fiber. I machined a round part that split down the middle, and then made a groove in one side that was 125 um deep and 1 mm wide, and laid the 8 125 um OD fibers side by side in the groove, smeared a little epoxy on it and then clamped the other piece over it. When the epoxy cured, I cleaved the fibers and polished the whole thing. I proved I could rig up a laser and sphere lens to light the fiber. I have no idea how much of the light actually made it into the fiber, or how efficiently the fiber radiated the light out the other end. So, the plotter worked well enough that I put it into use and never completed the 8-beam modification.

Yes, you'd have to deal with rotation of the fiber array, PLUS getting the magnification just right so the 8 beams all fell in the right raster lines. I can imagine test patterns that would help in making the adjustments.

Jon

--- In Homebrew_PCBs@yahoogroups.com, "cunningfellow" <andrewm1973@...> wrote:

>
>
> Long ago I saw some high contrast paper
> that went light yellow in the acidic
> rapid fixer. It had to be used with
> non-rapid (hypo) fixer.
>

Yeah, this SOUNDS like the reaction I am having! This IS high contrast film, too.

> Or it could be something that happens to
> the film when old.
>

Well, I hope not. I paid $350 for the Agfa film over a decade ago, and was worried that a new box would run $1000. I got this stuff for $9.99 on eBay, but the Ultrafine outfit sells their replacement for $86, which sounds really reasonable. But, if I can get the right chemistry, I hope I can get this stuff to work.

Jon

--- In Homebrew_PCBs@yahoogroups.com, lists <Stuartlists@...> wrote:

> It's quite an interesting phenomena.

> This is non-hardening and contains no acid to arrest development.
>
> Are you using a stop bath and are you rinsing well between processes?

I normally used a stop bath, but as the machine developers don't use it, I tried without stop bath, and had no different result. I also tried rinsing between developer and fixer, same result.

But, all this was with Kodak Rapid Fixer, really intended for print paper, I think. I have used this for years with the Agfa film with no problem.

I am trying to get the proper fixer from Kodak or somebody else who could guarantee their fixer works with this film. Hopefully I will hear from somebody Monday about getting something.

Jon

Clever!!



From: jmelson2 Sent: Monday, March 07, 2011 00:52



Well, I actually built the laser pattern component. What I did was get a
roll of communication optical fiber. I stripped one end of 8 fibers and
carefully scraped off the protective plastic cladding, just leaving the bare
glass fiber. I machined a round part that split down the middle, and then
made a groove in one side that was 125 um deep and 1 mm wide, and laid the 8
125 um OD fibers side by side in the groove, smeared a little epoxy on it
and then clamped the other piece over it. When the epoxy cured, I cleaved
the fibers and polished the whole thing. I proved I could rig up a laser and
sphere lens to light the fiber. I have no idea how much of the light
actually made it into the fiber, or how efficiently the fiber radiated the
light out the other end. So, the plotter worked well enough that I put it
into use and never completed the 8-beam modification.

Yes, you'd have to deal with rotation of the fiber array, PLUS getting the
magnification just right so the 8 beams all fell in the right raster lines.
I can imagine test patterns that would help in making the adjustments.
Jon



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

>Are you saying that exposing the plastic film backing to light sensitizes it, and the fixer "develops" that exposure into the pink color? Gee, I hope not.>

None of it particularly good news Jon.
I'm out of touch with PRD techniques.
Personally, I would treat all film as light sensitive. Always keep it under wraps and only work in 'safe' or subdued light. 'Safe' is not all it seems. I saw a film developed that had been lying dormant in a 'Box Brownie' camera for very many years. You may have seen the little red window on these old cameras which show the frame number on the backing paper. Light had gone through the red filter, through the backing paper and exposed the film to a red circle with a number in the middle. How safe is plastic film backing over long periods?

In the case of colour slide printing paper, the answer to your question is yes!
Stray light prior to developing/fixing may cause the pinkness you describe. The same applies to slide film although because of film speed vs paper speed, complete fogging usually occurs.

Artificial light and sunlight have different colour temperatures. Colour slide film is designed for daylight exposure and slides will have a red/pink colour cast if the film is used under artificial lighting conditions. What we see as a white light table lamp will be very red in the picture. A flashgun or blue filter over the camera lens will correct this.

The picture printing process is a positive print from a positive slide on light sensitive paper.
If stray light spills onto the paper prior to development, it will cause a pink colour cast to the finished print. In extreme cases where complete fogging occurs, development will turn the paper red.

As mentioned, I have Kodachrome colour slides from military activities in Yemen taken in 1964. After nearly 50-years, they are all quite pink. Kodak film products IMHO were never as stable as one would wish. They were first class when fresh, but over long periods the quality of dyes deteriorated. Ektachrome had a tendency to turn blue. Agfa products on the other hand seemed more stable. I have 40-year old Agfa movie film which is unchanged.

So, this could be the root of the problem. If the film has not already exposed to stray light, the answer might be to try working in very subdued light. If the film has been inadvertently exposed or the emulsion has deteriorated, there's not much anyone can do


Peter
MM5PSL

----- Original Message -----
From: jmelson2
To: Homebrew_PCBs@yahoogroups.com
Sent: Monday, March 07, 2011 5:39 AM
Subject: [Homebrew_PCBs] Re: Kodak PRD film ?

--- In Homebrew_PCBs@yahoogroups.com, "Peter Leybourne" <peter.ley08@...> wrote:
>
> I'm not familiar with Kodak PRD film, however, this is from my photography days.
> Kodachrome colour slide film has a very distinct colour cast as it ages. The processed slides turn pink! Your film will probably be old stock, albeit unprocessed which could have something to do with it.
>
> Exposing photographic slide film in a camera to capture an image is usually for only a fraction of a second - 1000th of a second sometimes. It follows, any handling of film must be done in total darkness otherwise it will fog.
>
> The same applies to printing paper although it's not so light sensitive. Small amounts of stray light on the paper before fixing will turn it pink. If that happens, the pink colour cast is 'fixed' and will not clear.
>
This is a special litho film with a sensitizing dye to make it red sensitive. It is designed specifically for use in a laser photoplotter. I have been using similar films, first from Kodak 15 years ago, and then two different Agfa films. This new experimentation is because I am down to the last 5 sheets of the Agfa film, and they are out of the business. I have the recommended safelight (a dark greenish-blue tube over a 48" fluorescent tube), they spec the same one as agfa did. This film is a good bit more sensitive than the Agfa film, but can still be handled for minutes under this safelight without fogging. Yes, this is pretty old film, but the Agfa film I have been using is over 12 years old, and still working fine. The PRD is about 9 years old, so age could be a factor.

Are you saying that exposing the plastic film backing to light sensitizes it, and the fixer "develops" that exposure into the pink color? Gee, I hope not. It is true that this film is mostly intended for automated processing, so most users could probably avoid this trouble. They also usually use a machine developer with hot developer and fixer.

My exposure is 5 microseconds (per pixel).

Jon



Reply to sender | Reply to group | Reply via web post | Start a New Topic
Messages in this topic (27)
Recent Activity: a.. New Members 15 a.. New Photos 1
Visit Your Group

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

Oh, the irony -- as subsequent posts revealed, I got the math wrong too! Instead of 6 revolutions per second, it should actually be 10.

--- In Homebrew_PCBs@yahoogroups.com, "Andrew" <a_wake@...> wrote:
>
> I'd say the math is wrong on both sides. :)
>
> IIRC, his drum is around 6.5" diameter, to achieve 20.48 inches circumference. He is generating 20480 bits per revolution, giving the 1000 bits per inch.
>
> The drum is rotating 6 times per second, so 6 * 20480 = 122880 bits per second. Somewhere in the earlier discussion, I think I saw that he is using 1 byte per bit (??? presumably to save the bit-shifting circuitry?), so that would work out to 122880 bytes/second. If he is using bit-shifting circuitry, so that each byte is 8 bits, then he can reduce that to 15360 bytes per second.
>
> Either way, it is a long way from 625000 bytes per second!!
>
> --- In Homebrew_PCBs@yahoogroups.com, "cunningfellow" <andrewm1973@> wrote:
> >
> > > Joh Elson wrote:
> > >
> > > Note that my photoplotter as it is cranks
> > > out a pixel every 5 us, so even as 8-bit
> > > bytes, that is 625,000 bytes a second.
> >
> > 16.384" drum
> > 1000 dpi
> > 600 RPM (10 r/s)
> > 1 bit per pixel
> > 8 bits per byte
> >
> > I get 20 kilobyte per second
> >
> > Am I missing something
> >
>

--- In Homebrew_PCBs@yahoogroups.com, "Peter Leybourne" <peter.ley08@...> wrote:

> So, this could be the root of the problem. If the film has not already exposed to stray light, the answer might be to try working in very subdued light. If the film has been inadvertently exposed or the emulsion has deteriorated, there's not much anyone can do

First, this is black and white LITHO film, not color film. Second, it has NOT been exposed to any stray light, and I am using the recommended safe light as per the Kodak spec sheet.

It is VERY VERY subdued, so after getting adjusted to the light, I can BARELY see what I am doing.

This film is only about 8 years old. I have been using Agfa film with no problem for years, my batch is now 13 years old.

I can clearly see the film develops fine, and only turns reddish-orange when put in the fixer.

Thanks,

Jon

Hi Jon,

I understand your problem here, and have worked in "the green
darkness".

You mentioned that the film is a proper black and white (no pink)
when it comes out of the developer, so we can eliminate base color,
anti-hallation backings/dyes, and so on.

I seem to recall running across this sort of thing a couple of times
over the past couple of decades (while doing things that materials
weren't designed for), but no specific solution springs to mind. It's
possible that the proprietary PRD fixer, assuming there was one, may have
had a component that stopped the "pinkage". A documented formula would be
useful.

In the meantime, I have one test and three, wait, *four* suggestions:

Scrape the back (non-photosensitive side) of the film to see if
there's a backing, and if so, if that's where the pink color resides.

1. Run an extended wash/rinse on the side of the film that holds the
pink. It may eventually clear.

2. Dip a piece of developed pink film in some vinegar. If it clears,
you may need a final trip through a stop bath.

3. If #2 does nothing, try an alkaline bath, like some baking soda
in some water to see if it clears. Sodium carbonate would be better, but
that's not a typical kitchen item.

4. Oh, also, try doubling the time in the fixer, I have a fragmented
memory about "pink" being a problem when Tmax films came out way back
when. The Tmax fixer was very strong and conventional fixers required
twice the time to clear.

I'm just assuming that you haven't tried these yet. I'd appreciate
hearing if any of this helps, it seems like you've been through a lot of
discussion here without getting anywhere.

-Gary

>First, this is black and white LITHO film, not color film. Second, it has
>NOT been exposed to any stray light, and I am using the recommended safe
>light as per the Kodak spec sheet.
>
>It is VERY VERY subdued, so after getting adjusted to the light, I can
>BARELY see what I am doing.
>
>This film is only about 8 years old. I have been using Agfa film with no
>problem for years, my batch is now 13 years old.
>
>I can clearly see the film develops fine, and only turns reddish-orange
>when put in the fixer.
>
>Thanks,
>
>Jon

On 06/03/11 06:24, jmelson2 wrote:

> First, I got a 670 nm 5mW laser diode, but it sprays the light in a
> big elliptical pattern.

Sounds about right for a semiconductor laser. Usually you get somewhere
between a 3:1 and a 6:1 aspect ratio for the spot (when really you want
1:1 or close to it).

> I then put a piece of aluminum foil with a large pinhole in it
> slightly in front of the sphere lens. I blacked the foil facing
> toward the film with marker pen to kill reflections.

That's a pretty neat way to reshape the spot, assuming you don't mind
the power loss from doing so. That said, it's probably pretty minimal on
any decent laser diode.

> I have a 13mm microfiche objective lens about 6" in front of the
> laser, and about 1/4" from the film. The whole assembly can be slid
> in and out for focus adjustment. This made the spot too large, so I
> put a double-meniscus lens in the middle of the tube. I don't have
> any specs on this lens, it is out of something I scrapped. I would
> guess this lens reduced the spot size about 2 or 3:1.

Hmm. Interesting. Well, I just ordered a couple of assorted lenses and a
collimator lens assembly from a local laser parts distributor. With a
bit of luck, these will work better than the single-lens focus assembly
I'm using now.

The trick will be in figuring out how small the spot actually is, and if
it's accurately focussed. There is no way I'm focussing it with a
photographic focus-finder, even on low power with a set of laser safety
goggles...!

BTW, this is all for another pet project of mine -- a laser photo
printer. Basically printing straight onto photographic paper using the
scanner assembly from a laser printer. Once you expose the paper with
the laser, you develop it as normal in the usual B&W chemistry (I've
been using Ilford paper developer and Rapid Fixer).

Although I might have to get some photoplotter film to play with... it's
just a shame it's so expensive. Well, not expensive in the conventional
sense, just expensive in that you have to buy it in rather large rolls...

Thanks,
--
Phil.
ygroups@...
http://www.philpem.me.uk/

Here is some additional info for cleaning up a laser beam:

Changing the shape factor can also be done with a front surface mirror at an
angle. Vary the angle until the shape is correct.

Typically a laser has side modes or spots. Shine it on a white wall and
the extra ones can be seen. To clean them up use a microscope objective
focusing the beam on a really small pin hole. On the other side there is
another objective also focused on the same pinhole and the output is a
parallel cleaned up laser beam. This assembly must be very stable and also
adjustable to focus & align the components.

Bertho



From: Philip Pemberton Sent: Tuesday, March 08, 2011 17:56

On 06/03/11 06:24, jmelson2 wrote:
> First, I got a 670 nm 5mW laser diode, but it sprays the light in a
> big elliptical pattern.

Sounds about right for a semiconductor laser. Usually you get somewhere
between a 3:1 and a 6:1 aspect ratio for the spot (when really you want
1:1 or close to it).

> I then put a piece of aluminum foil with a large pinhole in it
> slightly in front of the sphere lens. I blacked the foil facing
> toward the film with marker pen to kill reflections.

That's a pretty neat way to reshape the spot, assuming you don't mind
the power loss from doing so. That said, it's probably pretty minimal on
any decent laser diode.

> I have a 13mm microfiche objective lens about 6" in front of the
> laser, and about 1/4" from the film. The whole assembly can be slid
> in and out for focus adjustment. This made the spot too large, so I
> put a double-meniscus lens in the middle of the tube. I don't have
> any specs on this lens, it is out of something I scrapped. I would
> guess this lens reduced the spot size about 2 or 3:1.

Hmm. Interesting. Well, I just ordered a couple of assorted lenses and a
collimator lens assembly from a local laser parts distributor. With a
bit of luck, these will work better than the single-lens focus assembly
I'm using now.

The trick will be in figuring out how small the spot actually is, and if
it's accurately focussed. There is no way I'm focussing it with a
photographic focus-finder, even on low power with a set of laser safety
goggles...!

BTW, this is all for another pet project of mine -- a laser photo
printer. Basically printing straight onto photographic paper using the
scanner assembly from a laser printer. Once you expose the paper with
the laser, you develop it as normal in the usual B&W chemistry (I've
been using Ilford paper developer and Rapid Fixer).

Although I might have to get some photoplotter film to play with... it's
just a shame it's so expensive. Well, not expensive in the conventional
sense, just expensive in that you have to buy it in rather large rolls...

Thanks,
--
Phil.





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

I like this solution - the power loss is minimal (only the side lobes are cut off) and the complementary lenses convert all incoming power to outgoing power (less the lenses' reflection and absorption losses).

Simple, elegant, effective. Who needs more?

Donald.
--
*Plain Text* email -- it's an accessibility issue
() no proprietary attachments; no html mail
/\ ascii ribbon campaign - <www.asciiribbon.org>

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

> From: "Boman33" <boman33@...>
> To: "Homebrew PCBs" <Homebrew_PCBs@yahoogroups.com>
> Sent: Tuesday, March 8, 2011 6:32:47 PM
> Subject: RE: [Homebrew_PCBs] Re: Laser Photoplotter
>
> Here is some additional info for cleaning up a laser beam:
>
> Changing the shape factor can also be done with a front surface mirror
> at an angle. Vary the angle until the shape is correct.
>
> Typically a laser has side modes or spots. Shine it on a white wall
> and the extra ones can be seen. To clean them up use a microscope
> objective focusing the beam on a really small pin hole. On the other side there
> is another objective also focused on the same pinhole and the output is
> a parallel cleaned up laser beam. This assembly must be very stable and
> also adjustable to focus & align the components.
>
> Bertho
>

[snip]

--- In Homebrew_PCBs@yahoogroups.com, Philip Pemberton <ygroups@...> wrote:

>
> The trick will be in figuring out how small the spot actually is, and if
> it's accurately focussed. There is no way I'm focussing it with a
> photographic focus-finder, even on low power with a set of laser safety
> goggles...!
>

What I do is run the laser on a pulse generator with about a 1:1000 duty cycle. That gives a bright enough spot for focusing without too much light.

> BTW, this is all for another pet project of mine -- a laser photo
> printer. Basically printing straight onto photographic paper using the
> scanner assembly from a laser printer. Once you expose the paper with
> the laser, you develop it as normal in the usual B&W chemistry (I've
> been using Ilford paper developer and Rapid Fixer).
>

Yes, I can swap a plug-in board on my photoplotter to do grey scale.
It takes two consecutive bytes for a 12-bit DAC, so resolution drops to 500 DPI, then I repeat every raster line, to get 500 x 500 DPI.

I use Kodak digital sciences laser recording film, apparently made for laser recording of medical scans. I don't know if regular B&W paper is red sensitive.

Jon

On 09/03/11 06:36, jmelson2 wrote:

> I use Kodak digital sciences laser recording film, apparently made for laser recording of medical scans. I don't know if regular B&W paper is red sensitive.

It's not -- but it is blue sensitive! Hence the use of a BluRay LD.

How did you rig up the laser driver? High speed opamp to generate
constant current, then another one driving a transistor/resistor shunt
to short over the diode (and thus set the current through the diode)?

Thanks,
--
Phil.
ygroups@...
http://www.philpem.me.uk/

--- In Homebrew_PCBs@yahoogroups.com, Philip Pemberton <ygroups@...> wrote:

>
> On 09/03/11 06:36, jmelson2 wrote:
> > I use Kodak digital sciences laser recording film, apparently made for laser recording of medical scans. I don't know if regular B&W paper is red sensitive.
>
> It's not -- but it is blue sensitive! Hence the use of a BluRay LD.
>

I started building this thing in 1986, without any idea how to actually produce a beam that could be turned on and off. In 1996 I got it running with a 670 nm laser diode, which solved a bunch of problems.

> How did you rig up the laser driver? High speed opamp to generate
> constant current, then another one driving a transistor/resistor shunt
> to short over the diode (and thus set the current through the diode)?

The litho imaging driver is dead simple, a CMOS output controls an NPN transistor, with a resistor around 82 Ohms for current limiting.
You change the resistor to change the current. Pixels are now 5 us long, so extreme control of rise- and fall-time are not real critical.

I also have a grey-scale driver that has a 12-bit DAC that feeds an op-amp circuit. It compares photodiode current to the intensity value and programs a pass transistor to get that photodiode current.
Pixels in that mode are 10 us long, seems to work pretty well.

Jon

--- In Homebrew_PCBs@yahoogroups.com, "jmelson2" <elson@...> wrote:

>
>
>
> --- In Homebrew_PCBs@yahoogroups.com, lists <Stuartlists@> wrote:
>
> > It's quite an interesting phenomena.

Well, the problem is SOLVED! After MUCH calling around, I found the local graphic arts distributor, Xpedex, had a developer and fixer that they recommended for this film. It is their own brand, not Kodak. Well, it works! I didn't have time to fire up the photoplotter, but just partially fogged a snip of film with the red LEDs of my darkroom timer, developed it, then rinsed it and fixed it, and immediately I could see a difference, the unexposed part of the film slowly faded from white to clear, rather than to what looked brown under the safelight. When fully fixed and then washed, it is totally clear in the unexposed part.

This "fixer" obviously has acetic acid in it, you can smell it immediately when you open the bottle. So, I guess it is really stop bath plus fixer.

I will probably use up the rest of my Agfa film before recalibrating the photoplotter, but it is clear that this stuff is going to work!

And, I saved a bundle getting the film on eBay!
(Hmmm, if anyone else wants some, the guy has more rolls of it.
Look for "recording film". I'd buy it, but since I have 200 feet on this roll, that is probably enough for a decade.

Jon

Sorry, I chose this post as the most recent of yours to reply to; I used to dabble in trying various methods of DIY PCB's and I did try the plotter method with a 7440, trying to figure out which Staedler red pen will suffice,(argh!) but I couldn't do double-sided boards without figuring how to register the board in a way for the plotter to do the second side.

Even with a laser, I'm not sure if this an obvious issue or not a concern as you are only doing single-sided work.

For me, I find Sunday coupon pages printed on my old laser printer adequate enough but the real issue that eludes me from quick PCB fabrication is through-plating.

I am intrigued though, why did you stop going with the direct laser method? The pictures you provided were most impressive but perhaps you ran into the same issue with double-sided boards?

--- In Homebrew_PCBs@yahoogroups.com, mlerman@... wrote:
>
> I'm interested in the optics, too. If we could produce a spot of .001 inch or so, we could use it to remove paint resist from a pcb, just like a cnc router. Then etch and done. The plotter can be a simple xy - used ones with a resolution of .001 inch seem to be readily available.
>
> Mark
>
>

In article <il9rbf+r6fe@...>,
jmelson2 <elson@...> wrote:

> > > It's quite an interesting phenomena.
> Well, the problem is SOLVED! After MUCH calling around, I found the
> local graphic arts distributor, Xpedex, had a developer and fixer that
> they recommended for this film. It is their own brand, not Kodak.
> Well, it works! I didn't have time to fire up the photoplotter, but
> just partially fogged a snip of film with the red LEDs of my darkroom
> timer, developed it, then rinsed it and fixed it, and immediately I
> could see a difference, the unexposed part of the film slowly faded from
> white to clear, rather than to what looked brown under the safelight.
> When fully fixed and then washed, it is totally clear in the unexposed
> part.

Excellent!

> This "fixer" obviously has acetic acid in it, you can smell it
> immediately when you open the bottle. So, I guess it is really stop
> bath plus fixer.

Yup, most fixers in use are "acid-hardening".

Is it "slower" than the fixer you were using previously, which might
suggest Sodium rather than Ammonium Thiosulphate?

--
Stuart Winsor

Midland RISC OS show - Sat July 9th 2011

The problem with DLP is that larger boards are very inconsistent, and
even smaller ones give variable results. Right now I am working with
an HP 2200 that is very simple to modify to pass thick boards and
requires no MCU to simulate sensors, but it does not work
consistently. I am hoping to use it as a test bed to explore dlp
further and get the kinks out. Meanwhile I did modify an inkjet (see
<http://www.marksData.com>), which was fun, but I think dlp has more
potential. DS boards on a plotter should be easy - just make
registration holes on the Y axis with pins to locate the board. Do
one side, flip and do the others. That's the way my cnc router works,
though you can do it optically as well with a camera on the plotter head.

Mark


At 06:43 AM 3/10/2011, you wrote:

>Sorry, I chose this post as the most recent of yours to reply to; I
>used to dabble in trying various methods of DIY PCB's and I did try
>the plotter method with a 7440, trying to figure out which Staedler
>red pen will suffice,(argh!) but I couldn't do double-sided boards
>without figuring how to register the board in a way for the plotter
>to do the second side.
>
>Even with a laser, I'm not sure if this an obvious issue or not a
>concern as you are only doing single-sided work.
>
>For me, I find Sunday coupon pages printed on my old laser printer
>adequate enough but the real issue that eludes me from quick PCB
>fabrication is through-plating.
>
>I am intrigued though, why did you stop going with the direct laser
>method? The pictures you provided were most impressive but perhaps
>you ran into the same issue with double-sided boards?
>
>
>--- In Homebrew_PCBs@yahoogroups.com, mlerman@... wrote:
> >
> > I'm interested in the optics, too. If we could produce a spot of
> .001 inch or so, we could use it to remove paint resist from a pcb,
> just like a cnc router. Then etch and done. The plotter can be a
> simple xy - used ones with a resolution of .001 inch seem to be
> readily available.
> >
> > Mark
> >
> >
>
>
>
>
>------------------------------------
>
>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, "gmanca101" <gmanca101@...> wrote:

>
> Sorry, I chose this post as the most recent of yours to reply to; I used to dabble in trying various methods of DIY PCB's and I did try the plotter method with a 7440, trying to figure out which Staedler red pen will suffice,(argh!) but I couldn't do double-sided boards without figuring how to register the board in a way for the plotter to do the second side.
>
> Even with a laser, I'm not sure if this an obvious issue or not a concern as you are only doing single-sided work.
>
> For me, I find Sunday coupon pages printed on my old laser printer adequate enough but the real issue that eludes me from quick PCB fabrication is through-plating.
>
> I am intrigued though, why did you stop going with the direct laser method? The pictures you provided were most impressive but perhaps you ran into the same issue with double-sided boards?

Direct laser? I have never done anything with direct laser, if that means exposing resist directly on the board with a laser.

This photoplotter has been running since 1996 with minimal change. I have had to change films a couple times as one film became obsolete. Agfa is no longer in the silver film business, so I have just had to qualify another film.

I used to do double-sided boards all the time. What I did is generate the two films on the photoplotter, one in mirror image. I took a scrap of PCB material and a sheet of .060" Plexiglas and put it on a light table. Using a magnifier, I align the two films with the emulsion facing together, and glue them in alignment with rubber cement. I use Riston dry film resist fro Think and Tinker, who resell it at a VERY reasonable price. I have a Kepro dry film laminator machine with heated rubber rollers to laminate the film to the board. Oh, the board has to be cleaned with fine sandpaper or Scotchbrite and washed very well for the best dry film adhesion.
I then slip the board between the two aligned phototools and print for one minute on each side with a bank of fluorescent blacklights.
You wait for 15 minutes for the reaction to complete, then develop the resist with warm sodium carbonate solution for about a minute.

I don't have a through-hole process. I do have a CNC milling machine, and a system for putting an air bearing drill spindle on it, and have drilled boards that way. a LOT of setup time and tedious work to make one board.

With FPGAs, I can usually prototype anything I want now in a single chip, using a hacked over board left from some previous project, so I just don't do a heck of a lot of one-off boards anymore.

But, I use this entire setup to make solder stencils out of .003" brass shim stock, just like it was a double-sided PC board without the fiberglass substrate in the middle.

This process is now working extremely well, and I can make apertures down to .010" with excellent results.

I have a huge old Calcomp pen plotter here, I rigged a mercury short arc lamp with ellipsodal mirror and a small loudspeaker moving a brass vane as a light "switch" feeding an optical fiber. I mounted a set of lenses to the pen carriage and it exposed low-sensitivity litho film with a .010" beam. Complex boards took close to an hour to plot out, and the plotter was so dynamic it actually caused a lens to self-disassemble. I should have totally skipped this mess and just kept plugging away on the raster machine. The advantage was no software needed, just plot as if it was a pen plot on paper. The raster machine required a program to convert Gerber to raster, and that program is ten times more complicated than the plotter hardware.

Jon

--- In Homebrew_PCBs@yahoogroups.com, lists <Stuartlists@...> wrote:

>
>
> Yup, most fixers in use are "acid-hardening".
>
> Is it "slower" than the fixer you were using previously, which might
> suggest Sodium rather than Ammonium Thiosulphate?
>

Don't really know. It wasn't very obviously slower, though. About one minute in the fixer at a 1:4 dilution, and all the white was gone.
I left it in for two minutes just to be sure the fixing was completed.

Jon

--- In Homebrew_PCBs@yahoogroups.com, Mark Lerman <mlerman@...> wrote:

>
> The problem with DLP is that larger boards are very inconsistent, and
> even smaller ones give variable results.

I'm not sure what you mean by "inconsistent". I have done some tests on various laser printers where I printed a 1" grid of lines. I then turn one film over and lay it over the other one on a light table, so see how accurate the registration is. My experience is that almost every printer has at least a tenth of an inch error somewhere on the page! The stuff I am now doing with solder stencils requires .010" apertures, so the two films of the two sides need to line up extremely well. (I etch mirror image patterns in from both sides simultaneously to reduce sideways etching.) I actually have a problem because the program I wrote to convert Gerber to raster does the mirror-flip before plotting, and rounding processes in the program shift the apertures by one mil on the different sides. I can clearly see this as an offset between the apertures coming in from opposite sides. It doesn't really cause any problem in the solder stencil, but it is easily visible under the microscope. But, making these master artwork on a laser printer would be a fool's mission!

The boards these stencils are for have chips with lead pitch down to .4mm, and some of them are a foot wide! So, I need something with a LOT more accuracy.

Jon

In article <ilb297+bk6u@...>,
jmelson2 <elson@...> wrote:

> --- In Homebrew_PCBs@yahoogroups.com, lists <Stuartlists@...> wrote:
> >
> >
> > Yup, most fixers in use are "acid-hardening".
> >
> > Is it "slower" than the fixer you were using previously, which might
> > suggest Sodium rather than Ammonium Thiosulphate?
> >
> Don't really know. It wasn't very obviously slower, though. About one
> minute in the fixer at a 1:4 dilution, and all the white was gone. I
> left it in for two minutes just to be sure the fixing was completed.

Ok, just curious.

--
Stuart Winsor

Midland RISC OS show - Sat July 9th 2011

I see, yes DLP would have more potential, I also tried Epson ink blends but issues with the printheads made me abandon it. And when you look at all of the variables, adding another in regards to finding the correct ink versus universally-usable toner is just not worth the time.

The issue with the plotter was that I did not want to destroy the machine by turning it into a true x-y device since the plotter uses the paper for registration as it moves the paper on the y axis. It worked well as a single-sided board maker but to have to use two different methods is just impractical.

Is there any way to help figure out how to improve the DLP method? That's really the best bet and once the accuracy hurdle is climbed, double-sided becomes just an issue of mechanics.

--- In Homebrew_PCBs@yahoogroups.com, Mark Lerman <mlerman@...> wrote:
>
> The problem with DLP is that larger boards are very inconsistent, and
> even smaller ones give variable results. Right now I am working with
> an HP 2200 that is very simple to modify to pass thick boards and
> requires no MCU to simulate sensors, but it does not work
> consistently. I am hoping to use it as a test bed to explore dlp
> further and get the kinks out. Meanwhile I did modify an inkjet (see
> <http://www.marksData.com>), which was fun, but I think dlp has more
> potential. DS boards on a plotter should be easy - just make
> registration holes on the Y axis with pins to locate the board. Do
> one side, flip and do the others. That's the way my cnc router works,
> though you can do it optically as well with a camera on the plotter head.
>
> Mark
>
>
> At 06:43 AM 3/10/2011, you wrote:
> >Sorry, I chose this post as the most recent of yours to reply to; I
> >used to dabble in trying various methods of DIY PCB's and I did try
> >the plotter method with a 7440, trying to figure out which Staedler
> >red pen will suffice,(argh!) but I couldn't do double-sided boards
> >without figuring how to register the board in a way for the plotter
> >to do the second side.
> >
> >Even with a laser, I'm not sure if this an obvious issue or not a
> >concern as you are only doing single-sided work.
> >
> >For me, I find Sunday coupon pages printed on my old laser printer
> >adequate enough but the real issue that eludes me from quick PCB
> >fabrication is through-plating.
> >
> >I am intrigued though, why did you stop going with the direct laser
> >method? The pictures you provided were most impressive but perhaps
> >you ran into the same issue with double-sided boards?
> >
> >
> >--- In Homebrew_PCBs@yahoogroups.com, mlerman@ wrote:
> > >
> > > I'm interested in the optics, too. If we could produce a spot of
> > .001 inch or so, we could use it to remove paint resist from a pcb,
> > just like a cnc router. Then etch and done. The plotter can be a
> > simple xy - used ones with a resolution of .001 inch seem to be
> > readily available.
> > >
> > > Mark
> > >
> > >
> >
> >
> >
> >
> >------------------------------------
> >
> >Be sure to visit the group home and check for new Links, Files, and Photos:
> >http://groups.yahoo.com/group/Homebrew_PCBsYahoo! Groups Links
> >
> >
> >
>

We've hijacked this thread - sorry. This will be my last posting without changing the subject.

I'm working on it, and I think I might have a workable solution, but I need to spend a week or so on it. I'll let you know.

With regard to DS, it is purely a mechanical problem. The laser printer prints a fixed time after the paper sensor is activated, so it's purely a matter of having good mechanics. The printer I'm working with, an HP 2200, is mechanically very stable and easy to work with. I can also get used ones, with toner cartridges, for $60 including shipping.

Mark




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

>From: gmanca101 <gmanca101@...>
>Sent: Mar 10, 2011 6:08 PM
>To: Homebrew_PCBs@yahoogroups.com
>Subject: [Homebrew_PCBs] Re: Laser Photoplotter
>
>I see, yes DLP would have more potential, I also tried Epson ink blends but issues with the printheads made me abandon it. And when you look at all of the variables, adding another in regards to finding the correct ink versus universally-usable toner is just not worth the time.
>
>The issue with the plotter was that I did not want to destroy the machine by turning it into a true x-y device since the plotter uses the paper for registration as it moves the paper on the y axis. It worked well as a single-sided board maker but to have to use two different methods is just impractical.
>
>Is there any way to help figure out how to improve the DLP method? That's really the best bet and once the accuracy hurdle is climbed, double-sided becomes just an issue of mechanics.
>
>--- In Homebrew_PCBs@yahoogroups.com, Mark Lerman <mlerman@...> wrote:
>>
>> The problem with DLP is that larger boards are very inconsistent, and
>> even smaller ones give variable results. Right now I am working with
>> an HP 2200 that is very simple to modify to pass thick boards and
>> requires no MCU to simulate sensors, but it does not work
>> consistently. I am hoping to use it as a test bed to explore dlp
>> further and get the kinks out. Meanwhile I did modify an inkjet (see
>> <http://www.marksData.com>), which was fun, but I think dlp has more
>> potential. DS boards on a plotter should be easy - just make
>> registration holes on the Y axis with pins to locate the board. Do
>> one side, flip and do the others. That's the way my cnc router works,
>> though you can do it optically as well with a camera on the plotter head.
>>
>> Mark
>>
>>
>> At 06:43 AM 3/10/2011, you wrote:
>> >Sorry, I chose this post as the most recent of yours to reply to; I
>> >used to dabble in trying various methods of DIY PCB's and I did try
>> >the plotter method with a 7440, trying to figure out which Staedler
>> >red pen will suffice,(argh!) but I couldn't do double-sided boards
>> >without figuring how to register the board in a way for the plotter
>> >to do the second side.
>> >
>> >Even with a laser, I'm not sure if this an obvious issue or not a
>> >concern as you are only doing single-sided work.
>> >
>> >For me, I find Sunday coupon pages printed on my old laser printer
>> >adequate enough but the real issue that eludes me from quick PCB
>> >fabrication is through-plating.
>> >
>> >I am intrigued though, why did you stop going with the direct laser
>> >method? The pictures you provided were most impressive but perhaps
>> >you ran into the same issue with double-sided boards?
>> >
>> >
>> >--- In Homebrew_PCBs@yahoogroups.com, mlerman@ wrote:
>> > >
>> > > I'm interested in the optics, too. If we could produce a spot of
>> > .001 inch or so, we could use it to remove paint resist from a pcb,
>> > just like a cnc router. Then etch and done. The plotter can be a
>> > simple xy - used ones with a resolution of .001 inch seem to be
>> > readily available.
>> > >
>> > > Mark
>> > >
>> > >
>> >
>> >
>> >
>> >
>> >------------------------------------
>> >
>> >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
>
>
>