These films work perfectly with a white light source as well.
The photo plotting that I have accomplished uses a #222 flashlight bulb and ran for a year
with no complications in the output exposures.
I finally changed the bulb from guilt on the possibility of varied emission output over time.
It uses the pen down command of the plotter to fire the bulb.
The bulb has it's own lensing and launches perfectly into a cleaved POF (plastic optical fiber)
with the other end thinned and glued into the inside diameter of a stainless steel syringe barrel (to reduce wear).
just plot on the non-emulson side for scratch protection, if you're worried of the direct contact.
The direct contact was never an issue in testing, just polish the tip by plotting on paper for a few prints.
this light source costs 0.32 and you have 6 penlights in your lab already and works
with any XY table.
Bruce, the point is still FILTERS and 7 mil films for dimensional stability.
There is always numerous solutions, as I have proved with my exposure box. It is worth mentioning that Kodak ACCUMAX RED7 and FUGIFilm HG HPR-7 are specifically designed for exposure with a red laser. I think I will try the "Laser Schmaser" technique
However I must say that there are existing patents of exposing film with LEDs and fiberoptics, and that your proposed solution sounds nice.
635 - 660 nm is visible red and just about all visible red diodes will be in this range. You can probably use just about any pointer or module but plot times will vary with the available power.
You can probably use just about any pointer or module but plot times will vary with the available power.
I read somewhere that exposure can occur in uSecs, but as you say, it will all depend on available power. The more I investigate this project, the more interesting it all becomes. I believe it will be a serious challenge to get everything working and achieve nice output, but I also believe it is possible. I think the hardest part will be determining the DPI, applying it to the plotter, and providing the correct exposure time.
Earlier today, I had a discussion with another member about this project, and the subject of stepper resolution came up. Using 1/4-20 leadscrews and the GeckoDrive G251X drives, the resolution per step is 0.000025". While writing this response, I took a common DPI for laser photoplotters, being 4,000 DPI, and I devided that into 1". Without any type of microstepping, such as available with the G251X, a full step from a 1.8 step angle stepper motor would provide the resolution necessary to achieve 4,000 DPI. Now that is interesting.
I think the hardest part will be determining the DPI, applying it to the plotter, and providing the correct exposure time.
I'm not sure why determining dpi is hard but on the subject of exposure time, if you do a raster scan as opposed to cnc maching type paths then you can very easily have a constant head speed and then exposure is proportional to speed. It would then be very simple to find the correct exposure by writing lines with varying speeds.
Fine pitch leadscrews certainly give a lot of resolution but are a little slow, perhaps some resolution could be traded for speed because you could at the very least double the resolution. Or you could look at a belt drive with linear encoder. You could probably work out the rough speeds of the commercial printers using the printing areas, dpis and printing speeds.
Alternatively use a linear encoder to sync the laser to the movement and just use a stepper and belt for the motion, as long as the motion is reasonably smooth that would work great.
Actually, the way I see it, the DPI has nothing to do with the exposure time itself, because the DPI will be based on the image being rastered and the mechanical/electrical constraints of stepper resolution and the focus of the laser. As far as constant speed goes, this would be very desirable in a drum based solution, whereas in a linear solution, which I intend to build, to obtain maximum output, ramping of the motors will become a necessity, but exposure time could always be based upon the ramp. However I must agree that a constant speed solution would be a lot easier to build and program.
My main point was having the option for different DPIs like the commercial units. So let me rephrase, I believe the hard part will be focusing the laser to provide a specific DPI at the correct exposure time, based upon the output of the laser. In my opinion, this will take some serious experimenting. Additionally, exposure overlap of the lines must be considered. I believe the combination of the 1/4-20 leadscrews and the GeckoDrives will provide sufficient resolution to output some finely detailed graphics. Additionally, using 1/4-20 leadscrews and the GeckoDrives, my PulseTheStepPin driver should be able to move the gantry a minimum of 1"(X) X 1"(Y) every 1.3 seconds, however higher speeds should be obtainable. I believe overall that it will be a perfect solution, but focusing that laser to achieve DPIs will be the ultimate enemy.
Constant speed is no more or less a requirement for a drum or flatbed. You index one axis (linear or rotary) and then scan across with the other axis. Acceleration and deceleration can occur outside of the plotting area and besides with such a small load it could have a very steep ramp.
Do you really need to have changeable DPI? You are printing small films so you could just choose a high resolution. With a gantry you would be talking about having a z-axis or a bed that lifts to adjust the spot size.
Constant speed is no more or less a requirement for a drum or flatbed.
I would have to disagree. By utilizing a drum, you could simply set the motor at max speed for proper exposure, and just move the X axis line by line. Whereas with a flatbed table you must always change directions and either return the X axis to the beginning of the line, or reverse the image bits and expose on a return trip, either way the stepper motor must change directions and then either maintain a constant speed or ramp up. With a drum, it will be a simple step in one direction only.
Do you really need to have changeable DPI?
No, I don't really need different DPIs, but it is always nice to have options, and since the drive setup will be able to accommodate different DPIs, it would be nice to have.
With a gantry you would be talking about having a z-axis or a bed that lifts to adjust the spot size.
I agree that that spot size could be accomplished by raising or lowering the Z axis, but that is not the same as focusing the dot size with an objective lens. I think accuracy would be sacrificed.
Bruce, I think you are misunderstanding how the drum type is operated. You will increment the drum slightly then move the head across exposing selectively as you go, then increment a little again and then scan back across. This is absolutely identical to doing a raster type scan with a flatbed. It sounds like you assumed that the drum would rotate continuously a bit like a lathe, that would be a massive pain. The way they work is very much like a normal printer except the paper is wrapped around a drum.
I agree that that spot size could be accomplished by raising or lowering the Z axis, but that is not the same as focusing the dot size with an objective lens. I think accuracy would be sacrificed.
There are a few easy ways to change the spot size. Change the objective completely or change the beam diameter entering the lens which changes the effective aperture of the lens. The former is a pain, the latter reduces the amount of light available unless your means of changing the beam diameter is to change the collimator/beam expander. Although defocussing the objective slightly is not the same it would work and yes accuracy would be affected however that was kind of the point! All the energy is still delivered in the area and I doubt one type of spit will be any better than the other.
Graham
p.s. On the laser diodes Leon mentions, any laser pointer contains collimating optics. DVD players etc contain very cheap plastic focussing lenses so if they are good enough anything should work.
The drum type would definitely be easier to construct and configure, however, patents pertaining to flatbeds have only recently expired, I have not examined patents for drum type plotters, but I would imagine that some patents are still in effect.
I would be extremely surprised to find any patents covering the basic operation still in effect and unless you sell the devices or go in to mass production I would be even more surprised to find a patent holder who cared.
I have come across that web page a couple of times, and yes it is pretty cool.
From your example patent, it appears that drum plotters have been around for quite some time. I only searched for photoplotters, and the ones I came across were flatbed. I will have to do a little more research into that area.
I only found it because I looked at the wikipedia page and it mentioned the inventors of the first one (gerber instruments inc), very hard to know what to search for in general.
Normally when I find a patent of relevance, I will search all netted cited references. This will normally provide a good overview of the existing technology at that point in time.
Yesterday I took a little time out and tore apart a combination Phillips VHS/DVD player, and I must say that I am a little disappointed. The quality of the optical block looks a little cheesy to say the least, and no where near the quality of optical blocks that I have retrieved from CD players. However this laser is brand new because I never used the DVD player. I am now contemplating the removal of the laser diode and lense from the optical block and inserting them into a custom housing.
I finely got a chance to get a shot of a Gerber 24 Aperture Wheel. This was used on the Model 40 with a Halogen lamp.
The Model 40 was a 16" x 20" flatbed Photo Plotter using an Osram halogen bulb for exposure. The head was stationary, and the bed was driven in x and Y with stepper motors.
The ramp up and ramp down of the servos where tied to the lamp output. As the steppers ramped up, the lamp output increased. This prevented "bulbing" on the film.
If a constant light source, (as the LED being described), where to be used, a variable Neutral Density filter on a servo could be used to keep the light output consistent with the speed:
WOW that website wants a mint for their laser modules
Anyhow thanks for the photo. I have attached an altered photo to explain my interpretation of it, and this is how I interpret it:
Of course the center of wheel mounts to a motor or output shaft of some sort and the hole that is closest to the center hole is for an indexing pin to prevent misalignment.
The item with the red circle is for interchangeable apertures. The removal of two screws allows quick and easy replacement of any desired aperture.
The area enclosed by the blue circle contains a retaining spring which snaps into a groove around the circumference of an interchangeable aperture. This retaining ring retains a piece of glass and a piece of film that contains the desired image.
And the area enclosed by the yellow circle shows a specific visible aperture on film.
Nice to see an actual aperture wheel but why the analysis, are you still thinking of doing it that way?
Graham, I am convinced that raster processing is definitely the way to go, and I definitely will go that route. However, I must admit that the older technology has also caught my interest and that aperture wheel looks pretty darn cool. I like the whole concept and the way it is constructed.
Additionally, even though I have not done any thorough research into the subject, I think I might also be interested in Laser Direct Imaging. Therefore my new PCB driller might just become a three process type of equipment, which includes, PCB drilling, laser photoplotting, and laser direct imaging.
PublisonWOW that website wants a mint for their laser modulesAnyhow thanks for the photo. I have attached an altered photo to explain my interpretation of it, and this is how I interpret it:
Of course the center of wheel mounts to a motor or output shaft of some sort and the hole that is closest to the center hole is for an indexing pin to prevent misalignment.
The item with the red circle is for interchangeable apertures. The removal of two screws allows quick and easy replacement of any desired aperture.
The area enclosed by the blue circle contains a retaining spring which snaps into a groove around the circumference of an interchangeable aperture. This retaining ring retains a piece of glass and a piece of film that contains the desired image.
And the area enclosed by the yellow circle shows a specific visible aperture on film.
Bruce
The apertures are actually the polygon plastic shapes that were coated with an photographic emulsion and and exposed to create the the negative. Those are placed on precision pins in the aperture wheel. When the wheel was populated, according to the aperture list from the plot file, (Gerber File), a lines and pads test plot was run. Line apertures where usually run on the first 12 apertures, and the flash, (pads), where run on 13-24. Measurements was taken from the film to determine how far out of tolerance the results were.At this point, ND filters were introduced to compensate. These are the circular filters in this photo:
The retainer held a glass microscope slide cover, then the filter, and another glass cover over it. Then the spring clip held it all in.
The last photo shows the vertical pins on the aperture wheel that would lock the wheel into a pawl to position the wheel into it's final position.
There was also an overall ND filter below the lamp that could be changed as the lamp degraded.
May not be pertinent to the raster scanner you want to do, but I thought it would be of some use to you and others.
In your reply, I found the phrase "photographic emulsion", and I have been inclined to Google that phrase for quite some time, but it always slipped my mind.
One other question that maybe I missed in the threadb if you are using a drum; are you moving the laser assembly in the opposite axis to that of the drum or a fixed laser with a deflection mirror to move the beam? If using a mirror, you will need to compensate for the distortion in the spot at the drum *** the beam will be at an angle less than perpendicular to the drum as the beam moves farther away from the initial point of focus to the drum. Also if you use deflection of the beam, you will have to compensate for constant change in the linear velocity across the film (though not as much since you will be looking to produce exposed or not exposed rather than gray scale.
Beam. Deflection=greater scan speed and finer control (less mass etc)
Moving laser assy= cheaper easier slower maybe courser tollerances.
Maybe if you go the drum, see if you can find obsolete stuff in medical surplus forums. Look for the Kodak laser printerb drum scan with scan optics and visible red HeNe laser. Probably lots of surplus/expired film out there as well. Used standard developer chemistry as well.
Since I would be adapting it to my new PCB driller design, my initial plan was to have a flatbed, with the film traveling on both the y and x axises, and having the laser stationary to x and y, but adjustable on the z axis. However, the video that I linked to immediately before your post, demonstrates a successful implementation of Laser Direct Imaging. I just so happen to have a spare laser controller laying around. I may also attempt something similar to that gentlemens contraption.
Hard to see it until the laser gets turned on but it looks like the assembly has the optics to correct beam angle. It may take another listen or two to catch the whole concept. shame what he does to his optics with the smoke. The lens surfaces may be cleanable, but first surface mirrors as the octagon beam scan mirror may be harmed by the residue or cleaning attempt. Should have used that favorite Siemens training phrase "hear me when I say and believe me when I tell you".
By the way a possible source of laser scanners like the kodak may be dotmed.com, lots of goodies or leads there.
Comments
These films work perfectly with a white light source as well.
The photo plotting that I have accomplished uses a #222 flashlight bulb and ran for a year
with no complications in the output exposures.
I finally changed the bulb from guilt on the possibility of varied emission output over time.
It uses the pen down command of the plotter to fire the bulb.
The bulb has it's own lensing and launches perfectly into a cleaved POF (plastic optical fiber)
with the other end thinned and glued into the inside diameter of a stainless steel syringe barrel (to reduce wear).
just plot on the non-emulson side for scratch protection, if you're worried of the direct contact.
The direct contact was never an issue in testing, just polish the tip by plotting on paper for a few prints.
this light source costs 0.32 and you have 6 penlights in your lab already and works
with any XY table.
Bruce, the point is still FILTERS and 7 mil films for dimensional stability.
jr
There is always numerous solutions, as I have proved with my exposure box. It is worth mentioning that Kodak ACCUMAX RED7 and FUGIFilm HG HPR-7 are specifically designed for exposure with a red laser. I think I will try the "Laser Schmaser" technique
However I must say that there are existing patents of exposing film with LEDs and fiberoptics, and that your proposed solution sounds nice.
Bruce
Graham
I read somewhere that exposure can occur in uSecs, but as you say, it will all depend on available power. The more I investigate this project, the more interesting it all becomes. I believe it will be a serious challenge to get everything working and achieve nice output, but I also believe it is possible. I think the hardest part will be determining the DPI, applying it to the plotter, and providing the correct exposure time.
Earlier today, I had a discussion with another member about this project, and the subject of stepper resolution came up. Using 1/4-20 leadscrews and the GeckoDrive G251X drives, the resolution per step is 0.000025". While writing this response, I took a common DPI for laser photoplotters, being 4,000 DPI, and I devided that into 1". Without any type of microstepping, such as available with the G251X, a full step from a 1.8 step angle stepper motor would provide the resolution necessary to achieve 4,000 DPI. Now that is interesting.
Bruce
I'm not sure why determining dpi is hard but on the subject of exposure time, if you do a raster scan as opposed to cnc maching type paths then you can very easily have a constant head speed and then exposure is proportional to speed. It would then be very simple to find the correct exposure by writing lines with varying speeds.
Fine pitch leadscrews certainly give a lot of resolution but are a little slow, perhaps some resolution could be traded for speed because you could at the very least double the resolution. Or you could look at a belt drive with linear encoder. You could probably work out the rough speeds of the commercial printers using the printing areas, dpis and printing speeds.
Alternatively use a linear encoder to sync the laser to the movement and just use a stepper and belt for the motion, as long as the motion is reasonably smooth that would work great.
Graham
Actually, the way I see it, the DPI has nothing to do with the exposure time itself, because the DPI will be based on the image being rastered and the mechanical/electrical constraints of stepper resolution and the focus of the laser. As far as constant speed goes, this would be very desirable in a drum based solution, whereas in a linear solution, which I intend to build, to obtain maximum output, ramping of the motors will become a necessity, but exposure time could always be based upon the ramp. However I must agree that a constant speed solution would be a lot easier to build and program.
My main point was having the option for different DPIs like the commercial units. So let me rephrase, I believe the hard part will be focusing the laser to provide a specific DPI at the correct exposure time, based upon the output of the laser. In my opinion, this will take some serious experimenting. Additionally, exposure overlap of the lines must be considered. I believe the combination of the 1/4-20 leadscrews and the GeckoDrives will provide sufficient resolution to output some finely detailed graphics. Additionally, using 1/4-20 leadscrews and the GeckoDrives, my PulseTheStepPin driver should be able to move the gantry a minimum of 1"(X) X 1"(Y) every 1.3 seconds, however higher speeds should be obtainable. I believe overall that it will be a perfect solution, but focusing that laser to achieve DPIs will be the ultimate enemy.
Bruce
Constant speed is no more or less a requirement for a drum or flatbed. You index one axis (linear or rotary) and then scan across with the other axis. Acceleration and deceleration can occur outside of the plotting area and besides with such a small load it could have a very steep ramp.
Do you really need to have changeable DPI? You are printing small films so you could just choose a high resolution. With a gantry you would be talking about having a z-axis or a bed that lifts to adjust the spot size.
Graham
Sorry, I misinterpreted your post.
I would have to disagree. By utilizing a drum, you could simply set the motor at max speed for proper exposure, and just move the X axis line by line. Whereas with a flatbed table you must always change directions and either return the X axis to the beginning of the line, or reverse the image bits and expose on a return trip, either way the stepper motor must change directions and then either maintain a constant speed or ramp up. With a drum, it will be a simple step in one direction only.
No, I don't really need different DPIs, but it is always nice to have options, and since the drive setup will be able to accommodate different DPIs, it would be nice to have.
I agree that that spot size could be accomplished by raising or lowering the Z axis, but that is not the same as focusing the dot size with an objective lens. I think accuracy would be sacrificed.
Bruce
There are a few easy ways to change the spot size. Change the objective completely or change the beam diameter entering the lens which changes the effective aperture of the lens. The former is a pain, the latter reduces the amount of light available unless your means of changing the beam diameter is to change the collimator/beam expander. Although defocussing the objective slightly is not the same it would work and yes accuracy would be affected however that was kind of the point! All the energy is still delivered in the area and I doubt one type of spit will be any better than the other.
Graham
p.s. On the laser diodes Leon mentions, any laser pointer contains collimating optics. DVD players etc contain very cheap plastic focussing lenses so if they are good enough anything should work.
http://www.youtube.com/watch?v=Xj5TTDI-s5o
So yep, drum type makes even more sense. Continuous rotation then just increment the x axis one step at a time, of course no spiral needed, duh!
The drum type would definitely be easier to construct and configure, however, patents pertaining to flatbeds have only recently expired, I have not examined patents for drum type plotters, but I would imagine that some patents are still in effect.
Bruce
This is pretty cool: http://www.franksworkshop.com.au/Electronics/Photoplotter/Photoplotter.htm
Graham
www.google.co.uk/patents/US3984747
I have come across that web page a couple of times, and yes it is pretty cool.
From your example patent, it appears that drum plotters have been around for quite some time. I only searched for photoplotters, and the ones I came across were flatbed. I will have to do a little more research into that area.
Bruce
Graham
Yesterday I took a little time out and tore apart a combination Phillips VHS/DVD player, and I must say that I am a little disappointed. The quality of the optical block looks a little cheesy to say the least, and no where near the quality of optical blocks that I have retrieved from CD players. However this laser is brand new because I never used the DVD player. I am now contemplating the removal of the laser diode and lense from the optical block and inserting them into a custom housing.
Bruce
The Model 40 was a 16" x 20" flatbed Photo Plotter using an Osram halogen bulb for exposure. The head was stationary, and the bed was driven in x and Y with stepper motors.
The ramp up and ramp down of the servos where tied to the lamp output. As the steppers ramped up, the lamp output increased. This prevented "bulbing" on the film.
If a constant light source, (as the LED being described), where to be used, a variable Neutral Density filter on a servo could be used to keep the light output consistent with the speed:
http://www.newport.com/Circular-Variable-Metallic-Neutral-Density-Filters/141140/1033/info.aspx
Jim
WOW that website wants a mint for their laser modules
Anyhow thanks for the photo. I have attached an altered photo to explain my interpretation of it, and this is how I interpret it:
- Of course the center of wheel mounts to a motor or output shaft of some sort and the hole that is closest to the center hole is for an indexing pin to prevent misalignment.
- The item with the red circle is for interchangeable apertures. The removal of two screws allows quick and easy replacement of any desired aperture.
- The area enclosed by the blue circle contains a retaining spring which snaps into a groove around the circumference of an interchangeable aperture. This retaining ring retains a piece of glass and a piece of film that contains the desired image.
- And the area enclosed by the yellow circle shows a specific visible aperture on film.
BruceI now see the squares that contain the image.
Nice to see an actual aperture wheel but why the analysis, are you still thinking of doing it that way?
Graham
Graham, I am convinced that raster processing is definitely the way to go, and I definitely will go that route. However, I must admit that the older technology has also caught my interest and that aperture wheel looks pretty darn cool. I like the whole concept and the way it is constructed.
Additionally, even though I have not done any thorough research into the subject, I think I might also be interested in Laser Direct Imaging. Therefore my new PCB driller might just become a three process type of equipment, which includes, PCB drilling, laser photoplotting, and laser direct imaging.
Bruce
Graham,
You are correct about the price. Neutral Density filters seem to command a high price.
Looks I'm sitting on $988.00 USD worth of filters, (.1-.9, 1.0, 2.0, 3.0, 4.0), according to Edmund Optics:
http://www.edmundoptics.com/products/displayproduct.cfm?productid=2928
Actually I have two complete unopened sets, and about 20-25 open packages.
Jim
The retainer held a glass microscope slide cover, then the filter, and another glass cover over it. Then the spring clip held it all in.
The last photo shows the vertical pins on the aperture wheel that would lock the wheel into a pawl to position the wheel into it's final position.
There was also an overall ND filter below the lamp that could be changed as the lamp degraded.
May not be pertinent to the raster scanner you want to do, but I thought it would be of some use to you and others.
Jim
In your reply, I found the phrase "photographic emulsion", and I have been inclined to Google that phrase for quite some time, but it always slipped my mind.
For those that may be interested in creating emulsions, you may want to take a peak at this: http://unblinkingeye.com/Articles/Emulsion/emulsion.html
Bruce
http://www.youtube.com/watch?v=fi4P-Bwc6g8&feature=related
Beam. Deflection=greater scan speed and finer control (less mass etc)
Moving laser assy= cheaper easier slower maybe courser tollerances.
Maybe if you go the drum, see if you can find obsolete stuff in medical surplus forums. Look for the Kodak laser printerb drum scan with scan optics and visible red HeNe laser. Probably lots of surplus/expired film out there as well. Used standard developer chemistry as well.
FF
Since I would be adapting it to my new PCB driller design, my initial plan was to have a flatbed, with the film traveling on both the y and x axises, and having the laser stationary to x and y, but adjustable on the z axis. However, the video that I linked to immediately before your post, demonstrates a successful implementation of Laser Direct Imaging. I just so happen to have a spare laser controller laying around. I may also attempt something similar to that gentlemens contraption.
Bruce
By the way a possible source of laser scanners like the kodak may be dotmed.com, lots of goodies or leads there.
Frank