Shop OBEX P1 Docs P2 Docs Learn Events
What is the very best way to get a positive image for PCB making these days? - Page 4 — Parallax Forums

What is the very best way to get a positive image for PCB making these days?

124

Comments

  • idbruceidbruce Posts: 6,197
    edited 2012-04-19 05:48
    @T Chap

    I am aware of double layering, however this is what I want to eliminate. On your laser printer, print out a 300 X 300 pixel filled black rectangle and you will see a definite dense outline around the perimeter. If this dense outline could be changed for the whole image, double layering could become a thing of the past.

    Bruce
  • LeonLeon Posts: 7,620
    edited 2012-04-19 05:48
    I always reverse the bottom image.
  • T ChapT Chap Posts: 4,223
    edited 2012-04-19 07:10
    Even if you can eliminate the fringe issue, you cannot rely on one mylar for a home process, whether laser or ink jet. It is begging for problems. Some traces will get lost, maybe not every board though.
  • idbruceidbruce Posts: 6,197
    edited 2012-04-19 07:35
    T Chap

    It is not really a fringe issue. The issue is to get all printed data to the same density as the outline. Considering that the proposed solution is not readily available on the internet, unless someone has tried this proposed solution discretely, I do not see how it can be unequivocally stated that proposed solution would not provide a photo mask dark enough to proficiently create PCBs.

    Bruce
  • idbruceidbruce Posts: 6,197
    edited 2012-04-19 07:40
    And just for the record.... At this point, I have completed at least 100 exposures concerning PCBs and laser printed photo masks, so I am fully aware of the circumstances.
  • LeonLeon Posts: 7,620
    edited 2012-04-19 08:10
    T Chap wrote: »
    Even if you can eliminate the fringe issue, you cannot rely on one mylar for a home process, whether laser or ink jet. It is begging for problems. Some traces will get lost, maybe not every board though.

    I use a single transparency, and never have problems. Other people who use inkjet printers achieve similar results.
  • idbruceidbruce Posts: 6,197
    edited 2012-04-19 10:41
    To Those That Might Be Interested

    In an effort to create a more opaque photo mask in a single layer, I attempted to utilize PCL5 to manipulate the printer. Although I was eventually able to programatically communicate with the printer, at this point I have had no success, however it was an interesting learning process, and I will try again at a later date.

    Here is an interesting page for those of you who may have a PCL5 compatible printer: And here is an easy example of using PCL5: Bruce
  • idbruceidbruce Posts: 6,197
    edited 2012-04-19 19:02
    To Those That May Be Interested

    Upon further examination and consideration, I have now come to the conclusion that the darker rectangle surrounding the faded rectangle is not software related. I now believe it is just build up of melted toner that has gone to the outer edges.

    So now I wonder what the printed output would look like if the laser printer fuser temperature was reduced to a point of just adhering the toner to the paper or transparency.

    Bruce
  • idbruceidbruce Posts: 6,197
    edited 2012-04-20 04:02
    To Those That May Be Interested

    In reference to my last few previous posts, but more particularly the last post, I have decided to devote one more day to experimenting with my LaserJet 6L. For today's experiments, I will be exploring various options concerning the fusing of toner to transparency film with a laser printer.

    For my first experiment, I will attempt to completely disassemble and disconnect the fuser and roller assemblies, and allow the printer to exit a film with loosely positioned toner to see what it looks like untouched by further processing. If it looks good (better than the normal printing procedure), I will then bake the film to see if I can get the toner to adhere to the film with the absence of pressure. I believe this would be the best possible scenario.

    Another option is to reassemble the printer with the fuser and roller assemblies, and decrease the current available to the fuser for a cooler fusing process.

    Bruce
  • LeonLeon Posts: 7,620
    edited 2012-04-20 04:39
    The film you are using might be the cause of your problems. You should try a film that is made for PCB transparencies, such as Mega Electronics LaserStar:

    http://www.megauk.com/artwork_films.php

    It worked very well in my LaserJet II, many years ago.

    They will probably send you a sample if you ask them nicely. They let me have a sample of their JetStar film for testing.

    It's available in the USA from Farnell.

    I even got my laser printer to produce a transparency by using ordinary paper, and making it translucent with vegetable oil. It worked OK, but was a bit messy. It does illustrate how non-critical the standard photo-etch technique can be.

    I can't understand why you are having so many problems. You've been at it for nearly six months, and still haven't produced a single working board! :)
  • idbruceidbruce Posts: 6,197
    edited 2012-04-20 05:44
    Leon

    I got my film from a graphics art supply and it is supposed to be used for plate making in conjunction with laser printers, and it would work perfectly, if I was not trying to conquer copper pours. As it stands, considering the translucency of the toner within copper pours, if I expose for the pour, than the line art is improperly exposed, and if I expose for the line art than the pour is improperly exposed. It is really wierd. If I was to expose the top and the bottom at different exposure times, I am sure I could create the perfect sample.

    I know you have seen this file before, but here it is again. I can either expose the bottom perfectly or I can expose the top perfectly, but not both, all based upon exposure time. I am certain that if I had a laser photo plot, it would be a cake walk.

    If I expose for the top layer, the photo resist washes off of the copper pour on the bottom layer during developing, due to the translucency.

    If I expose for the bottom layer, the photo resist on the top first becomes fuzzy and then washes away while I try to remove the haze.

    I have perfect etchings for both the top and bottom layers, but not on the same sample chip. And of course the goal is for perfect exposure and perfect etching for both layers on the same sample chip.

    Bruce
  • idbruceidbruce Posts: 6,197
    edited 2012-04-20 05:47
    Leon

    Additionally, the instructions for the film are to use the lowest possible fuser setting, however, the LaserJet 6L does not provide any means for lowering the fuser temperature.

    Bruce
  • LeonLeon Posts: 7,620
    edited 2012-04-20 06:19
    That would be one reason for trying the LaserStar film. It doesn't have any such recommendation.

    I don't think that I'd have had a problem with those top and bottom patterns when I used a laser printer. I think that you might have a problem with light leakage, which doesn't arise with UV-sensitive material. Making two single-sided boards might show the cause of the problem.
  • idbruceidbruce Posts: 6,197
    edited 2012-04-20 06:26
    Leon

    Perhaps it is the fuser temperature, because a lower temperature would defintely result in a thicker toner layer on the film.
    You've been at it for nearly six months, and still haven't produced a single working board!

    You're funny!

    The fact is that I can create a working board any day of the week, but once again, I am striving for perfection. If I have to, I will go get the photo plot, but I truly want a LOCAL solution.

    Bruce
  • idbruceidbruce Posts: 6,197
    edited 2012-04-20 06:29
    Making two single-sided boards might show the cause of the problem.

    I am sure a lot of it pertains to the close proximity of the board to the light source, and of course there is always the possibility that one light is stronger than the other.
  • idbruceidbruce Posts: 6,197
    edited 2012-04-21 04:31
    Leon

    You better start practicing, because I am getting ready to start giving you lessons :)

    Bruce
  • LeonLeon Posts: 7,620
    edited 2012-04-21 04:43
    Have you fixed the problems?
  • idbruceidbruce Posts: 6,197
    edited 2012-04-21 05:08
    Leon

    For the last couple days, in addition to various exposure tests, I have also been experimenting with various mask materials. Last night I attempted another exposure with vellum, only this time I sprayed it with a clear coat. I exposed both sides at the same time, for 1 minute and 35 seconds. After developing the sample, it appeared to look pretty darn good on both top and bottom. Then upon closer inspection with a magnifying glass, I noticed that the traces appeared a little jagged, however the films produce nice and smooth lines. So I decided to give one of my thickest films with a matte finish another try, only this time, I put a clear coat on both sides of the film. With the exception of a small light bleed on one edge, the sample came out of the exposure and developing process looking quite "stupendous" :)

    This particular film with a clear coating on both sides in conjunction with a 1 minute and 35 second exposure time, as well as a developer mixture of 13 grams sodium hydroxide to 1 liter of water, appears to be the perfect combination.

    Over the last couple days, I have noticed light bleeds on one edge only, and always on the bottom layer. Once I figure out how this light is getting in, I should have the perfect process.

    Bruce
  • idbruceidbruce Posts: 6,197
    edited 2012-04-21 05:13
    It is noteworthy to mention that when I ordered new boards, they came in new packaging, and the protective covering seems a little more difficult to remove. Perhaps the film is creeping back on me during the exposure process. I will have to pay better attention after the next exposure to see if this is the cause.

    Bruce
  • idbruceidbruce Posts: 6,197
    edited 2012-04-21 06:23
    I would imagine that I should have a "beautiful" sample to photograph sometime later today. And when I say "beautiful", I mean it will be a "10" on a scale of 1-10, with 10 being the highest rating. :)

    Am I telling tall tales? :)

    I don't think so. :)
  • idbruceidbruce Posts: 6,197
    edited 2012-04-23 14:12
    Hello Everyone

    I just wanted to say that I have a very nice looking sample chip to show you guys, but I do not have any photographs yet. I will probably tin plate the sample before taking the photos.

    Bruce
  • $WMc%$WMc% Posts: 1,884
    edited 2012-04-24 19:13
    Hello Bruce
    '
    I'd like to see the PICs of the exposure when you get them ready to post. I have found copper pours the most difficult to do.
    '
    Had a freak storm come through FL.
    '
    Lightin strike has about put me in the dirt.
    '
    I hope to recover in a week or so.
  • idbruceidbruce Posts: 6,197
    edited 2012-04-24 20:05
    Walt

    The way I interpret your post, you must have got personally struck by the bolt, if that is the case, I am sorry to hear that and I hope you get better soon.

    As for the sample, it looks pretty darn good. The file attached in Post #102 represents the sample that I made. It does not have much of a copper pour, but there is a good chunk of solid copper on the back side with no pin holes. I will post photos soon.

    After all my testing, I have come to the conclusion that all my problems were either associated with improper developer temperature or an inadequate photo mask. I have finally achieved good temperature control for the developer and I am currently working on something to get better masks.

    Since I am assuming that you got struck, I will also assume that you have not felt good enough to read the forums. Since we last chatted, I have started several knew threads, and I think they have some very good content. You might want to check them out.
    New Project/New Experiment - PCB Etchant And Developer Heaters
    http://forums.parallax.com/showthrea...eloper-Heaters
    Bruce
  • $WMc%$WMc% Posts: 1,884
    edited 2012-04-25 19:36
    Hello Bruce
    '
    I wasn't hit by the strike, But my house was.The strike knocked out my PCs and the back-up usb drives that I had....A lot of data lost.
    '
    Now I see a use for Cloud...
    '
    I'll check out your posts
  • idbruceidbruce Posts: 6,197
    edited 2012-04-27 17:13
    Walt

    Glad to hear that you were not hit, but sorry about the data loss. I truly know what it is like to lose data.

    Bruce
  • idbruceidbruce Posts: 6,197
    edited 2014-07-01 19:48
    Hello Everyone

    It has been two years since this thread has seen any action, but since my current endeavors and problems apply to this thread, I thought I would make a post.

    In the last two weeks, I have attempted to make two boards. The first board turned out nice on the first try, but the second board was really quite disappointing, with the copper pours being highly faded, and I just did not waste my time attempting to etch it. So here I am, stuck again with the same old problem, being a poor positive image for exposure. So I fiddled around all day, toying with various printers, which left me quite frustrated, and having to face the fact that print shops no longer make positive films, made my situation even worse.

    A while back, Phil helped me set up the thermistor for a modified LaserJet 6L printer, and it worked pretty well, but has since become filthy with excessive toner buildup. To sum that up, I just got tired of taking it apart and cleaning it. So instead of cleaning and using the LaserJet 6L, I decided to use the LaserJet 2000D in an unmodified form. I began by testing various professional transparency films, but none of them even came close to achieving the film I desired. Being an experimental type person, I decided to see how many layers of printed films it would actually take to make a combined film that would be fairly opaque. Altogether, it took three layers of film with printed circuitry to form a pretty dense positive. I have heard about and even attempted doubling up films, but three of them....

    Taking the layers one at a time, I taped the top and bottom of each layer to a white layout board, aligning them as I progressed. I then trimmed one side to the final dimension and lifted all three layers and inserted scotch tape under the layers and folded it over the top of the layers. After doing this, I then done the other side the same way, and then finally the top and bottom.

    As it stands, the combined positive is fairly stout and pretty much opaque, but still allows light to pass through the transparent areas. To compensate for all these layers, I will undoubtedly have to increase my exposure time and experiment a little.

    Of course I am hoping this will be the end of my woes concerning positive films. I will let you know if it works or not.
  • TubularTubular Posts: 4,703
    edited 2014-07-01 20:32
    I've found if you accept a bit of pitting in your copper pours, generally one layer of film is sufficient, even if the laser printer isn't great. Two layers is better, particularly if using overhead transparency film.

    Also its worth mirroring the top layer before you print it, as this lets you place it directly against the copper pcb. That closeness stops light ducking under the substrate, blurring the definition a bit.

    And, are you printing on Mylar (polyester) - if not this is well worthwhile. It does seem to absorb a bit of the UV
  • idbruceidbruce Posts: 6,197
    edited 2014-07-01 21:02
    Okay....

    My normal exposure time is 1:30, and since I was going with three layers, I decided to bump it up to 2:30. After several minutes in the developer, not even a single trace could be seen and the resist was all in tact. I will now attempt a 5:00 exposure time. Hopefully I will get it in the ballpark before I eat up to much material.
  • idbruceidbruce Posts: 6,197
    edited 2014-07-01 21:04
    Tubular

    I would assume the film is polyester, although I am not 100 % certain. It is a milky white color.
  • TubularTubular Posts: 4,703
    edited 2014-07-01 21:09
    Yep that sounds like the one. Its good stuff, but I have found getting the toner against the pcb (mirror printing) helps with definition. Of course the pits are better defined too, but it might save you a layer

    I've also found 2 layers needs much longer exposure time (~double).

    I'm getting close to revisiting all this again with old HP laserjets, 4 of them to try at last count, vs more modern samsung, oki and lexmarks. I'd also like to try direct printing onto the adafruit copper clad flex (newly available) and see how that goes
Sign In or Register to comment.