Shop OBEX P1 Docs P2 Docs Learn Events
ATTN: PCBers SUBJECT: New website and exposure box design for your creative needs - Page 8 — Parallax Forums

ATTN: PCBers SUBJECT: New website and exposure box design for your creative needs

1456810

Comments

  • idbruceidbruce Posts: 6,197
    edited 2012-03-17 06:50
    In addition to the sample of above, I have attached another pdf of another sample that I will attempt to conquer in the next several days.

    Bruce
  • idbruceidbruce Posts: 6,197
    edited 2012-03-22 08:00
    My new printer (used from eBay) is supposed to be delivered sometime later today. I am very anxious to hook it up and see if it can output a nice photomask. At there very least, I am hoping that it will produce a much better mask than my LaserJet 6L. Sometime later today, I will post my opinion of the output. I am keeping fingers and toes crossed until then.

    Bruce
  • idbruceidbruce Posts: 6,197
    edited 2012-03-23 03:04
    For Those That May Be Interested

    I must say that overall, the LaserJet 2200D is a pretty decent printer for normal monochrome business printing, however, I am not that impessed with it's output at creaing fine line (4/4) for photo masks. At 1200 DPI, it is not much better than 600 DPI of the LaserJet 6L, but it is definitely a little better. When compared to the print out that Leon so graciously sent me, the output from the LaserJet 2200D appears to have much crisper lines, however, when it comes down to the 4/4 pattern, the output from Leon's 1200 DPI inkjet definitely has much better line separation. I am still uncertain whether this is a hardware or a software issue, because the LaserJet 6L had the same issue of line separation pertaining to a 4/4 pattern. Perhaps if I was to downoad the same PCB editor that Leon uses, and ask him for a copy of the file sent me, I could then determine whether it was hardware or software.

    Somewhere along the line of testing, I have become obsessed with achieving a 4/4 pattern, and I am certain that this has tremendously sidetracked me from properly testing and documenting the use of my exposure box in combination with the use of Philmore DATAK Premier Series Presensitized Positive Acting PCBs. While it is true that I was able to produce a nice 4 mil wire trace, that does not mean that the PCB film will consistently produce a nice 4/4 pattern. Since my last bout of testing, I have done a fair amount of research pertaining to trace width and trace separation. According to my understanding, PCB manufactures all have a set of rules pertaining to minimum trace widths and minimum trace spacing. These minimums are based upon their production capabilities, which generally equates to the quality of the film being used, as well as the exposure and etching equipment. From my research, I have ascertained that I am striving results that far exceed the minimums set by many board manufacturers. When I started this project, it was never my intent to exceed these minimums. My main goal for this project was to create an exposure box that would allow hobbyists and enthusiasts to quickly and easily create a double-sided PCB of decent quality. In order to achieve this goal, I have come to the conclusion that I must also establish a set of rules and guidelines.

    Establishing a rule or guideline pertaining to the use of printers for creating a mask, is definitely not something that I want to do, considering that there is wide difference in opinion pertaining to this issue. This difference of opinion is not uncommon and it has occured within this thread. Leon is a firm believer in creating photomasks with an inkjet printer, and I am a firm believer in using a laser printer. As I have learned the hard way (when comparing Leon's output to mine), there are pros and cons to both types of printers. Without further testing, it is currently my opinion that 1200 DPI laser printers output much crisper lines than a 1200 DPI inkjet printer, but the 1200 DPI inkjet printer has much better fine line seperation than a 1200 DPI laser printer. Either way, I am sure both types of printers are sufficient for the hobbyist or enthusiast that wants to create a PCB board. Considering that most people are probably not as dedicated to this process as Leon and I, I believe most people will not have a 1200 DPI printer at their disposal, however, I do believe that most people will have access to a printer capable of 600 DPI. With this in mind, I believe that at a bare minimum, a 600 DPI printer should be used for creating photomasks. Which brings about another question for Leon?

    @Leon - Is a 600 DPI inkjet printer sufficient at creating photomasks?

    Bruce
  • LeonLeon Posts: 7,620
    edited 2012-03-23 04:36
    When I used a laser printer for the artwork, I got the best results with Mega Electronics LaserStar film.

    I use Pulsonix:

    http://www.pulsonix.com

    The demo version will do what you want.

    I've tried my inkjet printer on 600 dpi, and the results weren't all that good. 600 dpi should be OK for 12/12 mil tracks, though. 1200 dpi gives excellent results, but the correct film (Mega Electronics JetStar Premium) must be used. Standard JetStar worked well with the Canon inkjet printer I used to use, I think that was 600 dpi. My current printer (HP 5940) wasn't expensive, but the ink cartridges are, they contain the print head. I've had the 5940 for several years, I think that they are still made.

    I don't see any point in going below 8/8 mil for PCBs made at home, perhaps with the occasional 6/6 mil track. I mostly use 10/10 and 12/12 mil
  • idbruceidbruce Posts: 6,197
    edited 2012-03-23 07:08
    @Leon
    I use Pulsonix:

    http://www.pulsonix.com

    The demo version will do what you want.

    I went to their website and attempted to download the trial verion, at which point they requested my personal information. I provided my information and they responded by saying they would send me an email with download information. Approximately 2 hrs. later, I still have not heard from them.

    Of course it is not your fault, but that upsets me, because they were persistent to make sure the form was completely filled out.
  • LeonLeon Posts: 7,620
    edited 2012-03-23 07:34
    I emailed them about the problem.

    I tried it and got an email immediately. Try my link:

    http://pulsonix.com/softwaredownloadrequest.asp?TV=Pulsonix&DC=211665.253473282&PC=cd959e2a090f54a06a687563660f2951
  • idbruceidbruce Posts: 6,197
    edited 2012-03-23 10:15
    @Leon
    I emailed them about the problem.

    I also emailed them
    To Whom It May Concern

    I recently attempted to download your trial version, at which point I was required to fill out a form pertaining to personal information about myself. After filling out the form with my personal information and submitting the form, I was instructed that download information would be emailed to me. It is now approximately 2 hours later and I still have not received an email with the proper instructions for downloading the trial version.

    Unless it is Pulsonix's goal to upset potential customers, this is a VERY BAD sales promotional technique.

    Between our two emails, the problem has been resolved. They sent me an email with an apology, a download link, and said they would look into resolving the problem. So I am not upset anymore :)

    This actually looks like some nice software. I have been exploring it for about an hour, and I can tell that it will take some time to get accustomed to it. In the hope of saving some time, I don't suppose you still have that test pattern handy do you? Or is it posted in an earlier post of yours? I will check.
  • LeonLeon Posts: 7,620
    edited 2012-03-23 10:30
    I've attached it.
  • idbruceidbruce Posts: 6,197
    edited 2012-03-23 11:09
    Leon

    I appreciate the effort Leon, but that confused me even more :) I am sure there is a way to scale it down, but instead I found two PDF files that you previously uploaded, and although it is not the exact same file that you just uploaded or sent to me, it does provide some insight. It now appears to be a software related issue to specific to Eagle, because printing the previous PDF upload with the LaserJet 2200D now provides clear and distinct lines for the 4/4 pattern.

    When comparing my current print out on paper to the paper print out that you sent along with the transparencies. The print out from the LaserJet 2200D looks much better. However, when comparing my current print out on paper to your transparencies, the output looks pretty equivalent between the two printers. Needless to say that your transparency print outs look far superior than you paper print outs. As for the LaserJet 2200D, I have not yet tested it for transparency printing.

    I am also going to print out the previous PDF with the LaserJet 6L and see how that compares.

    Whatever the case, I think I am just going to admit defeat on the 4/4 pattern and move on to a minimum pattern that works reliably in combination with my exposure box and at least a 600 DPI laser printer. If I am not mistaken, I believe I have already tried 4/4, 6/6, and 8/8 at 600 DPI without much success. So I believe for my next test, I will make a mask using the LaserJet 6L at 600 DPI with the following pattern and see what happens. If it fails at 600 DPI, I might attempt it at 1200 DPI with the LaserJet 2200D just for giggles.

    Bruce
  • LeonLeon Posts: 7,620
    edited 2012-03-23 11:24
    Did you download that file I attached to my last post? It's a Pulsonix PCB file and you should be able to open it by double-clicking it. Alternatively, open it from within Pulsonix.

    My paper printout would have been at 600 dpi (default printer setting). I only use 1200 dpi (with the HP Premium Inkjet Transparency setting) when actually creating the artwork on the JetStar film, I have a special setting for the printer in the CAM/Plot output settings. The film will always be superior to ordinary paper, because the latter absorbs the ink. If I used the paper used for printing photographs at 1200 dpi, the printout would probably be about the same as the Jetstar.
  • idbruceidbruce Posts: 6,197
    edited 2012-03-23 11:32
    Leon

    Yes, I did download the the file and opened it with Pulsonix, but when I tried to print it, it came out HUGE :) It would probably take me several hours to figure out how to scale it down to the proper size.

    Bruce
  • LeonLeon Posts: 7,620
    edited 2012-03-23 11:39
    Did you use the Output > CAM/Plot dialogue to print it? That should output it 1:1, by default.
  • idbruceidbruce Posts: 6,197
    edited 2012-03-23 12:04
    Leon
    Did you use the Output > CAM/Plot dialogue to print it?

    No I did not. Like I said, it would have taken me several hours to figure that out. Thanks for the tip, because it worked like a charm. There is definitely a problem with Eagle and the 4/4 pattern, because with Pulsonix, there is a definite seperation between the lines.


    As I mentioned in a previous post, the quality of your print out was much better on a sheet of transparency. I highly doubt it, but I wonder if a laser printer would also have better line definition on a transparent print out. I guess there is only one way to find out.

    Bruce
  • idbruceidbruce Posts: 6,197
    edited 2012-03-23 12:11
    Leon

    Now I wonder about the rest of the patterns in Eagle, even though they look fine to the naked eye, they may also be off ever so slightly.

    Bruce
  • LeonLeon Posts: 7,620
    edited 2012-03-23 12:15
    The CAM/Plot dialogue can look complex, but it's actually quite straightforward - you just need to select Windows for output. Mirroring isn't used with Gerber output, but the top layer needs to be mirrored for printed artwork. I think that file of mine was set up for PDF output.

    Laser printer output should be better on film, as I suggested for inkjet output. The paper fibres will introduce "noise", even with toner.

    You should be able to set Eagle up for accurate output. Perhaps you should ask on the Homebrew PCB group, lots of people there use Eagle. I think that only one person there uses Pulsonix, and he bought it on my recommendation for use where he works.
  • idbruceidbruce Posts: 6,197
    edited 2012-03-24 14:47
    I just got finished from disassembling, cleaning (including the laser optics), and reassembling the LaserJet 6L. Once again, it prints immaculate looking sheets at 600 DPI.

    So I decided to print out Leons test pattern with once again using Pulsonix in combination with the LaserJet 6L. And once again, there was a definite separation of the lines in the 4/4 pattern.

    FOR THOSE OF YOU USING EAGLE (INCLUDING MYSELF), PLEASE BEWARE THAT THE SPACING IS INACCURATE WHEN PRINTING POSITIVES, ESPECIALLY CONCERNING 4/4 PATTERNS

    As opposed to learning a new layout editor, I will probably continue to use Eagle until I find more substantial problems.

    Bruce
  • Wa_MoWa_Mo Posts: 17
    edited 2012-03-26 09:26
    Interesting thread.

    I am also thinking about making my own boards. So, I scoured the net and found these two items:

    http://quinndunki.com/blondihacks/?p=835
    http://www.youtube.com/watch?v=tWnfnt2rNO0

    which I found to be of interest.
  • idbruceidbruce Posts: 6,197
    edited 2012-03-28 16:16
    Hello Everyone

    Today is the day for some more experimentation :) I have several masks printed, and I am almost ready to try and create some more boards. However, today I am going to try a different method.

    Pertaining to some previous comments that I made, I have given up on trying to acquire the 4/4 pattern, and additionally my main goal has changed. My new goal is to be able to create high quality boards with photo masks being created with the minimum of a 600 DPI laser printer. The two main reasons for changing my main goal is as follows:
    1. Most people probably would not want to invest the money to obtain a high quality laser photo plot for a simple hobby circuit.
    2. Most people probably have access to a 600 DPI laser printer.
    In an effort to help me obtain my new main goal, I am now going to test with a minimum trace width of 10 mil, however, I have one sample mask that necks down one of the traces to approximately 6 mil.

    One of the main problems that I have had in the past is that my masks were not truly opaque. Since I will not be testing with a laser photo plot, I have decided to double up the transparencies to help create a more opaque mask. Basically two transparencies will be printed for each layer to be exposed, they will then be cut to size, and the two identical transparencies will then be aligned in respect to each other and bonded together by smearing glue along their perimeters. As an additional step to make the masks more opaque, the transparency of each mask that will be placed against the board face will be top-coated with Acrylic Krylon Crystal Clear Satin to blend and darken the toner from the printer. I am sure that this method will vary my exposure time slightly, but I think it will produce much better boards in the long run.

    I still have a little work to finish before I start testing, so testing may extend into tommorow. I will begin my testing by trying to achieve a high quality sample of the small image attached below. If and when I achieve this high quality sample, I will then proceed to make a much larger board. For more information pertaining to the larger board, please refer to this thread:
    Hopefully, I will soon have a new sample board and a new stepper motor driver board.

    Bruce
  • idbruceidbruce Posts: 6,197
    edited 2012-04-05 15:31
    Leon

    A while ago, I read somewhere that developer gets old within a 24 hour period, and that you should always mix new batches when developing boards. In your experience, is there any validity to this?

    Bruce
  • LeonLeon Posts: 7,620
    edited 2012-04-05 15:58
    I use sodium hydroxide. It lasts indefinitely and I've been using the same stuff for months.
  • idbruceidbruce Posts: 6,197
    edited 2012-04-05 16:47
    Leon

    That's what I thought. He must be using something different. However, when I started mixing sodium hydroxide, it did appear to get old overnight, so I have been mixing fresh batches all the time. I mixed up a new batch yesterday and there is still a lot left, so instead of discarding the leftover, I figured I would ask you.

    Thanks Leon

    Bruce
  • LeonLeon Posts: 7,620
    edited 2012-04-05 18:06
    What do you mean by "getting old"? How are you storing it? I keep mine in a plastic milk bottle.
  • idbruceidbruce Posts: 6,197
    edited 2012-04-05 18:11
    Leon

    I mean it just did not seem to develop as well 24 hours later as compared to when it was initially mixed. Instead of making 16 oz. batches, I have been making 12 oz. batches, and storing it in a glass jar with a plastic lid.

    Bruce
  • LeonLeon Posts: 7,620
    edited 2012-04-05 18:36
    Mine works just as well months after I make it up. There just isn't any way it can deteriorate, apart from by absorbing CO2, and that will be minimal. It should be at normal room temperature, about 20C, of course. It takes about 20 seconds to remove the resist, with continuous agitation.
  • idbruceidbruce Posts: 6,197
    edited 2012-04-05 18:54
    Leon
    I mean it just did not seem to develop as well 24 hours later as compared to when it was initially mixed.

    This could have been due to an improper exposure time, but I thought I timed it correctly, and after what I read, I assumed it was fact. Anyhow, it is good to know that that instance was probably just a case of improper exposure and that my mixed chemicals will remain useable.

    Thanks Leon

    Bruce
  • idbruceidbruce Posts: 6,197
    edited 2012-04-06 08:52
    For Those Of You Following Along

    For the last several days, I have been spending a lot of time working on another thread which discusses photo masks: As I have determined through experimentation, my biggest problem has been that my photo masks are not truly opaque, and as reflected in the other thread, I have been creating a tool and experimenting with fusing two transparency layers together in an effort to obtain more opaque photo masks. For my first experiments with double layer masks, I created two masks, a top and bottom layer for a 15/16" X 15/16" board. The PDF file attached below, represents the two masks that I created. After a little experimentation, it was determined that the double layer mask definitely needs a longer exposure time, which of course makes perfect sense.

    After a few tries, the results are beginning to look quite promising. My last experiment had an exposure time of two minutes as compared to 1 1/2 minutes for a single layer mask. However, it now appears that the double layer masks may still need additional exposure time.

    Depending upon my ambition, hopefully I will have a nice sample board later today which will reflect the file below.

    Bruce
  • idbruceidbruce Posts: 6,197
    edited 2012-04-06 16:22
    To Those That May Be Interested

    As mentioned in the previous post, my biggest problem with making PCBs has been the opacity of the photo masks, however, another problem still remains and it is quite significant. Philmore DATAK Premier Series positive acting boards require a developer temperature between 100 and 110 degrees fahrenheit. If the temperature of the developer is below this temperature range, the board will not develop, and if the temperature of the developer is above this temperature range, it will quickly dissolve the photoresist from the board surface. I have ruined many samples and many experiments have gone bad, simply because the developer was not within the specified range, and todays experiments and sample tests are no exception to these minor disasters. At a cost of $0.39 per square inch (including shipping), repeatedly destroying board material can become a little expensive. Considering that my samples are approximately 1" X 1", I have thrown many $0.39 experiments into the trash bin due to improper developer temperature.

    Considering that I believe I now have a solution for my photo masks, I now want to tackle the developer temperature problem, because I am tired of ruining board material. In addition to developer temperature, etchant temperature is also a concern. The etchant temperature for these boards should be with the same range as the developer. So by solving the problem for one situation, I also solve it for the other. So my very next goal is to a permanent solution to these problems. The samples will have to wait just a bit longer.

    Bruce
  • idbruceidbruce Posts: 6,197
    edited 2012-04-15 00:06
    To Those That May Be Interested

    Since the creation of the Exposure Box/Cylinder a certain someone (not mentioning any names) has continually claimed to have a better and easier process. However, this unnamed person has never accepted the challenge of producing a quarter size double-sided board that is well aligned on both sides. Now this same person continually battles all my ideas. I personally think he is jealous and incapable of producing such a small board that looks as good as the one provided below. :):):)

    Now that my PCB Prototyping Table is complete and now that I have much better control over the temperature of my chemicals, I can now start experimenting with the exposures of my double-layered photo masks.

    Let's see if he can keep up, shall we? :):):)

    Over the next several days, we will see which one of us has the better technology and setup, and who can put out the highest quality board.

    Let the competition begin!

    Bruce
    642 x 708 - 63K
    642 x 708 - 64K
  • LeonLeon Posts: 7,620
    edited 2012-04-15 01:48
    Let's see a real board, with 8/8 mil tracks, and drilled holes.
  • idbruceidbruce Posts: 6,197
    edited 2012-04-15 01:49
    You first Leon
This discussion has been closed.