Shop OBEX P1 Docs P2 Docs Learn Events
ATTN: PCBers SUBJECT: New website and exposure box design for your creative needs - Page 6 — Parallax Forums

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

1468910

Comments

  • idbruceidbruce Posts: 6,197
    edited 2012-02-09 05:06
    @Leon
    You should really be testing with the inter-track spacing equal to the track width, such as 8/8 mils. Just achieving narrow tracks with wide spacings is quite easy, but that isn't what people use when designing boards for maximum track density.

    I can't believe that all I get is criticism from you, but that is all I see. If it is so easy, I have provided the board file, let's see you make one just as nice or better than the samples I have provided.

    Bruce
  • LeonLeon Posts: 7,620
    edited 2012-02-09 05:14
    I'm just providing constructive criticism. Equal track width and inter-track spacing is what users will require, and is how most people test their PCB process.

    Few people here seem to be making their own PCBs, you should ask on the Homebrew PCB group for comments on your proposed system.

    I've been making my own PCBs for over 40 years, using a variety of techniques, so I know a bit about it.

    I'll import your Eagle file into Pulsonix and see how I get on with it. I might need to order some fresh PCB material. As I've said before, 8/8 mil is the best I can do routinely. 6/6 tracks looked a bit "lumpy" when I tried it, but there weren't any actual breaks.

    I just created the transparency. The 4 and 6 mil tracks aren't all that good under magnification, but the 8 mil track looks OK. All the tracks have somewhat rough edges, but that is par for the course with inkjets. It's better than a laser printer, but nowhere near as good as a photoplot.
  • idbruceidbruce Posts: 6,197
    edited 2012-02-09 06:06
    Leon
    I've been making my own PCBs for over 40 years, using a variety of techniques, so I know a bit about it.

    There is no doubt in my mind about that. This exposure box/cylinder was designed with the electronics hobbyist and enthuiast in mind, and it was never intended for professional circuit board designers. However, there is no doubt in my mind that if I had a real nice positive in my hand that I could achieve 6/6 and perhaps 4/4 wire traces.

    Bruce

    PS Don't worry about Leon, unless you like the challenge. I am sure you can do it.
  • LeonLeon Posts: 7,620
    edited 2012-02-09 06:22
    Yes, a good transparency is vital. I can send you one of mine for that test board of yours.

    What about that suggestion of mine that you try the Homebrew PCB group for comments?

    Here is a test pattern of mine (PDF), with 8/8, 10/10, 12/12 and 15/15 tracks.
  • idbruceidbruce Posts: 6,197
    edited 2012-02-09 06:45
    Leon

    Before I toss it in front of the Homebrew PCB group, I want to do some more experimenting now, and perhaps try the 6/6 and 4/4. Constructive criticism accepted. I appreciate the offer for the transparency, but as I mentioned in an earlier post, there is a print shop within an hours drive that can do digital to film for me. The only downside is the fact that he cannot utilize Gerber files. He wants both PDF and PS instead.

    I definitely cannot produce a 4/4 pattern with my printer, but I am now drawing up a 6/6 pattern, and I will see how that looks.

    Bruce
  • ctwardellctwardell Posts: 1,716
    edited 2012-02-09 06:52
    Bruce and Leon,

    Thanks for addressing my question about why you guys make your own boards, the turn around time certainly makes since.

    I used to make my own years ago using the dry transfers from Radio Shack, I haven't tried the photoresist method.

    Maybe I'll give it a go on my next project that doesn't require a lot of holes or vias.

    Thanks again,

    C.W.
  • idbruceidbruce Posts: 6,197
    edited 2012-02-09 07:11
    Leon

    For some reason, my printer is not equally spacing 6/6, looks like I will have to go to the printer. Over the last couple days, I have been doing a lot of research on photoplotters, and I am contemplating to try and build one. Considering that I will dealing with 3 X 4 boards or less, it shouldn't be too expensive or complicated. I believe CD Writers have a laser diode in the 670 nm range which would be ideal for Kodak AccuMax Film, however I am uncertain if the wattage would be enough. I seen that a Phil P was nosing around in the Homebrew PCB and was asking questions about a plotter. I wondered if it was Phil Pilgrim :)

    Bruce
  • LeonLeon Posts: 7,620
    edited 2012-02-09 07:53
    I can print 6/6 and 4/4 on my HP 5940, at both 600 and 1200 dpi. I've added them to my test pattern.

    It's intended for printing on A4, of course.

    I tried your pattern, but the PCB material was too old and didn't etch properly. I'll order some fresh stuff tomorrow,
  • Phil Pilgrim (PhiPi)Phil Pilgrim (PhiPi) Posts: 23,514
    edited 2012-02-09 09:35
    idbruce wrote:
    I seen that a Phil P was nosing around in the Homebrew PCB and was asking questions about a plotter. I wondered if it was Phil Pilgrim
    Nope, I haven't made my own boards in years. The etchant fumes started to rust all my tools, I've got stains on my countertop, and I finally decided it wasn't worth the mess, hassle, and mediocre results with no plated through-holes.. Plus, the local print shop doesn't have or use a flim printer anymore. Add to that the ready availability of professionally-made proto boards in a couple days for very little money, doing it myself no longer made any sense.

    -Phil
  • idbruceidbruce Posts: 6,197
    edited 2012-02-09 13:50
    @Phil

    I thought for sure it was you. It was one of those very technical questions about focusing the laser beam output through various optics in order to achieve a higher dpi :) But now that you mention it, I rarely see you ask a question:)
    The etchant fumes started to rust all my tools

    I have not heard that one before, but it makes perfect sense.

    Bruce
  • idbruceidbruce Posts: 6,197
    edited 2012-02-09 16:41
    To Whom It May Concern

    In a recent post, I mentioned the possibility of creating a photoplotter for making PCB masks. Here is a very interesting document which is titled "Gerber RS-274X Format User's Guide". This document is essential for anyone even considering the creation of a photoplotter for making PCB masks.

    http://mcglothin.us/RobotScrapbook/CommunityPcbMill2005/IntroductionToGCodes/rs274xc.pdf

    Bruce
  • idbruceidbruce Posts: 6,197
    edited 2012-02-10 18:11
    UPDATE - NEW EXPOSURE BOX/CYLINDER CHALLENGE

    Creating a decent photomask is of the utmost importantace. As Leon pointed out, my samples should include spacing equal to the trace widths. After pushing the LaserJet 6L to it's limits, I have discovered that I can only reliably produce a 8/8 pattern. So I have decided, to run a new test for the rear side of the sample PCB which will include an 8/8 pattern and two copper pours, having 0.008" isolation. I have attached front and rear images for the new sample to be tested.

    I should mention that the 8/8 pattern is not 100% crisp, but I will see what I can do.

    Bruce
    299 x 299 - 914B
    300 x 300 - 1K
  • LeonLeon Posts: 7,620
    edited 2012-02-11 09:54
    My fresh PCB laminate arrived this morning and I've just finished a little test board using the pattern I posted two days ago. I was quite surprised to find that 6/6 mil seemed usable, and 4/4 didn't have any breaks or shorts, but the tracks seemed a bit rough. However, when I removed the resist, even the 4/4 mil tracks seemed usable. Not bad for an inexpensive printer and about $40 worth of equipment! Total time was about 15 minutes (I watched TV during the 12 minute UV exposure).

    I'll post a photo when I get my camera set up tomorrow. I don't have a proper macro lens (they are rather expensive) but I have a 50 mm with a reversing adapter which should work quite well.
  • LeonLeon Posts: 7,620
    edited 2012-02-12 05:56
    Here is a photo of the test board. I couldn't find the reversing ring for my 50 mm lens so I had to use a zoom lens which has a (rather awkward) macro capability necessitating a distance of over a metre. The 6/6 and 4/4 mil tracks are actually a lot better than they appear in the photo.

    In the past, I've printed artwork twice size on a dot matrix printer, and had a transparency produced by a litho plate maker. That worked very well. A couple of years ago I worked on a project at ESIEE, a grande ecole near Paris, and they had a very nice photographic system (something like a giant enlarger) in their PCB department. They closed the facility a year ago and they'd have probably given me any of the equipment if I'd arranged for it to be removed; I've been offered some other stuff.
    1024 x 822 - 104K
    Test.jpg 103.8K
  • idbruceidbruce Posts: 6,197
    edited 2012-02-12 07:03
    Leon

    As for me, by the time I got finished preparing the positive, it did not look so great, but I attempted an exposure with it anyhow. After etching, I deemed that the sample was not worth photographing or posting, so I have decided to either find a more suitable printer or have a photoplot made.

    Bruce
  • LeonLeon Posts: 7,620
    edited 2012-02-12 07:11
    2400 dpi printers are made, and should give excellent results, but cost far more than my HP 5940. I usually use 10/10 and 12/12 mil, and can't see me ever needing to go below 8/8 on home-made boards, so I'm quite happy with my technique.

    I can send you my test transparency if you want to try it with your Datak process. It would be interesting to see how it performs.
  • idbruceidbruce Posts: 6,197
    edited 2012-02-12 07:35
    I can send you my test transparency if you want to try it with your Datak process. It would be interesting to see how it performs.

    Now that does sound interesting. I would definitely be interested in performing that test to see how the two methods compare with the same positive. Additionally, I have not attempted duplicating the sample exposures on my UV exposure box. It has been a while since I have used it.

    Anyhow, yes please send your positive to:
    Bruce Drummond
    144 North Michigan Avenue
    Hobart, Indiana 46342
    I look forward to seeing how it compares with different films and different exposure setups.

    Bruce
  • LeonLeon Posts: 7,620
    edited 2012-02-12 07:46
    I'll post it tomorrow. I'll print 2x2 of them, in case one gets scratched, and you can cut them apart. They are fairly robust, though.

    It is tiny, about 1"x1". Would it help if I made them 3"x4"?

    I just checked continuity of the 4/4 and 6/6 tracks. They are all OK.
  • idbruceidbruce Posts: 6,197
    edited 2012-02-12 08:08
    Leon
    I'll post it tomorrow. I'll print 2x2 of them, in case one gets scratched, and you can cut them apart. They are fairly robust, though.

    If we are going to do a side by side comparison, it would be nice to have the positive that you used.
    It is tiny, about 1"x1". Would it help if I made them 3"x4"?

    Actually 1 X 1 is almost perfect, since I have an exposure plate made for 15/16 X 15/16 and sample boards cut for 15/16 X 15/16. However I will need to trim 1/32 off of each side.

    Bruce
  • LeonLeon Posts: 7,620
    edited 2012-02-12 08:41
    I just dimensioned it, and it's 1.092" x 0.880" overall. I'll make the border 15/16" x 15/16" for you.

    BTW, the text line width is 5 mil, and that always comes out OK, when I put text on the copper like that.
  • pjvpjv Posts: 1,903
    edited 2012-02-12 11:15
    Hello Bruce;

    You seem to be the ultimate hobbyist/experimenter, so here is something that might be of interest to you.

    Years ago we used to print and etch our own PC boards, and to get the precision we wanted, we also had to make our own films; both positives as well as negatives. So to do this well enough, we puchased a Geavaart (sp?) line camera with a bottom (through) as well as top (reflective) light source for some number of thousands of dollars. Also we got a one pass feed through contact film processor which then yielded the positive or negative. The camera, about 36 x 30 inches, 4 ft tall, has a set of interchangable lenses, and can make prints ranging from 4:1 up to 1:4 down by hand cranked dials. A vacuum system ensured the film was in intemate contact with the platen. Being a professional machine, the results were reasonable.

    We no longer make our own boards, and really do not have much use for the camera and the developer anymore, so I woud offer this set to you, or anyone else for that matter for a price of $1,000 plus shipping and crating. This mechanism should far exceed the capabilities of what you are trying to put together at home.

    I must say that I'm unsure if the films we used or the processing bath chemicals are readily available today, being that we have not used them for many years now.

    I f you have some interest, I will get model numbers and take some pictures for you.

    Cheers,

    Peter (pjv)
  • idbruceidbruce Posts: 6,197
    edited 2012-02-12 12:44
    @pjv

    Peter, I was initially looking at this camera which is only an hour away from me, but it has a broken vacuum switch, and considering Kodak filed for bankruptcy, I opted not to: http://www.ebay.com/itm/Kodak-Imagemaker-5060A-Graphics-Arts-Camera-/250960986965?pt=LH_DefaultDomain_0&hash=item3a6e70c355

    As you mentioned:
    I must say that I'm unsure if the films we used or the processing bath chemicals are readily available today, being that we have not used them for many years now.

    Besides being strapped for cash at the moment, this subject would be of the utmost concern for me. I have noticed that years ago, all the print shops had access to similar cameras, and now the closest one to my vicinity is an hours drive away. Now that scares me. So to be honest, I would not be interested at the moment, but if my financial situation should change in the near future, perhaps through the sale of my patent, I would definitely reconsider and do some investigating to see if film and developer were obtainable.

    Anyhow, I would like to thank you for your interest and offer.

    Bruce
  • Martin HodgeMartin Hodge Posts: 1,246
    edited 2012-02-12 22:06
    This might be a little outside the discussion, but one can dream (and drool).
    http://www.fortex.co.uk/page-product-photoplotter.shtml
  • idbruceidbruce Posts: 6,197
    edited 2012-02-13 00:53
    @Martin Hodge
    This might be a little outside the discussion, but one can dream (and drool).
    http://www.fortex.co.uk/page-product-photoplotter.shtml

    I don't believe that you are outside the realm of discussion as it pertains this topic. When using any type of exposure box, it is always desirable to use the best artwork possible, especially when making PCBs, and in this case, a laser photoplotter is just the tool for the job. However, the photoplotter that you linked to is much to large for my needs and space available. And yes, I can assume that many people would just dream and drool, but you can also make up your mind to build a laser photoplotter, and that is exactly what I intend to do. You may want to take a peek at this thread and perhaps join in on the discussion - http://forums.parallax.com/showthread.php?137850-CD-Laser-Diode-And-Optical-Block-(Future-Laser-Photoplotter). In Post #38 of that thread, a light bulb went off inside my head and I had an epiphany. :)

    Now I just need to learn how to operate the laser diode and optical block.

    Bruce
  • LeonLeon Posts: 7,620
    edited 2012-02-13 02:45
    Martin,

    That's the Bungard photoplotter, that I mentioned on Bruce's other thread. They cost over £6,000 each here in the UK, and require a darkroom.
  • idbruceidbruce Posts: 6,197
    edited 2012-02-13 03:21
    Leon

    I could be wrong, but I believe that is just a knockoff (look-a-like) to the Bungard.
  • LeonLeon Posts: 7,620
    edited 2012-02-13 03:27
    Bungard obviously allows resellers to put their own labels on them. Mega does the same. I thought that Mega had developed it themselves until I came across the Bungard site.
  • idbruceidbruce Posts: 6,197
    edited 2012-02-13 03:52
    I guess they do look identical.
  • LeonLeon Posts: 7,620
    edited 2012-02-13 05:54
    I just had a thought. Once one gets down to 6/6 and 4/4 mil, 1/2 oz copper is really required. I (and you, presumably) am using 1 oz copper laminate.
  • idbruceidbruce Posts: 6,197
    edited 2012-02-13 06:26
    1/2 oz. ? I have seen 1 oz and 2 oz, but I cannot recollect ever seeing 1/2 oz. Anyhow, I have been using 1/16" 2 oz boards for my samples as mentioned in #149 and #110.
This discussion has been closed.