Connect with us

There as gota be a better way

Discussion in 'Electronic Design' started by Glenn Ashmore, Feb 26, 2004.

Scroll to continue with content
  1. How do you guys go about picking an enclosure for a PCB when you want
    the final part to be as small as possible? I have tried picking a box
    first and then laying out the PCB to fit and end up either unable to
    cram all the components into the space or with a lot of room left over.
    Then I tried optimizing the board and spent days finding a box that
    was close and then massaging the board to fit it.

    Is there a good strategy for putting together a complete package?

    --
    Glenn Ashmore

    I'm building a 45' cutter in strip/composite. Watch my progress (or lack
    there of) at: http://www.rutuonline.com
    Shameless Commercial Division: http://www.spade-anchor-us.com
     
  2. Ian Stirling

    Ian Stirling Guest

    Do a rough layout, then pick a box a little smaller?
     
  3. Experience is a cruel mistress.

    Best regards,
    Spehro Pefhany
     
  4. Tim Wescott

    Tim Wescott Guest

    This is the way that I've done it. Usually if you throw all the parts on
    the board you'll have a good idea of the area.

    This also works in complex systems when you need to negotiate board space
    with the mechanical engineers months before the circuit is actually done.
    You pick all the parts that you know you're going to use, assume that the
    circuit will grow by a factor of 2 (they always do), then start wrangling.
     
  5. Finding a nice box is often difficult. The smallest possible
    box is a custom made, ready made boxes are a compromise.

    If you *really* have set your mind to a particular box, it's
    often possible to jam it all in. Don't forget you can mount
    components on both sides of the PCB. Or simplify your design ;)

    I favour finding a box first, then make the PCB layout. You
    have more control over the PCB than over finding a perfect box.
     
  6. R.Legg

    R.Legg Guest

    Plan.

    If you know what you want to do - estimate real estate of components,
    on paper.

    This will give you an idea of what you need in terms of board area,
    process methods and material cost for any package that is determined
    to be suitable.

    RL
     
  7. Sporkman

    Sporkman Guest

    Depending on whether this is for a home project or a profitable
    enterprise, and whether it's going to be a one-time single-use or will
    occur in volume, some options exist in rapid prototyping that may be
    more practical than you think. I know of services that can give you a
    custom-fit, functional (quite durable) plastic box of shallow depth for
    around $250 including overnight shipping (turnaround 2 to 3 days
    typically). Then those same companies can create rubber molds for short
    production runs of boxes made of epoxy or urethane. They can finish 'em
    to make 'em look like high-quality consumer products or leave 'em
    rough. You can even get elastomer gaskets and EMI/RFI coatings. One
    service that does well for the durable prototype is QuickParts.com. One
    that is probably better for short-run and really pretty presentation
    and/or production units is Forcast3D.com. I like the first because I
    can send them STL files directly from my 3D CAD and they'll give you an
    automated quote in 30 seconds or less. The second doesn't have such an
    automated quoting system, but they have more options and better support
    for those who don't know what they're doing in terms of materials and
    processes. They usually get back to you within a couple of hours.

    ---------blatent self-serving blurb-------------
    I can design a nice little custom box for you in a blue flash (in 3D)
    and get you pricing too.
    ---------end of blatent advertising-------------

    Mark 'Sporky' Stapleton
    WaterMark Design, LLC
    Charlotte, NC
    http://www.h2omarkdesign.com
     
  8. Sporkman

    Sporkman Guest

    Correction . . . the latter rapid-prototyping company is Forecast3D.com,
    not Forcast3D.com.
     
  9. Ooooo... You got me all excited! But my projects are always a one off to
    maybe five off. Hard to justify the setup charges for that.

    I will remain tempted however because the 900Mhz remote for the light
    system needs a small waterproof box. It will also need a waterproof
    bezzel for dash mounting. That project has a very slight possibility of
    being commercially viable. I have it fit in a Serpac H-65 right now but
    that is a bit thick and far from waterproof.
    --
    Glenn Ashmore

    I'm building a 45' cutter in strip/composite. Watch my progress (or lack
    there of) at: http://www.rutuonline.com
    Shameless Commercial Division: http://www.spade-anchor-us.com
     
  10. "Spehro Pefhany
    And pretty expensive by the hour as well. ;-)
     
  11. YD

    YD Guest

    I usually try to make a first layout as compact as possible. With a
    mockup and other parts (transformers, jacks, whatever) I determine a
    minimum box size then buy the next larger. Then if needed revise the
    layout taking into account available space, mounting holes and so on.

    This works for simple circuits but often enough it's the front panel
    controls that determine box width and heigth, usually can't choose
    more than one or two depths for it.

    - YD.
     
  12. Active8

    Active8 Guest

    <snip>

    I don't know how complex your lighting control is, and sorry, I
    can't find the link I'm thinking of, but google found it for me.

    Simple 900 MHz chips for remotes and such exist and can fit in
    enclosures as small as a key fob. The link I found was for an
    Italian company and I'm sure there are others. They had some really
    nice stock enclosures. Key fobs, TV remote style, and this
    ellipsoid kinda deal, complete with buttons and battery
    compartment. Bezels for IR, too, though I'm not sure if an end
    bezel is what you want. It could be replaced with clear plastic.
    Waterproofing, if you can't get gaskets, could be accomplished with
    silicone and homemade gaskets.

    Someone mentioned planning. Last time, I got real lucky and found a
    box after much planning. The circuit was bigger than I wanted
    because of the human/machine interface, but after checking Serpac
    and Hammond and playing with their AutoCAD files, I managed (the
    luck part) to get a Hammond 9 V battery enclosure that worked.
    Jammed a mainboard with 2 other boards stacked with standoffs and
    ..100" headers in the stinkin' box with the 2 stacked boards butt
    against each other. I trimmed 'em a little just so they wouldn't
    rub or stress anything over temp extremes.
     
  13. I setled on the pre-FCC certified Linx tranciever modules and stubby
    antennas. A little bigher than absolutely needed but require no extra
    design work and I have got it working line of sight out to about 900'.

    Should come in real handy to blink the lights in a dark and crowded
    anchorage when I have no idea where the boat is. :)
    --
    Glenn Ashmore

    I'm building a 45' cutter in strip/composite. Watch my progress (or lack
    there of) at: http://www.rutuonline.com
    Shameless Commercial Division: http://www.spade-anchor-us.com
     
  14. Active8

    Active8 Guest

    LOL. Good plan. I tried getting a good visual reference of my
    parking spot in relation to a light standard at the fairgrounds and
    still kicked myself in the ass for not carrying the GPS receiver.
     
Ask a Question
Want to reply to this thread or ask your own question?
You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.
Electronics Point Logo
Continue to site
Quote of the day

-