Dangerous Prototypes

[font=Georgia:]yabp case

ian,

Someone asked whether I could design a Bus Pirate case similar to the "clamshell" version of the yaobls (Yet Another Open Bench Logic Sniffer) case (thread (http://http://dangerousprototypes.com/forum/viewtopic.php?f=23&t=2516)). I assume they meant Bus Pirate v3.x or Bus Pirate v4.x, but I'm starting to realize that these boards are not the same size. I am interested in the board geometry (board outline, mounting holes, keepout areas, etc.) and the locations of components that might interact with or affect the case design (connectors, jumpers, LEDs, push-buttons, etc.). Where do I go to find the latest version of the Eagle layout files for these designs?

Thanks for your time.

dsm

Private Message edited slightly for clarity.
[hr:][/hr:]
[quote author="ian"]v3.5 and v4 are in the Dangerous Prototypes SVn at google code. They are the same size but that will probably change. The headers are in about the same place, but they are not exactly the same.[/quote]
[hr:][/hr:]
Bus Pirate board geometry

ian,

I downloaded the Eagle brd and sch files for the Bus Pirate v4.0, v4.0a, v4.0b, v4.1, and v3.5 revisions from source (http://http://code.google.com/p/dangerous-prototypes-open-hardware/source/browse/#svn%2Ftrunk%2FBus_Pirate%2Fhardware) and downloads (http://http://code.google.com/p/dangerous-prototypes-open-hardware/downloads/detail?name=Bus.Pirate.source.r1242.v3.5.zip) and measured various board geometry features and certain component locations.

For the Bus Pirate v4.0 and v4.1 layouts, the board geometry and the component locations are exactly on a 25 mil grid (except for the USB connector which is shifted left 12.5 mils for rational reasons) (except for the push-buttons which are exactly on a 5 mil grid). So far, so good. But the "history.txt" file in the downloads directory says these boards don't fit the Seeed Studio case. Is this an existing or currently unannounced new case?

For the Bus Pirate v4.0a, v4.0b, and v3.5 layouts, the board geometry (board edges are exactly on a 1.0 mm grid) (mounting holes are exactly on a 0.025 mm grid) seems clear, but the component locations don't appear to be on any obvious "mil" or "mm" grid. What is the story here?

What I am trying to do is determine are the locations of any board geometry or components that would affect a case design.
For example, for Bus Pirate v3.5

board ~ lower left corner ~ nominal 0.000, 0.000 ~ mm
    board ~ upper left corner ~ nominal 0.000, 40.000 ~ mm
    board ~ upper right corner ~ nominal 65.000, 40.000 ~ mm
    board ~ lower right corner ~ nominal 65.000, 0.000 ~ mm

    mounting hole ~ lower left ~ nominal 3.175, 3.175 ~ mm
    mounting hole ~ upper left ~ nominal 3.175, 36.825 ~ mm
    mounting hole ~ upper right ~ nominal 61.825, 36.825 ~ mm
    mounting hole ~ lower right ~ nominal 61.825, 3.175 ~ mm

Since 0.1 mm is roughly 0.0039 inches,
specifying these dimensions to 0.005 mm is like specifying a board feature to 0.000196 inches accuracy.
This level of accuracy doesn't seem commensurate to the problem.

    connector ~ ExtPwr ~ pin one or center ~ ???
    connector ~ ICSP ~ pin one or center ~ ???
    connector ~ IO ~ pin one or center ~ ???
    connector ~ USB ~ center ~ ???

    push-button ~ Normal ~ center ~ ???
    push-button ~ Reset ~ center ~ ???

    LED ~ pwr ~ center ~ ???
    LED ~ usb ~ center ~ ???
    LED ~ mode ~ center ~ ???
    LED ~ vreg ~ center ~ ???
[/list]
In the past, board designs often used a "mil" grid (usually because of 0.1" grid connectors and 0.1" grid through-hole device packages) while package designers often used a "mm" grid. Given the number of high pin count metric device packages there is probably less of a need for boards to be on a "mil" grid any more, but it is useful to have some grid scheme to locate components (so locations are not arbitrary measurements that have to be specified at the highest resolution). I can work with either scheme, but currently the Bus Pirate board seems to use a hybrid design grid that I'm having considerable difficulty figuring out.

I realize that this a fairly minor issue and you have a lot on your plate.
Do you have anyone available to answer this type of question?
Am I missing something obvious?
Or is there a short, simple answer?

Thanks for your time.

dsm

[attachment=0]
Private Message edited slightly for clarity.
[hr:][/hr:]
[quote author="arhi"]Ian should correct me if I'm wrong but BP4.x ain't final yet hence not in production so changes are still very possible. I think it would be very useful if the pcb can be modified a bit to be more "case friendly".

Maybe it would be best if you, David, can give some pointers to what you think would be a best changes in 4.x pcb design to make it more case friendly, both for a custom case and for existing abs cases, and then Ian can implement those changes in Eagle pcb layout.

The BP I have (3.5b I think) is case friendly - the 4 mounting holes are ok, the usb is ok, only the connector I replaces with 90 degree headers but going from top is also more then ok.

The BP 4.1 comes with buttons, that is the major difference ... but they are not problematic imho ... the only changes I'd make on 4.1 are
- for ICSP connector I'd use the layout that is not 100% in line but every second pin to be with a little offset (like for e.g. on arduino) so that you can use icsp holes without actually soldering a header.
- I'd add a 1mm border of clean pcb on the top edge (where the led's are) so that it can easy slide in a tray case or can be easily clamped without blocking light from leds.[/quote]
[hr:][/hr:]
Bus Pirate board layout suggestions

ian,
arhi,

Okay, I'll add my $0.02 worth...

• Use a "mil" layout grid (imperial) for the board design. Use a "mm" layout grid (metric) for the package design.
• Use a relatively coarse grid (100 mil or 50 mil or 25 mil) for all board geometry features (board outline, mounting holes, keepout areas, etc.) and the locations of all components that might interact with or affect the case design (connectors, jumpers, LEDs, push-buttons, etc.). Lock these parts down so that they don't accidentally get moved later.
• Position most 0.1" grid through-hole components so that their pins are centered on the coarse grid.
• Place 80 mil (minimum) mounting holes 125 mils (both X and Y) (minimum) from the corners of the board.
  80 mils will allow a 2 mm self-tapping screw.
• If you use larger mounting holes, place them proportionally further away from the corners of the board.
• Make the board slightly larger than necessary so you don't have to redesign the case if the board design changes slightly.
• Come up with a standard case that is the right size for a variety of Dangerous Prototype designs.
  Design boards to fit this case. Worry about connectors later in the process - they can often be cut with a laser as needed.
  For example, how about a smaller case that fits boards roughly 1.5" by 2.5" for a Bus Pirate product?
  For example, how about a larger case that fits boards roughly 2.0" by 3.5" (the size of a standard business card in US)?
• arhi makes a good suggestion about offsetting the pins for the ICSP connector. Sparkfun once ran an article (http://http://www.sparkfun.com/tutorials/114) about the optimum offset of pins in a staggered array for this application.  The reverse scheme is to place the ICSP holes directly on grid in the normal manner, but offset the pins in the connector that temporarily plugs into this array of holes.
• arhi makes a good suggestion about leaving a keepout area on the edge of the board.  Use 50 mils for this keepout area.
• Move the LEDs slightly further away from the edge of the board to allow more room for lightpipes.
• Position the USB connector to slightly hang over the edge of the board to allow for a maximum wall thickness of 1.5 mm plus a small margin of error.
• Other than the right-angle USB connector, use vertical 4-wall shell connectors for everything else.
  These other connectors should exit from the top of the case.

Thanks for your time.

dsm[/font:]

Hey dsm,

It's pretty much just me on the design side :) I am trying hard to get someone full time to help out with engineering, but it will be a while. Your points are all valid, and any non-sense in the design is entirely my fault ;) I appreciate your investigation and feedback and will certainly add these updates to the routing queue. I'll also try to follow best practices in the future.

BTW: a first batch of v4 is in production. It will probably be ready in a few weeks.

[font=Georgia:]Which Bus Pirate Case?

ian,

When I looked at the various Bus Pirate source files, there seem to be three groupings:

BusPirate-SMPS.brd   20Apr11 ~ no mounting holes

BusPirate-v3.5-SOIC.brd   19Apr11 ~ MH3 @ 2434.055", 1449.803" ~ other components not on an obvious grid
BusPirate-v3.5-SSOP.brd   19Apr11
BusPirate-v3.5a-SSOP.brd   17May11
BusPirate-v4a.brd   06May11
BusPirate-v4b.brd   17May11

BusPirate-v4.brd   06May11 ~ MH3 @ 2.45", 1.3" ~ other components on an obvious grid
BusPirate-v4.1.brd   06May11

So when you say
[quote author="ian"]BTW: a first batch of v4 is in production. It will probably be ready in a few weeks.[/quote]
I still don't know which board dimensions to use since there are two "version 4" layouts!!
As you might imagine, I would prefer to not put a lot of time and effort into the wrong Bus Pirate case design.
Any idea which "version 4" layout is going into production?
Is there any simple way to get a bare "version 4" production board so that I can have something concrete to play with?

Thanks for your time.

dsm
[hr:][/hr:]
ian,
[quote author="ian"]It's pretty much just me on the design side :) I am trying hard to get someone full time to help out with engineering, but it will be a while. Your points are all valid, and any non-sense in the design is entirely my fault ;) I appreciate your investigation and feedback and will certainly add these updates to the routing queue. I'll also try to follow best practices in the future.[/quote]
I can imagine that what you are doing is a lot of fun part of the time and a lot of cruft the rest of the time.
Lately I've been asking you and arhi a bunch of questions in order to get around roadblocks that I'm running into that other people might also be encountering. I apologize for taking up so much of your time. I wish I could be more helpful.

Thanks for your time.

dsm[/font:]

v3.5 and v4a+ are sized for the 'seed case'  (which they can no longer source anyways, so I don't know why I keep talking about it). v4.1 is just a test I did. Here's the history.txt:

Bus Pirate v3.5- Two versions, SOIC and SSOP PIC chips. Same as v3b with reduced part count.

v3.5b - fixed header key to match seeed cable

v3.5c - Fixed missing 4066 control trace

Bus Pirate v4 - Original pre-production version sold at Seeed for $50.

Bus Pirate v4a - v4 resized to fit Seeed case. First production version.

Bus Pirate v4.1 - Small rerouting of v4, not yet resized to fit v4a case.

One thing is wrong, v4 is the going to be the first batch of 100, because that is what I had them make... Since they can't get the case and we will be changing the size yet again soon, I didn't see any point to mess around with an update for the initial batch of 100.

I'm sorry, there's no good answer. We (Seeed and I) are trying really hard to get everything in a nice case, starting with the Bus Pirate, so I keep changing the size for cases that are then no longer available. Everything will be changed again soon too.

[font=Georgia:]Bus Pirate board geometry

ian,

Thanks for getting back to me about the Bus Pirate board layout even though the board geometry scheme chosen was different than I had hoped. The locations of board geometry features and components that affect the case design for v4.0a are shown in the attached spreadsheet.

[attachment=0]
In the "no good deed goes unpunished" category, your response raises a couple short questions:

• What screw size were you assuming for the mounting holes? English or metric? 3.0 mm is 0.118", but after plating the mounting holes will probably be closer to 2.8 mm or 0.110. Plastic pins pins that are part of the case to position the board might be a possibility if you use snap fingers to hold the case together.
• Who is supplying the 12-pin shrouded header that you plan to use? As far as I can tell, 10-pin, 14-pin, 16-pin, and 20-pin shrouded headers are readily available, but the 12-pin variant seems to be fairly rare.  I would prefer to not guess about the header dimensions if at all possible.
• Is there a larger (about an inch high or so) DP-LOGO image or source available?

[quote author="ian"]We (Seeed and I) are trying really hard to get everything in a nice case, starting with the Bus Pirate, so I keep changing the size for cases that are then no longer available. Everything will be changed again soon too.[/quote]
Perhaps working with Seeed to get several standard injection-molded cases fabricated would be worthwhile. How about three or four cases designed to fit roughly golden-ratio-sized boards up to the maximum dimensions allowed by the free version of Eagle PCB layout? Seeed could suggest what board sizes would maximize their yield from whatever panel size they use. For example, the typical spreadsheet board yield model specifies

PanelSizeX
PanelSizeY
NumberOfLayers
LeftPanelMarginX
RightPanelMarginX
TopPanelMarginY
BottomPanelMarginY
RouterBitWidth [1]
BoardX
BoardY

I'll finish designing a couple yabp cases (even though this seems like a redundant activity if you're working with Seeed to get new cases designed and built) after the "-v1" clamshell yaobls case prototypes arrive back from Shapeways in couple days. From my viewpoint, I still want to figure out how to use the "draft angle" feature of the tool I've been using for the clamshell yaobls case designs.

Thanks for your time.

dsm

[1]    Unless the boards are scored and snapped - and Seeed doesn't look like they do that.[/font:]

[font=Georgia:]"clamshell" yabp case "-v2" images

ian,

I updated the "clamshell" yabp (Yet Another Bus Pirate) case design (see previous posting), uploaded the design to Shapeways, and ordered some prototype parts.

The "clamshell" yabp bottom (http://http://www.shapeways.com/model/347348/clamshell_yabp_case_bottom_dsm_v2.html) case can be used by itself to support the Developer use case.
See 31Oct11 posting for updated "-v3" design.
Features include the following:

• Designed to support Bus Pirate v4.0a boards (although it should work with some other Bus Pirate boards).
• Bus Pirate board secured between top case and bottom case.
• Top case and bottom case secured by four torx head plastite self-tapping screws through bottom case.
• Lip detail matches "chamshell" yabp top case.
• Improved dimensional model (derived from "clamshell" yaobls case design) for tighter tolerances.
• Thinner walls for lower cost.
• Openings in bottom for lower cost.
• Integral feet.

The "clamshell" yabp top (http://http://www.shapeways.com/model/347349/clamshell_yabp_case_top_dsm_v2.html) case can be used with the "clamshell" yabp bottom case to support the User use case.
See 31Oct11 posting for updated "-v3" design.
Features include the following:

• Designed to support Bus Pirate v4.0a boards (although it should work with some other Bus Pirate boards).
• Lip detail matches "chamshell" yabp bottom case.
• Improved dimensional model (derived from "clamshell" yaobls case design) for tighter tolerances.
  0.4 mm connector-to-case margin (top, sides, bottom) (USB connector).
  0.1 mm board-to-case margin (adds 0.2 mm overall).
  0.2 mm top connector opening side margin (adds 0.4 mm overall).
  0.4 mm top connector opening end margin (adds 0.2 mm overall).
  0.2 mm reveal between the top and bottom cases.
• Push-button lever plunger scheme (proposed by arhi) used for Normal and Reset push-buttons.
• Commercial acrylic light pipes used for PWR, USB, MODE, and VREG LEDs.
• Opening through the top case for the ExtPwr connector.
  Assumed to be a 2-pin vertical male header that mates to a 0.1" jumper.
• Openings through the top case for the ICSP connector.
  Assumed to be a 5-pin vertical male header that mates to a 0.1" receptacle.
• Opening through the top case for the IO connector.
  Assumed to be a 12-pin (2x6) vertical shrouded male header that mates to a polarized 0.1" 2x6 receptacle.
  Since 12-pin (2x6) shrouded male headers are comparatively unusual compared to ordinary 10-pin (2x5), 14-pin (2x7), 16-pin (2x8), or 20-pin (2x10) shrouded headers, one option would be to make the opening through the case large enough to allow a 16-pin (2x8) shrouded male header with four pins depopulated to be used. Defer for now.
• Thinner walls for lower cost.

[attachment=5]
[attachment=4]
[attachment=3]
[attachment=2]
[attachment=1]
[attachment=0]
Thanks for your time.

dsm[/font:]

---

:D now this is one nice box :)

I thought I post this before:

Nice one. I really like the 'buttons'

But I have lots of conductive material on my desk, so I don't like the bottom being open :) I would opt to have a pinout printed on the bottom (or is it reserved for a sticker??).

A sticker will solve the issue of stuff from desk getting into ...

btw, before I held similar (for obls) case in my hands I was afraid those holes would allow garbage from my desk to still make a short but actually it's not that easy as holes are not big and board is pretty far up so you really need a "specially wrong day" for things to go south :D

It always happens when it isn't a suitable moment (if ever :)) On the other side it will boost the sales of the buspirate ;)

/me is having a vision of 24h express delivery of BPs ;)

[quote author="dsm"]

• Opening through the top case for the IO connector.
  Assumed to be a 12-pin (2x6) vertical shrouded male header that mates to a polarized 0.1" 2x6 receptacle.
  Since 12-pin (2x6) vertical shrouded male headers are comparatively unusual compared to ordinary 10-pin (2x5), 16-pin (2x8), or 20-pin (2x10) shrouded headers, the opening through the case allows for a 16-pin (2x8) vertical shrouded male header with four pins depopulated.
  
  dsm[/font][/quote]
  
  Is it possible for the

IO connector opening to have brake away tabs in the opening so everyone can make it as big as needed for there BP.

So default an opening for the 10-pin connector and the option to extend this to 12-pin and 16-pin.
I got mine 2X6 shrouded pinheaders from ebay. (http://http://www.ebay.com/itm/ws/eBayISAPI.dll?ViewItem&item=270820725854)

[font=Georgia:]"clamshell" yabp case comments

Sjaak,

The holes are in the bottom of the case to reduce costs. One of the really powerful concepts about additive machining is that adding more features and details is less important than the amount of material used. The situation is reversed with subtractive machining where each additional feature and detail means more processing steps (and thus cost).

arhi and I have hashed this "bottom hole" issue out before in the context of the yaobls case. I am less concerned about random metal filings getting into the case than I am about random ESD hits during handling and transport.  Once you compare the size of the bottom holes (which are slightly smaller than those in the yaobls case) with size of your fingers, you'll see what I mean. If you really are concerned about random metal filings getting into the case, as arhi points out, you can always add a colorful, high-definition, laminated IO pin reference chart that covers the holes in the bottom of the case.

If someone really needs a completely closed case, deleting the holes and posting another case variant to Shapeways would take me less than ten minutes.

Thanks for your time.

dsm

[attachment=1]
[hr:][/hr:]
pietja,

I ordered some 12-pin shrouded male headers from the source on eBay that you reference. That source was not able to provide me with a datasheet with mechanical dimensions. 12-pin shrouded male headers are fairly difficult to find compared to common 10-pin, 14-pin, 16-pin, and 20-pin shrouded male headers. Rarity usually translates into higher cost and supply chain vulnerability. Designing a product around lucky eBay finds is generally not a good idea.

Further, there are some interesting functions that could be implemented using currently unused pins on the 64-pin PIC 24FJ256GB106 micro-controller. More functionality for the incremental cost of a few extra pins? Why not?

I looked at a variety of common 10-pin, 14-pin, 16-pin, and 20-pin shrouded male headers from volume commercial sources and then extrapolated to what the dimensions for a hypothetical 12-pin shrouded male header from the same sources should be. Generally the resulting model is to add 0.150" onto each end of the shrouded connector and 0.080" onto each side of the shrouded connector. To this I add the additional connector opening end margin and connector opening side margin mentioned earlier in this thread.

This "clamshell" yabp case design is just a prototype. Making case variants available for whatever size IO connector you suggest is not difficult. This is probably a better solution than adding break-away tabs. My real problem is that nobody has yet specified a part number for the IO connector from a volume source - making this whole "clamshell" yabp case design exercise fairly moot.

Thanks for your time.

dsm

[attachment=0][/font:]

---

dsm,

I found you a datasheet (http://http://www.icconnectors.com/003pdf2x429pag/00118.pdf) for every size shrouded male header.

As for availability, when the BPv4 goes into production at Seeed i think they will have a source for those 12pin headers like icconnectors (http://http://www.icconnectors.com) or someone else but that's something ian or the guys at Seeed can tell us.

btw. at work i order lots of connectors from icconnectors through a local distributor.

[font=Georgia:]yabp IO connector size

pietja,

Thank you, your information is quite helpful.

Width
Working back though the datasheet, the stated width of the connector is 8.80 mm. The datasheet is inconsistent about the height of the alignment ridges, but use the larger dimension or 9.05 mm. For symmetry, increase the width of the connector to 9.30 mm or 0.366". Subtracting 0.2" for the receptacle envelope leaves 0.166" or an 0.083" offset (which is slightly larger than the 0.080" offset I mentioned) on each side of the receptacle envelope. Alternately, we could use an asymmetric opening with an 0.083" offset on the left side of the receptacle envelope and an 0.073" offset on the right side of the receptacle envelope.

Length
Working back through the datasheet, the stated length of the 12-pin shrouded male header is 22.98 mm or 0.905". Subtracting 0.6" for the receptacle envelope leaves 0.305" or an 0.152" offset (which is slightly larger than the 0.150" offset I mentioned) on each end of the receptacle envelope.

Still...
While the icconnectors web site does carry a suitable 12-pin shrouded male header, I don't see any prices or a simple way to order small numbers of parts online (unlike Digi-Key or Mouser or Newark). I assume icconnectors (which is based in Taiwan) mainly deals with industrial volume users. While 12-pin shrouded male headers clearly exist, they are still fairly rare compared to common 10-pin, 14-pin, 16-pin, and 20-pin shrouded male headers. Perhaps ian will be able to talk Seeed Studio into carrying the correct connector, but it still seems like an odd connector choice for an open source project.

Thanks for your time.

dsm[/font:]

Another beautiful case, the buttons are great. I will send you a freebie of the first batch of v4 so you can check it out. Please PM me your address.

[font=Georgia:]"clamshell" yabp case "-v2" update

The "-v2" design "clamshell" yabp prototype case parts ordered 01Oct11 from Shapeways arrived back 20Oct11.
For a list of the differences from the previous "-v1" design "clamshell" yabp case, see my 02Oct11 posting.
I expect that the "-v3" design "clamshell" yabp parts will be the final iteration before the design is released.

Unfortunately the v4.0a Bus Pirate board ian was going to send has yet to arrive, so the images use a bare
v3.5a-SOIC DIY Edition Bus Pirate board (link (http://http://dangerousprototypes.com/2011/09/26/testing-pcb-sales-bus-pirate-v3-5a-soic-on-sale-now/)). Note that only the board outline and the mounting holes
match between these two revisions - all of the connector, LED, and switch positions are different.

[attachment=0]
[attachment=1]
[attachment=2]
[attachment=4]
[attachment=3]
clamshell-yabp-case-bottom-dsm-v2
Proposed changes for the "-v3" design include the following:

• The case-bottom was fabricated in a white polished SLS nylon material which leaves a fairly smooth surface
  but allows some details to be slightly abraded away. Use unpolished SLS nylon or alumide in the future.
• The 0.9 mm lip width on both the case-bottom and the case-top allows 0.2 mm (~ 0.008”) of play overall.
  Based on results from the clamshell-BC-case design, slightly increase the case-bottom lip width to 1.0 mm
  which will only allow 0.1 mm (~ 0.004”) of play overall.
• Refine the USB-connector-lip-detail slightly.
• Refine the bottom-mounting-hole-boss and screw-head-well details slightly.

clamshell-yaobls-case-top-dsm-v2
Proposed changes for the "-v3" design include the following:

• The case-top was fabricated in a white polished SLS nylon material which leaves a fairly smooth surface
  but allows some details to be slightly abraded away. Use unpolished SLS nylon or alumide in the future.
• Keep the case-top lip width as 0.9 mm (see above).
• Refine the top-mounting-hole-boss and top-mounting-hole details slightly.
• Keep the 0.2 mm reveal between the case-bottom and the case-top.
• Because the SLS nylon material in slightly flexible, the push-button plunger scheme works well.
  Based on results from the clamshell-BC-case design, slightly increase the push-button plunger length to approach within 0.5 mm of the push-button (assumed to be an ALPS SKQGADE010 tactile push-button) (data (http://http://www3.alps.com/WebObjects/catalog.woa/E/PDF/Switch/Tact/SKQG/SKQG.PDF)) (Mouser (http://http://www.mouser.com/)) (catalog (http://http://www.mouser.com/catalog/catalogUSD/644/1722.pdf)).
• Based on results from the clamshell-BC-case design, refine the normal-cantilever and reset-cantilever details slightly.
• The holes for the LED light-pipes are in the correct position.
  Based on results from the clamshell-BC-case design, increase the size of the LED light-pipe holes
  in order to use LPC020CTP (data (http://http://vcclite.com/led-light-pipes/lpc-4mm-round/)) (Mouser (http://http://www.mouser.com/)) (catalog (http://http://www.mouser.com/catalog/catalogUSD/644/170.pdf)) light-pipes since this series offers more length options.
  Secure the LED light-pipes with a drop of glue or a retaining grommet.
• Currently the space-above-PCB is set to 6.2 mm to allow slightly longer case screws.
  Based on results from the clamshell-BC-case design, decrease the space-above-PCB to 5.0 mm.
• This case design currently uses 2.0 mm by 6.0 mm T-6 torx head plastite case screws.
  Torx head screws are used since a Torx head is typically smaller than a pan head for the same screw size.
  While a 6.0 mm screw is a bit too short to hold well, using four screws partially makes up for this deficiency.
  I suggest that the distance from the edge of the board to the mounting holes should be increased when the
  Bus Pirate board is updated.

If there are any additional case features anyone wants, please make your suggestions known soon.

Thanks for your time.

dsm[/font:]

---

The first batch just came into stock, we ordered your yesterday. Unfortunately this will not be the final version of the PCB, but it will help with placement I hope.

[font=Georgia:]"clamshell" yabp case "-v3" images

I updated the "clamshell" yabp (Yet Another Bus Pirate) case design (see 21Oct11 posting), uploaded the design to Shapeways, and ordered some prototype parts. This case is designed to support Bus Pirate v4.0a boards.

The "clamshell" yabp bottom (http://http://www.shapeways.com/model/369509/clamshell_yabp_case_bottom_dsm_v3.html) case can be used by itself to support the Developer use case.
The "clamshell" yabp top (http://http://www.shapeways.com/model/369510/clamshell_yabp_case_top_dsm_v3.html) case can be used with the "clamshell" yabp bottom case to support the User use case.

[attachment=5]
[attachment=4]
[attachment=3]
[attachment=2]
[attachment=0]
[attachment=1]
The four LED light-pipes are assumed to be LPC020CTP parts (data (http://http://vcclite.com/led-light-pipes/lpc-4mm-round/)) (Mouser (http://http://www.mouser.com/)) (catalog (http://http://www.mouser.com/catalog/catalogUSD/644/170.pdf)).
The four case screws are assumed to be 2.0 mm by 6.0 mm T-6 torx head plastite screws from Small Parts (link (http://http://www.smallparts.com/)), but purchased on Amazon (link (http://http://www.amazon.com/Plated-Steel-Tapping-M2-0-4-Length/dp/B00186T8TO/)).

Thanks for your time.

dsm[/font:]

---

[font=Georgia:]new Seeed Studio enclosures

As of 30Oct11, Seeed Studio (link (http://http://www.seeedstudio.com/depot/accessories-enclosure-c-178_184.html)) is carrying some new plastic enclosures that may be of interest.

General Plastic Case

• ACC131N5B ~ 58 x 35 x 15 mm ~ white matte ~ $2.95 ~ snap fingers. HF-N-5
• ACC131N7B ~ 100 x 60 x 20 mm ~ light grey shiny ~ $3.50 ~ snap fingers. HF-N-7
• ACC13148B ~ 65 x 38 x 22 mm ~ black matte ~ $2.95 ~ snap fingers. HF-N-48
• ACC131N9B ~ 70 x 44 x 28 mm ~ light grey shiny ~ $3.50 ~ snap fingers. HF-N-9
• ACC13N32B ~ 102 x 63 x 39 mm ~ light grey shiny ~ $4.50 ~ snap fingers. HF-N-32
• ACC13134B ~ 95 x 75 x 43 mm ~ light grey shiny ~ $4.50 ~ screws. HF-N-34

Clear Plastic Enclosure

• ACC1319TB ~ 222 x 145 x 55 mm ~ clear top / light grey bottom ~ $19.50 ~ six screws ~ gasket. HF-H-23T
• ACC1313TB ~ 115 x 99 x 55 mm ~ clear top / light grey bottom ~ $12.50 ~ four screws ~ gasket. HF-H-3T
• ACC101A1B ~ 100 x 68 x 50 mm ~ clear top / light grey bottom ~ $4.50 ~ four screws ~ gasket. HF-H-1T

[hr:][/hr:]
Comments

• All of these enclosures appear to be manufactured by Shenzhen Hongfashunda Mold Co., Ltd. (link (http://http://www.hfjxc.com/En/index.asp)).
  The Seeed Studio General Plastic Case(s) appear to be SHM's Universal Plastic Enclosure(s).
  The Seeed Studio Clear Plastic Enclosure(s) appear to be SHM's Plastic Waterproof Boxes(s).
• The ACC131N5B case appears similar to the THM202E1B case that Seeed Studio used to sell for the older Bus Pirate v3 (aka THM180C4M) except that it is missing openings for the USB connector and the shrouded 2x5 header.
• The type of plastic is unspecified - which may be a problem if you want to laser cut your own openings.
• The board size each case can accommodate is somewhat smaller than the case dimensions specified.
  For example, case-dimension minus (2 * wall-thickness + case-to-board-margin) equals board-dimension.
• Unlike the other cases, the ACC101A1B has been carried by Seeed Studio since 29Dec10 as the Seeeduino Stalker v2 enclosure (and appears to be a comparative bargain).

Would a person be better off

(a) redesigning a board to fit an existing case, or
(b) designing a new case (with suitable openings) to fit an existing board?

Thanks for your time.

dsm[/font:]

These boards are cheap enough to make a person adapt the board in order to fit them but .. none is "ideal" for 5x5 and 10x10cm pcb's :( ... as for material, I believe transparent ones are PET or Nylon (not really resistant to fire) while the opaque ones are HDPE or Nylon

They do list inner dimensions when you click on details

I'm glad to see new case options. I got the aluminum free in their giveaways.

Why are prefab boxes always so ugly? I think clear would be the minimum for DIY electronics (see the LEDs and silk, show off to friends). Clear color tinted materials, with interchangeable tops of different colors, would be way cool. Nice rounded corners that are easy to hold, sides optimized for button/connector holes, pops apart for debugging with the bottom still attached for isolation, something with a modicum of design sense. Instead it's always beige/black/offwhite.</rant>

prefab boxes are not always ugly - the dirt cheap prc boxes are ugly but that's another story ... I have bunch of plastic boxes I ordered few years from rapidelectronics and they look great for e.g. ... locally I can also get plastic boxes, some look good, some look professional, and some are ugly ... price is different too :)

[font=Georgia:]new Seeed Studio enclosures comments

arhi,

[quote author="arhi"]These boards are cheap enough to make a person adapt the board in order to fit them...[/quote]Agreed, but even inexpensive cases still need openings for connectors, LEDs, push-buttons, etc.
An inexpensive case that needs expensive post-processing is no bargain.

[quote author="arhi"]...none is "ideal" for 5x5 and 10x10cm pcb's[/quote]Agreed, except most of my designs are typically more rectangular and not square.  I looked at nearly all of the enclosures on the SHM web site and none are a perfect match for the Dangerous Prototype projects that I've been tracking.

[quote author="arhi"]...as for material, I believe transparent ones are PET or Nylon (not really resistant to fire) while the opaque ones are HDPE or Nylon.[/quote] I'm just hoping the cases are not made out of ABS since ABS fumes can cause problems for laser cutters.
Update - I sent an email to SHM asking them what plastics they use for the HF-N-5 and the HF-H-1T.

[quote author="arhi"]
They do list inner dimensions when you click on details.[/quote] I noticed this, but as of this date, I also noticed that the Seeed Studio descriptions for these products have some minor dimensional inconsistencies.

Thanks for your time.

dsm

[hr:][/hr:]
ian,

[quote author="ian"]Why are prefab boxes always so ugly?[/quote]
I haven't a clue about why this might be, but I tend to agree.
A clear or smoky top would allow the electronics and LEDs to be seen (without the expense of LED light-pipes).

And why are "universal" plastic cases so damned thick?
The recent yabp and yaobls case designs are only 11.5 mm thick (excluding connectors protruding from the top surface
and feet protruding from the bottom surface), but not one of the Seeed Studio or SHM cases is as thin.
As far as color is concerned, the laser-sintered nylon material can be dyed other colors (link (http://http://www.shapeways.com/forum/index.php?t=msg&th=5507)).

Thanks for your time.

dsm[/font:]

[quote author="dsm"]An inexpensive case that needs expensive post-processing is no bargain.[/quote]

The cheaper the case the easier is to post process in my experience :D. You just need a nibbler and you are done :D The more expensive cases use 2.5+ mm thick plastic and they are hard to nibble.

[quote author="dsm"]I'm just hoping the cases are not made out of ABS since ABS fumes can cause problems for laser cutters.[/quote]
ABS is tad easier to hand drill/cut/sand then nylon/hdpe/ldpe/pp/pet, but is lot more expensive then those polymers hence I really doubt they made these cheap ones out of ABS :D (IIRC ABS pellets are 10 times more expensive then PP for e.g.) .. the "real" electronics boxes should be made out of material that is fire proof, but none of these cheap ones are :(

The clear tops are hopefully pet, but they might went with acrylic too. I really hate acrylic as you can't easily process it by hands (you need laser cutter or at least a milling machine) ... PET is on the other hand easy to work with

[font=Georgia:]"clamshell" yabp case "-v3" update

The "-v3" design "clamshell" yabp prototype case parts ordered 31Oct11 from Shapeways arrived back 16Nov11.
For a list of the differences from the previous "-v2" design "clamshell" yabp case, see my 31Oct11 posting.

Unfortunately the v4.0a Bus Pirate board ian was going to send has yet to arrive, so the images use a bare
v3.5a-SOIC DIY Edition Bus Pirate board (link (http://http://dangerousprototypes.com/2011/09/26/testing-pcb-sales-bus-pirate-v3-5a-soic-on-sale-now/)). Note that only the board outline and the mounting holes
match between these two revisions - all of the connector, LED, and switch positions are different.

[attachment=4]
[attachment=3]
[attachment=0]
[attachment=1]
[attachment=2]
Thanks for your time.

dsm[/font:]

---

Unfortunately the v4.0a Bus Pirate board ian was going to send has yet to arrive,

I'll find the order number and track it.

[font=Georgia:]"clamshell" yabp case "-v3" update ~ II

The good news is that the sample v4 Bus Pirate board that ian was going to arrange for Seeed Studio to send finally arrived 21Dec11 (Wednesday). Thank you ian and Seeed Studio.
The bad news is that the sample v4 Bus Pirate board that arrived was not the 65 mm x 40 mm Bus Pirate v4.0a board size that my "clamshell" yabp case "-v3" was designed to fit.
The good news is that the sample v4 Bus Pirate board that arrived is the new 60 mm x 37 mm Bus Pirate v4 board size.
Well, no, apparently not. The sample v4 Bus Pirate board that arrived seems to be 63.50 mm x 36.83 mm.
The bad news is that the component locations on this sample v4 Bus Pirate board are unlikely to be the same as the production v4.0c Bus Pirate boards.
The good news is that if I can identify the correct "brd" CAD data for this sample v4 Bus Pirate board, cranking out the data for the correct replacement Shapeways SLS 3D printed case should only take me an hour or so.
The bad news is that replacement Shapeways SLS 3D printed case will only fit this v4 Bus Pirate board layout.
The good news is that anyone else who received this v4 Bus Pirate board will be able to buy their own copy of the case from Shapeways.

Does anyone know which version of the v4 Bus Pirate this board might be?
How about BusPirate-v4.brd [r1131]?

The Seeed Studio anti-static box is marked as follows:

Bus Pirate v4

[for developers]
TES102D2P
10/20/11
[/list]
[attachment=1]
[attachment=0]
Nominal archive of v4 Bus Pirate hardware data (link (http://http://code.google.com/p/dangerous-prototypes-open-hardware/source/browse/#svn%2Ftrunk%2FBus_Pirate%2Fhardware%2Fv4)).

Thanks for your time.

dsm[/font:]

---

Hi DSM,

I'm glad it finally got there. It only took two orders :)

How about BusPirate-v4.brd [r1131]?

Yes, based on your info (and my BPv4 production version) I believe that is the board they produced. The v4a is fit for a case that Seeed had, but then they decided another size would be better so we did not make that version.

[font=Georgia:]"clamshell" yabp case (v4fd) "-v1" images ~ I

I created a variant of the "clamshell" yabp case design to fit the sample v4 Bus Pirate board that ian arranged to be sent to me.
I am assuming that this board was based on the BusPirate-v4.brd [r1131] layout data.
This design has been uploaded the Shapeways site and I've ordered some prototype parts.

Note that this board is only intended to work with a specific preliminary v4 Bus Pirate board intended for developers.
See recent previous messages in this forum thread for more details.

Instead of the screws I had been using, this version of the "clamshell" yabp case uses neodymium magnets to hold the top case and bottom case together. This change should make it easier to open and close the case (no lost screws or stripped screws holes). The same configuration can now support both the Developer use case and the User use case (see the yaobls discussion thread for more details). The four 2.8 mm PTH mounting holes will need to be enlarged with a 3.2 mm (0.126") or a 1/8" (3.175 mm) drill before this scheme will work. I believe that the top and bottom conducting layers are at the same potential (GND) so no insulation will be required around the magnets, but there remains the question of whether the removal of the PTH top-to-bottom conduction path at the mounting holes will affect the electrical behavior of the circuitry.

[attachment=1]
[attachment=0]
[attachment=4]
[attachment=2][attachment=5][attachment=3]
The four LED light-pipes are assumed to be LPC020CTP parts (data (http://http://vcclite.com/led-light-pipes/lpc-4mm-round/)) (Mouser (http://http://www.mouser.com/)) (catalog (http://http://www.mouser.com/catalog/catalogUSD/644/170.pdf)).
The eight 3.0 mm diameter by 3.0 mm long neodymium magnets were ordered off eBay.

My main conclusion from this exercise is to not design a case until I have a board example in hand.

Thanks for your time.

dsm[/font:]

---

[font=Georgia:]"clamshell" yabp case (v4fd) "-v1" images ~ II

There are some additional "clamshell" yabp case (v4fd) "-v1" images that may be of interest.
Note that this board is only intended to work with a specific preliminary v4 Bus Pirate board intended for developers.
See recent previous messages in this forum thread for more details.

[attachment=5]
[attachment=0]
[attachment=4]
[attachment=3]
[attachment=2]
[attachment=1]
The top case and bottom case are defined to be different materials for clarity in the section views.
The board and all components on the board are defined to be in the same material (sorry about that).
The light-pipes are defined to be transparent enough to see through (note the LEDs in the assembled top view image),
but colored enough so they don't disappear in the exploded images.
The LEDs are so close to the edge of the board that the light-pipes used cut into the side of the case.

Thanks for your time.

dsm[/font:]

---

As always, it looks really fantastic. I also think the magnets are a great connector idea, I really like them.

[font=georgia:]additional comment

[quote author="dsm"]The four 2.8 mm PTH mounting holes will need to be enlarged with a 3.2 mm (0.126") or a 1/8" (3.175 mm) drill before this scheme will work. I believe that the top and bottom conducting layers are at the same potential (GND) so no insulation will be required around the magnets, but there remains the question of whether the removal of the PTH top-to-bottom conduction path at the mounting holes will affect the electrical behavior of the circuitry.[/quote]
I'm not quite out of the woods yet with this design because of the above board modification, although the top and bottom layers are still tied together by means of PTHs at the connectors and several vias scattered around the board.

[hr:][/hr:]
The BusPirate v3.5e and 4.0c designs both have non-PTH 3.2 mm mounting holes (which I firmly believe is the correct choice).
The top and bottom layers in both designs will be tied together by means of PTHs at the connectors and several vias scattered around the board.

Thanks for your time.

dsm[/font:]

[font=Georgia:]"clamshell" yabp case (v4fd) "-v1" update

The "-v1" design "clamshell" v4fd yabp prototype case parts ordered 07Jan12 from Shapeways arrived back 18Jan12.
This case is intended to fit the sample v4 (for developers) Bus Pirate board that ian arranged to be sent to me.
Due to other tasks, I wasn't able to get around to slightly modifying the board and assembling the case until this evening.

Note that this board is only intended to work with a specific preliminary v4 Bus Pirate board intended for developers.
See recent previous messages in this forum thread for more details.

Case Assembly Procedure

• Use an ordinary wooden toothpick to make sure any loose nylon powder left over from the Shapeways SLS 3D printing process is removed from the magnet wells in the top and bottom cases. The reason for this step is to make sure that the magnets seat deeply enough in the magnet wells.
• Press two 3.0 mm diameter by 3.0 mm long neodymium magnets into the two magnet wells
  at the USB end of the bottom case with the "north" pole of the magnets facing out of the magnet wells.
• Press two 3.0 mm diameter by 3.0 mm long neodymium magnets into the two magnet wells
  at the non-USB end of the bottom case with the "south" pole of the magnets facing out of the magnet wells.
• Press two 3.0 mm diameter by 3.0 mm long neodymium magnets into the two magnet wells
  at the USB end of the top case with the "south" pole of the magnets facing out of the magnet wells.
• Press two 3.0 mm diameter by 3.0 mm long neodymium magnets into the two magnet wells
  at the non-USB end of the top case with the "north" pole of the magnets facing out of the magnet wells.
• Make sure the magnets in the bottom case are pressed in as far as possible.
• Make a temporary precision spacer the same size as the v4fd Bus Pirate board out of an ordinary manila file folder.
• Place the temporary precision spacer between the magnets in the top case and the bottom cases (make sure the top case and the bottom case are properly aligned) and press each pair of top and bottom corners together firmly.
• Open the case, remove the temporary precision spacer, and set the top case and the bottom case aside.

The orientation of the magnets is intended to make sure that (a) the top and bottom case can't be assembled wrong and (b) to allow top and bottom cases from different production runs to be assembled correctly. The neodymium magnets are sufficiently strong that you can determine the "north" [1] and "south" [2] poles of the magnets by just letting the individual magnets roll around on a smooth table. The temporary precision spacer allows a 0.2 mm spacing between the top and bottom magnets.

Board Modification Procedure

• Enlarge the four corner mounting holes with a 3.2 mm bit and a drill. I used a #30 bit (3.2639 mm) because I don't have a 3.2 mm bit (link (http://http://bobmay.astronomy.net/misc/drillchart.htm)). The reason for this step is to allow enough space for the 3.0 mm diameter neodymium magnets (plus 0.2 mm margin) to be positioned through the board mounting holes.
• Deburr the top and bottom of the mounting holes to eliminate any rough edges. I suggest a deburr tool with a single cutting edge to reduce chatter.  Because of the tight space between the 12-pin shrouded male header and the mounting holes, I used a different process for those two locations and managed to muck up a corner of the board (see photographs). Sigh...
• The sides of the board were smoothly routed and were the expected dimension. The ends of the board were slightly rough and longer than expected. I used a palm sander with 400 grit sandpaper to sand 0.3 mm off the ends of the board and to round over the corners of the board slightly (approximately 0.5 mm radius).
• Blow off any fiberglass dust or metal particles.  Wash the board under running water and dry with a cloth.

Place the v4fd Bus Pirate board in the bottom case with the magnets in the board mounting holes.
The top and the bottom case will come together with an obvious "snap" and the resulting closure seems to be fairly robust.

Comments

• All connectors, LEDs, and push-buttons can be accessed from the exterior of the case.
• The LED light-pipes were left off for the photographs, but do fit properly. For the particular light-pipe I'm using, the LEDs were so close to the edge of the board that the top case had to be modified in order to make room for the light-pipe. (Comment ~ Please don't place components that close to the edge of the board.)
• The nylon material used for the case is slightly flexible which allows for a push-button cantilever scheme that pushes a plunger to activate a push-button. (Comment ~ The plunger length is made for the kind of low profile push-button shown the photographs. Please don't change the push-button height without some thought.)
• The USB photograph shows the 0.2 mm "reveal" between the top and bottom case.
• The board is located at the top of the bottom case in order to reduce the affect of draft angle if the case were to be injection molded.

[attachment=5]
[attachment=1]
[attachment=4]
[attachment=3]
[attachment=2]
[attachment=0]
Thanks for your time.

dsm

[1] "north" pole refers to the end of the magnet that is attracted to the earth's north magnetic pole.
[2] "south" pole refers to the end of the magnet that is attracted to the earth's south magnetic pole.[/font:]

---

Very nice. The magnets make such nice enclosures. You comments about placement are noted, thank you.

[font=Georgia:]Bus Pirate v3.5e or Bus Pirate v4c?

I'm pretty happy about the way the design eventually tuned out, although there were too many design iterations along the way.

[quote author="dsm"]My main conclusion from this exercise is to not design a case until I have a board example in hand.[/quote]
Any chance getting a Bus Pirate v3.5e board or Bus Pirate v4c board?
Bare boards would be okay, but loaded boards would be better.

Thanks for your time.

dsm[/font:]

PCBs are on the way. I do not have them yet, but I will mail them when I do. They will probably be unpopulated, though I am happy to send completed hardware when it is in production.

[font=Georgia:]comment

ian,

Thanks very much.

Having some DP6037 size boards would make it easier to design an "alternate SOB size case design" enclosure with a clear acrylic top and bottom plus an improved magnetic case closure scheme.

Just for fun, I sent you one of the assembled Bus Pirate v4fd cases described above. It should arrive in couple days.
The magnet wells were tight enough so that the magnets stayed in place without any glue, but loose enough so that the magnets could be pressed into place by hand. The neodymium magnets were ordered off eBay and were relatively inexpensive.

I also included a small icosahedron (generated from three golden rectangles at right angles to each other). I fabricated this design to get a handle on the strength of nylon objects printed using the Shapeways SLS 3D printing process versus typical FDM 3D printing. The edges are 2.0 mm in diameter and the skeletonized golden rectangles are 1.0 mm thick.  Try squeezing the icosahedron at two opposing vertices (not the edges) and you'll see what I mean.

Enjoy!

Thanks for your time.

dsm[/font:]

Very cool. Thank you so much. I look forward to seeing it. We are going to finalize the PCB footprints this week and I will send them out for samples. Vimark is going to add the keepout, USB placement, and edge keepout as suggested.

[font=Georgia:]did the package arrive?

The package was sent via USPS first class parcel post on Monday, 30Jan12.
I should have added tracking so I wouldn't have to ask this question, but did the package arrive?

Thanks for your time.

dsm[/font:]

No, not yet. It usually takes about 5 days, so I'm sure it will show up soon.

Just to check - you did send it to my .nl address?

[font=Georgia:]D'oh!

ian,

See private message.

Thanks for your time.

dsm[/font:]

I know I am resurrecting a long dead thread, but was this case ever finalized? I am the proud owner of a BPv4 but am in dire need of a case and would love to have one of this style.

If it was finalized, where can I buy it or download the files to have one made?

It's not being sold by Dangerous prototypes, but DSM has posted his designs on Shapeways and made them available for anyone to buy...

http://www.shapeways.com/model/431530/c ... sm-v1.html (http://www.shapeways.com/model/431530/clamshell-yabp-case-v4fd-bottom-dsm-v1.html)
http://www.shapeways.com/model/431532/c ... sm-v1.html (http://www.shapeways.com/model/431532/clamshell-yabp-case-v4fd-top-dsm-v1.html)