Parallax P2 Eval Board

cgraceycgracey Posts: 10,477
edited October 26 in Propeller 2 Vote Up0Vote Down
This is not completely done, but here's what it looks like. We will build 100 of these with the Amkor-packaged chips.

P2_Eval.png
1568 x 1559 - 308K
«13456710

Comments

  • 292 Comments sorted by Date Added Votes
  • P2_Eval2.png
    2015 x 1632 - 467K
  • That looks like a great board.

    Oh, now it's the waiting game... tap, tap, tap... Thanks Chip, and Parallax!

    Do not taunt Happy Fun Ball! @opengeekorg ---> Be Excellent To One Another SKYPE = acuity_doug
    Parallax colors simplified: https://forums.parallax.com/discussion/123709/commented-graphics-demo-spin<br>
  • That looks fantastic !!!

    Super itchy to get on the "sign-up sheet".... suuuuper itchy....

    j
  • Excellent board shape, no mistaking what it is.

    Purple solder mask would be good.
  • I would make the peripheral headers female instead of male, so we can use OTS jumper wires to connect it to a breadboard, just like the nice big propeller dev board. I love that dev board, it's just so awesome, would love to have something like that for a P2.

    Don't 'spose you could put a 16MB RAM chip on their too, so we can use this for bootstrapping some stuff (compilers, overlays, frame buffers, etc).
  • Looks good. Will work.

    I've been using keyed IDC headers for things like this lately. Saves me from having to think about doing it backwards...
    The caps beside the header would interfere with that.
    But, this can work if I'm careful...
    Prop Info and Apps: http://www.rayslogic.com/
  • 2X usb powered I guess... I think that's 5W. Should be enough...
    Prop Info and Apps: http://www.rayslogic.com/
  • Actually, I guess I'd just to a board-board connection without any cable...
    Having the connectors on the edge makes that easier...

    Might have to shape the board to clear ground post if want to use two headers..
    Prop Info and Apps: http://www.rayslogic.com/
  • I see a lot of what looks like power option pins for the VIO.

    Is using other than 3.3 V really a viable options?
    Prop Info and Apps: http://www.rayslogic.com/
  • jmgjmg Posts: 12,616
    edited October 26 Vote Up0Vote Down
    cgracey wrote: »
    This is not completely done, but here's what it looks like. We will build 100 of these with the Amkor-packaged chips.
    Looks good, is there a SCH, as it's not easy to track everything

    eg USB connectors appear to both have FTDI parts ? - so none wire direct to P2 - yet P2-USB testing will surely be quite important ?
    Also missing is a HDMI connector ?
    A board this size, has a ton of room.

    However, I can see somethings have got lost, and some risky moves ...

    * P2D2 has options for Crystal and oscillator, and Clipped Sine Oscillators. That is very easy to allow, yet here it is missing ?
    I would suggest supplying these boards with a high stability Oscillator fitted, to show off P2 features. (That means Clipped Sine)
    Also note P2D2 was improved to have a multi-footprint Oscillator, to cover the range of common osc sizes.

    * 5V on all the IO headers ?! - one slip and that's a dead P2!! *POOF* I'd call that far too risky, on limited supply Engineering Samples.

    Other comments:

    Switching regulators look to still use Schottky diodes, and have a BOM of 23 parts, for each one. Result is a large PCB footprint.
    Surely a more compact/ lower BOM solution can be found, so as to not scare customers away with the complexity to merely power a P2 ?
    It does not need to be an inductor-included switcher like on P2D2, as those are more expensive, and tend to place size ahead of efficiency.


    What package is used for the flash ? - it appears to be TSSOP8, but that is a rare package in Flash.
    Usual Candidates would be SO8 (big, but universal) and moving to DFN, and DFN2x3 seems to be a sweet spot, now available 4M/8M/16M sizes.
    The DFN2x3 can route quite snugly, right next to a P2 TQFP, see the attached file.

    Suggestion: Allow for both DFN2x3 and SO8 flash packages. Even if you never fit the 2x3 package, having it there shows what is possible.

    The DFN2x3 helps keep the 'applied footprint' small for P2, which will be important for winning sales.


    357 x 267 - 23K
  • Rayman wrote: »
    I see a lot of what looks like power option pins for the VIO.

    Is using other than 3.3 V really a viable options?

    IIRC Chip tested VIO down to just above 2V, but I think those VIO power options are more for Lower noise Power islands, for Analog work.
  • Like the design, but would have liked pin headers rather than pins. Too easy to clip something on and short something out.

    Like to use my pin jumper wire and not worry about shorting something.

    Mike
  • Rayman wrote: »
    I see a lot of what looks like power option pins for the VIO.

    Is using other than 3.3 V really a viable options?

    I think there's possibly a couple of 3v3 options selected by jumper - a local LDO regulator vs the switcher

    That header can also be used to measure local current, or apply a non 3v3 voltage
  • Rayman wrote: »
    I've been using keyed IDC headers for things like this lately. Saves me from having to think about doing it backwards...
    The caps beside the header would interfere with that.

    You mean the boxed headers ?
    Moving the caps, to allow a boxed header should be very easy to do.



  • cgracey wrote: »
    This is not completely done, but here's what it looks like.

    The pin labels on the inner headers are very confusing in their placement.

    ie do they tag the solder pad along side them, or the pin hovering above them ?
    Most PCB labels are solder-pad related, and tag the nearest part.

    Can those labels be moved, to avoid this ambiguity - eg if they place outside the solder pad, they tag the pad, and pin.

  • Tubular wrote: »
    I think there's possibly a couple of 3v3 options selected by jumper - a local LDO regulator vs the switcher

    I wonder, is that regulator stable, with no load cap ?
    They do not look that easy to replace / change ?
  • I just ordered a bunch of female-to-male jumper wires from JameCo. I will use them to connect this board to solderless breadboard(s).

    Re-inventing the wheel is not a waste of time if, when you are done, you understand why it is round.
  • I just ordered a bunch of female-to-male jumper wires from JameCo. I will use them to connect this board to solderless breadboard(s).

    Parallax has a bunch of jumper wires available:

    https://www.parallax.com/search?search_api_views_fulltext=Jumper+Wires
    Infernal Machine
  • jmg wrote: »
    Tubular wrote: »
    I think there's possibly a couple of 3v3 options selected by jumper - a local LDO regulator vs the switcher

    I wonder, is that regulator stable, with no load cap ?
    They do not look that easy to replace / change ?

    The load caps are at the pins - every supply pin has one
  • Tubular wrote: »
    The load caps are at the pins - every supply pin has one

    .. provided they are connected.
    It looks to me like the jumper option, isolates the IO pins from the local regulator, giving it no load C at all ?

  • ooooh, shiny :smile:

    I'm really looking forward to getting my hands on one of these!
    garryj
  • T ChapT Chap Posts: 3,902
    edited October 26 Vote Up0Vote Down
    Very nice! Can you put at least 2 holes in it. Also are all the pins on a .1” grid so it would be easy to mate a top board onto it.
  • iseries wrote: »
    Like the design, but would have liked pin headers rather than pins. Too easy to clip something on and short something out.

    Like to use my pin jumper wire and not worry about shorting something.

    Mike

    We could put female sockets, instead of male headers, no problem. we've gone back and forth on that issue.

    I hate the size of shrouded connectors, but they would inhibit 5 volt mistakes.
  • Wow, it's almost alive!
    I want one :smile: .
  • Phil Pilgrim (PhiPi)Phil Pilgrim (PhiPi) Posts: 22,093
    edited October 26 Vote Up0Vote Down
    Not sure about those diagonal connectors. They're cute, but somebody might want to put the board in an enclosure, and those make it impossible. Also, you must have mounting holes in the corners. No PCB exists in isolation. You really have to leverage the design to make use of other stuff that's already out there, like standard enclosures. So my advice would be to forget cute and make it usable.

    -Phil
    “Perfection is achieved not when there is nothing more to add, but when there is nothing left to take away. -Antoine de Saint-Exupery
  • The Flash is an SOIC-8.

    We will install a 16MB chip there with DDR mode. They are $1.25 in volume. Digi-Key has 100k+ in stock.
  • Mounting holes?
    Melbourne, Australia
  • Next to the VIO headers are local, quiet 3.3V LDO regulators. They will be useful for quiet analog on sets of 8 pins, each.

    Those jumpers can also be opened up for current measurements, from either the main VIO supply, which is adjustable, or the local 3.3V regulator.

    The main 1.8V and 3.3V switching regulators are both adjustable, up and down. the top two jumpers can also be removed for current measurements on either main supply.
  • Jmg, you noted that those switchers look kind of complicated. What would be simpler that could supply 3A and be adjustable?
  • T Chap wrote: »
    Very nice! Can you put at least 2 holes in it. Also are all the pins on a .1” grid so it would be easy to mate a top board onto it.

    Yes, all of the 2x6 headers are on the same 0.100" grid.

    We will put some holes on it.
Sign In or Register to comment.