Shop OBEX P1 Docs P2 Docs Learn Events
Gadget Gangster Propeller Platform USB - Going Away? - Page 3 — Parallax Forums

Gadget Gangster Propeller Platform USB - Going Away?

135

Comments

  • RaymanRayman Posts: 14,842
    edited 2012-03-29 16:31
    I'd like to know more about how the contact current rating scales with voltage...
    The switch I've been using is rated at 0.25 A at 125 VAC... This one is rated at 0.3 A at 6 VDC and 0.1 A at 30 VDC.
    Ideally, I'd like a switch that can handle the allowed USB load of 0.5 A at 5 VDC.
    Does my 125 VAC switch handle this? Well it does, but maybe it won't last the spec'd number of cycles?
    Or, does the 125 VAC rating mean it can handle a lot more at 5 V? I'm not sure...

    Anyway, I hope you find a good SMT switch, because I may use it, if it works...
    I think I'd leave off the switch entirely, like with the Quickstart... Or else, just leave the pads to mount it, unpopulated...
  • cavelambcavelamb Posts: 720
    edited 2012-03-29 16:33
    Worth the added cost - in my humble opinion.
    :)
  • RaymanRayman Posts: 14,842
    edited 2012-03-29 17:34
    Took another stroll through Digikey and I think Martin picked out a good switch.
    You pay a hefty price to get something rated for more that 300 mA it seems...
  • jazzedjazzed Posts: 11,803
    edited 2012-03-29 19:20
    Rayman wrote: »
    Took another stroll through Digikey and I think Martin picked out a good switch.
    You pay a hefty price to get something rated for more that 300 mA it seems...

    I've used the through-hole version of that switch. It seems pretty solid except that the top plate can pop off.
  • Martin HodgeMartin Hodge Posts: 1,246
    edited 2012-04-02 10:57
    Attached is a first draft Diptrace layout.
    Changes made:
    • Power switch to SMT
    • Reset switch to PTH. (Holes for either straight or right angle.)
    • Added 150 ohm resistors to RX/TX between Propeller and FT232.
    • Added single 'activity' LED to USB circuit.
    • Changed uSD card pull-ups to SIP socket.
    • Added reset pins to lower header.
    • Adjusted signal traces to 10/10 mil
    • Enlarged a few power traces.
    • Increased via sizes on power nets.
    • Added pads for SMT electrolytic caps while keeping holes for TH caps.
    ppusb.zip
  • TubularTubular Posts: 4,706
    edited 2012-04-02 14:19
    Martin (not Jazzed),
    Why not add the reset pin between P31 and Vss, as well as/instead of P0 and Vss?
  • Martin HodgeMartin Hodge Posts: 1,246
    edited 2012-04-02 14:34
    Tubular wrote: »
    Jazzed,
    Why not add the reset pin between P31 and Vss, as well as/instead of P0 and Vss?

    Tubular (not Pedward,)
    The idea was to provide a little bit of keying to help prevent boards from being plugged in backwards. Nothing is set in stone at this point.
  • pedwardpedward Posts: 1,642
    edited 2012-04-02 14:44
    Pedward,
    The idea was to provide a little bit of keying to help prevent boards from being plugged in backwards. Nothing is set in stone at this point.

    LOL! The PropKey must've been on your mind!
  • TubularTubular Posts: 4,706
    edited 2012-04-02 14:49
    LOL, sorry Martin got your post mixed with the one immediately above it. It's early here, no coffee yet.

    Ok, so keying is perhaps a good idea. Is there a reason to put the reset pin near P0, rather than P31?
  • RaymanRayman Posts: 14,842
    edited 2012-04-02 15:16
    Reset by P31 is better isn't it? Let's you plug in a PropPlug, if you fry the USB chip...
  • Martin HodgeMartin Hodge Posts: 1,246
    edited 2012-04-02 17:40
    Rayman wrote: »
    Reset by P31 is better isn't it? Let's you plug in a PropPlug, if you fry the USB chip...

    Oh that's right, we talked about that already. Durr :innocent:
  • pedwardpedward Posts: 1,642
    edited 2012-04-02 18:16
    I got a question, right now I don't bring 28-31 out to the pin header on the PropKey because those are consumed by peripherals. Now that I have a uSD socket, there are 4 pins used by that, with pullups. Do you think I should expose the 4 uSD pins on the socket/header or the 4 other pins?

    EDIT: oooh, I got an idea, I'll bring the 4 uSD pins out under where the uSD plugs in and expose the other 4 on the standard header. If you use a uSD, you don't use the pins, but if you use the pins, the uSD isn't plugged in!
  • TubularTubular Posts: 4,706
    edited 2012-04-02 18:35
    pedward wrote: »
    I got a question, right now I don't bring 28-31 out to the pin header on the PropKey because those are consumed by peripherals. Now that I have a uSD socket, there are 4 pins used by that, with pullups. Do you think I should expose the 4 uSD pins on the socket/header or the 4 other pins?

    EDIT: oooh, I got an idea, I'll bring the 4 uSD pins out under where the uSD plugs in and expose the other 4 on the standard header. If you use a uSD, you don't use the pins, but if you use the pins, the uSD isn't plugged in!

    That makes sense. And sometimes you want to mount a SD or uSD socket near the edge of an enclosure. You could see if its possible to take vss/vdd too to make it really easy
  • jazzedjazzed Posts: 11,803
    edited 2012-04-02 20:08
    If i was a diptrace user I might be able to look at this. Print to PDF ?
    Attached is a first draft Diptrace layout.
    Changes made:
    • Power switch to SMT
    • Reset switch to PTH. (Holes for either straight or right angle.)
    • Added 150 ohm resistors to RX/TX between Propeller and FT232.
    • Added single 'activity' LED to USB circuit.
    • Changed uSD card pull-ups to SIP socket.
    • Added reset pins to lower header.
    • Adjusted signal traces to 10/10 mil
    • Enlarged a few power traces.
    • Increased via sizes on power nets.
    • Added pads for SMT electrolytic caps while keeping holes for TH caps.
    ppusb.zip

    The middle three words seem to be missing. <-- edited.
  • TubularTubular Posts: 4,706
    edited 2012-04-02 20:34
    Here's a pdf. The scaling is slightly less than 100% (at least when I print it out. Believe the correct board size is 2.8" x 2.5")

    Jazzed, what do you mean about the middle three words missing?
  • jazzedjazzed Posts: 11,803
    edited 2012-04-02 21:00
    Tubular wrote: »
    Here's a pdf. The scaling is slightly less than 100% (at least when I print it out. Believe the correct board size is 2.8" x 2.5")

    Jazzed, what do you mean about the middle three words missing?

    Thanks Tubular. Seems my brain is on holiday today ....
  • pedwardpedward Posts: 1,642
    edited 2012-04-02 22:25
    Hi guys, I did some layout changes to the PropKey, to include the p28-p31 pins on the layout. I'm really on the fence about this change, since the previous layout is a lot cleaner and less crowded. I was hoping you guys might be able to take a quick comparison look and give me your thoughts in that thread.
  • Martin HodgeMartin Hodge Posts: 1,246
    edited 2012-04-05 21:36
    Minor update. Moved the reset header to the proper location and connected the uSD socket to 16-19.


    (This PDF is sans copper pour for clarity.)
    ppusb_20120406.pdf
  • pedwardpedward Posts: 1,642
    edited 2012-04-06 00:05
    With your layout change the pinouts/usage are identical between the PropKey and what you've done. Did you move the uSD to 16-19 so the first 16 port pins are contiguous?
  • RaymanRayman Posts: 14,842
    edited 2012-04-06 00:08
    Martin, can I talk you into moving SD back to P0..P3? I think all of my shields use the upper row of pins...
    Actually, I bet all existing shields leave P0..P3 alone, expecting SD there...
    Also, you'd have software incompatibility with all the existing Prop Platfrom USB code...
  • pedwardpedward Posts: 1,642
    edited 2012-04-06 00:28
    It could be routed both ways, with a jumper block to select the pins. I felt there was a real tangible benefit to having p0-p15 available and contiguous, for various interfacing reasons.
  • g3cwig3cwi Posts: 262
    edited 2012-04-06 02:19
    If there is a re-design going on can I suggest:

    1) the inclusion of a protection diode on the power input. A Schottky diode will have a nice low forward volt drop. Reason: I blew a board up due to a faulty walwart that dumped 12V AC into it.
    2) moving the crystal further from the regulators. Reason: radio hams are using this board for transmitting and the regulators heat the crystal causing it to drift.

    Cheers

    Richard
  • pedwardpedward Posts: 1,642
    edited 2012-04-06 02:42
    g3cwi wrote: »
    If there is a re-design going on can I suggest:

    1) the inclusion of a protection diode on the power input. A Schottky diode will have a nice low forward volt drop. Reason: I blew a board up due to a faulty walwart that dumped 12V AC into it.
    2) moving the crystal further from the regulators. Reason: radio hams are using this board for transmitting and the regulators heat the crystal causing it to drift.

    Cheers

    Richard

    Who is doing transmitting with the Prop? The crystal drift should only be a problem with direct modulation.
  • g3cwig3cwi Posts: 262
    edited 2012-04-06 02:46
  • Martin HodgeMartin Hodge Posts: 1,246
    edited 2012-04-06 09:08
    @Pedward. I suspected there would be some controversy over moving the SD card. I was a bit hesitant myself. I had been mulling over using this (or something similar) and a single jumper.

    The options I see are
    1. Leave it alone
    2. Switch it permanently to 16-19
    3. Use the 4PDT analog switch
    4. Use a 3x4 block of solder jumpers in a 4PDT arrangement.
      (Cheap, but more difficult for the user and requires extra step in mfg to set a default.)
    5. Use a 3x4 male header with 4 movable shunts.
      (Huge, may not be room.)
    @g3cwi, wouldn't it be a better solution to move the heat away from the crystal instead of the other way around? A simple external power supply fed into the V33 header would accomplish that. The diode is a good idea, I have a stock of those too.
  • RaymanRayman Posts: 14,842
    edited 2012-04-06 09:22
    In case you missed it, I vote #1 (leave it on P0..P3).

    As an aside, what I would do actually, is leave off the pull-up resistors. That way, it doesn't interfere if somebody want's to use P0..P15 for somethings special.

    Or, I see you have the pull-ups as something that could be soldered in after the fact...
  • pedwardpedward Posts: 1,642
    edited 2012-04-06 10:26
    For the xtal, I'd recommend simply making it a dual package layout so you could put a half-size can oscillator next to the chip.
  • Martin HodgeMartin Hodge Posts: 1,246
    edited 2012-04-06 10:56
    @Rayman, I was planning on putting a sip socket at the RP1 location so that the pull-ups can be removed. However maybe just leaving it unpopulated would be better. I've never had any problems without pull-ups.
  • RaymanRayman Posts: 14,842
    edited 2012-04-06 11:00
    Somebody said (cluso maybe?) it's actually better without the pullups because there's a way to detect the presense of the card using one of the pins...
  • jazzedjazzed Posts: 11,803
    edited 2012-04-06 11:34
    Minor update. Moved the reset header to the proper location and connected the uSD socket to 16-19.


    (This PDF is sans copper pour for clarity.)
    ppusb_20120406.pdf

    Excellent. Without these changes, may as well not do another board.
Sign In or Register to comment.