P2 Eval PCB Revision

13»

Comments

  • IMHO, anyone wanting all these options should just build their own pcb.
    My Prop boards: P8XBlade2, RamBlade, CpuBlade, TriBlade
    Prop OS (also see Sphinx, PropDos, PropCmd, Spinix)
    Website: www.clusos.com
    Prop Tools (Index) , Emulators (Index) , ZiCog (Z80)
  • jmgjmg Posts: 13,609
    Cluso99 wrote: »
    IMHO, anyone wanting all these options should just build their own pcb.

    Using that Logic, there would never be an Eval Board at all :)
    Surely the purpose of an Eval Board, is to easily Evaluate most of the uses ? - and to encourage broad usage of the P2, in calsses and Labs and ... anywhere ?
    Plenty of Eval's out there have unstuffed footprint options, where the designers had the foresight to think how it might be used..

  • jmg wrote: »
    Cluso99 wrote: »
    IMHO, anyone wanting all these options should just build their own pcb.

    Using that Logic, there would never be an Eval Board at all :)
    Surely the purpose of an Eval Board, is to easily Evaluate most of the uses ? - and to encourage broad usage of the P2, in calsses and Labs and ... anywhere ?
    Plenty of Eval's out there have unstuffed footprint options, where the designers had the foresight to think how it might be used..
    +1
    Propeller Object Exchange (last Publications / Updates) --- Oldbitcollector's guest map
    JustForMe
  • samuellsamuell Posts: 323
    edited 2019-06-21 - 10:50:57
    jmg wrote: »
    samuell wrote: »
    If I recall correctly, the terminations are done by the P2 itself, and no external resistors are needed.
    The P2 does not have 50 ohm style termination resistors, and certainly not on the Xtal pins.
    ...
    50 ohm resistors on the midst of a crystal oscillating circuit, really? Are you serious? X100 is a crystal, not an oscillator. When you mentioned "termination resistors" I though you wanted to convey "bias resistors", but you really meant "termination resistors". You'll dampen the oscillation with those. Anyway, the P2 has the necessary components inside, capacitors and resistors, to drive the crystal.

    Furthermore, in order to do what you want, you'll have to take out the crystal entirely, and then pray that the signal you inject is accepted by the P2 and doesn't damage it. Not everyone has clock generators sitting at home, by the way. But I think you could sacrifice your own P2 Eval board. So, scrape four pads on the ground plane for a coaxial connector, take out the crystal and then run a wire with a 50 ohm resistor in series from the center of the coax to the XI pin of the P2.
    jmg wrote: »
    samuell wrote: »
    Any way, you have plenty of frequency combinations to choose from with a single crystal frequency. The dividers and PLL allow a vast number or clock speeds.
    ..for some values of 'vast'. However, even with the set of valid PLL integers, Xtal precision is quite poor, on the frequency quality scale.
    So, you are talking about jitter? Because that is the only possible issue that I see, and any system that has a PLL in it will be able to show some degree of jitter. Is jitter problematic? Have you made some measurements, by the way?
    jmg wrote: »
    Cluso99 wrote: »
    IMHO, anyone wanting all these options should just build their own pcb.
    Using that Logic, there would never be an Eval Board at all :)
    ...
    Sorry, but that argument is as flawed as it can be. The purpose of the Eval board is to evaluate the chip. Thus, you add the most typical options. If you add too many unnecessary options you'll get a camel, which is a horse designed by committee.
    jmg wrote: »
    Cluso99 wrote: »
    IMHO, anyone wanting all these options should just build their own pcb.
    ...
    Surely the purpose of an Eval Board, is to easily Evaluate most of the uses ? - and to encourage broad usage of the P2, in calsses and Labs and ... anywhere ?
    Plenty of Eval's out there have unstuffed footprint options, where the designers had the foresight to think how it might be used..
    Nope! Plenty of evaluation boards only implement the typical application circuit. The only exception to this, that I'm aware of, is the CP2130EK, which has lots and lots of jumpers, and yet it hasn't empty footprints. That is why you have connectors on the P2 Eval. You also don't want to remove features that impede a proper evaluation (eg. replacing the crystal with an external coaxial connector to inject a clock, which would make it mandatory to have a clock generator in order to evaluate the P2).

    Kind regards, Samuel Lourenço
  • jmgjmg Posts: 13,609
    samuell wrote: »
    jmg wrote: »
    samuell wrote: »
    If I recall correctly, the terminations are done by the P2 itself, and no external resistors are needed.
    The P2 does not have 50 ohm style termination resistors, and certainly not on the Xtal pins.
    ...
    50 ohm resistors on the midst of a crystal oscillating circuit, really? Are you serious? X100 is a crystal, not an oscillator. When you mentioned "termination resistors" I though you wanted to convey "bias resistors", but you really meant "termination resistors". You'll dampen the oscillation with those. Anyway, the P2 has the necessary components inside, capacitors and resistors, to drive the crystal.
    ? I think You need to read the post I was replying to.
    Your post also said this "If I recall correctly, the terminations are done by the P2 itself, and no external resistors are needed."
    So I'm not following your many claims now...
    samuell wrote: »
    So, you are talking about jitter? Because that is the only possible issue that I see, and any system that has a PLL in it will be able to show some degree of jitter.
    Err nope. Jitter is only one issue, that dictates you use a higher PFD . You did see the discussions on PFD vs Jitter ?
    The other issue are
    * Xtal precision and temperature drift are poor
    * PLL has only small integers, and with the 'keep PFD high' rule, that constrains what you can ask for.

    Anyway, P2D2 already has a VCTCXO oscillator footprint, and Chip has said the VCTCXO is on the list, so side discussions are moot.
    The parts are cheap, precise, and widely available.
  • Ok, by termination resistors I meant bias resistors. Anyway, I can't make any conclusions without seeing a schematic. Lets see how the P2 behaves with a VCTCXO. Probably, a 1.8V part would be ideal.

    Anyway, I want to ask Chip if an oscillator can be used instead of a crystal. Many micros allow it, and I've done it before. Never encountered a case where I couldn't inject a clock on the XI leg of a micro that expects a crystal, but one can't be sure.

    Kind regards, Samuel Lourenço
  • jmgjmg Posts: 13,609
    samuell wrote: »
    Lets see how the P2 behaves with a VCTCXO. Probably, a 1.8V part would be ideal.
    That depends on if it is Cliped Sine, or CMOS out.
    I have already tested clipped sine VCTCXO fine with P2 (even without removing the Xtal from PCB, as MHz is far enough separated ). As mentioned that needs AC coupling, for the clipped sine > 0.8V drive.

    CMOS models should be 3v3, as the Xtal osc is 3v3 domain. A 1.8V CMOS VCTCXO model would not work very well, unless it was AC coupled.

  • samuell wrote: »
    Ok, by termination resistors I meant bias resistors. Anyway, I can't make any conclusions without seeing a schematic. Lets see how the P2 behaves with a VCTCXO. Probably, a 1.8V part would be ideal.

    Anyway, I want to ask Chip if an oscillator can be used instead of a crystal. Many micros allow it, and I've done it before. Never encountered a case where I couldn't inject a clock on the XI leg of a micro that expects a crystal, but one can't be sure.

    Kind regards, Samuel Lourenço

    Yes, you can inject a clock into XI. Just select crystal oscillator mode %01. These are bits D[3:2].

    %00 = XI/XO disabled
    %01 = XI/XO enabled, 1M-ohm feedback, +0pF/pin
    %10 = XI/XO enabled, 1M-ohm feedback, +15pF/pin
    %11 = XI/XO enabled, 1M-ohm feedback, +30pF/pin
  • jmgjmg Posts: 13,609
    Tubular wrote: »
    Ah ok. Well thats a better outlook then

    Am I right in saying normally these things appear on Mouser/Digikey well before stock is available? I've seen that with the FPGAs. Not so sure about LDO regs

    Avnet now at least show part codes for NCP187AMT330TAG 3,000+$0.2089, stock is more elusive... (but the Automotive one NCV8187 has samples button)
  • jmgjmg Posts: 13,609
    Tubular wrote: »
    Ah ok. Well thats a better outlook then

    Am I right in saying normally these things appear on Mouser/Digikey well before stock is available? I've seen that with the FPGAs. Not so sure about LDO regs

    .. Just up on Digikey... :)
    https://www.digikey.com/product-detail/en/NCP187AMT330TAG/NCP187AMT330TAG-ND
    Shows 3000 arriving in 1 week ....
Sign In or Register to comment.