Shop OBEX P1 Docs P2 Docs Learn Events
Propstick Crystal — Parallax Forums

Propstick Crystal

RumpleRumple Posts: 38
edited 2012-11-23 23:00 in Propeller 1
I'm looking at the Propstick USB schematic and I see that the 5 MHz crystal is connected between the XO and XI pins on the Prop. Below that I see that the Prop's XO pin is labelled NC (No Connection, I 'spose).

Hmmmm...

But my question is actually this: If I wanted to run a Propstick from a single-ended 5 MHz oscillator instead of a crystal, what would be the correct way to connect it?

Thanx for reading my post.
Tim

Comments

  • PublisonPublison Posts: 12,366
    edited 2012-11-22 17:02
    The Propstick USB has a socketed crystal that can be removed. An external signal can be connected to XI.
    Below that I see that the Prop's XO pin is labelled NC (No Connection, I 'spose).

    There seems to be a discrepancy in the schematic that I never noticed before. The section for just the crystal shows a connection to XI and XO, but on the chip, XO is not connected.

    http://www.parallax.com/Portals/0/Downloads/docs/prod/prop/32210-PropStickUSB-1.2.pd


    XO can be left disconnected depending on the CLK Register settings. (Page 28-29)


    http://www.parallax.com/Portals/0/Downloads/docs/prod/prop/Web-PropellerManual-v1.2.pdf
  • RumpleRumple Posts: 38
    edited 2012-11-22 17:58
    Yes, I've looked at that schematic a hundred times too and never noticed that. I've used the Propstick for dozens of things. But...here's the issue: I have to build a Prop circuit that'll start and work at -50F. The Propstick won't do that because of the EEPROM, the XTAL, the FT232 and possibly for other reasons. But, after pondering this, I've come up with some workaround rationalizations:

    EEPROM: This chip is rated for -40F. But it's only used at startup...I'm not logging any data. I'll use it and pray that it works at -50F, in case this product ends up in, like, Antarctica.

    XTAL: I can't find a 5 MHz crystal that claims to be good at -50F, but I can find lots of 10 MHz oscillators that do. I can divide it by two with a 4013 and, according to the above, inject it into the IX input. BTW, most TTL chips only claim to work above 32F...

    FT232: No problem, offboard programmer.

    OTHER: Dunno. Build and test, build and test, repeat as necessary.

    Does anyone see any potential potholes here?

    Thanx.
  • PublisonPublison Posts: 12,366
    edited 2012-11-22 18:04
    I just looked at the Diptrace layout from the product page, and X) goes to the correct pin on the prop but is not brought out to Pin 31 on the PROP-USB, so it's connected internal on the board.
  • PublisonPublison Posts: 12,366
    edited 2012-11-22 18:18
    Rumple wrote: »
    Yes, I've looked at that schematic a hundred times too and never noticed that. I've used the Propstick for dozens of things. But...here's the issue: I have to build a Prop circuit that'll start and work at -50F. The Propstick won't do that because of the EEPROM, the XTAL, the FT232 and possibly for other reasons. But, after pondering this, I've come up with some workaround rationalizations:

    EEPROM: This chip is rated for -40F. But it's only used at startup...not logging data(right?) I'll use it and pray that it works at -50F, in case this product ends up in, like, Antarctica.

    XTAL: I can't find a 5 MHz crystal that claims to be good at -50F, but I can find lots of 10 MHz oscillators that do. I can divide it by two with a 4013 and, according to the above, inject it into the IX input (right?)
    BTW, most TTL chips only claim to work above 32F...

    FT232: No problem, offboard programmer.

    OTHER: Dunno. Build and test, build and test, repeat as necessary.

    Does anyone see any potential potholes here?

    Thanx.


    BTW, I am not a Parallax employee.

    I can attempt to answer the question about the 10MHz oscillator. A 10 MHz on XI should work just fine with a PLL of X8 (80 MHz). Been used many times.

    I can't answer the questions about the cold temps. My installs have been in more friendly environments. :)

    The specs on the Propeller USB are:

    Operating temp range: +32 to +158 °F (0 to +70 °C)
  • RumpleRumple Posts: 38
    edited 2012-11-22 18:31
    Yes, the Propstick is out (although I've never tested its practical temperature range). So I'll go with discrete components. The claimed temp range of the Prop chip is -67 to +257 °F. No problem there.
  • RumpleRumple Posts: 38
    edited 2012-11-22 18:34
    BYW, Publison, thanx for the input. Happy holiday.
  • Phil Pilgrim (PhiPi)Phil Pilgrim (PhiPi) Posts: 23,514
    edited 2012-11-22 18:38
    Publison wrote:
    A 10 MHz on XI should work just fine with a PLL of X8 (80 MHz).
    That's true at room temperature. I don't know that It's ever been put to the test at temperature extremes.

    BTW, the "NC" in the PropStick USB schematic is definitely a misprint. It should be labeled "XO".

    -Phil
  • PublisonPublison Posts: 12,366
    edited 2012-11-22 18:38
    Rumple wrote: »
    Happy holiday.

    You as well. I'm stuffed....:)
  • PublisonPublison Posts: 12,366
    edited 2012-11-22 18:45
    Phil Pilgrim (PhiPi) wrote: »

    BTW, the "NC" in the PropStick USB schematic is definitely a misprint. It should be labeled "XO".

    -Phil

    Phil,

    I checked the Diptrace layout, and I don't see a trace going out to Pin 31 on the Prop-USB. X0 does go directly to the Prop pin 29
  • Phil Pilgrim (PhiPi)Phil Pilgrim (PhiPi) Posts: 23,514
    edited 2012-11-22 18:45
    BTW, you can drive the XI input directly with 80 MHz, without involving the PLL. Here's an 80 MHz oscillator that runs from 3.3V and down to -55F:

    -Phil
  • Phil Pilgrim (PhiPi)Phil Pilgrim (PhiPi) Posts: 23,514
    edited 2012-11-22 18:51
    Publison wrote:
    I checked the Diptrace layout, and I don't see a trace going out to Pin 31 on the Prop-USB. It does go directly to the Prop pin 29
    The "NC" on the carrier board's pin 31 may well be a no-connect. I was referring to the "NC" on the Prop part of the schematic to the right of that. The crystal would be non-functional without a connection to XO.

    -Phil
  • PublisonPublison Posts: 12,366
    edited 2012-11-22 19:04
    Phil Pilgrim (PhiPi) wrote: »
    The "NC" on the carrier board's pin 31 may well be a no-connect. I was referring to the "NC" on the Prop part of the schematic to the right of that. The crystal would be non-functional without a connection to XO.

    -Phil

    You'r correct Phil, I was just trying to convey that information to the OP per his original message. I also did not understand why the X0 was not brought out to Pin 31
  • Phil Pilgrim (PhiPi)Phil Pilgrim (PhiPi) Posts: 23,514
    edited 2012-11-22 19:11
    Publison wrote:
    I also did not understand why the X0 was not brought out to Pin 31
    It's probably to discourage off-board crystal attachment or attempts to drive external circuitry from XO, both of which could lead to problems. The same has been done with the connector on the QuickStart, BTW: XI, but no XO.

    -Phil
  • PublisonPublison Posts: 12,366
    edited 2012-11-22 19:12
    Phil Pilgrim (PhiPi) wrote: »
    It's probably to discourage off-board crystal attachment or attempts to drive external circuitry from XO, both of which could lead to problems. The same has been done with the connector on the QuickStart, BTW: XI, but no XO.

    -Phil

    Great info as always. Thanks!

    Jim
  • RumpleRumple Posts: 38
    edited 2012-11-22 21:02
    So I can inject 80 MHz into Prop-Q44 pin 28, leave pin 29 open ,and do


    CON
    _xinfreq=80_000_000 'use external 80 MHz oscillator
    _clkmode=XINPUT 'use external 80 MHz oscillator

    and it'll work?
  • Phil Pilgrim (PhiPi)Phil Pilgrim (PhiPi) Posts: 23,514
    edited 2012-11-22 22:57
    Yup.

    -Phil
  • Tracy AllenTracy Allen Posts: 6,664
    edited 2012-11-23 20:45
    There should be no need for the external oscillator--crystals rated to -55°C are readily available in the HC49 package. For example Mouser and Digikey both carry the ECS-080-20-4X-DU, in 4, 8 and 10MHz. I don't see 5MHz in distribution, but it is within the range of the possible.

    In a way, it is specsmanship. Even if a given crystal would most likely start up fine and operate at -55°C, its -40°C rating may be the crossover point for +/- 30ppm stability. AT cut crystals have a cubic temperature dependence. The crystal rated to for the wider temperature range may be speced at +/-100ppm stability over temperature. Of course there may be other issues having to do with operating temperature range, for example how the crystal is packaged. You'd best go into it with a call to the crystal manufacturer.
  • Phil Pilgrim (PhiPi)Phil Pilgrim (PhiPi) Posts: 23,514
    edited 2012-11-23 21:59
    Tracy, I nearly made the same post as you after searching DigiKey. But DigiKey erred in their specs. According to the datasheet, the storage temp goes down to -55C, but the operating temp goes only as low as -10C (+14F).

    -Phil
  • Tracy AllenTracy Allen Posts: 6,664
    edited 2012-11-23 22:44
    Phil, the storage temperature applies to all grades, but the part number for this particular xtal pegs it at a wider operating temperature range. I hate to be left reading between the lines, searching for the invisible ink. For a project that is going to be left out all alone in Antarctica, I'd certainly want to have a talk with the factory. What is the QC level?
    ECS-080-20-4X-DU
    D : +/-100ppm
    U : -55 – +125°C (operating temperature is implied but not stated)

    Screen shot 2012-11-23 at 10.26.30 PM.png


    Digikey also carries the ABLS-5.000MHZ-L4Q-T in HC49 (smt style)
    L : -55 – +125°C operating
    4 : 30ppm at 25°C
    Q : 100ppm over range

    Screen shot 2012-11-23 at 10.39.42 PM.png
    Screen shot 2012-11-23 at 10.26.30 PM.png
    964 x 223 - 45K
    Screen shot 2012-11-23 at 10.39.42 PM.png
    1006 x 492 - 101K
  • Phil Pilgrim (PhiPi)Phil Pilgrim (PhiPi) Posts: 23,514
    edited 2012-11-23 23:00
    Tracy,

    Even after rereading the datasheet, I'm not sure what to make of the discrepeancy between the "Operating Conditions" and the "Custom Options." I'm sure you're correct, and maybe it's just a matter of post-production binning, But, as you recommend, I would definitely contact the manufacturer before committing to that crystal.

    -Phil
Sign In or Register to comment.