Shop OBEX P1 Docs P2 Docs Learn Events
using watch crystal for propeller — Parallax Forums

using watch crystal for propeller

hinvhinv Posts: 1,252
edited 2007-04-20 20:40 in Propeller 1
Hi,

I was wondering if anybody has used a 32.768k watch crystal for the propeller.
I am not familiar with crystal specs. If it would work, it would be good for time stable low power projects.

Thanks,
Doug

Comments

  • LoopyBytelooseLoopyByteloose Posts: 12,537
    edited 2007-04-20 07:58
    It should work, but it is a lot slower than the 8Mhz mode. You won't have video, but it would make doing timing routines quite simple.

    It will complicate you RS-232 interface though.

    by the way, you can supposedly run the Propeller down to zero hertz or near zero.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    "If you want more fiber, eat the package.· Not enough?· Eat the manual."········
    ···················· Tropical regards,····· G. Herzog [noparse][[/noparse]·黃鶴 ]·in Taiwan
  • LeonLeon Posts: 7,620
    edited 2007-04-20 08:40
    It might be difficult to get the oscillator to work with a 32768 Hz crystal; the feedback capacitors are quite critical for proper operation, and the built-in ones might not be suitable.

    Leon





    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Amateur radio callsign: G1HSM
    Suzuki SV1000S motorcycle
  • Graham StablerGraham Stabler Posts: 2,507
    edited 2007-04-20 09:22
    On a similar note what is the smallest 5Mhz crystal that could be used with the propeller, I've some designs I'd like to build at absolute minimum size

    Graham
  • LeonLeon Posts: 7,620
    edited 2007-04-20 11:16
    Really small 5 MHz ones are difficult to find because of the crystal size for that frequency. 8 MHz and up crystals can be very small, though. ACT makes a small cylindrical crystal (ACT31SMX) for 3.5 MHz - 10 MHz that is 9.8 mm x 3 mm. It'll probably be expensive.

    Leon

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Amateur radio callsign: G1HSM
    Suzuki SV1000S motorcycle
  • Graham StablerGraham Stabler Posts: 2,507
    edited 2007-04-20 11:59
    I've not really looked into it yet but presumably you could run the prop with a higher f crystal and lower the PLL to reduce the crystal size, that might be worth looking into.

    Graham
  • parts-man73parts-man73 Posts: 830
    edited 2007-04-20 12:24
    How about using the internal oscillator?

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Brian Meade

    "They who dream by day are cognizant of many things which escape those who dream only by night" - Edgar Poe
  • rjo_rjo_ Posts: 1,825
    edited 2007-04-20 12:25
    Graham,

    http://en.wikipedia.org/wiki/Image:ChipScaleClock2_HR.jpg

    It isn't a crystal... but it is a clock... and it is small.

    Rich
    606 x 599 - 95K
  • LoopyBytelooseLoopyByteloose Posts: 12,537
    edited 2007-04-20 12:39
    What you really need to start with is the level of accuracy you want.

    If you want small, a resonator is smaller than a crystal. But it is usually a bit less accurate. It is fine for RS-232 serial.

    The little half height crystals are nearly as small as the clock crystal. And I really wonder if the Propeller would have trouble using a clock crystal without capacitors. You should just try it. It won't harm the Propeller. It will either start or won't start.

    Timekeeping for extreme accuracy requires crystals which include an oven to keep the temperature stable.

    Higher accuracy requires that you have an external connection [noparse][[/noparse]like the internet] to an Atomic Clock source.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    "If you want more fiber, eat the package.· Not enough?· Eat the manual."········
    ···················· Tropical regards,····· G. Herzog [noparse][[/noparse]·黃鶴 ]·in Taiwan
  • bambinobambino Posts: 789
    edited 2007-04-20 13:09
    I was wondering about the possability of using a regular prop setup with the 5Mhz crystal and then hooking up a 32k crystal to a couple of pins and use the counter on the input for a clock. Any one see any flags on that working?
  • rjo_rjo_ Posts: 1,825
    edited 2007-04-20 13:11
    Kramer

    I have a feeling that this is for a frequency sensitive micro-robotics application... and the problem isn't in the time keeping but in the paradigms we are all using... or in my case "trying to use."

    There are biological clocks... but nothing like 80_000_000. The brain doesn't seem to use any absolute signals at all... everything gets turned into differentials, which are compared to a genetic moduli and presets. Even recognition seems to be based on this pattern.

    Rich
  • Tracy AllenTracy Allen Posts: 6,658
    edited 2007-04-20 20:40
    For some kinds of "time stable low power projects" I would opt for using the heartbeat output of a Real Tiime Clock, at 1 second or 0.5 second, monitored by the Prop running on its internal oscillator at 20khz. RCslow is not accurate, but with an external accurate 1 second time base, the prop can count and pin down the nominal 20khz to a definite value and use that to calibrate time accurate slow tasks. Kick into high speed mode briefy for the tasks that require it.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Tracy Allen
    www.emesystems.com
Sign In or Register to comment.