Shop OBEX P1 Docs P2 Docs Learn Events
maximum clock speed — Parallax Forums

maximum clock speed

cristofcristof Posts: 6
edited 2007-06-15 00:44 in Propeller 1
I was just wondering what the maximum possible clock frequency is on this chip. I was looking at the manual and it looks like you can get 16x the crystal frequency and that the maximum crystal speed you can put into the XINPUT pins is 128MHz, but I am sure that you can't get the 16x PLL to work on that high of a frequency! Can anyone tell me what is the absolute fastest clock rate at which I can run my cogs? And how can I do it?
Thanks!

Comments

  • AnubisbotAnubisbot Posts: 112
    edited 2007-06-14 20:35
    Hi,
    the hiest i heard of 100 Mhz or actulay 96 when you use a 6Mhz crystal with 16x

    And i heard about a 180Mhz at 5V with 16x but constantly cooled with liquid nitrogen or something...

    I want to try it in a fun project wit a Mineral oil cooled chip (hole electronics submerged)

    Anubisbot
  • Paul BakerPaul Baker Posts: 6,351
    edited 2007-06-14 20:49
    Anubisbot said...
    Hi,
    the hiest i heard of 100 Mhz or actulay 96 when you use a 6Mhz crystal with 16x
    correct
    And i heard about a 180Mhz at 5V with 16x but constantly cooled with liquid nitrogen or something...
    not correct, the chips fail somewhere around 114 MHz when actively cooled with freezing aerosol.
    I want to try it in a fun project wit a Mineral oil cooled chip (hole electronics submerged)

    Anubisbot
    We make no assurance any chip will operate beyond 80MHz.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Paul Baker
    Propeller Applications Engineer

    Parallax, Inc.
  • Bob the Builder on a C64Bob the Builder on a C64 Posts: 112
    edited 2007-06-14 21:19
    Why does the chip have to overheat and burn if you manage to make it go at 114 mhz and up?

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Realize that I am really a mad scientist··· and


    Don't forget it!

    http://raydillon.com/Images/Illustration/GameArt/WildIsle/WildIsle-Ink-ScientistClose.jpg

    ·
  • Paul BakerPaul Baker Posts: 6,351
    edited 2007-06-14 21:46
    The chip doesn't burn up when it is run at 114MHz, it just simply stops running. The freezing aerosol was used to cool the chip down because colder chips run faster.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Paul Baker
    Propeller Applications Engineer

    Parallax, Inc.
  • cristofcristof Posts: 6
    edited 2007-06-14 22:40
    Thanks for the reply Paul, and good luck to you Anubisbot in your efforts.
  • cristofcristof Posts: 6
    edited 2007-06-15 00:02
    OK, now that that question is out of the way, here comes a fun one: I am trying to use the propeller to drive an analog to digital converter, and this device requires that I send it a clock signal. I want to run it at 20MHz, but when I use the counters on my Propeller, I can only get a 1.92V peak to peak wave at that speed. Slowing the toggle down to 10MHz gives me 2.84V peak to peak, but I can only get the optimal 3.3V at 6-7MHz. Any suggestions? I am not sure how lenient the ADC will be in terms of the clock frequency, but I'm about to try it and see what happens...
    Thanks!
  • cristofcristof Posts: 6
    edited 2007-06-15 00:04
    Here is the code that I am using:

    ctra := %00100_000 << 23 + 1 << 9 + 13 'Toggle pin 13
    frqa := $40_000_000 'Toggles every 4th clock (20MHz)
    dira[noparse][[/noparse]13] := 1 'Set APIN to output
    repeat

    Post Edited (cristof) : 6/15/2007 12:09:01 AM GMT
  • rokickirokicki Posts: 1,000
    edited 2007-06-15 00:14
    Are you sure about your measurements? Is your scope set up to show such high frequencies?
    I know mine isn't. You need a pretty high-frequency scope to accurately show a 20MHz square
    wave (I'd start with at least a 100MHz scope, and even here I'd only expect to see a rough
    approximation of the shape).
  • mirrormirror Posts: 322
    edited 2007-06-15 00:28
    1.92V is less than 3dB down (actually about 2.5dB) on 3.3V. It's a bit scary when you think of it like that. Especially when you think that audio bandwidth (and other filter bandwidths) are defined in terms of the -3dB points.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    It's not all that hard to count the number of grains of sand on the beach. The hardest part is making a start - after that it just takes time.·· Mirror - 15 May 2007
  • cgraceycgracey Posts: 14,133
    edited 2007-06-15 00:40
    On our 1.5GHz scope, a high-to-low or low-to-high I/O pin·transition·appears to take 750ps, which is just a limitation of the scope. The I/O pads SPICE simulate at 250ps.·Your test setup·is likely attenuating signal amplitude at·higher frequencies and·giving you misleading readings.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔


    Chip Gracey
    Parallax, Inc.
  • cristofcristof Posts: 6
    edited 2007-06-15 00:43
    The scope I am using is the Tektronix TDS 220, which is 100 MHz with 1 GS/s sampling rate, and the wave doesn't look anything like a square wave, it looks more like a triangle wave, but you're right, that could easily be the fault of the scope. I will just try plugging it in and see if it works, but right now I am still writing up the control code for the ADC.
  • cristofcristof Posts: 6
    edited 2007-06-15 00:44
    Thanks Chip! Should we still get a full peak to peak of 3.3V at 20MHz then?
Sign In or Register to comment.