Shop OBEX P1 Docs P2 Docs Learn Events
100Mhz Prop at 3V (not 3.3V) Possible? — Parallax Forums

100Mhz Prop at 3V (not 3.3V) Possible?

KyeKye Posts: 2,200
edited 2010-08-25 21:12 in Propeller 1
The title sums up my question. I need to run a whole system at 3V to make interface easier between all said devices in the system.

Comments

  • LeonLeon Posts: 7,620
    edited 2010-08-25 06:37
    The Propeller hasn't been properly characterised for 100 MHz operation. You might have to to select devices that will run at that voltage, and test them to see if they work over the expected temperature range.
  • BradCBradC Posts: 2,601
    edited 2010-08-25 07:08
    I have a propscope that runs the prop at 100Mhz on 3v. I can crash it repeatably in less than 5 seconds. Then again, I can also crash it repeatably at 50Mhz...
  • Bobb FwedBobb Fwed Posts: 1,119
    edited 2010-08-25 09:34
    According to datasheet v1.2 page 32, you can run 100MHz, 3V up to 75C or so.

    EDIT: That is, it will run for at least a minute under forced air conditions. I would bet 50C would be fairly stable, and practical under most hobbyist conditions.
  • LeonLeon Posts: 7,620
    edited 2010-08-25 09:48
    That's just based on a sample. The Propeller hasn't been properly characterised at that speed and voltage, AFAIK. It might be OK for hobbyist applications.
  • Phil Pilgrim (PhiPi)Phil Pilgrim (PhiPi) Posts: 23,514
    edited 2010-08-25 10:10
    Kye,

    What peripherals are you using that require 3V logic? (If it's the XBee, which is spec'd at 3V, it will also run at 3.3V.)

    -Phil
  • KyeKye Posts: 2,200
    edited 2010-08-25 12:31
    I'm using a new Omnivision camera, the 99XX series. It runs at 1.8 volts normally but can be take up to 3 volts at max. Absolute max is 4.5 volts. The electrical characteristics section says it was tested and shown to work between 1.2 and 3.0 volts.

    I noticed that 3V is in the propeller's range of operation so... it would be easier just to make the voltage ranges match rather than add a voltage translator.

    Also, this is a comercial product I'm working on. So, if you think it would be stable then I will not go this direction.
  • LeonLeon Posts: 7,620
    edited 2010-08-25 12:58
    You could make sure by asking Parallax for their view on a professional application running at 3V and 100 MHz. I'd be interested to hear what they say. I wouldn't be surprised if that graph disappears when they issue professional documentation.
  • Phil Pilgrim (PhiPi)Phil Pilgrim (PhiPi) Posts: 23,514
    edited 2010-08-25 13:17
    Key,

    Why not drive the camera chip using a buffer from the 74LVC family (e.g. 74LVC1G125, 74LVC1G14, etc.)? These will run with a Vdd as low as 1.65V, yet accept input levels of 5V or more.

    -Phil
  • KyeKye Posts: 2,200
    edited 2010-08-25 13:27
    Mmm, voltage translators may get expensive due to their own cost and increased board routing. The goal for this project is to only have circuits on the front and back of the PCB.
  • Phil Pilgrim (PhiPi)Phil Pilgrim (PhiPi) Posts: 23,514
    edited 2010-08-25 13:34
    The 74LVC1G14 is little more than $.06 in quantity and comes in a minuscule SC-70 package. If you need more than one channel, multiple channel versions (e.g. 74LVC2G14) are available. If this is going to be a professional product, it's best not to cut corners.

    -Phil
  • LeonLeon Posts: 7,620
    edited 2010-08-25 13:39
    There's the cost of another supply, of course. That could add $5 to the manufacturing cost.
  • Phil Pilgrim (PhiPi)Phil Pilgrim (PhiPi) Posts: 23,514
    edited 2010-08-25 14:42
    An LP3990-1.8V regulator costs $.45 in quantity and is stable with a cheap ceramic output cap.

    -Phil
  • kuronekokuroneko Posts: 3,623
    edited 2010-08-25 17:06
    Leon wrote: »
    You could make sure by asking Parallax for their view on a professional application running at 3V and 100 MHz. I'd be interested to hear what they say. I wouldn't be surprised if that graph disappears when they issue professional documentation.

    The PropScope doesn't count as professional application?
  • LeonLeon Posts: 7,620
    edited 2010-08-25 17:20
    It's not running at 3V, is it? That makes a big difference.

    Has the Propeller actually been characterised for 100 MHz operation? I don't believe that it has, so 100 MHz isn't suitable for professional applications.
  • kuronekokuroneko Posts: 3,623
    edited 2010-08-25 17:27
    Leon wrote: »
    It's not running at 3V, is it? That makes a big difference.

    Has the Propeller actually been characterised for 100 MHz operation? I don't believe that it has.

    The schematic says that it's running at 3V (and 100MHz). The used regulator is a fixed 3.0V type.
  • LeonLeon Posts: 7,620
    edited 2010-08-25 17:32
    It's intended for use by hobbyists, so it doesn't matter if the Propeller is running outside its specification.
  • potatoheadpotatohead Posts: 10,261
    edited 2010-08-25 18:37
    LOL!!!

    Sorry, but I think that's all kind of funny. I wonder who properly characterizes that professional? (yes folks, I'm here all night!!)
  • BigFootBigFoot Posts: 259
    edited 2010-08-25 20:34
    Our latest PoS terminal runs at 100 MHz and we have over 87,000 hours
    of test time with the 6.25 MHz crystals. No problems so far, we hope
    to sell thousands of them every year.

    I don't know about running the propeller at 3.0 volts though. You could
    run the processor at 3.3 volts and generate the 3.0 volts with a Schotky
    diode drop.

    Russ
  • Mike GreenMike Green Posts: 23,101
    edited 2010-08-25 20:45
    Leon,
    The Propeller has been characterized with a 100MHz system clock at a 3V supply voltage. There's a description of the test applied and a sample of Propeller chips (as described) has passed the test at ambient temperatures at or below 75C. The Propeller datasheet shows the maximum system clock speed over well beyond the military temperature range and at supply voltages from 3.0V to 3.6V. If you don't like how the testing was done, I'm sure, if you had a large enough order, that Parallax would arrange for testing to your requirements.
  • LeonLeon Posts: 7,620
    edited 2010-08-25 21:12
    A few chips were tested for a minute, that is not characterization as is performed by most semiconductor manufacturers. It's a lengthy and rather expensive process. If the Propeller was characterized for 100 MHz operation, the data sheet would give the maximum speed as 100 MHz, rather than 80 Mhz, and it would be sold as a 200 MIPS device.
Sign In or Register to comment.