• State TI Thinks Resolved
  • Locked Locked
  • Replies 3 replies
  • Answers 1 answer
  • Subscribers 33 subscribers
  • Views 551 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

TCAN4550-Q1: Crystal Lower limit

Manu Manu
Prodigy 30 points

Part Number: TCAN4550-Q1
Other Parts Discussed in Thread: TCAN4550

Ho all,

Is it possible to use a 6 mhz crystal? Even though I have restrictions on possible CAN baudarates, are the rest of the features operational? In particular, is the spi working with this frequency?

  • 0
    TI__Mastermind 35980 points

    Hi Manu,

    I am assuming you want to use a 6 Megahertz cyrstal (MHz) and not a 6 millihertz (mHz) crystal, if so, yes that is possible.  The only restriction you will have is that the SPI frequency must be less than the Crystal frequency by about 1MHz.  This is due to the clock synchronization circuitry that allows the data to cross from the digital core clock domain to the SPI clock domain without generating errors.

    Therefore if you use a 6Mhz crystal, the maximum SPI frequency you could use would be approximately 5Mhz.

    However on the odd chance you did want to use a 6mHz clock frequency, I don't think it would work at all because that would be too slow.  But I know that is not what you meant and it was just a typo!

    Regards,

    Jonathan

  • 0 in reply to Jonathan Nerger
    Prodigy 30 points

    Thanks for your answer.

    'mHz' it's just a typo. I meant MHz. 

    I have to look for what the problem is, in fact on the crystal pins I have 1.2V fixed and the crystal does not seem to generate the clock. After about 4 minutes in the same pins i found 0V. Do you have any ideas that can help me diagnose and solve the problem?

    Thanks in advance.

  • 0 in reply to Manu Manu
    TI__Mastermind 35980 points

    Hi Manu,

    Regarding your first question about the crystal not oscillating, there may be a voltage issue, or schematic/component issue.  If you would send me the schematics I can review them and look for any errors or adjustments that may be needed. If you can share them on this forum that would be great, but if you would prefer to not share them publicly, let me know and I will send you an email.  But check the following:

    1.  Make sure the Vsup voltage is greater than 6V as seen at the device pin.  If you have a reverse polarity diode the voltage drop of this diode must be accounted for.  But the TCAN4550 needs approximately 6V to ensure the internal 5V LDO can create the full 5V and not have an undervoltage issue.  The internal LDO voltage is brought out on the VCCOUT pin and the 5V voltage rail can be measured here to see if the digital core is properly powered.
    2. Make sure your VIO voltage is either 3.3V or 5V depending on your MCU's digital IO signal levels.  The VIO rail also powers the Crystal Oscillator circuit, so this supply needs to be at the proper voltage levels, have proper decoupling capacitors, and be stable in order for the crystal oscillator to function.
    3. Check the crystal's load capacitance values and adjust them to match the recommended values.  The Crystal Oscillator in the TCAN4550 is a Pierce Oscillator type and it requires that the Crystal's Inductance and Load Capacitance be properly balanced to create the oscillation.  If the load capacitors are too small or too large, this can prevent oscillation.

    Regarding your observation of the 0V after 4 minutes, this is an easy one and the result of a device feature.  The TCAN4550 has a Failsafe Feature with 3 mechanisms that can cause the device to enter Sleep Mode which effectively turns off the device.  Refer to sections 8.4.2 and 8.4.5 of the datasheet for more information.  But what you are seeing is the Sleep Wake Error Timer timing out.

    Upon power up, a power on reset or wake event from sleep mode causes the TCAN4550 to enter standby mode.  This starts a four minute timer, tinactive, that requires the processor to either reset the interrupt flags or configure the device to normal mode.  This feature makes sure the node is in the lowest power mode if the processor does not come up properly.  This automatic mode change also takes place when the device has been put into sleep mode ad receives a wake event, Wake Up Pattern (WUP) on the CAN bus, or Local Wake Up (LWU) event on the WAKE pin.  To disable this feature for sleep events register 16'h0800[1] (SWE_DIS) must be set to one.  This will not disable the feature when powering up or when a power on reset takes place.

    So, during your initial development, I recommend you disabling this feature since you will be having long delays while you run tests on the various features.  You will have to disable this feature through SPI every time you power on or reset the device since this feature is enabled by default and the registers are volatile and do not retain their values when not powered.   Or you can configure the device to Normal Mode within 4 minutes.  If the device does go to sleep, all you need to do is toggle the WAKE pin to wake it back up.  However, the register values will have also been reset.

    Regards,

    Jonathan