• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Subscribers 34 subscribers
  • Views 193 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: Support Request –High-Temperature Issue

Imene Torchi
Prodigy 10 points
Part Number: TCAN4550


Tool/software:

Hello,

In our project we are using the TCAN4550RGYT.
We are facing an issue at higher ambient temperatures: the CAN communication stops working once the temperature reaches around 80 °C.

This behavior was not observed on our previous prototypes; it only appears with the latest batch of prototypes.

Could you please help us investigate this issue and identify the possible root cause?

Thank you in advance for your support.

Best regards,

Imène 

  • 0
    TI__Mastermind 33900 points

    Hello 

    Can you provide a schematic for review that includes your clock/crystal circuit?  Almost all, if not all, high temperature related CAN communication errors have been the result of a shifting parasitic capacitance due to the higher temperature that changes the total capacitive load on the crystal.  This leads to larger oscillation amplitudes which cause the lowest level of the waveform on the OSC2 pin to drop below the comparator threshold (90-150mV) that is used to detect and configure the device for single-ended clock mode typically done by connecting the OSC2 pin to ground.  

    To avoid this shift, some optimization of the clock circuit is needed to create some margin to accommodate temperature variation and component tolerance of all the components in the circuit (TCAN4550, crystal, capacitors, etc.).  The TCAN4550 devices with the stronger transconductance amplifiers within our distribution curve will source more current to the crystal and create a larger oscillation waveform which will be more susceptible to a mode change.

    To prevent this, a series "dampening" resistor is recommended between the amplifier output (OSC1 pin) and the crystal to limit the current or "dampen" the oscillation amplitude.  A value of 30-50 ohms is a common value when needed.  If this is not available, then increasing the value of the load capacitance on the crystal's two load caps will also absorb this current as well as offset the reduced parasitic capacitance resulting from the high temperature shift and stabilize the circuit.

    Please see the TCAN455x Clock Optimization and Design Guidelines Application Report (Link) for more information.

    Best Regards,

    Jonathan

  • 0 in reply to Jonathan Nerger
    Prodigy 10 points

    Hello Jonathan,

    Please find attached the schematic of the CAN1 transceiver.


    The CAN_1C_P and CAN_1C_N signals are directly connected to the connector. The board includes four CAN transceivers:

    • CAN1 and CAN3 configured as transmitters,

    • CAN2 and CAN4 configured as receivers.

    During testing, I connected CAN1 to CAN2, and CAN3 to CAN4.
     At 25 °C ambient, communication works correctly between 1 Left right arrow 2 and 3 Left right arrow 4.

    We have made some modifications:

    • For CAN1, we replaced resistor R517 with a 33 Ω value.

    • For CAN3, we also replaced the corresponding resistor with a 33 Ω value.

    Could you please confirm if we should also change the resistor values for all CAN channels?

    Currently:

    • Communication between CAN3 Left right arrow CAN4 is stable and works up to 100 °C.

    • Communication between CAN1 Left right arrow CAN2 usually stops around 80 °C, but sometimes it remains functional up to 100 °C, while at other times we still observe interruptions.

    Please let me know if you need any additional information.
    Thank you in advance for your support.

    Best regards,
    Imène

  • +1 in reply to Imene Torchi
    TI__Mastermind 33900 points

    Hi Imène,

    I would recommend replacing R517 with 33 ohms on all CAN channels to ensure all devices in both the transmit and receive configurations are stable.  

    The device has an Automatic Gain Control (AGC) circuit that will try to adjust the amount of current flowing to the crystal so that the oscillation peak-to-peak amplitude is approximately 1Vpp.  When the amplitude exceeds 1Vpp, the AGC will try to reduce the current and reduce the amplitude.  However, it can't be reduced to zero and there is a minimum amount of current that will always be sourced.  If the amplitude exceeds 1Vpp with the AGC at the minimum level, then the TCAN4550 will be unable to regulate the amplitude and there is a risk of switching to the single-ended clock mode.

    For devices that do not necessarily require the 33 ohm series resistor, this will not be an issue because the TCAN4550 device is and will still operate within the AGC's range of regulation and will simply adjust the current to accommodate the additional resistance. 

    As long at this resistance isn't increased to such a large value the AGC is now operating beyond it's max limit where it can't source enough current to start and sustain oscillation, then you should be fine.  A 33 ohm resistor is a common value I've seen used.

    Best Regards,

    Jonathan

  • 0 in reply to Jonathan Nerger
    Prodigy 10 points

    Hello Jonathan,

    We have replaced the resistors with 33 Ω on all CAN channels, and the issue has been resolved. Thank you very much for your help.

    Best regards,


    Imène

  • 0 in reply to Imene Torchi
    TI__Mastermind 33900 points

    Hi Imène,

    Thanks for letting me know.  I'm glad I could help.

    Best Regards,

    Jonathan