• State Resolved
  • Locked Locked
  • Replies 8 replies
  • Answers 1 answer
  • Subscribers 33 subscribers
  • Views 154 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

TCAN4551-Q1: Bypassing VSUP on TCAN4551/4550

Leon Hermans
Prodigy 130 points
Part Number: TCAN4551-Q1
Other Parts Discussed in Thread: TCAN4550

Tool/software:

Hello -

I understand the UVLO startup requirements of the TCAN455x (>5.9V worst case to assure proper startup).  Also, understand that you can 'bypass' VSUP by applying ~5.2V or greater to the VCCout terminal on the TCAN4550.

A couple related questions:

1)  Can you also bypass the LDO by applying 5.2V or greater to the VCCfltr terminal on the TCAN 4551?  And that way the CAN bus power will be derived from that terminal rather than VSUP?

2)  Before Vio is applied, is this the proper figure to use for Isup? ~25-42uA?

3)  Assuming Vio is applied, SPI active, and CAN bus with ~55 ohm loading, VSUP = 24V, VCCout/VCCfltr applied with external 5.2V, can you help us to estimate Isup?  I don't see a figure or way to calculate that.

Thank you & Happy Thanksgiving,

Leon

  • 0
    TI__Mastermind 22360 points

    Hi Leon,

    Due to the US holiday our expert supporting this device will get back to you next week.

    Regards,

    Sean

  • 0
    TI__Mastermind 29260 points

    Hi Leon,

    Leon Hermans said:
    I understand the UVLO startup requirements of the TCAN455x (>5.9V worst case to assure proper startup).  Also, understand that you can 'bypass' VSUP by applying ~5.2V or greater to the VCCout terminal on the TCAN4550.

    I don't know where you got this understanding, but it is incorrect and VSUP is required.  The VCCOUT terminal is the output of the internal 5V LDO that supplies the CAN FD transceiver and other sub-circuits.  There are other circuits in the device that operate directly off of the VSUP pin.  By treating the VCCOUT pin as a supply "Input" pin can damage the device.

    Leon Hermans said:
    1)  Can you also bypass the LDO by applying 5.2V or greater to the VCCfltr terminal on the TCAN 4551?  And that way the CAN bus power will be derived from that terminal rather than VSUP?

    No! You should not supply an external voltage to either the VCCOUT pin on the TCAN4550 or the VCCFLTR pin on the TCAN4551.  The TCAN4551 device still has the internal 5V LDO that requires a capacitor on the VCCFLTR pin for stability to supply the CAN FD transceiver, but it differs from the TCAN4550 devices in that it can't source additional current to an external circuit such as a MCU.

    Leon Hermans said:
    2)  Before Vio is applied, is this the proper figure to use for Isup? ~25-42uA?

    Yes, if VSUP is supplied, but VIO = 0V, then this would be the correct supply current.  The VIO supply sources the Clock circuit (OSC1 and OSC2) so the device will not have a valid clock to operate the digital core or CAN FD controllers until VIO is supplied and the clock is operational.

    Leon Hermans said:
    3)  Assuming Vio is applied, SPI active, and CAN bus with ~55 ohm loading, VSUP = 24V, VCCout/VCCfltr applied with external 5.2V, can you help us to estimate Isup?  I don't see a figure or way to calculate that.

    I can't provide any help with calculations that include an external 5.2V supplied to the VCCOUT/VCCFLTR pins.  The Datasheet provides maximum supply current specifications for both 60 ohm (typical) and 50 ohm (worst case) bus termination loads.  Because there is a difference of 10mA between the 60 and 50 ohm cases, the 55 ohm load would be in between these levels with a max level of 85mA.  During a Recessive bit, the driver is not active so the termination resistance difference is not a significant factor when determining the max current levels. 

    Regards,

    Jonathan

  • 0 in reply to Jonathan Nerger
    Prodigy 130 points

    Hi Jonathan 

    thanks for your reply. Perhaps I phrased it incorrectly.  I understand VSUP must be supplied with 5.9V or greater.  However, several messages on E2E led me to believe you can put an external 5.2V or greater on VCCOUT and thus reduce the current draw in VSUO significantly. Here is one example:

    e2e.ti.com/.../tcan4550-bypass-the-vsup

    could you let me know if, after considering it further, you are able to assist in estimating the ISUP current when an external 5.2V is applied to VCCOUT. 
    thank you

  • +1 in reply to Leon Hermans
    TI__Mastermind 29260 points

    Hi Leon,

    Thank you for pointing out the e2e thread you are referencing.  I had not seen that post, but I understand the premise of your question.  However I take exception to the content of this post and respectfully do not agree that this is possible, or recommended.  In my opinion, this post should be taken down to avoid future confusion.

    The TCAN455x devices were designed to be completely supplied from the VSUP pin with exception of the clock circuit (OSC1/2) which uses the VIO supply pin.  The device uses the 5V LDO to step down the VSUP voltage and then additional LDOs step the 5V down to additional internal voltage rails to operate the digital core and to be used for the nWKRQ pin etc.  This 5V LDO can also be used as a supply for other circuitry for through the VCCOUT pin in the TCAN4550 version of the device.

    This 5V LDO can't be disabled and the under-voltage threshold for the VSUP pin is set to ensure there is enough overhead voltage to create a regulated 5V from this LDO. 

    The device uses this internal 5V LDO as part of it's Power On Reset (POR) and mode control functions.

    The Device implements a Hardware or Software Reset by disabling this internal 5V LDO which in turn disables the other internal digital LDOs creating a Power On Reset event.  Forcing the device to sink current from an external 5.2V source could prevent the device from needed Reset Events.

    The Internal 5V LDO also is enabled/disabled for mode changes, particularly when entering and exiting Sleep Mode.  Forcing the device to sink current from an external 5.2V source could prevent the device from entering the low power Sleep Mode, and have other mode changes.

    The device uses the 5V LDO to create the 2.5V bias voltage for the CAN signals.  Bypassing this LDO with a 5.2V source will also have impacts to the CAN bus voltage levels.  Guaranteed current limits in a fault condition would also be bypassed and there could be large current flow through the external 5.2V supply.

    All of my comments are just from a theoretical consideration of your question and the possible problems with this request.  I do not have specific test data from trying to bypass the internal 5V LDO with and external supply, but these are some reasons why I think this is problematic.  I also do know know what type of possible leakage paths could exist by trying to back feed 5.2V when the device is trying to disable the internal 5V LDO and whether there could be any threat of damage to the device.

    I hope my concerns with this approach are clear.  But specifically to your question about to calculate or estimate the Isup current, I don't have any information to share. 

    Most of the device operates off of one of the internal LDOs.  Because the 5V LDO is the main core LDO, likely most of the device current could be estimated to flow through the 5V supply.  Some circuits related to the INH, WAKE, and nWKRQ pins operate directly off of the VSUP.  The CAN transceiver can be used as a wake-up source so it also has some circuitry that operates off of VSUP in Sleep Mode.  However, I don't know how to isolate the amount of current that would flow through VSUP instead of an external supply.  If you would still like to pursue this, I would suggest a direct measurement would be required to determine this.

    Regards,

    Jonathan

  • 0 in reply to Jonathan Nerger
    Prodigy 130 points

    Hi Jonathan, thanks so much for the detailed reply. We will abandon considering back feeding of an external 5.2V source into VCCOUT.  (FYI we have a 24V master supply and a single buck that steps this down to a middle voltage, which ideally is less than 5.5V due to the input limitations of several other ICs.  We have settled on making this mid voltage 6V and then using an LDO to further step down below 5.5v). Sadly , we need 1.8V VIO so need to use the TCAN 4551.  

    one final question: are you able to share if any upcoming devices on the roadmap for any integrated CAN controller+transceiver which will offer an integrated BUCK style supply regulator from VSUP or the ability to feed external 5V supply to the device?

    We are close to finalizing design for a product family using the TCAN4551 but not planning production for almost 1 year, so if there is a more optimal  product coming soon we want to consider that  

    Sincerely 

    Leon

  • +1 in reply to Leon Hermans
    TI__Mastermind 29260 points

    Hi Leon,

    We do not have a buck version of the TCAN455x device, but we do have a new device in development that might be a better option for you depending on your timeline.  It is very similar to the TCAN4551 but can operate with a VSUP range of 4.5V to 36V and also offers a VIO range of 1.8V to 5V.  The internal 5V LDO is removed and replaced with an additional VCC supply pin that supplies the CAN transceiver with 5V from an external source.   This allows both the VSUP and VCC to be supplied together from a single external 5V source.

    So, if I understand your preferred solution, you would be able to use 5V for your middle voltage, and then you would just need to supply the VIO pin with 1.8V.  But the large power dissipation from the internal 5V LDO in the TCAN4551 is removed, and it does not need to be replaced with a Buck.

    The registers would be similar but not 100% the same.  However, it would still have backwards compatibility with TCAN4550/1 software drivers that have been developed.

    What is your timeline on finalizing your design?  Would such a device fit your application?  If you do not want to post that in this forum let me know and we can switch to direct messages or email communication.

    Regards,

    Jonathan

  • 0 in reply to Jonathan Nerger
    Prodigy 130 points

    Hi Jonathan 

    this sounds like it could work. I’d like to share some schematics by email and discuss further.  Could you please send me your contact info?

    thanks

  • 0 in reply to Leon Hermans
    TI__Mastermind 29260 points

    Hi Leon,

    I'll send an email to the address linked in your myTI E2E profile.

    Regards,

    Jonathan