Interfacing CAN transceiver with MSP432 series.

Thanks for your quick response.

What about Signal and Power Isolation in case of Transceiver in  TCAN4550-Q1 .

Hello,

You could use digital isolators like ISO7741 and ISO7761 in order to isolate the digital interfaces between the TCAN4550 and your MCU.  For generating an isolated power rail, a device like SN6505 could be useful.

Regards,
Max

Any reference design?

could be useful to understand.

Thank you.

I'm thinking of using MCP2515 with ADM3055E Since it provides both Signal and Power isolation.

Any opinion on this?

Thanks.

Hi Max,

I was going through this reference design:

Automotive RFCMOS 77GHz radar module reference design with object data output over dual CAN FD

I have few question on the design:

* TCAN4550RGYRQ1 used in this design doesn't have any Input Isolation nor Power isolation. can you justify me why Isolation is not been considered here?

* How good it is to use Integrated Isolation to the External?

Thanks.

Hello,

Many CAN interfaces operate without additional galvanic isolation, and it was not considered a system-level need for this particular reference design.  Isolation is typically most beneficial when communication needs to occur across power/ground domains or when high enough currents may flow through a shared ground connection that voltage offsets between nodes exceed what the communication interface is intended to support (i.e., the rated "common mode" range of the interface).

If isolation is needed, though, it could be added via use of the digital isolator devices I mentioned.  These would introduce galvanic isolation on the digital lines such as the SPI lines between the TCAN4550 and the MCU.

Regards,
Max

Thank you.

I was expecting your opinion on this: "

I'm thinking of using MCP2515 with ADM3055E Since it provides both Signal and Power isolation.  ".

That's one option to consider.  If you wanted to do something similar using TCAN4550 you could make use of the test mode that allows the general purpose IOs to be repurposed to represent the internal TXD/RXD signals (allowing it to interface with an external CAN transceiver).  You could then use an isolated CAN transceiver like ISO1042 and derive a separate power supply or make use an a digital isolator with integrated power generation like ISOW7821 and pair it with a non-isolated transceiver like TCAN1042.  Here is an application note describing both options in further detail:

http://www.ti.com/lit/an/slla386/slla386.pdf

I believe the ISO1042 option with SN6505 for isolated power rail generation would give the overall smallest solution size.

May I ask why you are using isolation in this application?

Regards,
Max

Okay Thank you.

** But I've one question in all of the data sheet, When they say MCU Tx and Rx to transceiver in the block diagram, It's CAN controller Rx and Tx to the Transceiver right? Not the normal MCU UART right?

* In my case I'll also be using RK3399 Processor which SPI operates at 1.8V in that case I need level shifter to interface with TCAN4550 right?

* Since these will be used in Industrial and Automotive application, Isolation is always better option when in case of RS232/485/CAN.

* I believe TCAN4550 doesn't have built i Isolation.

Thanks and Regards.

Yes, the TXD and RXD lines connecting to a transceiver would be coming from a CAN controller (whether that is one integrated in an MCU or a discrete one).  The CAN protocol requires additional controller functions (bitwise arbitration, bit stuffing, error monitoring, etc.) that wouldn't be supported via a standard UART port.

Yes, you would need a level shifter to interface TCAN4550-Q1 to a 1.8-V processor's SPI port.  However, the upcoming TCAN4551-Q1 device would be able to support VIO = 1.8 V without the level shifter.  That may be something to consider.

I understand that isolation is beneficial in many cases, but there are also many industrial and automotive interfaces (using CAN, RS-485, etc.) that operate without it.  You may want to consider whether this is a real application need or merely something nice to have, since a solution without isolation would be substantially simpler.  (You are correct that there is no isolation implemented internally to TCAN4550-Q1).

Regards,
Max

Thanks for the clarification.

When can we expect the marketing release of TCAN4551-Q1.

I agree with you that solution without a Isolation would be a simpler but to be on the safer side I feel it's better to have.

Will TCAN4551-Q1 have internal isolation? ( A quick confirmation, CAN FD is compatible with CAN 2.0(a/b) right? )

If I'm using discrete solution meaning(considering Isolation is must), Controller->CAN Controller-> CAN Transceiver would you recommend use of Isolator at each stage? and what in case of Integrated CAN controller Transceiver?

Thanks.

Hello,

Do you mind if I contact you via email?  I'm not able to disclose product release schedules via this public forum.

TCAN4550-Q1 will not have internal isolation, though.  For the discrete solution you mentioned, I don't see a benefit to introducing isolation at each stage if you are just trying to achieve robust communication (i.e., there are no specified safety-related standards you are targeting).  A single isolation barrier should be sufficient.

Max

Sure. I've sent you a friend request.

Haven't heard anything after that.

Waiting for your reply.

Sorry for the delay - I've just sent you a message.  Please let me know if you did not receive it.

Max

Thanks.

Can you be bit specific and detail on this statement of data sheet:   "  When using a VIO of 1.8 V an external clock of the same voltage should be used and not a crystal.".

Thanks and regards,

Lakshminarayana.

Okay.

Thank you.