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.
Tool/software:
Hello,
We try to do an CAN interface that should be bidirectionnal (but with isolation). We use for that an ISO1042 and the TCAN1057 IC.
In one direction, we generate a signal on TCAN1057 CANH/CANL. We have the communication but with S pin of TCAN1057 open. It works fine.
In the other direction, when we generate a signal on CANH,CANL on ISO1042, it doesn't work with this setup and we need to pull down the S pin of TCAN1057.
It's not really fit with the datasheet or we miss something? Do we need to use a pull down on S pin and not connected it to ground directly?
I attached the schematic.
Thank you for your support
Note: r603, r604=22ohms (small filter)
The S pin has an integrated pull-up. If you leave this pin high, the TCAN1057A-Q1 will not transmit anything. You can connect it directly to ground.
Please note that simply connecting two transceivers together does not work because the circuit will never go out of the dominant state.
Dear Clemens,
Many thanks for you quick answer. We found this link Isolating an off-the-shelf CAN device with back to back CAN Transceivers - Interface forum - Interface - TI E2E support forums
that answer to our issue.
Good thanksgiving.
Hello,
Is this schematic will work?
Schematic seen in Isolating an off-the-shelf CAN device with back to back CAN Transceivers - Interface forum - Interface - TI E2E support forums
Do we need to have isolated power_supplies for any iso1042? In our board first iso1042 is connected to +5v_ni and +5v_i.
We could power supply second ic to +5v_I?
Best regards,
A second idea will be to use two monostable functions to drive s pins. with a tcan1057 - iso78xx - tcan1057 architecture.
It could work?
Hello Christophe,
Please allow us until Monday to get back to you on this request, as we have a holiday today.
Best,
Michael
Hello Christophe,
Thank you for your patience. From my understanding both of the above schematics will need additional logic in-between the transceivers to ensure that the CAN bus does not get stuck in the dominant state as Clemens said earlier. If one side of the CAN bus saw a dominant and the RX pin would go low and the second transceiver would stay low. Thus, creating a bus lockout condition.
Do we need to have isolated power_supplies for any iso1042? In our board first iso1042 is connected to +5v_ni and +5v_i.
We could power supply second ic to +5v_I?
The answer here will depend on how important isolation is to your system. In general, you only need one supply and power the logic side, in-between logic and second transceiver off of the isolated supply.
Best,
Andrew
Hello Andrew, Clemens,
I find the ti evaluation board: TIDA-01487 but I don't find a web to buy it (we are in a hurry and we want to confirm at least one solution): https://www.ti.com/lit/df/tidrsa3a/tidrsa3a.pdf?ts=1733312571493
I think that it is really what we need. Do you have a link to buy this evaluation board?
thank you for your support.
Hello Christophe,
The schematic is from the reference design TIDA-01487 reference design @ TI.com. Unfortunately, there is no physical board or orderable part for this design since reference designs are for reference only. The link does have all the design files needed to send the board to a manufacturer to build it independently. The other option would be to order samples of each EVM separately and validate each IC on its own EVM or connect them together.
The "additional logic in-between the transceivers to ensure that the CAN bus does not get stuck in the dominant state" mentioned previously is shown in 2.4.3 CAN Bus Arbitration Logic and tested in 3.2.1 CAN Communication
Best,
Andrew