• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Answers 2 answers
  • Subscribers 32 subscribers
  • Views 161 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.

SN65HVD233: Some questions in CAN bus

Nestor Groel
Prodigy 10 points
Part Number: SN65HVD233
Other Parts Discussed in Thread: ESD2CAN24-Q1, , ESD2CAN24

Good morning, excuse me for asking these questions but it is the first time that I am going to use CAN in a design and I want to eliminate some doubts:

1) With the SN65HVD233D can I use CAN in 24 volts? I see that the data sheet indicates +-36 volt protection but I don't know if it is enough to say that it is tolerant to that voltage on the bus.

2) If the CAN line to be installed already has terminators at both ends (with split to the GND of the bus), is it necessary to include it in the boards ? I think it is not necessary but I would like to confirm it because I don't know which PCB would be the first and last and if the bus would be extended in the future.

3) The SN65HVD233D will be powered by a source completely isolated from the bus voltage, can I use the GND of the bus (isolated from the GND of the chip) to connect the PES1DCAN diodes? I think it would still fulfill its function as an ESD protector for the CAN lines, but I don't know if it would affect the SN65HVD233D in any way.

4) In some designs a 1Kohm resistor is used between the MCU and D and T of SN65HVD233D, in others instead pull up resistors are put on both lines, in other cases both are done (the datasheet even adds a C to GND for line D) and in other cases nothing is added. What is the Texas recommendation?

Sorry for my bad english and newbie questions. Thank you

  • 0
    TI__Genius 15345 points

    Nestor,

    No worries on the questions that is what we are here for Slight smile!

    1. The +/- 36V Bus fault protection is the maximum voltage that can be applied to the bus. Now the device can survive a continuous short to 36V but it will degrade the overall life of this device. But as long as you operate within the absolute maximum conditions of the device you won't do any permanent damage to it. Now it would be up to you as a system designer to run some tests and to see what is the maximum transient that your battery could apply to bus if there was an accidental short. If you find that it would be greater than 36V we have external devices like ESD diodes (ESD2CAN24-Q1) or CAN devices (TCAN33xH-Q1) with higher bus fault tolerances that can survive larger shorts.

    2. If there is already termination on the cable for the CAN bus you should not install additional termination resistors. Now the most optimal configuration for communication is to have the termination resistors as close to the CANH/L pins as possible. But that is just a recommendation and not a requirement. What is a requirement is that each CAN device only drives a 60 ohm load. All CAN devices are designed to drive a 60 ohm load and if your load is too small (by adding additional termination) then the CAN device will not be able to drive a proper differential voltage and you will not be able to communicate. That is why you should only have two 120 ohm resistors (add in parallel to 60) one at each end of the CAN bus. This is just meant to help with reflections between the two farthest CAN devices communicating with each other on the bus.

    3. First I recommend looking at the ESD2CAN24-Q1 diodes. They offer improvements over the competitor protection diodes like lower capacitance and higher ESD protection. Now, how are you going to power the device separately from the bus voltage? The CAN bus is powered off of the VCC pin of the SN65HVD233. If you have a schematic I would be happy to review it. The ESD diodes should be placed close and along the signal path so connecting them to the bus ground is fine. But I am unsure how you are separating the ground of the chip and the bus. Internally they will be connected.

    4. What is the T pin on the device you are referring to? I assume you meant the R pin. I usually recommend adding series resistors close to the transmitter of the D and R pins. For the D pin that is the MCU and for the R pin that is the CAN device. The reason for these series resistors is to help with reflections when your traces become very long on the PCB. I have not seen people use 1K series resistors for this purpose. They are most likely using those resistors to disconnect the D/R pin from the MCU and not for impedance matching.

    Pull up resistors can be used to bias the pins to a default state at power up. 

    Either way, both of these resistors are not required and it will be up to you as the system designer. You could always include them in the schematic and if you find out during testing that you don't need them you could remove them or populate with 0 ohm resistors.

    Your English was not bad at all. I didn't even notice. Let me know if you have any more questions.

    Best,

    Chris

  • 0 in reply to Chris Ayoub
    Prodigy 10 points

    THANKS CHRIS for your fast answer!!!!

    My issue was resolved but I wait to hit the button to see if you have any comment to my answers, THANKS

    1) I'm planning this device as part of a network of "pseudo PLC" in automation of modern greenhouses. In fact I plan to use and recommends 12 volt in the bus, but 24 volt it's more common in PLC and electro valves environments, then I'm designing the devices to support up to 24 volt as a "normal max". In both cases would be uses power sources with tight voltage control, MAX 10% in any case. If the client expose the board, designed to use in 12 volt, to more than 30 volt would must to buy some new units :-).

    2) Perfect!

    3) ESD2CAN24-Q1 OK! I have a few of pesdcan that I had plan to use in the prototypes, but actually I need to buy for production and then buy ESD2CAN24 now and use it in prototypes too. THANKS

    About the power source isolation. I'm now in the design of the schematics, I had plan to use a isolated DC-DC to generate the 3.3 volt for the MCU plus conditioning circuits for improve the robustness of the unit a little. All the inputs and outputs are isolated (optically in general). I had plan to use the SN65HVD233 connected to 3.3 volt of micro controller and abuse of the differential nature of CAN to float the signal "without" reference to the bus GND. Because the good ESD ratings of the transceiver and ESD protection added I had think to take the risk and cut the cost not using a Isolated transceiver or isolate it from the MCU.

    I thinking now that floating CAN_L and CAN_H can generate a spike far away of the specs, mmmm.

    But the fact that the isolated DC-DC it's by far more expensive than a simple switching step down and your opportune question, obligates to me to think in not isolate the power source of the Vbus and add instead some ESD and filter to better robustness of the device.

    What do you think about that? In my country (with the costs really tighten by the economical situation) and the greenhouse business that it's not very prone to pay for "new"technology, a lower cost improve the possibility of success of the project.

    4) T pin it's my mistake, yes, it's the R pin. In this PCB the MCU distance from the transceiver will be less than one inch, I think then that series resistors are not necessary but maybe the pull up and down. As you suggest I add this to the schematics and not populate if they are not necessary. I not know the behavior of the MCU pins when it's on reset or power on process, but with first prototype I can evaluate the behavior and not disturbance of the bus. Nevertheless, two pull resistances are really near zero cost and perhaps it's better to add it.

    I really appreciate your answers, I used to work with RS485 but all the benefits of CAN solution tempted me to try it in this project. Thanks again

    Néstor

  • 0 in reply to Nestor Groel
    TI__Genius 15345 points

    Néstor,

    Nestor Groel said:
    But the fact that the isolated DC-DC it's by far more expensive than a simple switching step down and your opportune question, obligates to me to think in not isolate the power source of the Vbus and add instead some ESD and filter to better robustness of the device.

    I agree with this logic. Isolating the power supply of the device is fine. I think your concern here is that when another CAN device is communicating on the bus it may have a different ground potential than the device with the isolated DC/DC supply. So will see a sort of "ground shift" between the two devices. We design our devices to be able to handle this exact case. Our devices come with  a common mode range, this is exact same concept as common mode range on RS-485 devices. The SN65HVD233 has a common mode range of -7-12V. So as long as the ground shift is within that range the device can still communicate effectively. 

    Now the ideal set up here would be to use an isolated CAN device so that you can tie all the bus grounds together. This would offer the most robust communication. But we don't have any isolated 3.3V CAN devices. All of our ISO CAN devices are 5V. So if you wanted to use the isolated ground you would have to find a way to reference the grounds to each other and make sure that their ground shift is not larger than the common mode range.

    I think the most robust and easiest solution would be to add the extra protection and filtering to the 3.3V signal so that you don't have to isolate it from the rest of the system. Having all your CAN devices referenced to the same ground will be much easier and will offer much more reliable communication. 

    I believe that was your only question but if I missed one let me know.

    Best,

    Chris

  • 0 in reply to Chris Ayoub
    Prodigy 10 points

    Thanks Chris, all perfect, really a LUXURY the quality and speed of answers in this support forum. THANKS AGAIN