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.

SN65HVD230: SN65HVD230D: CANH CANL Wave is abnormal.

Part Number: SN65HVD230

Hi TI expert,

We are using SN65HVD230,  and here is some test results and we have some questions, please help to check.

Block:

Test 1: Vref connect to CANH, CANL via 3.3K resistance

CH1:CAN_H、CH2:CAN_L 、f1:CAN_H-CAN_L

dominant: CAN_H:2.8V CAN_L:1.0V

Recessive: CAN_H:1.8V CAN_L:1.8V

the question is:

1. By connecting the Vref  to CAN BUS via resistance, the Recessive level is pulled down to 1.8v, which is not consistent with the requirement of ISO11898 of 2.0-3.0v. Meanwhile, the datasheet describes typ2.3v and there's no  MIN and MAX value;

2. The MIN of dominant CANH in the manual is 2.45v, which is not matched with the 2.75v of ISO11898. The margin of 2.8v and 2.75v is very small.

3. From dominant to recessive, the waveform will have a slow process

4. However, ISO11898 requires vysm0.9 ~1.1, which is not satisfied here.

Test 2: Remove 3.3K resistance between Vref and CANH, CANL.

CH1:CAN_H、CH2:CAN_L 、f1:CAN_H-CAN_L

dominant: CAN_H:2.8V CAN_L:1.0V

Recessive: CAN_H:2.2V CAN_L:2.2V

Here are the questions:

1. Without connecting CAN BUS  to Vref via resistance, the Recessive level is 2.2v, which is consistent with the demand of ISO11898 from 2.0v to 3.0v

2. The MIN of dominant CANH is 2.45v, which is not matched with the 2.75v of ISO11898. The margin of 2.8v and 2.75v is very small

3. From dominant to recessive, there will be a slowing process in the waveform (different from test 1).

4. However, ISO11898 requires Vysm0.9 ~1.1, which is not satisfied here either.

5. Here, when CAN L changes from recessive to dominant, there is a return sulcus.

This problem has been bothering us for a long time and we hope to get your reply as soon as possible.Thank you very much.

Cyan

  • Cyan,

    Thanks for bringing your questions to E2E. While 3.3V CAN transceivers are compatible with transceivers compliant to ISO11898-2, 3.3V CAN transceivers are not compliant with ISO11898-2 themselves. One of the main reasons is for what you've pointed out, the devices can't meet the recessive voltage level while simultaneously meeting the symmetry requirement.

    Also, none of the images posted are showing on my side, is it possible to send them directly to my email? You can find it by clicking on my username.

    Regards,

  • Some images were not uploaded successfully.  I will send them again.

    "While 3.3V CAN transceivers are compatible with transceivers compliant to ISO11898-2, 3.3V CAN transceivers are not compliant with ISO11898-2 themselves."

    -----I still can't understand the meaning of this sentence, could you elaborate on it.

    Hi TI expert,

    We are using SN65HVD230,  and here is some test results and we have some questions, please help to check.

    Block:

    Test 1: Vref connect to CANH, CANL via 3.3K resistance

    CH1:CAN_H、CH2:CAN_L 、f1:CAN_H-CAN_L

    dominant: CAN_H:2.8V CAN_L:1.0V

    Recessive: CAN_H:1.8V CAN_L:1.8V

    the question is:

    1. By connecting the Vref  to CAN BUS via resistance, the Recessive level is pulled down to 1.8v, which is not consistent with the requirement of ISO11898 of 2.0-3.0v. Meanwhile, the datasheet describes typ2.3v and there's no  MIN and MAX value;

    2. The MIN of dominant CANH in the manual is 2.45v, which is not matched with the 2.75v of ISO11898. The margin of 2.8v and 2.75v is very small.

    3. From dominant to recessive, the waveform will have a slow process

    4. However, ISO11898 requires vysm0.9 ~1.1, which is not satisfied here.

    Test 2: Remove 3.3K resistance between Vref and CANH, CANL.

    CH1:CAN_H、CH2:CAN_L 、f1:CAN_H-CAN_L

    dominant: CAN_H:2.8V CAN_L:1.0V

    Recessive: CAN_H:2.2V CAN_L:2.2V

    Here are the questions:

    1. Without connecting CAN BUS  to Vref via resistance, the Recessive level is 2.2v, which is consistent with the demand of ISO11898 from 2.0v to 3.0v

    2. The MIN of dominant CANH is 2.45v, which is not matched with the 2.75v of ISO11898. The margin of 2.8v and 2.75v is very small

    3. From dominant to recessive, there will be a slowing process in the waveform (different from test 1).

    4. However, ISO11898 requires Vysm0.9 ~1.1, which is not satisfied here either.

    5. Here, when CAN L changes from recessive to dominant, there is a return sulcus.

    This problem has been bothering us for a long time and we hope to get your reply as soon as possible.Thank you very much.

    Cyan

  • Hi Eric,

    I have add the images above, could you please help to check it again?

    Thanks.

    Cyan

  • Hi Eric,

    I have added images above. please help to check it, thanks very much.

    Cyan

  • Cyan,

    Thank you for your patience, TI was on holiday on 04/10/20 hence the delayed response.

    When I saw the SN65HV230 transceiver is compatible with devices that meet the ISO11898-2 standard, but not compliant to the standard themselves, I mean that the 3.3V CAN transceivers can never fully comply with the ISO11898-2 standard. They are compatible because they still can work on a CAN bus with 5V CAN transceivers which are compliant with the ISO11898-2 specification.

    3.3V CAN transceivers cannot comply with the ISO11898-2 because of how they are designed and the limitation due to using 3.3V as the main supply. The ISO11898-2 standard is written for 5V CAN transceivers in an automotive environment, and thus has specific voltage level requirements for the dominant and recessive levels with a 5V supply. Because 3.3V CAN transceivers are using a 3.3V supply, and they are designed to be interoperable with 5V transceivers, the recessive voltage requirement, and thus the Vsymmetry requirement from the ISO11898-2 cannot be met.

    The "charging" waveform that is seen at each recessive level is caused by the fact that the recessive level is set to higher than Vcc/2 to meet the ISO11898-2 requirement of the recessive level being between 2 and 3 volts. When a dominant bit is driven by the 3.3V transceiver, the common-mode level drops by a few hundred millivolts, and once the device goes back to a recessive state, the voltage will charge back up to the recessive-level bias. More information on this is available in the datasheet in section 11.3.1.2.1. 

    All of what you are seeing is expected from all 3.3V CAN transceivers, not just 3.3V CAN transceivers from Texas Instruments. These devices weren't meant to meet the ISO11898-2 standard because that standard is written for 5V CAN transceivers. There is also a bit more information in this FAQ.

    Please let me know if you have any other questions.

    Regards,

  • Thanks very much, Eric.

    That's very clear.