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.

SN65HVD1050 CAN Transceiver Start up problem at low Temperatures

Other Parts Discussed in Thread: SN65HVD1050, SN65HVD251, SN65HVDA1050A-Q1, SN65HVDA542-5-Q1, SN65HVDA1040A-Q1

For a better EMC peformance we wanted to change to the CAN Transceiver SN65HVD1050. But a start up test at about -25C showed that the CAN Transceiver works sometimes with a delay of up to 800ms. That means during this time our product is not able to receive or transmit any CAN message. If we change to the CAN Tranceiver SN65HVD251 the problem does not exsist. The S port for silent mode is always low. We power the CAN transceiver with 5V and a fast settling time. At room temperature the CAN Tranceiver works as expected. Is this a known problem or is there a workaround for this issue?

  • Hi Peter,

    Please try the SN65HVDA1050A-Q1 device.  It was optimzied for re-start up at cold temperatures.  There isn't a known initial start up issue on the SN65HVD1050 but your specific power supply on Vcc may have some Vcc ramp that may be similiar to the re-start up.  If you have a scope shot of the ramp on Vcc during cold temp start up plus the TXD (even just a square wave in) and then either CANH/L diff pair or RXD on the same screen to verify a few things we could look at it.  But short term a test with the SN65HVDA1050A-Q1 or even the SN65HVDA542-5-Q1 would answer several questions.

    - Scott

  • Hi Scott,

    thanks for your answer. I will try the automotive type of the SN65HVD1050. But for me it is still not clear why the standard type of the SN65HVD1050 did not work. 

    I think the scope shot makes a little more clear my problem.

    SN65HVD1050 normal start up.

    SN65HVD1050 faulty start up at low temperature of -25C

    This effect does not happen always, but  about 1 out of 20 start ups did not work. The SN65HVD251 works always.

    Thanks in advance.

    Peter

  • Has there been any update on this issue Peter described?  I believe we are seeing the same thing, but at room temperature.  I'm not 100% sure, but I believe the issue only happens at startup.  I was able to get this issue to happen 20 times in a row (power cycled our board each time), but left the CAN transceiver attached to a powered CAN network (2.5V).  

  • Hi JR,

    Does switching to the newer HVDA1050A version of the device fix the issue? The "A" version fixes a bug related to VCC brownout conditions, but as Scott mentioned the power ramp at cold may be similar enough to trigger the same behavior. Testing with the newer device would at least let us know whether this is a factor or if there is something else going on.

    Best regards,
    Max
  • Max:

    We were seeing this issue when the vehicle starts, so very possibly a VCC brownout issue.  We haven't tested the 1050A.  If we build any more hardware with the 1050x, we will use the 1050A.  We are switching to the 1040A for bus wake up feature.

    Can you confirm that the 1040A doesn't have the brownout issue?

    For our current units in the field, is there anything we can do in software to help the 1050 become operational?

  • Hi JR,

    Max is actually traveling in Japan this week so to help him out i am going to jump in here. The SN65HVDA1040A-Q1 and the SN65HVDA1051A-Q1 have both a fix for the brown out issue and improved ESD performance for the automotive industry. When using these devices in the future I recommend using these A version to avoid any possible hardware issues.

    As far as your existing deployed hardware, since these are physical layer devices there is really no fix you can offer in software to work around this issue brown out issue. The only fix for the original (non A) devices is to make sure that the brown out condition never occurs. Please let me know if you have any more questions?
  • This is urgent as we are in the middle of building up thousands of boards, and we've halted production.

    Thanks for the info.  As for the software fix, we saw one device that wasn't working, but as we started to transmit, it started operating correcly after 200 microseconds, but that was only once we saw that issue go away.  Does the chip normally stay in the non-operational state until a power cycle?

    Also, I just noticed there is more than one "A" version.  There is the "1050", the "1050A", and the "A1050A".  So does the "1050A" have the fix?  Or is it just the "A1050A"?  Also, above you mention the "A1051A", so I want to make sure that the "A1050A" and "A1051A" both have the fix, since from the above it looks like only the "A1051A" might have it.

  • Hi JR,

    Typically when there is an issue like this the only fix is a power cycle which gets the part back into its normal operating condition. I have not explored this problem on the SN65HVD1050 extensively so I cannot say confidently that waiting some period of time (i.e - 200ms) after a brownout will alleviate the glitch.

    Sorry about the miscommunication above. There is no SN65HVDA1051A, this was a typo. The SN65HVD1050 has the brown out issue. The SN65HVD1050A and SN65HVDA1050A both implement the fix. If you notice in the datasheets for the 1050A and A1050A there is a difference in ESD performance between the two. The ESD protection on the SN65HVDA1050A was enhanced and tested for IEC ESD. This is specified in the datasheet for the SN65HVDA1050A in the ESD section. Again though both the SN65HVD1050A and SN65HVDA1050A implement the fix for the brownout issue. Please let me know if you have more questions.
  • Hi JR,

    Sorry for the multiple messages on this topic but I wanted to make sure you had the right information. I just confirmed with my design team that a PCN (product change notification) was issued several years ago regarding the SN65HVD1050A. The SN65HVDA1050A has superior performance to the SN65HVD1050A so the PCN went out to all of our customers buying the SN65HVD1050A stating that we would no longer be making the SN65HVD1050A and the silicon would be moving to the SN65HVDA1050A permanently. So long story short is the SN65HVD1050A and SN65HVDA1050A are the same device. Please let me know if you have any more questions?