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.

CAN Transceiver SN65HVD234

Other Parts Discussed in Thread: SN65HVD255, SN65HVD234

Dear sirs

In our project we have planned to use a SN65HVD234D CAN Transceiver. In the features list in the datasheet the following feature is stated: "Unpowered Node Does Not Disturb the Bus". Am I right that we can switch off the power of SN65HVD234D if it is not used and the bus is not affected for the other participants. Could you please confirm that. Best regards Thomas Lochmatter

  • The HVD234 has low unpowered bus leakage on the order of 90uA at the normal CAN bus voltage levels.  Thus inserting a single of a small number of unpowered nodes will not generally add much loading to the operating network and thus allow for normal operation for those nodes still powered.  Additionally the device has glitch free power up/down behaviour that will not put a glitch on the bus that would corrupt data as it is powered and unpowered.  These features were design for the option to hot plug a module in and out of a CAN network.  If you have many nodes that will be unpowered then the leakage would start to add up as the HVD234 wasn't designed with a large number of unpowered nodes on the bus, but just one or two as they hot swap.   If you need large numbers of un-powered nodes on the bus you may want to look at the SN65HVD255/6/7 CAN transceivers that took the unpowered bus leakage to below 5.5uA for this different type of unpowered application. 

    -- Scott

  • Hello Scot,

    we actually use the SN65HVD234 in a huge amount of circuits and our manufacturer told us that some devices was damaged (burned with broken plastic housing) while do the initial test in circuit. This happens for some devces - the most work without problems. I can not figure out if it's a design mistake or a fabrication charge problem. The device is powered with 3.3 V stable and there is only an interconnection between 2 CAN devices directly with short way (some centimeters).

    The enable pin is directly set to 3.3 V and Vcc is blocked with 100 nF capacitor, RS pin is grounded with 10 k ohms.

    Can you give me a direct application engineer contact to discuss this problem?

    Thank you very much.

    Michael.

  • If you are seeing burned or broken plastic housing you are submitting the devices to conditions leading to massive EOS (Electrical Over Stress).  Normally you would pull EN pin the 3.3V via a pull up R in the 4.7 to 10k ohm range, not directly to the supply.  To you have TVS (transient) protection devices on the bus pins of the HVD234?  Putting oscilloscope probes on the bus pins and IO pins during these production tests and looking for un-expected transients is what I would do first to see if there are conditions outside the absolute maximums of the device that are not expected but occurring during the testing and manufacturing.

    -- Scott

  • Hello Scott,

    if I see datasheet, the EN input has a resistor devider internally (page 12); next there is no special limit for this pin (page 14) - only H level. And finally the graph of some characteristics (see figure 17, 18 a.s.o) shows EN=VCC and VCC=3.3V. There is no description about using a additional resistor. If you see the test circuit (figure 11) there the pin EN is connectied to VCC without any resistor.

    So the qiestion is: Does it resolve the problem of sometimes burned HVD234 driver if I add this resistor? There is a massive redesign procedure and if this is not the solution I would have a problem.

    Michael.

  • There are limits on the EN pin in the Absolute Maximum Ratings table (-0.5V to +7V).  Figure 11 is for a automated protection test, not a general recommendation for board and system level design.  A series resistor with the EN pin would help provide protection to the pin and limit the current into it if the voltage ratings were violated and the ESD protection cells would conduct. 

    Have you sent any failed units back via TI quality channels and had FA done to help isolate what part of the device failed?  Have you put an oscilloscope on the pins of your system and looked for un-expected transients? 

    You may want to take a look at our CAN EVM, nominally populated with the SN65HVD255 5V CAN XCVR but the EVM supports 3.3V CAN XCVRs as well.  Information may be found at:  http://www.ti.com/tool/sn65hvd255devm  . 

    This EVM has pads on the PCB for series protection of pins, TVS protection of the bus pins, and other filtering and protection concepts that may be useful for your redesign.  The EVM is populated with 0 ohm series resistors and no external protection devices so you may evaluate a device directly, but if you want to try out protection and filtering concepts you add them as described in sections 2.2 and 2.3.

    -- Scott

  • Hello Scott,

    "Have you sent any failed units back via TI quality channels and had FA done to help isolate what part of the device failed"

     

    --> How can I do this? I tried to ask a FAE but your support line told me to ask farnell or Mouser for help questions. How can I sent via TI quality channels? A link?

     

    Thank you.

  • If you are purchasing via distribution the devices will have to be returned to TI via the specified return mechanisms set in place with the distributors.  I'm not sure what those are.  They then feed back into TI. 

    -- Scott

  • Is there a direct way to TI quality?

    Michael.

  • I don't know of one for devices purchased in distribution, I will check.

    -- Scott

  • I have confirmed you will need to contact the distributor to send the parts for FA. The distributors want to handle it this way. They should be able to provide the FA request form and details as needed.

    -- Scott