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.

SN65HVD75: electromagnetic susceptibility

Part Number: SN65HVD75
Other Parts Discussed in Thread: THVD1450

I have dug through the datasheets looking for some kind of information on whether this transceiver can handle sitting next to a big neodymium magnet. We  are operating next to one and occasionally (depends on position of the magnet) it appears to stop working. we have ruled out a few other possibilities and this chip appears to be the culprit. Any hint on if this chip can take magnetic abuse or not?

Thanks beforehand,

  • Hi Joshua,

    We haven't done any testing on this to my knowledge, but I can't think of any reason this device would be susceptible to magnetic fields.  The internal metallization uses copper (including the bond wires), so there shouldn't be mechanical issues due to magnetic forces.  In terms of electrical effects, there aren't any internal components that are sensitive to magnetic fields and the die dimensions are so small that there shouldn't be substantial loop areas for inducing currents.

    What behavior are you seeing?  Do certain magnet positions correspond to a loss of communication?  If so, can you tell if it is related to the transmitter or receiver function?  (Probing the various pins of the device to observe on an oscilloscope would be very helpful here.)  Do the issues resolve after the magnetic field has been changed (i.e., by varying the magnet position), or do they seem to be permanent?

    Regards,
    Max

  • Max, 

    This is a mag lev application and we have a very large neodymium magnet passing very close to the transceiver. At a certain position the transceiver stops transmitting and we get a a dc voltage about half the supply (very clean). This is a transmit only application. we have routed out to O-scope the signal, our microcontroller stays awake the whole time but the signal out will stop at certain positions then resume once the magnet's position has changed. 

    Thank you,

  • Joshua,

    Interesting.  So the VCC/2 DC voltage you are seeing is present on both the A and B signal lines simultaneously?  Do you see any other strange behavior at the time (like increased ICC, if possible to measure)?  Can you please let us know what components are present on the bus lines - for instance, termination resistances, failsafe biasing resistance (pull-ups/pull-downs), ESD protection components, etc.?  Is any noise coupling onto the DE pin as far as you can tell?  Does the issue occur with static magnetic fields (i.e., magnet held in a position), or is it more related to the movement of the magnet creating a dynamic field?

    Regards,
    Max

  • Max,

    yes it in on both signal lines. No other strange behavior with this instance, we had another odd thing going on with the power supply a couple of weeks ago but that problem wasn't visible while this was going on. here is the cutout of the relevant parts of the schematic

    R17 is currently not populated and RX,RE,TE,TX are going directly into a micro controller.

    The issue seems to happen with static fields. What do you mean about the DE pin (the RE or TE pin)?

    We are currently running more tests to try to figure this out. Let me know if you want us to test anything in particular

  • Thanks for the additional info.  Regarding the schematic - is there a decoupling capacitance on VCC that isn't shown here?  Typically we would recommend 100 nF to ground on this pin.

    I was asking about pin 3 before (it's called TE on your schematic but DE in the product datasheet).  I was wondering about that since you mentioned you see a VCC/2 output.  You could see this on the A/B lines if the driver were disabled (pin 3 at logic low state) and there were no other external biases applied to the bus, since the device will weakly bias these I/Os to about that level.  The only other way I could see this part giving a VCC/2 output on both lines would be if somehow both the high-side and low-side drivers on the A/B lines activated simultaneously.  If this were the case, though, you'd see large ICC currents and the device would heat up. (Have you seen anything like that?)

    If you could provide some waveform captures showing the issue it might help spur more ideas.  (It can be tough to describe the analog behavior of different signal waveforms fully just via text.)  Ideally we would be able to see a case where pin 3 is at a logic high but the A/B output pin states do not properly correspond to the TX pin input so that we could isolate the issue just to the behavior of the IC.

    I'm also curious - does power cycling clear the issue?  Is it observed consistently across units/systems, or does it seem to affect only a percentage?

    Max

  • Max,

    Yes we have a 0.1uF decoupling on the power line located right next to it. 

    I actually have great news. the engineer that designed this board just found the likely culprit. He put on the wrong chip, on the board he needed the 5v rated one not the 3.3V. The magnets were probably causing power fluctuations to exceed absolute max ratings of the chip. 

    thank you for looking into this for me and I will let you know if the new chip shows the same problems (which I doubt)

  • Thanks for the update.  Hopefully that is the root cause.  Even with the proper chip if there are strong power fluctuations that cause the VCC voltage to dip, the transceiver may shut itself off temporarily (there is typically an under-voltage lock-out implemented to prevent errant operation when not fully powered).  So, that may be something to check next if issues continue.  The decoupling capacitance will help with this to some degree, but may not be sufficient.

    By the way, in case you are interested in a similar part that is rated to work from VCC = 3 V to 5.5 V (so you don't need to keep track of multiple transceivers) we have recently released THVD1450.  It should give better performance in high-noise environments as well.

    Regards,
    Max

  • Hi Joshua,

    I just wanted to check back in - did swapping the chips resolve the issue?

    Regards,
    Max

  • It did, thank you for checking up. We have not been able to recreate the problem with the appropriate chip

  • That's good to hear!  Thanks for the update.  If any further issues pop up in the future please don't hesitate to make a follow-up post.

    Max