• State Resolved
  • Replies 2 replies
  • Answers 1 answer
  • Subscribers 34 subscribers
  • Views 53 views
  • Users 0 members are here
Support feedback
Options
Options
Related

Original question:

TCAN1145-Q1: why do I still need standby mode since we already have sleep mode

TCAN334G: CAN tranciever standby mode pulling bus low

Anthony Desmarais
Prodigy 10 points

Part Number: TCAN334G
Other Parts Discussed in Thread: TCAN334, TCAN3404

Hi, we have developed an automotive device that uses the TCAN334 or the TCAN3404 (alternate device).
This device when not in use needs to be in a low power mode and in this case we pull the STB line of the chip high.

On my test bench I have two devices setup to communicate together on the CAN bus, both using this same tranciever.

On testing on our bench I see that when one device is put to sleep, the common mode voltage level gets pulled to ground despite the other device still being powered on and actively transmitting on the can BUS. 
The differential voltage is still good enough for the logic levels to be correct so communication is still possible but the waveform on a scope looks very odd (with CAN H and CAN L always low outside of data transmissions).

My question is if this is intentional behavior and in a real world environment (on a vehicle CAN bus) woudl we see this or should more devices on the CAN bus bring the common mode voltage closer to 1.9V.

Thanks in advanced.
Anthony

  • 0
    TI__Guru* 76022 points

    Hi Anthony,

    Yes, that should be the case I.e., when the device enters standby (STB high), the output driver stage should be disabled and the bus pins should go to a high-impedance state. Hence, the transceiver no longer actively biases CANH / CANL to the nominal recessive voltage (~1.9 V). On a 2-node bench with only standard termination (two 120 Ω resistors), the recessive-state common mode should now be determined by the one remaining active transceiver's internal bias and the resistive termination network. 

    I.e., With only a single active driver on a lightly-loaded bus, the internal biasing of one transceiver is often insufficient to hold both lines at 1.9 V. This should result to both lines sitting near ground in the recessive state, similar to your observations. 

    Communication can still work due to the differential voltage (CANH-CANL), rather than the absolute levels. I.e., ISO defines dominant as ≥ 0.9 V differential and recessive as  ≤ 0.5 V differential, with the common mode spec being -2V to +7V. Therefore, as long as your differential is within spec, the logic levels should be correct regardless of where the common mode sits. Hence, your scope may be  showing you something that looks wrong visually but may still be electrically valid and a real vehicle CAN bus should look much better because a typical vehicle CAN bus includes 10-30+ nodes where multiple nodes may remain awake during the standby mode and each active transceiver further contributes to the recessive-state bus biasing, pulling the common mode back toward ~1.9 V. Furthermore, many automotive designs use split termination; center-tapped 60Ω+60Ω arrangement with the tap connected to the 1.9 V bias which should actively hold the common mode near the 1.9 V, thanks. 

    Best Regards,

    Michael. 

  • 0 in reply to Michael Ikwuyum
    Prodigy 10 points

    Thanks for the detailed explanation Michael, I suspected as much but it is always good to get the answer from the experts.