• State Resolved
  • Locked Locked
  • Replies 6 replies
  • Answers 1 answer
  • Subscribers 32 subscribers
  • Views 773 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

Isolated CAN BLDC Driver

John Easton
Prodigy 200 points

Hello,

Hopefully somebody can help me with Bus Powered CAN Nodes/Isolation. I want to control 3 no. 24V BLDC Motors each with their own PSU via CAN.

I am intending to run a separate Logic 24V supply along the CAN bus to power each CAN node (via regulator on each node). Bus length will be 8-10 metres long.

The microcontrollers on each node will be isolated from the motor drivers they are controlling.

My question is, will I need to use isolated CAN transceivers & DC/DC converters?

Thank you in advance for any adviceBus Powered can Nodes Model (1).pdf

  • 0
    TI__Guru 62910 points
    Hi John,

    Since all the CAN transceivers share a common power/ground connection and are located close to one another, I don't see any reason that isolation would be required. Isolation is typically used if there is a possibility of large potential differences between grounds of transceivers on a network (since these could lead to common-mode voltage offsets that exceed CAN's standard -12 V to +12 V range) or if a galvanic barrier is required for safety reasons.

    Regards,
    Max
  • 0 in reply to Miguel Robertson
    Prodigy 200 points

    Thank you for your advice Max, much appreciated. Would you mind looking over another block diagram if I post it?

    Regards

    John

  • 0 in reply to John Easton
    TI__Guru 62910 points
    John,

    I don't mind at all. Just so you know, I work primarily with TI's industrial interface devices (RS-485, CAN, IO-Link, etc.), so I could help out with questions centered around those products. If you had questions on a totally different topic you may want to create a new post in the appropriate forum to make sure the right people see it.

    Regards,
    Max
  • 0 in reply to Miguel Robertson
    Prodigy 200 points

    Hi Max,

    Would you mind giving your advice on a few items please?

    1: I'm basing my design on the TCAN1042. I have 2 CAN connectors on my board so I can loop in and out, Should I put the TVS device/caps on both the connectors or just 1?

    2: Is there any problem with putting Leds on the RXD & TXD pins? (5V-2K2 resistor-TXD  and 5V-2K2 resistor-RXD) 

    3: Would you recommend allowing for a choke? I have read some papers that say inductive spikes can be an issue for the transceiver, and may require a TVS device between the transceiver and choke to clamp this.

    Thanks in advance

    John

  • 0 in reply to John Easton
    TI__Guru 62910 points
    Hi John,

    I would put the TVS diodes on the board at a point near where the CANH and CANL lines of each connector meet. Having them at each connector wouldn't be a problem in terms of signal quality or protection, but would likely be overkill. We do generally recommend placing the diodes as close as possible to the point where the transient strikes couple onto the PCB (e.g., external connectors), so having it at the junction point of both buses should be the best compromise.

    It's OK to use LEDs on the logic I/O pins, but be careful to keep the currents within the recommended ranges for each pin. For the RXD output pin, for example, the pull-up current should be no more than 2 mA in order to keep the low-level output voltage below 0.4 V (see the "VOL" spec's test condition).

    If you have a reason to suspect that the design will have EMC issues later on, then it is usually safest to include pads for a common-mode choke on the PCB. The TCAN1042 has excellent EMC performance and is able to meet major automotive OEM requirements without using a choke, so most likely it would not be used (and the pads could be populated with shorts or 0-Ohm resistors). It can be useful to have the option of populating a choke, though, just in case it is needed to solve an unforeseen issue (presence of unusually high noise sources, antennas formed by trace routing/cabling, etc.).

    You are right that if a choke is populated, inductive spikes can occur - especially for worst-case fault conditions like a short-circuit of the bus lines of a switching driver to a power supply. Even in this case, though, you most likely wouldn't need additional TVS diodes between the choke and the TCAN1042. The device has a particularly wide absolute maximum range on the CANH and CANL pins, and so these inductive spikes can be tolerated without triggering the device's internal transient protection circuitry. The only exception might be if you are using a very high-inductance choke (>100 uH) or if there is a risk of shorting to a very high voltage (>48 V).

    Best regards,
    Max
  • 0 in reply to Miguel Robertson
    Prodigy 200 points
    Thanks for the in depth reply Max, much appreciated!