• TI Thinks Resolved

ISO1050: communication fail with 9 nodes at 1Mbps

Expert 7045 points

Replies: 3

Views: 127

Part Number: ISO1050

Hi, my customer is using ISO1050 as below: when CAN_RH and CAN_RL is connected together, the 120ohm terminal resistor is connected.

When they test with 9 nodes at 250kbps, the communication is normal. They only connect 2 120ohm termination resistor in.

When they test with 9 nodes at 1Mbps, the communication failed in some of the nodes.

Are there any problem with the circuit above?

How can we find if the problem is caused by hardware or software?

  • Hi Howard, 

    The TVS diode may be causing some issues. Looking at a SM6T6V8CA datasheet I found online, it looks like the capacitance might be on the order of a few nF, which may be causing an issue at higher speeds.

    Also it looks like the breakdown voltage can start as low as 6.45V - which is inside the common mode range of the CAN standard (+/-12V). 

    An easy way to test this theory is depopulate the TVS diodes on all of the nodes and re-test at 250kbps and 1Mbps. If it is in fact the TVS diode that is causing the issue, I do have an application note that discusses how to choose a TVS diode and offers some recommendations that in the 10's of pF junction capacitance. 

    Do you have a wiring diagram for how the nodes are connected? This might also give us some hints on what is going wrong. 

    It would also be helpful to see oscilloscope waveforms for the different nodes so we can see what the ISO1050 is seeing that it is not able to work at 1Mbps. 

    I also want to quick point out that we do have a new isolated CAN transceiver that can work at even faster speeds - the ISO1042. Along with faster speeds, it also offers a wider common mode, wider standoff voltage, and isolation rating. 

    Best regards, 

    Dan

  • In reply to Dan Kisling:

    Hi Howard, 

    I haven't heard from you in a few days. Were you able to determine why the nodes were not able to communicate at faster data rates?

    Best regards, 

    Dan

  • In reply to Dan Kisling:

    Hi Howard, 

    I haven't heard from you in a while so I am going to assume you were able to resolve your issue. If this is not the case please feel free to reply or start a new thread and I would be happy to assist you further. 

    Best regards, 

    Dan