I am trying to identify the proper pin configuration for the Aux. power connection. Please help me by identifying the pin position and assignment on a (4) pin Auxiliary power connector. Answered by Matt Kuzel, Chairman of the Physical Layer SIG,
e-mail: kuzel@voyager.net .
A217) There are no specifications in DeviceNet for connectors or pinning for auxiliary power. A few people have suggested that it may be a good idea for DeviceNet to document the popular choices. A number of big users, like GM, have specific requirements. For now, you should contact your customers, or the end users of your equipment, to get this info.
I am developing a DeviceNet device that will be powered by both the DeviceNet power bus and the vehicle battery. The current design has optical isolation between the battery and DeviceNet powered circuits. However, the product fails our electrostatic discharge test. A possible solution is to add some components between the battery- and DeviceNet- traces on the board. One solution calls for several transient voltage suppressors, while another calls for some capacitors and a high-valued resistor. However, I^m concerned because these will create a potential current path between the two grounds. The DeviceNet information that I have indicates that there can be no current. Does this mean absolutely none, or is there a target number (i.e. 500uA, 1mA...?)?
Answered by Matt Kuzel, Chairman of the Physical Layer SIG, e-mail: kuzel@voyager.net .
A221) The first big thought is: Why are you using isolated physical layers? I am not sure how big the vehicle is, but, automotive uses of CAN are all un-isolated as far as I know. Part of this is because the chassis can often be used as a return path. Another part is because the distances are so short, that no significant signal return voltage drops can develop. I would think that this might be true for your application too. I can also see reasons to want to separate comm power from actuator power -- primarily noise immunity. Just curious. Regarding your question, there is really not a good answer. This issue is a current work topic in the Physical Layer and Conformance SIGs (Special Interest Groups) There is presently a DSE (Draft Standard Enhancement) in process within the Physical Layer SIG to clarify the isolation testing. Also the conformance SIG has put in place a requirement that device vendors document that a test has been done. Right now the test is "advise only", not pass/fail, and does not specify a current limit. It was proposed to use EN60950 as a test spec. This calls out AC, rather than DC. I do not have a copy, so I am not sure yet of the pass/fail current threshold, or, if there even is one in that spec. My guess is that a threshold in the 1 to 10 mA range is likely. BIG CAUTION: Realize that transient suppression components may not survive the hi-pot testing.