We have a question that customers pose to us regularly. The power supply for the DeviceNet nodes is specified as to level, ripple, how they are to be wired in the network, and a host of things. However, we have yet to find a discussion on whether it is recommend (or not) to use a second, isolated supply for applications (in our case it would be solenoid valves). We recommend to our customers, due to the potential for electrical noise during the valve operation and the current load potential to use an isolated supply (from the DeviceNet power) for the valves. Is there a guideline or requirement set by DeviceNet or the scanner companies for application power? Answered by Matt Kuzel, Chairman of the Physical Layer SIG, e-mail: kuzel@voyager.net and Ray Romito, DeviceNet Trainer, e-mail: raromito@ra.rockwell.com . A287) DeviceNet does not deal at all with what we usually call "auxiliary" power supplies. About the only related area is that in the Physical Layer it is stated that the spec allows the use of such supplies, provides examples of how to ground isolate when they are used, and discusses some behavior in regard to power-up sequences. Your recommendation to your customers is the safe thing to suggest. Keeping network power free of transients is a very good thing. Can the one power supply be used to provide power to 2 separate DeviceNet networks? Answered by Matt Kuzel, Chairman of the Physical Layer SIG, e-mail: kuzel@voyager.net A307) The simple answer is yes. There are some possible complications. First is that you may want to fuse the two nets separately to fault isolate them from each other. Second is the 100 VA limit for Class2 Power Limited Sources as described in the National Electric Code. If your installation must meet this Code, then the available 100VA (about 4A at 24V) is the total for the two networks. This may be a problem if your power busses are heavily loaded. The separate fusing can help, but to get the full 100VA on each network, the NEC requires special fuses that have physical sizes that are different for different fuse currents. This is to prevent someone from being able to simply "plug in" a higher current fuse and exceed the power limit. I have a DeviceNet application where I have 30 devices installed in a 72" high x 108" wide wall. These devices are spread out evenly on three horizontal rows of DIN Rail. Can I bring in a trunk to three-drop lines and further branch to about 10 devices from each drop line? Documentation I see conflicts as to whether it is trunk/drop line only or trunk/drop line/drop line device branches? Answered by Matt Kuzel, Chairman of the Physical Layer SIG, e-mail: kuzel@voyager.net A310) There is no problem with a branching drop line. Any branching structure is OK. Any number of devices ( 64 of course) on a drop line is OK. The only restrictions are: - the 20-ft. maximum distance from the trunk line tap to the farthest device transceiver on the branching drop line. - the cumulative drop line usage on the network. From the size of your wall, and assuming that you use three branching drops, each with 10 devices, it does not sound like either of these restrictions will be violated. This also assumes that you do not have drop line elsewhere on the network that exceed the restrictions.