发表于:2002-10-22 10:03:00
楼主
Refer to the Miswiring Protection circuit in Figure 9.4 (Volume 1, Chapter 9, Release 2.0). A rough estimate of the current required for the transceiver is 100mA - 70mA for the Philips PCA82C250 (in dominant mode) and 25mA for optos. If V+ is 24 volts, this means the regulator power is (24 - 5)V * 0.1A = 2 watts. This is a problem. I don^t have room big power parts. Can a lower value of V+ be used. What about 12 volts?
Answered by Matt Kuzel, Physical Layer SIG Chairperson,
e-mail: kuzel@voyager.net
A120) These current and power levels are peak levels. Average current (which is much lower) should be used for establishing power capability of components. Many designers only consider recessive current on the transceiver, since it is so rarely dominant. A possible worst case might be the device continuously re-trying a message because it is the only one on the network. An example of this would be simply powering the device up by itself. The device will continuously retry the duplicate MAC ID check request message. One way to handle the power with smaller parts is to divide the power by using a pre-regulator or voltage follower before the 5 volt regulator.
Last year we installed a DeviceNet system with a scanner and six nodes of block I/O. It was working fine for 11 months and then shut down for 3 months. When it was turned back on, the 24 VDC supply was shorted on the scanner. After replacing the scanner, it appears that all nodes have failed. Whatever caused the damage on the scanner has destroyed every node on the network. Do you have any idea what may have caused this?
Answered by Bob Law, DeviceNet Expert Troubleshooter,
e-mail: rdlaw@ra.rockwell.com
A178) The only thing I can say is that somehow a higher voltage, maybe 110 VAC, must have gotten into the network. This could have been a catastrophic failure of the power supply or someone may have accidentally touched a HOT AC wire to one of the conductors of the network. If someone did this while working on an unrelated circuit in the cabinet, it would be human nature not to mention it. Even with the network powered down the components would be damaged and when you power up the system some time later, you find out the network is toasted.
The only other case of this kind of damage being done to our scanners was when a colorblind electrician put 110VAC on the power pair of the network.