On Thu, 2006-05-18 at 16:50 +1200, David Mann wrote: > On May 18, 2006, at 1:17 PM, Nick Rout wrote: > > > 5m, thats the info i was trying to remember while chatting to Don. He > > was busy trying to splice usb plugs onto cat5 cable to make a usb > > extender, seems a recipe for disaster perhaps. > > It might work: the impedances are 90 ohm [USB] and 100 ohm [cat5], > both +/-15%. Even if the particular cat5 cable being used falls out > of the USB spec, it shouldn't matter too much.
The slight impedance mismatch is unlikely to cause any problems. The max USB length of 5 m is determined by the voltage drop along the cable (you need a min of 4.4 V at the slave) and the maximum propagation delay of 30 ns required by the USB protocol for switching the bidirectional bus. Using CAT5 may cause grief since it is unshielded and USB (unlike Ethernet) does not use transformers to reject common mode pickup. What you may find is that sometimes it will work then other times it won't. Michael.
