If you are referring to Elcom, they went belly up. My wife was a design engineer there for a few years, and I have their products in my home. In all, they work pretty good. I have the stereo system, video system, phone/modem system and the data I/O system. Frequencies of operation were from 2 to 25Mhz. Higher frequencies are nicer since it can operate as a single ended RF transmission, though passing FCC is tougher (I did a consulting gig their). Someone asked earlier about viewing X10 signals and the like. I traditionally use a spectrum analyzer capacitively coupled to the power line (choose you caps so that your powered delivered at 60hz is within the capability of the analyzer). This allows you to easily see the frequency of interest and all the noise generators. You can also use a scope if you bandpass filter... there is so much junk on the powerline that you will have great difficulty trying to observe the 120khz signals. The largest problems I have had with powerline signal transmission comes from low impedance loads and the fact that they vary (power line impedance can vary from below 5 ohms to over 100ohms). As mentioned earlier in the thread, these are usually caps directly across the power line. Not as much of a problem at high frequencies since you can operate beyond their resonance. However in the 10khz to 2mhz range they are murder. I get around them by using a Panamax surge suppressor that has a PI filter on the input. I remove the cap on the AC side, leaving the inductors, load side cap and varistors. The inductors have enough reactance to look high enough in impedance at the frequency of operation. This works great if the offending device plugs in. If it is hardwired in, I try to add 100uh inductance in series with the hot leg. A slick method to overcome the varying impedance is to monitor the received signal, and adjust the transmission power to obtain a satisfactory CNR automatically. The power will just be high enough to allow reliable communication.This can allow excessive transmission power, but when these systems are tested at an FCC lab, they are on a filtered circuit so the transmission powers will be very low. (i.e. CHEATING) A slightly different method employs analyzing the reflected power, and automatically biasing the output driver to obtain satisfactory match. This method will not overcome noise generators. Scott F. Touchton Harold M Hallikainen To: PICLIST@MITVMA.MIT.EDU Subject: Re: [PIC]: Sending data over power wiring Sent by: pic microcontroller discussion list 10/29/01 04:28 PM Please respond to pic microcontroller discussion list I've always thought so, but remember that company that was going to give us broadband internet over AC wiring since the infrastructure was already in place? What ever happened to them? Harold FCC Rules Online at http://hallikainen.com/FccRules Lighting control for theatre and television at http://www.dovesystems.com ________________________________________________________________ -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics