To: TSCM-L@yahoogroups.com From: "James M. Atkinson" Date: Mon, 19 Sep 2005 14:30:26 -0400 Subject: Re: [TSCM-L] NLJDs Good afternoon, You can do the same thing with a local TV or FM radio station, or cellular/pager station. All you really need is a signal that is -50 dBm or greater, but you could make due with one that is -70 or -90 with usable results. I have been able to do this with an FM radio station in the 95 MHz range, several UHF TV stations (gotta love the new HDTV allocations), 150 and 930 Mhz pagers, and the really loud 1.7-1.9 GHz band. The most fun I have had like this was UHF TV station in the 600 MHz range that was so strong it was hard not to pick up harmonics all over the room to well over 10 GHz. Out in the rolling countryside where you can see sheep and corn it is not likely you will get this to work effectively, but in most urban (DC, NY, Dallas, Boston, LA, etc) and suburban environments you will be in-like-Flynn. If you are in a rural area, it is really not going to be worth the effort in trying to get it to work. What you have to do is use your SA to find a really hot carrier that is fairly evenly distributed around the room (ie: a 500 MHz HDTV signal), then tune two antenna's, one for 1000 MHz, and one for 1500 MHZ, pass the signal through a very sharp bandpass filter with say a 3% bandwidth such as a 5 pole K&L Microwave Tunable bandpass filter (at 1 GHz the bandwidth would be 30 MHz wide or less), but if you re a filter design geek you can get this to less then .5%. Then run the signal into a very good quality ultra low noise amplifier like a Sonoma, and feed it into your spectrum analyzers, or into a pair of decent receivers that you have some super narrow IF bandwidth filters on them such as a 1.5 kHz add on filter on a Icom R-8500 (tuned to the 2x and 3x harmonics). The only thing that will bite you is that the low cost receivers do not go much above 2 GHz without mangling the signal, and most start having significant problems over 1 GHz which will stop you from really using an illuminating frequency above 660 MHz. Try to stay with a spectrum analyzer that has "discontinuous spans", this way you can use ONE spectrum analyzer for the whole project, plus a briefcase with your amplifiers and filters that your going to have on site anyway. This way all you really need to add is a power divider that takes the two signals from the filter/amplifier string and joins it into a single signal that you can view on the same S/A, although using two S/A's does have it's benefits. You will also need an antenna that you can tune to the 2nd/3rd harmonics, and a PCB etched dipole with tuning cap may fit the bill for you. A system like this though will depend on an illuminating signal that you can not control, and one which will strongly vary in amplitude as you move around the room. SO your best bet may be to actually supply your own signal, but if you use an active transmission you are more likely to annoy the FCC. There is an upside to using a local station to supply your signal though... sensitivity. You can run at much higher sensitivity, with more low noise amplification, better filters, and less interference as you are only dealing with the 2nd/3rd signals and not the illumination signal. In fact you want to knock down the illumination signal as much as possible so that it does not rise above the DANL of the SA. It is desirable to knock down/bandpass filter every signal other than the 2nd/3rd signals so that they do not rise above the DANL by anymore than 3 dB. A 200 dB bandpass filter with super narrow bandwidth would be nice, but you are going to pay dearly for it, and every time the weather changes you will have to realign it. It's easier to use a commercially available purpose built NLJD, but you will always get higher performance out of using laboratory grade test equipment and either using a signal native to the area (ie: TV station), or creating your own signal, either in the form of a simple frequency, or a swept signal from a sweep generator. On the other side of the coin the commercially available NLJD will be much smaller, and much lighter than a 45 pound, $50,000 microwave spectrum analyzers, 8-10k in filters, 6k in amplifier, and assorted antenna's and cables. Some eavesdropping devices used by intelligence services will sometimes use a strong RF signal that is native to the area for "power", and will use the re-radiation of the signal to carry their eavesdropping signal. An example of this a simple non-linear junction that moves slightly when exposed to sound, or one that shifts the re-radiated signal ever so slightly in relationship to either phase or frequency (phase is better). Don't forget, you also need a rubber mallet and a block of wood as well. If you want to get fancy up can build a solenoid or pneumatic actuator into the head/antenna to thump the wall for you, just don't put a dent in the sheet rock. -jma