Stu, It looks to me as if you have covered almost all of the bases, and have eliminated every one of the typical causes of desense. However, there is one possibility that has not been mentioned: Leakage inside the Micor radio.
Even with a careful duplex conversion, there are several "sneak" paths for extraneous signals to enter the Micor receiver. One ingress point is the tiny slots around the RCA plug at the input to the helical resonator block. A wrap of metal tape around that plug will seal the plug. Additional bypass capacitors and ferrite beads on DC supply leads may help. Try using a separate DC power supply for the receiver. Moreover, I suspect that the receiver needs more shielding. 73, Eric Lemmon WB6FLY -----Original Message----- From: Repeater-Builder@yahoogroups.com [mailto:[EMAIL PROTECTED] On Behalf Of Stu Benner Sent: Friday, January 25, 2008 1:54 PM To: Repeater-Builder@yahoogroups.com Subject: RE: [Repeater-Builder] Desense Problem on 222 MHz Repeater Thanks to everyone for the replies so far. Please see comments embedded below. I think that they address most of the comments, questions, and recommendations posed by all to this point. If you suspect IMD between the channel 12 DTV transmitter and your Tx carrier, work on attenuating the DTV signal. Using cavity notch filters to reject the entire 6 MHz of DTV isn't too practical, so instead, try adding pass cavities on your Tx before the duplexer input. That will help determine if the IM is originating in your PA. An isolator *may* help, but with channel 12 being the better part of 20 MHz away (about 10%), it may not afford full protection -- isolators don't have infinite bandwidth. Likewise, finding a 250 watt 220 MHz isolator may not be easy. I think I have some 220 isolators that came off a combiner (Sinclair), but doubt they're good for 250 watts judging by their size. [Stu] Agree, notches don't work well for that plus we don't have handy any cavities that will tune there. Have used up to two BP cavities in the TX path along with a 2-stage isolator. No difference in desense is observed. If you suspect a "rusty bolt" mix, use an alternate antenna for testing. If nothing else, try a quarter-wave whip (suitable for operation at your 250 watts TPO), even if it's just temporarily mounted on the tower (be sure it's at a sufficient height to prevent desense due to close proximity to the repeater itself). [Stu] We're presently on split antennas. One is at about 80 ft., the other is at about 15 ft. This improves the desense on the order of 6 dB. Another good possibility is IM in your receiver front end (or preamp, if you're using one). Again, pass cavities are your friend here. Attenuate the channel 12 signal as much as possible and see if it makes a difference. Have you looked at what sigs are reaching your receiver input on a spectrum analyzer? With 15 dB of desense, you should be able to see the culprit(s); it's not like they're going to be buried in the noise if it's causing 15 dB of desense. [Stu] The desense is significant with or without a preamp. Worse with but I can't find my notes to quote numbers. Used up to 2 BP cavities on RX with no perceptible difference in desense. Have also installed a DCI 4-pole filter on RX and TX with no effect. Have looked at the receiver input with a spectrum analyzer. The most significant signal is the one FM broadcast transmitter at the site. Running power down on it or turning it off has no effect on the desense. Our TX signal at our RX input is consistent with our measured duplexer isolation (about -88 dBc or -34 dBm). Within several hundred kHz of the RX frequency there are no detectable narrowband signals. Even that 1 dB of desense would give me some agita. I'd verify that the duplexers are properly tuned and the transmitter is clean before even starting down any other paths related to the channel 12 issue. IIRC, the Telewave cavities have adjustable coupling. If necessary, sacrifice a little extra loss for additional rejection if necessary. [Stu] I tuned the duplexer myself with a network analyzer and the transmitter looks clean. I have coupling set where I get about 1 dB through loss and the notches are at about 88 dB on TX and about 90 dB on RX. I also assume you're using all known-good interconnect cables (no foil+braid or other cables not suitable for duplex operation). [Stu] All cables are either Heliax or double braided. Are you using a Polyphaser or other type of surge arrestor? If so, try bypassing it. I've seen gas discharge tube type surge arrestors become noisemakers after absorbing a strike. [Stu] Yes but there is no difference in desense when it is removed. Has the VSWR changed at all on your antenna? If so, it could indicate water in a connector or the harness which will cause all kinds of grief, including wideband noise. [Stu] The problem has existed since the repeater was installed. It exists whether we duplex on a DB264 at 80 feet or a G7-220 at 15 feet, both fed with Heliax Finally, does the desense change appreciably if you vary transmitter power output (it probably will). Do you any have desense when running on just exciter power? [Stu] The desense is roughly proportional to transmit power. Barely perceptible at 20W (exciter only) with split antennas. A little worse at 20W (exciter only) using one antenna. Have tried both tube-type and solid state amplifiers at various power levels. --- Jeff WN3A
