Batboard

I have a significant repeater de-sense issue and am open to suggestions. First a run down on the system-

Motorola GR300 (Two GM300's tied together) on VHF. Using a Linker Controller. Transmit power is about 40 watts and the repeater is well cooled with a fan that runs continuously. Additionally, I tied the transmitter PA cooling fins to the receiver cooling fins with a aluminum stock for additional heat transfer. The repeater is using a 6 can Sinclair duplexer system that was aligned professionally using a top of the line tracking generator and the tech is VERY experienced with duplexers. Also located in the building is a UHF repeater which shares a common feedline and antenna with the VHF repeater(dual band Hustler) fed through a diplexer. The antenna is near the top of a 60 ft tower fed with 7/8" hardline. There is a polyphaser in the building which connects directly to the diplexer module. The repeater also shares the building with an FM translator transmitter running about 25 watts into an antenna slightly above the repeater antenna. That is the only other equipment operating in the building. About 100 ft from the repeater is another translator building for TV translator. Unknown what frequencies or power levels involved there. The VHF repeater has been in place for about 5 years and has experienced issues before. I was able to trace down the source of that problem and that is no longer a problem (transmitting equipment removed). Also experienced de-sense in the past and moved the antenna further away from the tower and that seemed to resolved the problem but the problem has now returned.

Here is what I have done thus far to try and isolate what is causing the desense-
(1) Checked antenna and cabling with AEA antenna analyzer. (SWR about 1.4:1 on VHF)(better on VHF
(2) Removed diplexer module and polyphaser and gone directly into the hardline. (no effect)
(3) Checked tightness on all duplexer harness lines and flexed lines while the repeater was up with no effect on desense.
(4) Installed a dummy load on duplexer output. Desense is almost completely gone.
(5) Installed a temporary VHF groundplane antenna with a short run of CommScope LMR400 completely independent of the existing antenna on the tower. (no effect)
(6) Turned the GR300 transmitter off and hooked a standalone transmitter with separate power supply to the transmit line going into the duplexers
to test to see if the GR300 transmitter might be dirty. (no effect)
(7) Installed a completely different repeater which is running now. (no effect)
(8) Turned the FM translator off very briefly and then back on. (slight effect)

Right now the receive sensitivity is about -95dBm to-100dBm and I am at a loss as to what is going on. I am considering the possibility of a nearby TV translator that is generating spurs on my input frequency but have not taken a spectrum analyzer to the site yet. I have seen receive sensitivity as good as -118dBm and better in the past.

Am I missing something? Suggestions are welcome.

Have you tested receiver sensitivity with transmitter off?

radiokid wrote:Have you tested receiver sensitivity with transmitter off?

Yes. Receive sensitivity is about -123dBm (little less than .2uv). Also tested it on the bench with transmitter into a dummy load. Receiver still at -123dBm.

Andy

1.4:1 on the end of 60 feet of feed line isn't that great. Remember that any feed line losses will attenuate the reflected signal, reducing the VSWR. It sounds like you may have a bad antenna.

Agreed - vswr of 1.4 is a return loss of 15db which is just barely acceptable in a working system, and could be an indicator of an antenna problem. However, in his first post he said he's tried a separate antenna on a length of LMR with no change in the noise figure. So, he may have two problems - elevated site noise *and* a bad antenna. Not uncommon. Totally worth the effort to climb the tower, inspect the antenna, inspect the connection, sweep the line with a dummy load at the top, and possibly substitute the antenna.

Finding generalized site noise can be a task, but it's not impossible. You need a reference receiver, a reference directional antenna, a signal generator, an iso-tee, and some cables. A handful of 3, 6, and 10db antenuators is handy to have. Essentially, it's a fox hunt for noise. Your ref rcvr can be a portable scanner. Since you're looking for noise, having a broadband radio with poor rejection, like most scanners, is a plus.

Perform a standard site noise test on your repeater to verify the problem is occurring - measure the direct sensitivity of the rptr rcvr with the signal gen, measure again with the iso-tee inserted terminated to a dummy load (the difference is the insertion loss of the iso-tee), and then measure once more replacing the dummy load with the antenna line (the difference is the site noise). For example, direct sense is -120db to open squelch, -70db to open squelch through the iso-tee into a dummy load (yielding insertion loss of 50db), and -60db to open squelch with the antenna connected (yielding site noise / desense of 10db).

In my experience, 10db of site noise in the 150-160 band, and 3-5db in the 450-470 band at an urban site is normal these days. I don't see reduced numbers until I get out into the woods. Any kind of population centers with a direct line of sight to a tower seem to elevate the noise by a few db these days. It's terrible, but it's all around us, and there's nothing we can do about it.

Once you've verified the noise is present on the rcvr of interest, perform the test again on your reference rcvr to establish a baseline. If you get similar numbers (ie: 10db of noise) on the ground with your ref antenna in your hand, then it's all around you - it's either trash from one of the transmitters on site (like a LP FM station running barefoot - no filters), or IM from poorly isolated transmitters, poor site grounding, rusty fences, etc, or you're in the beam path of another site that is landing trash on you. These are the toughest to find, and you'll need the cooperation of everyone to get it resolved. You may even have to file an FCC complaint to get people to the table.

But, if your ref rcvr and ant get a greatly reduced noise figure, that suggests the noise is at primary antenna elevation, perhaps not on site, and you can use the directional antenna to triangulate on the source. Put your ref rcvr on the primary antenna to see if it gets the same noise figure as your repeater to verify the noise is up in the sky, and probably not on site. If it agrees, then pack up your stuff, go someplace nearby, and try to DF the noise. This is why you really want to ensure your antenna isn't the noise maker by inspecting, and substituting it first. Otherwise, you will be spinning your wheels.

As you approach the noise source, you will have to increase the signal gen to open your rcvr, and when you're very close, you will have to insert some attenuators in the antenna line. In the end, you may still have to file a FCC complaint to get action. But, you'll have located the noise, and can give that info to the enforcement officer so they can confirm it.

I'd question the Hustler antenna as well. Hammy-grade fiberglass base antennas typically have terrible PIM rejection.

+1 on the desense test with an iso-tee. I did one over the weekend on a 2-meter repeater and measured 15+ dB of noise on the input freq.

Noise floor measurements have been a routine record for me for at least the last twenty years. I write fwd/rev & sense/desense on blank Avery return address labels, and stick them on the station. Especially important during the initial turn up to establish the baseline. It gives a rough idea of trending over time on future visits.

That measurement can help find an antenna gone bad especially on lines you normally cannot put a vswr meter on like the master rx with tower top amp and filter that passes Anritsu return loss test. ie: 6db of desense recorded during installation, but the latest measurement says desense is now less than 1db. Something changed up top. Probably not the main line, polyphaser, or jumpers since the Anritsu says RL > 20db, and DTF says the end of line is about right. You can't see beyond the TTA. So, it's time to bring in a crew, and open it all up.

Thanks to all for the wealth of knowledge you are willing to share. One thing I have learned is the repeater antenna is directly above the translator antenna with a separation of about 15 ft. The translator is showing 24 watts out on the status display. I may try moving the repeater antenna over to the other side of the tower to see if that affects it. Thinking perhaps there is some side lobe on the FM antenna. The FM antenna is a CP two element yagi with the each of the elements "bent" in the middle at about 90degrees and then spaced 90 degrees between the active and passive element, sidemounted on the tower. The site engineer described it as an "elephant IUD". I have not been back to the site since posting the original message. I do have an iso-tee of sorts. It is a "T" with a male "N" on one side and a female "N" on the other with a BNC with a variable slider on the BNC leg to adjust attenuation for whatever is receiving the signal for evaluation (spectrum analyzer, etc). I have been using a Telewave 44A wattmeter with a -40db sampling port to inject a signal into the repeater.
This repeater has worked extremely well in the past. I live about 9 miles from the repeater site which is about 800 ft higher than my house and I could hit it with a 280mw HT using an 8" antenna from my basement.
The antenna is relatively new but that doesn't mean it couldn't have been recently damaged by wind but there is nothing obvious from the ground.

Again, thanks for the responses and I will keep you guys posted.

Andy

Update on my desense issues. I located the source of the RF incursion and it is the 25 watt FM translator (98.5mhz) in the building. I also found some other issues as well. This time I took a spectrum analyzer with me as well as a couple of portable frequency counters. The building is hot with RF from the translator, however the transmit antenna is not that far away. Additionally, I found a defective polyphaser and I reseated the hardline connector. Apparently the polyphaser has a female "N" connector on both ports. Someone in the past screwed a male PL259 directly into and did an excellent job of spreading the brass feelers inside. I had never paid close attention to that before today, I started running SWR tests through each element of the system. I think I had tested it before but perhaps it was making just enough contact to read correctly. I did the same thing with the diplexer and found it was defective. Again, I think I checked it but have no idea why it read total failure today. Another thing I did was switch some cables around and replace some with different lengths. This was a suggestion from a knowledgeable friend and seemed to help. I also lowered the hardline cable end to near the floor and tied the duplexer in there. Bottom line, I am now at about -120dBm sensitivity and there is no detectable desense on the repeater. I need to return to the site and install a new polyphaser and diplexer but at least now I think I am making progress. Again thanks to all for your great suggestions and assistance.

Andy

Good job! I'd initiate contact with the FM station's engineer and let them know what's going on. Sounds like they need to install a low-pass filter.

You can also knock their signal down with a tuned coax stub on the RX line between the duplexer and receiver if it becomes an issue again. Not the sexiest solution but it works.

Did anybody run the math for transmitter RF products and harmonics of the 25W FM translator output? FM channels are +/- 100 kHz modulation, therefore the 25W "noise cloud extends from 147.600 to 147.900 MHz, and that assumes the TX never overdeviates on peaks.

This is not directly related, but will give it here as a possible source of a problem.

A number of years back (like about 1975), I had a UHF repeater on a commercial frequency and a 2 meter ham repeater in the same building. It seemed every time the UHF repeater came up, it jammed the 2 meter receiver. I spent several days trying to figure out what was causing the problem. I got desperate and started going through the multiplier stages of the UHF transmitter and the multiplier stages of the 2 meter receiver with a calculator. Then I managed to hit the correct numbers. Had a direct hit from one of the multiplier UHF transmitter stages and one of the 2 meter receiver multiplier stages. They only way to resolve the problem was to buy a new receiver crystal for the 2 meter Motorola receiver with the opposite injection. That solved the problem.

So locating noise problems on repeaters is not that simple. You go through the steps for the easy reasons. Then when you come up empty, you have to start looking at other possibilities.

Have never found repeater receiver noise problems to be found easy. It takes some real careful looking and keeping records of what you did and what you found. In some cases, it may be something as simple as rusted bolts or brackets that are loose on the tower.

I will give one more case that had me going for several months. Had a public safety location on a fire tower. There where 3 VHF repeaters on the fire tower. Only when it was sort of damp out, like when we had fog conditions did 2 of the repeaters have an issue. When it was raining or dry out the problem wasn't there.

One day It was foggy and I just happen to be driving past the site while the problem was there. Went up the hill to the fire tower. Turned on my portable to the the repeater with the problem. As I was climbing up the stairs on the fire tower, the noise was changing while both repeaters were keyed. If I stopped and stood still, the noise would go away. Started to look at the stairs. What I found was the flat cross members used for support on the stair hand rails were rubbing each other. So I kicked each end of the flat straps to provide a space so the didn't rub in the middle where they crossed and touched each other. The problem went away.

So don't overlook the obvious. Think outside the normal box when your looking for noise on a receiver.

Jim

Is the Sinclair a band reject (single T connection to each cavity) or band pass (two connections to each cavity) design? You can post the model if you wish as I'm familiar with most of their designs. If it is the common band reject type you should be aware that these have very little out of band rejection. You might expect just 6 or 8 dB insertion loss between your receiver port and then antenna port at 108 MHz. This means your GM300 will see most of the FM transmitter signal, that needs fixing before you worry about broadband noise from the FM transmitter. A single band pass cavity in the receive line might work wonders.

An update for those who might be interested.
The desense is now gone. I have not been up to the site for any measurements but receiver performance is back to being great. The Chief Engineer for the site is a good friend and he told me awhile back he was doing some renovation and overhauling at the site as well as installing another translator.
I traced the problem down to a low power translator. My friend was up there last week (I was out of town), and he did some extensive grounding on hardlines, in the building, etc. Apparently there were some ground loops that needed attention and tying it all together solved the problem. This confirmed what I felt all along that my equipment was OK.

Good job!