Up one level
Back to Home
Measuring Effective Sensitivity
By Chris Boone, WB5ITT
HTML'd and edited by Mike Morris WA6ILQ
Print this Page

Effective sensitivity is all that matters, because that's what your users are experiencing.

Think about this one: Performance on the bench has no meaning when you put your repeater receiver into the cabinet at its permanent site.

No, I don't mean that you shouldn't tweak that receiver for the best it can do, but don't expect your pride and joy that measures 0.11uV for 20dB quieting to actually work that well once you hook it up to the rest of your repeater, and at the repeater site.

Here's the situation: Local site noise is going to drown out weak signals. The measurement of local noise is called the noise floor. If the existing receiver sensitivity is down at the level of the noise floor then adding a preamp just raises the noise floor and actually makes the situation worse.

A note from Scott Zimmerman N3XCC:

A final thought on a raised noise floor: It can be thought of as being similar to the astronomers nemesis: light pollution. Out here in the country where I live, on a clear night, I can see tons and tons of stars including the Milky Way. Go toward a small town and you can only see a small fraction of those stars. If you head for the big city (say New York City) you'd be lucky to pick out a dozen stars while standing in the middle of Times Square. Why? Because the ambient light (the noise floor) makes it hard to see (hear) those distant stars (stations).

Here's a writeup from Chris Boone WB5ITT on how to accurately measure the actual, real, usable sensitivity your receiver has in your repeater's environment:

How to determine the effective (or usable) sensitivity of your repeater receiver.
Copyright © by Chris Boone, WB5ITT

This 3 step procedure (short and easy) tells you how much noise you have at the repeater site on channel and how effective the receiver will hear against this noise.

The procedure uses the following equipment:

1) An adjustable level signal generator on the repeater receiver channel.

2) An isolated T ("iso-T") pad (you can take an ordinary UHF coax T and remove the pin from the male end of the T; they usually unscrew, cut and file it down, then screw it back in. This gives you a T connector but the male pin will not mate with a female now) or use a 40dB isolation slug plugged into a Bird Wattmeter or similar.

3) A 50 ohm dummy load (any size will do, even a 1 watt resistor on a coax connector that will mate with your iso-T or wattmeter above; you will not be transmitting into it.) Do not use a 47 Ohm resistor, instead use two 100 ohm resistors in parallel.

Step 1:

Hook up your signal generator to the receiver through a double shielded jumper directly. Adjust the generator to where you have a decent (but not totally quieting) signal while listening to the receiver audio. If you have access to a SINAD meter, set the generator for exactly 12dB SINAD reading. If not and you can monitor the IF limiter or S-meter, adjust the generator for a low but stable reading on the meter but still have some frying in the audio. Now, note your generator level (for example, let's say it is 0.25uV, which is a typical sensitivity).

Step 2:

Now disconnect the signal generator and connect the non-isolated ends of the T-pad inline to the receiver and the 50 ohm dummy load. (if you are using the wattmeter pad method, place the wattmeter inline as if you were measuring power from a transmitter, except your transmitter will be the receiver). Connect the generator to the isolated connector of the T. Now, check the receiver again and adjust the generator for the exact same results you got in step one. Because of the loss of the iso-T or the pad, you will have to increase the generator to compensate. This difference in levels is the isolation of the pad. Remember this number (commercial pads are usually 20-40dB).

Step 3:

Now replace the 50 ohm load with the antenna and feedline. As you remove the 50 ohm load, the receiver will go quieter. This is normal because the system is unbalanced right now. When you connect the antenna, you will likely hear the signal go noisier than before OR it will go away all together. Now, readjust the signal generator again for the exact SAME reading results as above. When you reach that point, note the difference in generator level now and what it was with the 50 ohm load attached. This is the amount of noise you had to overcome - your site noise floor for that frequency. Take that difference and add it to your original receiver sensitivity in the first test. This answer is your effective (or usable) receiver sensitivity for that frequency. This is the real-life number that your users experience every day.

A working example of this is:

A repeater receiver on the bench with signal generator directly into it hears 0.25uV for 12dB SINAD. At the site, with the iso-T inline and the 50 ohm load on one end, the receiver on the other and the signal generator connected to the isolated end of the T, the receiver now shows 24.0uV is needed to get back to 12dB SINAD. This means the iso-T or pad has 40dB of isolation.

When the antenna is substituted for the 50 ohm dummy load, the signal generator is turned up to 100uV of output to get back to the 12dB SINAD point. Taking into account the 40dB pad, we now know the effective sensitivity of the receiver on-site is 100uV - 40dB or 1uV of effective sensitivity. This shows the noise floor at your repeater site is 10dB worse than the receiver sensitivity on the bench. No wonder those HTs have problems getting in!

On lowband VHF and 6 meters, it is not uncommon to see the effective sensitivity be 100-200uV without a Noise Blanker and 10-20uV with a properly aligned NB. This is especially true if your site is near a location of noisy high voltage electric lines or a substation (and believe me, I know from personal experience). Your results will vary according to your site and band. If you operate a 220 MHz or 440 MHz repeater, you will likely find not much difference at all between on-air and on-bench sensitivity.

I once had a 2 meter repeater site that produced a 5.0uV effective sensitivity with a receiver that measured 0.25uV on the bench. Oddly enough, the site was extremely quiet at 220 MHz, but a poor performer with a 2 meter repeater. A move to a new site suddenly made 100mW handhelds full quieting where 30watt mobiles had been noisy before at the old site

Comments or questions can be emailed to cboone /at/ earthlink /dot/ net

73, Chris WB5ITT

Notes from WA6ILQ:

Don't even bother trying to measure the effective sensitivity if you have any desense.

Keep in mind the effective sensitivity at a site can be a dynamic value (unless the environment is a broadcast site with a constant RF environment).

You can add some additional attenuation to the iso-T, and make it somewhat variable, by adding a barrel connector on the male port that is isolated. Screw the barrel in and out to vary the attenuation.

Secondly, some sites are almost unusable due to the digital / HDTV mixes and intermod. As of late 2006, the UHF noise floor on Mount Wilson, north of Los Angeles, was approaching 4uV due to all of the new Digital / HDTV transmitters that were going on line. Where analog TV has two major energy spikes per channel, an HDTV transmitter uses the full 6 MHz bandwidth full time. Two (or more) HDTV transmitters feeding a multiport combiner feeding an antenna on a rusty tower can trash across a broad chunk of spectrum across an entire site. Now do this at a 5,300 foot high mountiantop site that has primary and backup transmitters (usually on separate antennas on their own towers) on TV channels 2, 4, 5, 7, 9, 11, 13, 22, 28, 32, 34, 36, 44, 48, 50, 52, 58, plus over two dozen FM broadcast transmitters (and their backups) all within a circle of land that is 3/4 mile in diameter. Some of the towers are as tall as 900 feet. And the signals from several dozen transmitters are combined into two antenna structures.

Update - Analog channels 2, 4 and 5 have gone dark with the cessation of analog broadcasting. For the first time in 50 years 6m SSB DX is actually usable in Los Angeles.

Back to the top of the page
Up one level
Back to Home

Artistic layout, and hand-coded HTML © Copyright 2008 by Repeater-Builder.

This web page, this web site, the information presented in and on its pages and in these modifications and conversions is © Copyrighted 1995 and (date of last update) by Kevin Custer W3KKC and multiple originating authors. All Rights Reserved, including that of paper and web publication elsewhere.