RF noise measurements using an isolated-tee bridge

A method exists for measuring the noise level on an antenna system using a communications analyser (signal generator and SINAD meter*); an isolated-tee bridge (or simply called an iso-tee) and a small dummy load.

An iso-tee is basically a three way coaxial device with an input; output and a third, loosely coupled, connection – the input and output are connected directly to one another whereas the third (generator) port is air (capacitively) coupled to the input-output line.

Once can be built quite easily, instructions and ideas can be found on the internet.

Figure 1 - the iso-tee

Method

To measure the noise level, firstly ensure that the antenna is correctly matched. This can be done by measuring the VSWR. If the SWR is too high, the noise level measurement will be out due to the difference in the antenna and dummy load impedances.

If testing the noise level on a repeater, the transmitter must be connected to the antenna or to a high power dummy load.  Remember that disconnecting the receiver from the duplexer may affect the duplexer’s internal impedance so be sure to terminate the open leg of the duplexer during these tests.

Now connect the iso-tee as shown in figure 2, and generate a 1 kHz tone modulated signal (at 60% of RSD)** on the repeater receive frequency at -50dBm, and slowly turn the RF level down until 12dB SINAD* is seen on the communications analyser. Note this level as A. Remember that the RF level will not be a true reflection of the receive sensitivity due to the iso-tee.

Figure 2 - the reference level setup (A)

Now connect the iso-tee as shown in figure 3, and again generate a 1 kHz tone modulated signal; (at 60% of RSD)** on the repeater receive frequency at -50dBm, and slowly turn the RF level down until 12dB SINAD* is seen on the communications analyser. Note this level as B. Once again, the RF level will not be a true reflection of receiver sensitivity due to the iso-tee coupling losses.

Figure 3 - measuring with the antenna connected (B)

The difference between B and A is the noise level. Anything less than 6dB is acceptable. Anything higher indicates a RF noise problem on the site. It will be necessary to make the measurement for B over several minutes, you may notice that this level jumps around a lot – this indicates that the source of the noise is not always present (i.e. another repeater is keying up and generating the noise in question).

What next?

The next – obvious – question is how does one solve RF noise problems on a site? The first step is to locate the source of the noise. In my experience, the source of the noise has usually has inadequate filtering on the transmitter. More information on how to improve filtering can be found here.

* If testing a MOTOTRBO repeater, a 1011Hz voice test stream would be generated and instead looking for a 12dB SINAD level, a 3% BER would be the reference target.

** It is assumed tests are made in analogue mode. If no digital communications analyzer is available, the repeater would need to be set for analogue operation.