Friday, July 18, 2014

Installing BG7TBL noise source as poor's man tracking generator

After finishing the modification to the BG7TBL noise source, and put it in a shielded box, it was time to place it inside the spectrum analyzer, so I could use it as a poor's man tracking generator.

It was very easy. There is a lot of free space inside the analyzer, so I placed it just in the place where the original tracking generator would be placed. I used a small rigid coaxial cable to connect the output from the noise source to the free output BNC in analyzer's front panel. The 12 volts where picked from the main +12V volt line, available also in the front panel.

The noise source installed inside the IFR-A7550 spectrum analyzer

And now, I can access the noise source simply using a small piece of coax:

Accessing the noise source

Was it worth it? Absolutely!

Simply connecting the noise source with the analyzer's input shows differences between pieces of coaxial, and even variations using (cheap) adapters:

Insertion loss of a 1 meter long RG-58 coax cable with two PL259 connectors and two SO239-BNC adapters. The ripple is indication of impedance mismatch.

But of course, there are other things more interesting to test, for example, a 70cm band notch filter:

A 70cm band notch filter

Or my Diamond MX-2000 triplexer:

Diamond MX-2000 low pass filter response (HF/50MHz).

Diamond MX-2000 band pass filter response (144 MHz)

Diamond MX-2000 high pass filter response (432MHz)

But not only passive elements, it can measure also active elements, like amplifiers, but in this case, an attenuator is needed between the noise source and the amplifier, otherwise, you can overdrive it. I tested a cheap TV amplifier using a 40dB attenuator between noise source and amplifier's input. Analyzer reference was set with this attenuator in line:

Response of a cheap TV antenna amplifier. Note is UHF only and the presence of GSM signals between 900 and 1000MHz. Amplifier's case was plastic.

The amplifier has 0dB gain at 280MHz, and rises up to the maximum, about 28dB, around 625 MHz. It seems to be usable over 400 MHz.

Nice, isn't it? Not bad for a $25 accessory plus some soldering. It turns a simple spectrum analyzer into something much more useful.

Miguel A. Vallejo, EA4EOZ

1 comment:

  1. I found that, if I run the noise source at about 7 Volts, instead of the specified 12, the output is lower, but a lot flatter,