Follow up of my previous blog

OK. I thought I should give a shot to the standard observations that I posted in my previous blog.

I posted these two things:
1. Increasing the attenuation by 10dB increases the noise level by 10dB.
2. Increasing the RBW by 10dB also increases the noise level by 10dB.

The first observation was right with my experiment with the Rohde and Schwarz FSQ8.

For the second one, what I have seen on this million dollar baby is something different.

I terminated the RF input with a 50 ohm resistance and found the noise level increased to around 6dB instead of 10dB for a 10dB change of RBW at 2382MHz to 2482 MHz.

I started with 50MHz RBW (Max that I could go on the FSQ 8) and saw about -75dBm noise level. I made sure my VBW is auto mode.

When I reduced the RBW to 5MHz, I saw the noise floor is about -80dBm.

I went ahead and reduced the RBW further down to 500Khz and then I saw the noise floor dropping down to -85dBm.

So, for a factor of 100 decrease in frequency, the Noise floor reduces by 10dB.

10 log(50/5) = 10dB change in the frequency, there is a 5dB change in the noise floor.

For 10 log(50/.5) = 20dB change in the frequency, there is a 10dB change in the noise floor.

The detector I was using was an rms detector.

I shall continue to find out the math behind this theory and get back in my next blog.

so long..