A CLOSE LOOK TO LOW COST NOISE GENERATORS

General discussions
glovisol
Posts: 662
Joined: Thu May 10, 2018 6:42 pm
Location: Piedmont, Italy

Re: A CLOSE LOOK TO LOW COST NOISE GENERATORS

Post by glovisol » Fri Apr 12, 2019 6:49 am

Sdrom 33 you surely do have a keen eye!! Yes, the pic in the previous post showed I was using a switch mode PSU. Now, if you look at the pics I recently posted here:

https://www.sdrplay.com/community/viewt ... 3&start=70

there is strong interference at low frequencies and I wrote that this was "Broadcast stations stealing into the measurement" I sincerely apologise for this mistake: this interference was caused by strong LF RADIATION BY THE PSU, as I discovered some time ago. If you look a the first plots for the LF low pass filter here:

https://www.sdrplay.com/community/viewt ... f=5&t=3943

you note strong interference here as well and I was compelled to mark the plot with a red line: even then I was using the switch mode PSU! The reason why it took me such a long time to discover the problem was that the PSU noise, being at LF, does not show up as soon as you sweep filters at a few Mhz and beyond.

I might have misled and/or convinced some readers to use a noisy PSU for the job and I sincerely do apologise. In effect a conventional, AC transformer-series regulator, 9 V supply MUST be used for feeding the noise generator. I have built one myself and its excellent results can be seen in the very clean, last plots here:

https://www.sdrplay.com/community/viewt ... 3&start=70

Pics of the Noise Generator with its new and CLEAN (forgive the pun!) 9 V. supply are shown below. I am going to correct the original posts showing the noisy plots.
Attachments
conventional PSU 1.JPG
conventional PSU 1.JPG (106.59 KiB) Viewed 10307 times
Conventional PSU 2.JPG
Conventional PSU 2.JPG (108.62 KiB) Viewed 10307 times

Reason: No reason

glovisol
Posts: 662
Joined: Thu May 10, 2018 6:42 pm
Location: Piedmont, Italy

Re: A CLOSE LOOK TO LOW COST NOISE GENERATORS

Post by glovisol » Sat Apr 20, 2019 3:31 pm

The multiple trace capability of the new Spectrum Analyser software Ver. 1.0a overcomes the major problem related to low cost Noise Generators mentioned some time ago by Roger in this thread: namely the unlevelled noise power output that changes with frequency. This problem is now solved and the correct frequency response of the device under test can be easily obtained without any correction, as shown in the screen below, picturing the test of a low frequency low pass filter.

- Connect the attenuated output of the Noise Generatot to the RSP-1A start the Spectrum Analyser and store main trace. This is the frequency response of the Noise generator.
- Now connect the device under test between attenuated Noise Generator and RSP-1A and enable trace 1. The correct frequency response is at a glance the difference betwee the two traces.
Attachments
Noise Generator reference.jpg
Multiple trace capability of SA software
Noise Generator reference.jpg (325.57 KiB) Viewed 10229 times

Reason: No reason

Post Reply