I am new in the Arduino world and I found my experiments using various components really interesting, a pragmatic approach in the electronic world which makes you having fun in the meanwhile you understand things, in order to understand more about it I would like buy and use a spectrum analyzer - to measure the levels on the wires I am working with and, maybe in the future, from some antennas. Do you know an amateur model which can help me on that ? Or do you know some other tools I can use for achieve the goal ?
Good question indeed, I am working with an arduino Uno - which frequency do you think I am gonna work with ? I am looking for something which could result in a cheap and goos start, by time I will get my owm idea and I can understand what I want/need better than I am doing right now. :*
I think you're talking about using the RSSI signal to map the spectrum - the
cheap and cheerful approach to spectrum analysis. True spectrum analysers
are awesomely expensive, because they have to have better RF performance
than the circuits they are measuring, and are used to measure spurious signals,
noise levels, harmonic amplitudes, all in the face of strong carrier signal.
With modern spread-spectrum and frequency-hopping modulation techniques its
quite challenging to see a signal since there's no fixed carrier.
The Arduino is fast enough and the ADC has enough resolution for basic audio frequency analysis or a spectrum analyzer "effect", but I wouldn't try using it to build a "measurement instrument".
There are two approaches - You can use FFT to do it in software, or there's the MSGEQ7 chip that filters the audio into 7 bands and sends a time-multiplexed varying DC signal to the Arduino. Or, you could build your own analog filters if you want more frequency bands or better frequency resolution, etc. If you search, you can find example projects.
For radio frequency use, I suppose you could use the Arduino as a controller for your spectrum analysis and to handle the user interface & display, etc., but the high frequency stuff would have to be done some other way.
There are several stores, Amazon, Ebay, Alibaba.. to name a few that sell SDR's based on the RF280 chip. They are called DVB-T and some nice freeware programs to make a 24 - 26 to ~1700 MHz Radio. Google SDR...
However extracting the specific information you seem to desire.. (You've not really done a good job of specifying exactly what you want to do..) There is also a channel scanner for the 2.4 GHz ism band radios that might be a beginning...
Yes I agree I have not completely clear in my mind what I'd like to achieve.
I have a software background but I've been fascinated by eletronics since I started to work - in the embedded field, well ..
as a first experiment I would like to attach an SPI device and observe the signal going through the various lines which "play" the role (Clock, Chip Select, Data ..).
My starting point was this article:
That was the spark made me to desire to analyze the signal going out from my board as an excuse for starting to study electronics and improve my understanding.
Otherwise, what I'm putting together is an RTL-SDR (a digital TV receiver, USB, repurposed as a software defined radio) with the oscillator replaced with something a lot more stable, and a frequency converter so it can receive below 50MHz.
And this USB extension to get the RTL-SDR key away from my noisy computer:
I also moved the RTL-SDR into a metal case to provide a bit of shielding.
It isn't as good as an actual network analyzer aka spectrum analyzer, as the bandwidth is only about 3.6MHz, but better than nothing. My primary purpose is to use it as a receiver.
Rigol and Siglent are two names with decent scopes under $400, and even under $300.
You want a 50MHz or 100MHz dual trace scope with 1Gsps. Don't mess about with those scopes that only capture at 2x or a little better than the bandwidth.
You want a scope that has a bandwidth at least 10x greater than the highest frequency square wave or pulse wave that you'll be measuring. A square wave is the fundamental frequency plus all the odd harmonics (3, 5, 7, 9, etc) in ever decreasing amplitude. To get a reasonable picture of a square wave, you need up to the 9th harmonic, in general. So 10x the bandwidth. Add to that, the response does not drop off like a brick wall at the analog bandwidth - it gradually drops as it gets near the bandwidth rating, then drops off faster beyond that.
So 10X is really a minimum.
With a 16MHz Arduino, you aren't going to be looking at the 16MHz clock, generally, but signals put out by the Arduino. So a 50MHz or 100MHz scope should be good.
The Rigol DS1052E is a 50MHz dual trace 1Gsps scope that can be upgraded via a firmware installation to 100MHz per trace. I bought mine from DX.com a few years ago. They have some newer models with some improvements. Dave Jones of EEVBlog illustrates some of them in one of his YouTube videos (I cannot look them up for you at work).
If I were buying the Rigol now, I'd probably get the $400 50MHz four trace, it has a larger screen and a lot of display and software improvements.
