EEVblog Electronics Community Forum
Products => Test Equipment => Topic started by: teemum on April 23, 2017, 05:25:13 pm
-
I would need to test some stereo FM tuners and to do that I would need a stereo coded-test signal. It would be enough to have a non-modulated test signal generated, I can then feed this to my RF-signal generator EXT-modulation input.
I have a Rigol DG1022 function generator and I was wondering if I could generate a test signal with this. Does anyone have any ideas if I could use this somehow to generate a stereo coded test signal?
-
A "non'modulated " stereo signal is pretty boring.
There is 19KHz to turn on the FM stereo pilot LED on and maybe soem residual 38KHz artifacts (that you really don't want there i the first place).
Things get much more nteresting WITH a modulated signal
Back to your original question the short answer is "not sure" if you can do it :)
It should be possible if you know how to tell the Generator to do it.
Take a look at this application note from Tektronix although it doens't go into FM stereo, it's not competetly useless :) : http://www.tek.com/document/application-note/arbitrary-function-generator-offers-versatile-and-convenient-solution-am-f (http://www.tek.com/document/application-note/arbitrary-function-generator-offers-versatile-and-convenient-solution-am-f)
For hobby use it's way more fun to have an actual FM Stereo generator :)
-
What I ment by "non-modulated" was that the test signal doesn't need to be FM-modulated. Of course the 38kHz stereo signal have to be modulated.
-
To test the "stereo" part of an FM tuner you'll need a stereo signal and you either feed it directly into the tuner's FM stereo decoder (as if it was received and demodulated from the FM carrier) or
you modulate with it an FM oscillator and the tuner received is if it was a radio station over the air.
Not sure which you are trying to do.
While you are finding out how to tell the generator to produce the waveform, getting on of those Sig lighter FM transmitter thingies used in cars to play iPods over the radio might be a simpler and faster solution.
-
According to these guys: https://www.ap.com/technical-library/testing-rds-with-apx/ (https://www.ap.com/technical-library/testing-rds-with-apx/)
So how does one create the necessary waveform? An enterprising engineer equipped with a powerful signal processing tool like Matlab™ could build an MPX waveform with RDS encoding from scratch. That would involve applying a pre-emphasis filter1 to the left and right audio channels, manipulating them into (L+R) and (L-R), adding the 19 kHz pilot tone and properly modulating the (L-R) signal on the 38 kHz subcarrier. And that’s just to encode the stereo audio without RDS. To add RDS, you would need to work through the encoding scheme laid out in the RDS or RDBS standard, create the necessary signal, modulate it with the 57 kHz subcarrier, and then add it to the multiplexed waveform. This is certainly possible, but it would be a tremendous amount of work!
-
Adding RDS to the signal might be a big work, but I don't need it. I can import excel files to the Rigol Ultrawave-software, so maybe I could create something. The DG-1022 can handle only 4096 points of data in the waveform, but maybe this is sufficient if keep the signal simple.
For example if I create signal where there is 1kHz for left channel and 2kHz for right channel. I think it is enough if this 4096 point data has one cycle of 1kHz signal and 2 cycles of 2kHz signal. The modulated stereo signal would be on bandwidth between 23kHz and 56kHz. So I would need 56 cycles for this 56kHz signal. I guess this 4096 points of data is more than sufficient for this (73 samples per cycle).
The math would be something like this. All signals defined between -1 .. 1.
L(t) = 1kHz signal
R(t) = 2kHz signal
P(t) = 19kHz pilot signal
M(t) = 38kHz carrier
The final signal would be L(t)+R(t)+0.1*P(t)+X*(L(t)-R(t))*M(t).
Could this work? The only thing I don't know is the constant X.
-
To see something meaninful on a scope you only need one chanell (left or right) other wise you end up with a nice jumbled up waveform on the scope.
As for the 38Khz carrier, it's supposed to be modulated by the difference of the two chanels, and supressed , so only the sidebands are transmitted.
Your formula doesn't have enough cos symbols in it :)
Take a look on page 8 here : https://www.silabs.com/Marcom%20Documents/Resources/FMTutorial.pdf (https://www.silabs.com/Marcom%20Documents/Resources/FMTutorial.pdf)
-
@teemum:
If you basically need a test signal, why don't you just look at one of the mp3 FM modulators that are available to listen to mp3's on SD cards or straight stereo line in signals on older car radios? You will find many such gadgets for little money on eBay. RF quality may vary but usually stereo modulation is quite good.
Cheers,
Tom
-
For ideas, have a look at this: https://github.com/jontio/JMPX
-
I managed to create some waveforms with Rigol Ultrawave. Thanks DimitriP for the good document, I got the necessary information from it.
I have attached the Ultrawave waveforms. One is sending 1kHz to Left channel and one 1kHz to right channel. The 19kHz pilot signal is also included.
Of course the DG1022 will generate only audio frequency signal. You have to then feed the signal from Rigol output to the FM modulator / signal generator.
For the left channel the waveform is like 0.45*sin(1kHz)+0.1*sin(19kHz)+0.45*sin(38kHz)*sin(1kHz)
And for the right channel it is like 0.45*sin(1kHz)+0.1*sin(19kHz)-0.45*sin(38kHz)*sin(1kHz)
-
:-+
If you manage to come up with L-R and feed it to a scope it will look pretty interesting.
-
The signal looked in the scope same as it was in Ultrawave-software.
I feed the signal from DG1022 to HP8656B signal generator EXT-modulation input and used it to FM-modulate 100Mhz carrier with 75kHz deviation. Then I feed this signal to stereo FM tuner and both tuner audio outputs were connected to two scope channels. With those signals I attached to my previous post, with another one I saw 1kHz signal in one channel and with another signal 1kHz in another scope channel.