In January of 2021 srtate1 asked for the setup to measure THD on an FM tuner. I’m exploring that measurement now, and the responses to srtate1 back in 2021 were in ways incomplete and somewhat off-topic.
I have a suitable signal generator and know the standard test conditions for measuring FM tuner THD. What is somewhat puzzling is the actual connection between the tuner and the QA403. It’s pretty clear the tuner outputs go to the QA403 inputs. It seems there is no need for using the QA403 outputs, so there is no connection made there, as this apparently is not a loopback test. From my limited experimentation and observation of screenshots related to the original 2021 posts, the generator(s) must be turned on for the QA403. If they are not, there are no THD measurement results displayed. This seems odd, as there are no connections between the QA403 outputs and the FM tuner under test.
Again, the issues are: How are the connections made? Must the generators be on? If so, what are their settings? What other settings are recommended for optimizing this test? Thanks for any guidance.
Hi @Buddy_Dave, you don’t need to have the generator on to measure THD. Below I am measuring the DDS generator on my my scope with the DDS generator set to 1.5 kHz
You can see the generator is off. But look at the other array, showing the green “F”. That shows you where the system believes the fundamental is.
This is where you tell the analyzer how to select the fundamental. if you have to set to use the Sig Gen1, then yes, the generator must be on. But if you tell it to use the channel peak, then you can input any frequency, and the highest peak will be deemed the fundamental. If you are dealing with a situation where you must specify the frequency precisely, you can enter it and tell the system to use that.
I had gotten most of the way there prior, but now I know more about the Fundamental Selection and use of the Generator for this task.
Are there any other settings that would help optimize my measurements? FFT size, Window, for example? (I’m doing typical vintage stereo gear measurements.)
Hi @Buddy_Dave, since your generator frequency isn’t locked to a bin center, I’d use Flat Top window since it has a bit more bandwidth. And probably around 16 or 32k fft would be a good place to start. If the FFT is too big your screen updates are slow. If too small, you might see DC error leaking into lower frequency bins. A 32K FFT will give an screen update rate of 1/(32k/48k)=1.5 updates per second, which is probably comfortable for any alignments you might need to do.
