When it come to measuring it takes about a minute for the everything to settle down when I change the input 10 db so that's fine if I need to run to the bathroom but my ACMVM is immediate and accurate enough. I assume more $$$ is faster.
One minute doesn’t sound credible for your settings. When I replicated your case, the response time of the FFT was 1 second at the most.
The speed of measurement has nothing to do with $$$, but with the effort that goes into a measurement.
A simple AF-millivoltmeter measures wideband average voltage, calibrated in RMS (valid for pure sine waves only) and its response time depends on the lower frequency limit, because the time constant of the averaging has to be long enough to average out several periods of the lowest input frequency.
The modern DSO on the other hand does frequency selective true RMS measurements – and not just one, but a multitude of them at the same time. Luckily, we just need to know some basics and then be able to configure the measurement so that it suits our needs.
In your example the FFT bandwidth is unnecessarily wide for audio: 2.5 MHz. At the same time, the frequency step is very small, just 19.07 Hz. That means that your FFT has to process 2.5 MHz / 19.07 Hz = 131095 bins for every single scan. This is in accordance to what is displayed on the screen: 262144 points … divide that by two and you get the number of bins. In other words, 131095 individual 19.07 Hz wide sections of the total bandwidth are processed during one single scan. Do you still think the instrument is slow?
And it’s so easy to speed it up considerably. For example:
Limit the FFT length to 16 k and speed up the timebase to 1 ms/div.
Now you get an effective sample rate for the FFT of 2.5 MSa/s, thus reducing the bandwidth to 1.25 MHz, which should still be plenty for AF. At the same time, the frequency step has increased to 152,59 Hz, which is meaningless as long as your benchmark is a wideband measurement.
Keep in mind that changing the
display parameters, i.e. vertical reference level and scale as well as horizontal start/stop frequencies does
not alter the measurement in any way. It is just the way how you look at your data, including extreme zoom views.
A few simple rules:
• The upper frequency limit of the FFT is half the effective (FFT) sample rate.
• The lower frequency limit is determined by the reciprocal of the timebase divided by the number of horizontal divisions. 1ms/div corresponds to 1000 Hz / 14 ~ 71.42 Hz. The lower you get here, the longer the timebase and the smaller the frequency step, where both will considerably slow down the measurement.
• Make sure the upper frequency limit as defined above includes all possible input frequencies (including harmonics and spurs), not just the ones you are interested in. Use the horizontal display settings to define center frequency and span to zoom into narrowband structures.
• Always use flattop window for best amplitude flatness and accuracy in the passband and be aware that the -3dB bandwidth of this particular window is roughly about 4 frequency steps. In the example above, the -3dB bandwidth would be about 600 Hz.