[SOLVED] How do I know if a spectrum analyzer is enough to measure an specficic DAC?

I'm trying to measure the distorsion of the 2nd and 3rd harmonics of a specific DAC (DAC37J82) with an spectrum analyzer R&S®FSH4 (also tried with MDO4104C). But I'm getting way worst results than the ones specified in the DAC Datasheet (tried at 50, 150, 250 and 350 MHz), I know how to theoretically work with the frequency domain, but I'm new to spectrum analyzers.

The DAC is very fast, so I think that this may be the problem. How can I determine if the spectrum analyzer is enought to measure this DAC?

For example if if the spurious response of the spectrum analyzer is:

• 2nd and 3rd harmonic distortion (9 kHz to 100 MHz )< -57 dBc (< -65 dBc, typical), with auto settings on, signals 10 dB below reference level, and reference level ≤ -15 dBm

Does it mean that for a response that is better than that (the DAC is at 80dBc) I shouldn't be able to measure it? I'm actually really confused with the signs as the DAC datasheet measure the distorsion with positive dBc, but I usually work with negative values.

Datasheets:

• DAC37J82
• **broken link removed**
• MDO4104C

Sorry if i'm being too confusing with the explanation
And Thank you for the help.

With only around 60 dBc harmonic suppression, you need a filter to reduce the fundamental frequency magnitude between DAC and signal analyzer. DAC linearity is mostly measured with two-tone IM measurement, but problem is essentially the same, some filter is needed.

I haven't totally understood you, do you mean an attenuator? What is a filter going to do?

you are looking at 3 harmonics

at, say, 50 MHz, the fundamental or first harmonic is 50 MHz
the second harmonic is 100 MHz
the third harmonic is 150 MHz

the fundamental carries the majority of the energy
FvM is saying that you need a filter to reduce the fundamental,
then you can get a better measurement of the harmonics.

so you need a filter to remove the first harmonic.

Oh ok, now I get it. I'm not really sure if I have a filter that high in frequency, but thank you for the help. I will try to do that.