The TI E2E™ design support forums will undergo maintenance from Sept. 28 to Oct. 2. If you need design support during this time, contact your TI representative or open a new support request with our customer support center.

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

DAC38J84EVM: SNR Measurement of DAC38j84

Part Number: DAC38J84EVM
Other Parts Discussed in Thread: DAC38J84


I'm measuring the SNR of the DAC38j84 by applying a single tone (const=0x0000, fout=400MHz) at the input with NCO (on the evaluation board with GUI evaluation software).

Settings: Mixer Gain=0dB, Coarse DAC Gain=15, 8 x Interpolation, fdac=2.457GHz. No external LPF.

Here is what I got on the spectrum analyser: 

- LF to fdac/2 (2.4576GHz/2=1.2288GHz)

I got SNR = FS - noise floor (displayed) - process gain (10log(fdac/2/RBW)) = -1.81 - (-76) - 10log(1.2288GHz/220kHz) = 36.7dB !?

However when I applied this equation to get the "theoretical" SNR for 400MHz which is calculated as: FS -(NSD + 10log(fdac/2) = -1.81- (-153dBm/Hz@400MHz + 10log(fdac/2)) = 60.3dB......

When I reduce the SPAN of the spectrum analyser to 100MHz centered @400MHz, I got: 

- 350MHz to 450MHz

SNR = FS - noise floor (displayed) - process gain (10log(fdac/2/RBW)) = -1.49 - (-85) - 10log(1.2288GHz/10kHz) = 32.6dB !?

while again the theoretical gives: SNR = -1.49- (-153dBm/Hz@400MHz + 10log(fdac/2)) = 60.6dB...

I'm wondering why there is such a difference between the theoretical one and the measured one?

Thanks in advances,



  • XL,

    You have the spectrum analyzer attenuation very high (14 and 20dB) and are not subtracting the spectrum analyzer noise.

     TI calculates SNR from NSD * BW. NSD is measured in a band using the spectrum analyzer noise subtraction.