The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Design >> Analog Design >> How to measure SFDR of a DAC
https://designers-guide.org/forum/YaBB.pl?num=1067923133

Message started by ram on Nov 3^rd, 2003, 9:18pm

Title: How to measure SFDR of a DAC
Post by ram on Nov 3^rd, 2003, 9:18pm

Hi,

How to measure SFDR of a DAC?

Please guide me on that...

Regards
ram

Title: Re: How to measure SFDR of a DAC
Post by Paul on Nov 4^th, 2003, 12:39am

Hi ram,

you can extract the SFDR from an FFT plot of your output signal. The input signal is an ideal sinusoidal signal and the SFDR is given by the difference between your fundamental and the dominant harmonic spur in the output spectrum.

Notice that nonlinearities in the DAC are most important for large input signals, so you should use a close to full scale input signal to measure the real SFDR. SFDR is probably also dependent on the frequency of your input signal.

Paul

Title: Re: How to measure SFDR of a DAC
Post by ram on Nov 4^th, 2003, 1:48am

Thanks paul, :)

Which are the techniques we can use for betterment of SFDR....

Regards
ram

Title: Re: How to measure SFDR of a DAC
Post by Paul on Nov 4^th, 2003, 2:27am

It depends on the type of DAC you are using (delta-sigma can exhibit spurs for different reasons). Typically harmonics are due to non-linearities, so you must take care of the linearity of each block.

If SNDR is THE specification to meet, you can also cheat and specify a lower input signal range. At lower signals, the distorsion and SFDR should be better. But your SNR will suffer from this.

Paul

Title: Re: How to measure SFDR of a DAC
Post by ram on Nov 4^th, 2003, 3:24am

Thanks agian :) :)

The DAC that I am using is current steering differential DAC.

If the swing of the DAC and SFDR are the important specs to meet, what are the design cares to be taken to obtain better SFDR numbers.

Regards
ram

Title: Re: How to measure SFDR of a DAC
Post by Paul on Nov 4^th, 2003, 6:49am

Sorry, I'm not familiar with current-steering DACs. I would still expect the linearity and matching of the current steering element (diff pair) to be one of the important parameters. You should be able to find some papers on current steering DACs and design practices on the web.

Paul

Title: Re: How to measure SFDR of a DAC
Post by als on Aug 29^th, 2004, 1:10am

Ram,

Using the FFT a good method to non-linearity of a curent -switched D/A Converter. One comment would be that depending on the input frequency[from the digital code] and the clock rate, compensation may be necesary. A DAC acts like a zero-hold circuit, the frequency response includes a sinc function. So as the input frequency approaches the clock rate, the magnitude of the DAC output is reduced. More simply for an DAC to get a constant output level, the input level needs to increase as a function of frequency to cancel the effect of the sinc function.

Things to improve linearity are to increase the output resistance of the current source cell, minimize capacitance of the current source cell, minimize capacitance on the output node. Typically in current-switched designs, the code-dependent value of the DAC output impedance is a major source of non-linearity. Increasing the impedance reduces the effect of the "code-dependence", i.e., the DAC output is more "linear".