generating high frequency triangle wave using AD9834

Hi Frans,

I don't think it would still look like a triangle if we push the waveform's fundamental frequency at 36 MHz because it is very near to the nyquist frequency. It is best to let at least the 5th harmonic of the waveform to be less than the nyquist frequency in order to get a proper looking triangular wave. Having that, it would only be around 7 MHz of frequency.

Best Regards

Louijie

hi LouijieC,

thanks for your respond..

do you have any recommendation of DDS chip which can create a triangle/sawtooth wave at frequency 36MHz?

to accomplished the 5th harmonics, i think that the sampling rate should be more than 360MHz..

in that case, can i use AD9953 (400MSPS) to generate triangle waveform?

thank you

Best regards,

frans

Frans - All of the DDS parts we have running at >100MSPS rely on an angle to amplitude algorithm which doesn't allow us to directly implement anything other than a sinusoid.  That said, it is possible to fake the DDS out a little if it has amplitude sweep capability by setting the a phase offset word so that the DDS engine is stuck at 90 degree (assuming it's doing a sin(x) function), then performing amplitude sweeps.  Note, though, that you lose some of the functionality (phase tuning, e.g.) with this approach.