0
\$\begingroup\$

DDS and arbitrary function generator are digital function generator, and both of them are generate sample from memory, so how to differentiate DDS and arbitrary function generator in term of the operation? and waht is the advantages of DDS over arbitrary as DDS quite popular right now.

\$\endgroup\$
  • 1
    \$\begingroup\$ you might get better answers if you ask us to compare and contrast two specific parts, I think you might risk closure for "too broad" with the question in its current form \$\endgroup\$ – vicatcu Apr 17 '14 at 14:48
2
\$\begingroup\$

They are the same thing. A DDS generator without arbitrary waveform capability will have a fixed waveform lookup table stored internally. A DDS generator with arbitrary waveform capability will have a programmable lookup table.

A DDS generator without arbitrary capability is a little cheaper, but these days the difference is quite minimal or in some cases non-existent. The difference in pricing of test equipment is mostly historical, i.e. they can charge you extra for the arbitrary waveform function.

\$\endgroup\$
  • \$\begingroup\$ then why DDS function generator not called as arbitrary function generator? \$\endgroup\$ – user40353 Apr 17 '14 at 16:20
  • \$\begingroup\$ Because it can't generate an arbitrary waverform, but only a (few) specific one(s). \$\endgroup\$ – Chris Stratton Apr 17 '14 at 19:26
  • \$\begingroup\$ As I said, the waveforms are fixed. The AFG allows you to program in any waveform you like. \$\endgroup\$ – user Apr 17 '14 at 21:01
0
\$\begingroup\$

The DDS and AWG are almost 2 different thing.

  • AWG works on the principle of sampling while main concept of DDS is fetching waveform from memory.
  • AWG can generate complex waveform (LTE, xxxQAM...etc). DDS almost always deal with simpler waveform. Complex signal which are more common includes chirp signal. (It is technically possible to generate complex signal with DDS for special purpose application, but this is not too common for some reasons)
  • One big advantage of DDS is inherently, there are phase continuity and repeatedbility after frequency hopping. This is impossible for analog PLL synthesizer. While it is possible with AWG, it is much more painful to implement.

enter image description here

There is memory within DDS consisting of sine wave (in fact it is just a quarter of sine wave). By varying how fast you move along the sine wave, you get different frequency. (Try to explain as simple as possible)

enter image description here

Above is a block diagram in Keysight website describing DDS of Keysight UXG Signal Generator.

\$\endgroup\$
0
\$\begingroup\$

Consider a part like the AD9910, in its most common use it runs as a DDS outputting a sine, but it does have on chip ram and there is are ways to configure it for arbitary waveform output.

I would note that AFG usually has a much bigger ram to allow lower frequency synthesis as a DDS will rely on the phase wrapping to generate low frequency outputs which does not work if you are for example trying to generate something complex at low frequency because the temporal aliasing only works for a sine wave.

If you have a 1GHz clock and say a 10 bit (1K word) LUT you can generate arbitary waveforms down to 1MHz, but for anything lower then that you will be limited because the clock is actually too fast (And the reconstruction filter wrong if you just slow the clock). Such a small LUT is fine for sine generation (Especially if you also do some taylor series correction) but is a serious limit in an AWG.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.