I wondered if it was possible to ensure that the Phase -shift occurs in passing to zero volts out on a single sinusoidal output of an AD9850 ?
Thanks in advance.
DDS is usually paired with an LPF in most applications because of the images that it produces along with the fundamental frequency. The images were produced because of the DAC's sample and hold output. Aside from the LPF, I suggest to AC couple the output to reject DC. This can be done through a series capacitor or, at certain cases, with a transformer.Through this, given an interval of time, your output would settle at a a mean value of zero for a purely sinusoidal signal.
Unlike on some of our DDS parts, this doesn't contain an auto phase accumulator feature. What i could recommend is to setup a timer in your microcontroller to precisely measure the time interval of the sinusoidal signal. Then issue the phase transition at that specific time - at the zero crossing.
I have test this application note:
For modulating the amplitude by a cosine filter. but the problème was that on a simple sin output, the 0v is approximative 20mV and then, i have persistant harmonics. maybe have you an idea for get a real 0v?
Olivier - Do I understand correctly that you would like to implement a phase shift, but essentially have that phase shift triggered when the sinusoidal output is either at 0 or 180 degrees? That's an interesting problem. I cannot think of any of our DDS solutions which will automatically provide that. This link Zero-Crossing FSK/PSK Modulator: Analog Dialogue: Analog Devices will take you to a paper which talks about how to utilize a second DDS or second DDS channel to indicate when your primary channel is crossing zero phase, this could help you accomplish your goal
I am glad that this subject can someone interested.
Also, I think you understand where I'm coming. Indeed, a 180 ° phase shift causes many harmonic. To avoid these harmonics, the literature often speaks of a cosine filter that is neither more nor less than the amplitude modulation are between maximum and a minimum and to make the phase shift 0 ° to 180 ° or conversely at this moment.
Unfortunately, as you have noted, it remains the 20mw or 20db compared to the maximum. I see these well 20db between the main lobe and the first lobes armoniques.Out I need at least 50db difference for proper transmitted signal.
If I could do this amplitude modulation in addition to the passage of this phase shift 0V I think I can arrive at a correct result.
The other option I am considering is to modulate the output signal of the DDS with a video operational amplifier that will allow me to reach a value reel 0V.
For information, I modulates the "DDS" by sending an appropriate voltage that is a sine between 0 and 180 ° on the input RSET behind 3,9kohm of resistance. Also, I use only one output which I couple has a high pass filter 1kHz.
So if you have an paper about this, i am interisting!
Thank you in advance.
Retrieving data ...