OMAP L138 problem application

Said,

Just an idea:  if your sine wave is above the fc for the DAC's anti-aliasing filter, it will be attenuated.

Frank,

Thank you for your reply , I thought it was this in the beginning but in the datasheet of the DAC nothing mentions this .

And the amplitude of my signal increases with the frequency :

It starts with 180 mV for 10KHz up to 1V for 100KHz and it saturates to 1.5V for the high frequency ( from 200KHz ).

Said,

What DAC are you using, and what's your sampling frequency?

Donald

Donald,

I am working with the Omap L138 kit which includes the User interface board with the DAC ( DAC900 10 bit ) and the sampling frequency is 165 MSPS , but in the configuration of the uPP ( Universal parallel port ) I can configurate this frequency and in my project I work with 7.5 MHz..

Said.

Said,

Thanks. I'm not familiar with that DAC but according to the datasheet it has a current output. Could it be that the circuit you are driving has a frequency response that is giving the results you describe? i.e. different output amplitude at different frequencies

In another thread you discuss generating a sine wave and show a 'scope trace. It appears that you are trying to output just one cycle and are getting spikes in the output. It's a little hard to see from the 'scope trace but it looks to me as if the spikes might be quite normal and expected transient responses where your sinewave starts and stops abruptly. The DAC is quite likely to have a slightly underdamped transient response and this might be exacerbated by the circuit that you are driving. For a continuous sinewave output, or other waveforms that do not contain significant discontinuities, you wouldn't be troubled by the transient response. If you tried to generate a squarewave you'd probably see the 'overshooting' transient response most clearly.

HTH

Donald

Thank  you Donald for your reply ,

Yes of course the  DAC's output  is a current but there is a Transformer in the board which convert Current/Voltage , but when I check the data (byte per byte in the buffer ) it's always  between 0 and 1024 (10 bits) for all the frequencies.

You are right for the square wave , when I try to generate it  I have two peaks so I think it's a problem of the ' overshooting' as you mentioned.

But really I don't know how I can resolve this problem because I worked with the C6713's DSP and I didn't have those problems , I think the problem might come from the configuration of the uPP.  

Said

Said,

Another idea (FWIW):  was the DAC and DAC's output circuit for the C6713 implementation the same as the output circuit for the L138 implementation?  Maybe there is an impedance matching problem, resulting in ringing.

Frank ,

The implementation is completely different and also the functioning between the two DAC is not the same ,  the C6413's DAC has a series transfer data and the Omap's DAC has a parrallel transfer data .

Here is the schematic around the DAC , may be some components have an effect on the DAC .