Quadrature Modulator problem [hlp]

[No one]

Hello Everyone;
We have a quadrature modulator. In a test, the carrier frequency was 200MHz, and I and Q frequencies about 7MHz (I & Q were taken from two D/A converter). In output of I/Q modulator we had harmonics in 393MHz and 407MHz and ... that were greater than the specs in datasheet. What is the problem?

 

[m.salehi]

Hi
I think that I/Q frequency must be equal to carrier frequency.
but of u is not correct.
Regards.

 

[No one]

m.salehi wrote:

I think that I/Q frequency must be equal to carrier frequency.
but of u is not correct.

But I think that this is not correct, I&Q may be in different frequency respect to carrier.

 

[itzikhaim]

Hi No one !

Becuase your mixer is nonlinear device, it produces (infinite) harmonics of sums and subtractions : (n × fc) ± (m × f in_quad)
In your case :
(2 × 200MHz) + (1 × 7MHz) = 407 MHz
(2 × 200MHz) - (1 × 7MHz) = 393 MHz

Try to bandpass or low pass your output of the modulating mixer !!!

Good luck !

 

[biff44]

Are the specs in the data sheet "guaranteed maximum", or "typical"? If "typical", then there is your answer.

The spurious out of the modulator might be load sensitive. Try passing the modulated output through a 3 dB attenuator first, and then measure spurs again.

 

[melc]

The IQ signals comming from D2A are generated synchronously with the same sampling clock ? If yes, try to low pass filter the I and Q outputs of the D2A near the Nyquist (using analogic or FIR filter in data passed to D2A, or both). This is much easy than to filter after the modulator. If no, synchronise first those two D2A (usually for such purpose there are used dual syncronised D2A).