IQ modulator DC bias problem?

[cybq]

I am now developing a project which involving a IQ modulator. The problem is the modulator don't work when the four I,Q inputs(I+,I-,Q+,Q-)are 1.5V DC bias at the same time. But it's clear that it work with all inputs 1.5V DC bias in the datasheet. Why?

 

[rfsystem]

The effective signal is than

V(I+)-V(I-)=0

V(Q+)-V(Q-)=0

so the modulator should generate no output signal. Only spurious!

 

[toonafishy]

How are you applying the bias to the IQ terminals? If you connect a 1.5VDC power supply to each terminal, the low impedance of the power supply will short out the signal coming from the baseband. You should scale your baseband D/A signal so that the average bit value produces 1.5VDC.

 

[cybq]

I bias the I , Q inputs with resistor voltage divider from 3V power supply.
I think what RFsystem said is right, the 1.5VDC is also the part of modulate diffrence signal.

 

[IICCEE]

why not use a auto adaptor mode?