AD9854 IQ modulation? IQ modulation IC? ,phase 1 diff then phase 2, phase offset

[SArepairman]

So, I am having trouble understanding something. I am looking for some kind of chip that can operate at relatively low frequencies (0-200 MHz is great!).

I am interested in having it generate two sine waves, one with an offset from the other one. I.E. they are both say 1MHz but one of them has a tuned offset, so one is 0 degrees and the other one could be adjusted from 0-180 degrees. Like figure 8 here:


The AD9854 is a IQ chip, meaning it has two outputs, but is it capable of doing this? Or do I need two DDS chips like the picture shows?
I read the datasheet for the AD9854, it says that in the single tone mode the I and Q outputs are always 90 degrees out of phase but I am unclear if the other modes allow you to adjust the phase offset of Q relative to I.

If it is not capable can anyone recommend a chip that can do this?
ad9854 datasheet: http://www.analog.com/en/rfif-components/direct-digital-synthesis-dds/ad9854/products/product.html

[dannyf]

Not sure about the particular chips but you can program phase offsets for (most/all?) adi dds chips. So two of them will work for sure, assuming that they can go up to 200Mhz - many of them cannot.

[SArepairman]

Not sure about the particular chips but you can program phase offsets for (most/all?) adi dds chips. So two of them will work for sure, assuming that they can go up to 200Mhz - many of them cannot.

right but a single chip?

[cosmos]

I also read it like you say, the I and Q are fixed at 90 deg phase.

It can shift both in time by the same amount (two registers for it even) but the relative phase stays the same.
It can amplitude modulate the I and Q independently so you can reverse the phase relation to -90 deg.  by using negative amplitude on one branch but they are still 90 degrees apart. 

Two circuits as you sketch would do it, and there are then several ways to cause the adjustable shift.

If you add an external analog amplitude modulation to both outputs (n*I and (1-n)*Q , n = -1 to 1 ) and sum them (an upconversion mixer modulated with a DC basically) you should be able to create any phase you desire on the output.
Since you then still have the original I and Q you can do what you asked for.
(if it can fulfil your specs or is simpler than having two chips is another matter)

[KerryW]

They describe it as "Quadrature", which means "90 degrees apart".  They also describe the outputs as "Sine and cosine", which again implies a fixed 90 degree separation.

You might look at the AD9958, it seems to allow phase adjustment between the 2 outputs.

[SArepairman]

They describe it as "Quadrature", which means "90 degrees apart".  They also describe the outputs as "Sine and cosine", which again implies a fixed 90 degree separation.

You might look at the AD9958, it seems to allow phase adjustment between the 2 outputs.

yea i actually found the ad9958 a bit after I made the thread but its a ball buster because it requires 1.8 and 3.3 volts! 
but then using two chips requires external amplitude control on one of the chips.
'

but the more I think about it the more it makes sense that any of these DDS chips should have its own power regulator, so even if I had a 5V chip I would still probably need to give it a independent LDO, so I guess I just need an extra 1.8 V regulator, which is not that bad.


I also assume the AD9958******** allows you to vary the phase of two signals of various frequencies since it has independent cores that are synchronized., which is nice
**made a typo here said ad9854!