I'm designing an instrument and I have to provide a programmable sinusoidal signal source from 1 to 180Mhz, using the AD9958 DDS, with a level of about 1Vp-p on a 50ohm load (single ended). Unfortunately I cannot use a RF balun transformer because of heigth constraints and the power supply available is 5 V.
The solution I have in mind is using only one of the DAC outputs terminated with 100ohm resistor, connected to a 7th order Cauer filter (220Mhz / 100ohm), and a wide bandwidth current feedback opamp (AD8000 or OPA691) in non-inverting configuration with the necessary gain. For sure, the input stage of the amplfier is 100ohm terminated.
Do you think it is the best approach to the problem? Do you think that using other values for the termination impedance (e.g. 200ohm) will have negative effect on the performances? (This will allow reducing the gain of the output stage)
I was thinking in an alternative approach using the current feedback opamp (still a single one) used in differential mode, connected to two filters (one for each end) and connected to the differential DAC outputs. My concern (and my understanding) is that the current feedback opamp does not work well on this differential approach.
An improvement to the latest approach is adding a second opamp to implement a better current feedback based differential amplifier solution, but this adds extra cost to the design and I'm not sure if the signal performances will be much better than the initial approach to justify the added cost.