RF preamp, or no RF preamp?

Question

Frequency downconverters are required to test devices that operate in the range of microwave and mmwave. I have noticed that some companies use a signal chain that has an RF preamp, and some do not, opting to use an IF amplifier to combat the conversion loss of the mixer.

Double sideband mixing is used in these downconverters:

• Option 1: RF signal in → RF preamp → mixer → IF out
• Option 2: RF signal in → mixer → IF amp → IF out

In radio systems, I have never seen a down conversion block without an RF preamp. Is option 2 better?

Block diagrams can be found at down converter vendor sites, like this one from Renesas and this one from Virginia Diodes.

The above links show one example of each.

---

Edit
My original post was edited, and perhaps my intent of the question may be misunderstood. To clarify, why would a preamp not be used, and specifially, not used in a downconverter to make noise figure measurements? I can't see why a downconverter to make NF measurements would not have an RF preamp. The major problem I see is when using a NF analyzer, where you calibrate with no device-under-test, the ENR gets buried by the mixer's high noise figure, casuing a bad calibration.

Answer 1

Which option is better depends entirely on the circumstances, and it's not a simple tradeoff.

The major points that I can think of are

1. You can't get the best possible noise figure (or temperature) going straight into a mixer.
2. If you precede a mixer with an amplifier, you reduce your system's ability to handle strong signals (your 3IM distortion point, and every other measure of distortion handling, gets worse).
3. If you're building a radio system, you've got a lot of motivation to put a preamp at the antenna; you can design your receiver for that.

If you're doing laboratory measurements of sources that are plenty strong enough to overcome the system noise without a preamp, then the decision practically makes itself.

If you're building a system, or selling receivers, then you need to think carefully about all possible tradeoffs, and design what you think is best (and then hope that reality agrees with you).

For the specific case you're citing -- using downconverters in a noise figure test set -- it may be the expectation that in addition to inserting a downconverter into the mix, you would also supply a low-noise amplifier. If the test setup is modular already, this makes sense because you would choose a preamp that matches your downconverter choice, and that is consistent with whatever cost vs. accuracy tradeoff you're making with respect to the noise temperature/noise figure you expect out of the device you're testing.