Balanced amplification increasing noise of LNA?

I have found in another thread that In real life the NF of a balanced amplifier is approximately 25% greater than the NF of a similar single amplifier.
Now considering the building of an LNA for receiving purposes (radio astronomy) is it good to use such a configuration or should I go for a single source follower?

Any proposals for low noise configuration on 1.320-1.520GHz? (1.420 middle of the band)

Higher NF is due to the extra insertion loss added by the input balun transformer.
Balanced amplifiers have advantages (and disadvantages) compared to single amplifiers. Just search and find which is your best fit for application.

I am wondering if I use a wilkinson divider/combiner, will the noise be improved in comparison with a hybrid 90 ?

Two way Wilkinson have 3 dB insertion loss.

Here you can find some low insertion loss baluns:
Anaren Microwave - microwave components, subassemblies, LTCC, ceramic processing, Wireless, Space, and Defense Electronics, Amitron

Hi, vfone,

The Anaren low insertion loss Balun is also more than 3dB. the real insertion loss from anaren should be 3dB + the insertion loss from anaren datasheet.

Yes, you can use wilkinson divider/combiner. The noise will be improved very little (0.01dB) compared to hybrid 90.

If your power divider insertion loss 3.5dB, the additional insertion loss is 0.5dB. So your system NF will increase 0.5dB. Keep additional insertion loss as low as possible.

Anaren baluns from the link that I posted have low insertion loss (below 3dB). Check before any replay.

Balanced amp have gained popularity in I.C. RF receiver on a chip to combat intra-chip spur coupling by using their common mode rejection and reduced ground loop problems.

If you are doing a discrete LNA you are better off with single ended design.

Look at some of the discrete LNA chips targeted for GPS receivers. You should be able to get some in the 0.5 db NF range.

I think the insertion loss said by vfone not include 3dB of power divider by 2; but the insertion loss by me include 3dB. It's just interprete differently. Attached is an Anaren datsheet 3A512, the insertion loss from datasheet is 0.3dB.