[SOLVED] LNA matching at 200Mhz

[DavidZlavan]

Hello,

I want to use this LNA -> **broken link removed** for a design. Datasheet shows matching for 0.9GHz but I want to use it at 200MHz.

Which procedure do you use in these cases?

I tried to measure the LNA ADS model without matching and then apply a LC matching similar to the one they show me for 0.9GHz and tried to match that impedance to 50ohms at 200MHz but it didn't work.

Thank you in advance,

 

[mtwieg]

"Normal" procedure would be to select another device which is specified for your operating frequency.

What do you mean by "it didn't work?"

 

[E-design]

The data sheet has S-Parameters you can use for 200 MHz. You will have to be careful since the device is unstable below 600 MHz, so you will need to compensate to get K > 1.0 at 200 MHz.

 

[DavidZlavan]

The device is supposed to work from 50MHz to 3GHz but you have to change the configuration to use it in the frequency you want. The datasheet give you the configuration for 0.9GHz and 1.9GHz. With 0.9GHz configuration it can be used for other frequencies but not for 500MHz and below, poor matching, instable etc.

Would I have to find matching at 50 ohms? Because I remember something about LNAs normally working to another impedance that optimize gain or noise figure at that frequency


Moreover;

it says in page 11/16 mag=0.05 and ang=-137.18, that is -26dB of S11 at 0.9GHz but looking at the S param at pàge 12/16 S11 is -13dB!!!!
Also I simulate the model with this exact same network and S11 is -7.36dB!!!!!!

 

[BigBoss]

The amplifier may principally work but it will not give wanted performance because I guess it's been internally optimized for mentioned frequencies but it's try to worth by proper matching and of course a bit cut-and-try..

The s-parameter of the device only is tabled on page 12 but the matching circuits reflection coefficients have been given in the black-boxes.
The difference is coming from there..

 

[DavidZlavan]

I think I am not clear about my question, excuse me for my english.

First step to design a matching network would be to know at wich impedance it has to be the LNA to properly work at 200MHz and in this case it seems it is not given in the datasheet.

second 20 of this video shows what I mean, the impedances for some frequencies of a LNA, then the matching network is designed at the rest of the video

https://www.youtube.com/watch?v=ZUpuvQqK7ho

 

[DavidZlavan]

View attachment 04012016_resultados mga52589 para 70 to 200 .pdfView attachment MGA53589_70MHz_to_200MHz.pdf

Here the response of the manufacturer and the simulated with ADS model I have.

My conclusions:

It was indeed possible to make it work with different matching values.
It was needed a RC feedback for stabillity
Model still differs from measurements from manufacturer
Matching is not as good as expected and there is a trade off with NF and IMR
If I had to design the Matching network by my own I still don't know how I would do it. Maybe the IC is though to work at 50ohms always and feedback let you work at 200Mhz with stability.

@mtwieg poor matching, about -9dB. I expected about -20dB. Manufacturer solution gives me -8dB so I guess it "worked".

 

[mtwieg]

I think I am not clear about my question, excuse me for my english.

First step to design a matching network would be to know at wich impedance it has to be the LNA to properly work at 200MHz and in this case it seems it is not given in the datasheet.

second 20 of this video shows what I mean, the impedances for some frequencies of a LNA, then the matching network is designed at the rest of the video

https://www.youtube.com/watch?v=ZUpuvQqK7ho

This video shows an example where the source/load impedance are directly given, which makes matching network design easy. The datasheet for your device does not give this impedance, so how are you designing the matching networks? I hope you account for the fact that the device has poor reverse isolation at your operating frequency...