About how to roughly measure LNA NF contour - RF Cafe Forums

RF Cafe Forums closed its virtual doors in late 2012 mainly due to other social media platforms dominating public commenting venues. RF Cafe Forums began sometime around August of 2003 and was quite well-attended for many years. By 2012, Facebook and Twitter were overwhelmingly dominating online personal interaction, and RF Cafe Forums activity dropped off precipitously. Regardless, there are still lots of great posts in the archive that ware worth looking at. Below are the old forum threads, including responses to the original posts. Here is the full original RF Cafe Forums on Archive.org

-- Amateur Radio

-- Anecdotes, Gripes, & Humor

-- Antennas

-- CAE, CAD, & Software

-- Circuits & Components

-- Employment & Interviews

-- Miscellany

-- Swap Shop

-- Systems

-- Test & Measurement

-- Webmaster

duplicity
 Post subject: About how to roughly measure LNA NF contour
Posted: Mon Apr 19, 2010 11:05 am 
 
Lieutenant

Joined: Mon Apr 19, 2010 10:51 am

Posts: 2

Hi RFCafe senior engineer,

Good Day.

I have a serious issue about LNA NF measurement, and really need your kindly support.

What I want to do is try to draw LNA NF and Gain circle by measurement, and compare with simulation.

Attach related file, especially NF circle.

Image

So, I set up a simple measurement, as shown below:

Here, I got several problem,

1. What value of inductance I should use with power choke inductor, refer to application circuit?

2. How I define output M.N.? refer to application circuit?

3. If I do a simple output conjugate matching, should I turn on LNA, if yes, is power choke a part of matching network?

4. Here comes important problem,

4.1 What I can measure is total system(50ohm) NF, and I want to determine LNA's NF by equation.

4.2 For special case, that's Гin1= Гout1= Гin2, the FriiS equation is O.K.

4.3 But actual situation is, Гin1≠ Гout1≠ Гin2, right? So how to modify FriiS equation.

Image


 
  
 
Itay
 Post subject: Re: About how to roughly measure LNA NF contour
Posted: Tue Apr 20, 2010 3:19 pm 
 
Lieutenant

Joined: Tue Apr 20, 2010 2:35 pm

Posts: 1

Hello duplicity,

I will try to answer on some of your questions:

Quote:
1. What value of inductance I should use with power choke inductor, refer to application circuit?

The inductor value should give you a reactance of several hundreds ohms (at least 300 ohms) in your lowest frequency of operation. remember that as the frequency increases, XL increases as well.

In addition, you should choose the inductor such that its SRF (Self-Resonance Frequency) would be at least twice higher than your maximal frequency of operation in order to avoid a capacitive reactance.

Quote:
How I define output M.N.? refer to application circuit?

Usually for LNA, you define the output matching network as one for conjugate match. This is because the input matching for noise figure degrades the return loss at the input and reduces the overall gain (mismatch), and as a result you loose power.

Quote:
If I do a simple output conjugate matching, should I turn on LNA, if yes, is power choke a part of matching network?

You should design the output matching only after your input mtaching network is fully designed and connected to the LNA, because the input matching network will affect your S22. The reason for that is that the amplifier is not unilaterlal, i.e. there is a certain feedback from the output to the input S12.

The choke (power inductor as you named it) if properly chosen, should slightly influcnece on the output mathcing network. Actually, it isolates between the supply voltage to the matching network. Usually from RF signal point of view, it only slightly increases the gain.

So your output matching newtok should be designed to the conjugate of the modified S22 (With the influence of the input matching network).

You should turn on the LNA, in order to measure the output power.

____________

In general, when designing matching networks, pay attention to the stability of the amplifier.

It is very important to check that your required reflection coefficients are located within the stable region.

You can perform stability simulation or calculate K (Rollet's Factor). For unconditional stability K>1


 
   
 
duplicity
 Post subject: Re: About how to roughly measure LNA NF contour
Posted: Thu Apr 22, 2010 10:56 am 
 
Lieutenant

Joined: Mon Apr 19, 2010 10:51 am

Posts: 2

Quote:
In addition, you should choose the inductor such that its SRF (Self-Resonance Frequency) would be at least twice higher than your maximal frequency of operation in order to avoid a capacitive reactance.

I will kind in my mind, thanks for your lesson learn.

But actually, the reference design is around 3nH(LQP03TN3N0B04), whose SRF is over 6GHz, and impedance is only 38.46ohm at 1990MHz.

Image

Quote:
You should design the output matching only after your input mtaching network is fully designed and connected to the LNA, because the input matching network will affect your S22. The reason for that is that the amplifier is not unilaterlal, i.e. there is a certain feedback from the output to the input S12.

But how about isolation?! The input M.N. should not influence S22 load.

How i tine the S22 M.N. is tep by:

1. S11 should be ready first.

2. tune choke to get higher gain.

3. tune second componet for conjugate matching.

Quote:
You can perform stability simulation or calculate K (Rollet's Factor). For unconditional stability K>1

Image

Thank you for your kindly reply.


 
   
 
IR
 Post subject: Re: About how to roughly measure LNA NF contour
Posted: Thu Apr 22, 2010 2:57 pm 
 
Site Admin
User avatar

Joined: Mon Jun 27, 2005 2:02 pm

Posts: 406

Location: Germany

Quote:
But actually, the reference design is around 3nH(LQP03TN3N0B04), whose SRF is over 6GHz, and impedance is only 38.46ohm at 1990MHz.

3nH to my opinion is too low. The SRF is OK.

From my experience: I designed an amplifier for 2.7GHz and I chose a choke of 8.2nH.

Quote:
But how about isolation?! The input M.N. should not influence S22 load.

How i tine the S22 M.N. is tep by:

1. S11 should be ready first.

2. tune choke to get higher gain.

3. tune second componet for conjugate matching.

As I mentioned, the isolation is not infinite, i.e. the amplifier is not unilateral (ideal).

There is S12 which has a low but not negligible value.

I would perform the following steps - simulation only.

1) Design the input matching network for the required gain and NF (You have to find the intersection point of the desired Noise and gain circles) with the output matched to 50-ohm.

2) After that, plot S22 of the amplifier (With the input matching network).

3) Design the output matching network for congugate match. Also add the choke value and see how it affects the output matching and gain.

4) Tune both the input and output networks for optimal performance.

_________________

Best regards,

- IR

Posted  11/12/2012