How to Measure the Bandwidth of an LNA?

  • Thread starter Thread starter Robismyname
  • Start date Start date
  • Tags Tags
    Measure
AI Thread Summary
To measure the bandwidth (BW) of the RF2472G LNA, a Vector Network Analyzer (VNA) like the Agilent/HP 8720ES can be used, focusing on the S21 parameter to determine the -3dB points. While the S21 measurement provides insight into the gain across frequencies, RF LNAs typically have a usable frequency range that extends beyond the 3dB rolloff point, which is not the primary metric for these devices. Input and output matching networks are crucial in limiting the low end of the frequency response to prevent unwanted saturation and maintain gain at the desired frequency. For narrowband applications, the overall circuit's 3dB bandwidth will be defined by external components, while broadband applications may require gain across a wider frequency range. Manufacturer data can also provide valuable insights for characterizing the LNA.
Robismyname
Messages
14
Reaction score
0
My question is two fold.

1) I was asked to measure the BW of a LNA (RF2472G). Is this done with a VNA. I have an obsolete model in my lab Agilent/HP 8720ES.

2) Exactly how would you measure the BW of the LNA in general? Using the VNA? Is this a S11, S12, S22, S21 parameter? please help.
 
Engineering news on Phys.org
I believe it's the -3db points on both side of the S21. This is power wave, it's the BW of the two -3db power point. I am not familiar with the VNA you have, but all you have to do is to plot the S21 in the rectangular form.
 
As youngman stated, bandwidth would be the S21 measurement. However, be aware that RF LNAs generally do not use the 3dB bandwidth as a metric as you would with an op amp. Their frequency range is an indication of the range over which they are usable, and this is generally well beyond its 3dB rolloff point.
For example, your device's advertised range is DC - 6GHz. Its gain at 6GHz is around 10dB and the low freq gain is over 20dB.
 
True, but usually you are measuring the whole circuit including the input and output matching. Or else you just read off the S-parameter file and get the answer, no need for VNA.

In my experience, we always limit the low end in the input/output matching network so you don't get more gain in the frequency range out of the frequency of interest. There's no reason for passing the lower frequency as you don't want to have unwanted frequency driving the LNA into saturation and affect the gain at the frequency of interest. At least that's what I did, making it a bandpass as much as possible.
 
You are right, if this amp is to be used in a narrowband application it will be part of an overall circuit that will have a well defined 3dB bandwidth governed by some external network. (I actually use these parts in broadband applications where I need gain from MHz to GHz). Maybe the OP can chime in, it sounded like he was trying to characterize the part by itself (and, as you mentioned, that data is freely given out by manufacturer).
 
Thread 'Weird near-field phenomenon I get in my EM simulation'
I recently made a basic simulation of wire antennas and I am not sure if the near field in my simulation is modeled correctly. One of the things that worry me is the fact that sometimes I see in my simulation "movements" in the near field that seems to be faster than the speed of wave propagation I defined (the speed of light in the simulation). Specifically I see "nodes" of low amplitude in the E field that are quickly "emitted" from the antenna and then slow down as they approach the far...
Hello dear reader, a brief introduction: Some 4 years ago someone started developing health related issues, apparently due to exposure to RF & ELF related frequencies and/or fields (Magnetic). This is currently becoming known as EHS. (Electromagnetic hypersensitivity is a claimed sensitivity to electromagnetic fields, to which adverse symptoms are attributed.) She experiences a deep burning sensation throughout her entire body, leaving her in pain and exhausted after a pulse has occurred...
Back
Top