Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Register Log in

LNA differential as single ended

Status
Not open for further replies.
C

ccw27

Full Member level 5
Joined
Oct 13, 2004
Messages
267
Helped
14
Reputation
28
Reaction score
6
Trophy points
1,298
Activity points
2,558
I am simulating a differential LNA as single-ended by connecting one of the input to AC ground. When running Pnoise I see noise contribution from transistors in the AC ground path. How is that possible?
 

you can use an idea balun to simulate the differential LNA as single ended
then it will just work correctly
 

More infos:

The LNA has common mode rejection or not?
Do you take differential output or single end one?
The noise contribution is relevant or negligible?

Mazz
 

Indeed. If one input is AC coupled to ground, the differential is excited with a differential signal plus a common mode signal of the same amplitude. At one node, both signals have opposite phase and cancel each other out, at the other node both signals have same phase and add. Therefore it is better to use a balun to separate the influence of common mode and differential mode signals.
 

I think I should clarify. I would like to obviate the need of a balun so for a differential input LNA, I connect one of the input to AC ground whereas the other input is matched to 50 ohms. From the measurement I found out that the AC grounded input also needs to be connected to a 50 ohms to get the correct gain. Without the 50 ohms I get Gain of differential - 6 dB, with 50 ohms I get the theoretical Gain of differential - 3dB. Any ideas why I need to connect the AC ground input to a 50 ohms?
 

Maybe your AC grounded input is not correct.
are you sure this is the AC grounded, or it is a true ground?
 

One of the input of the LNA is bypassed using a capacitor to ground, the other one is matched to 50 ohms. The gain is dependent on whether I add a 50 ohms in series with the bypass capacitor. Without 50 ohms I see Differential Gain -6dB, with 50 ohms I see Differential Gain -3dB.
 

I'm sorry, but the question is not clear.
In first post you talk about simulations then you say measurements. Anyway, it is normal to see contribution of noise from the ac ground path if you do in the LNA the single end to differential conversion.
Are you talking about simulations only or also real measurements? Just to understand if we should discuss about possible issues coming from real environment (ocb, grounding ...) or from a math (simulation) point of view.
Mazz
 

I am talking about both simulation and measurement. However, I don't see this in spice simulation. So the gain and NF seem to vary with the impedance connected on the AC ground path. I believe the causes are mutual inductance, coupling between the differential inputs and non-ideal ground. Any more thoughts on this?
 

Simulation level: you have to compare the Gain always in the same way: from input port to differential output.
So, supposed that you have good matching at input port in both cases (differential and single ended) you are connecting a balun with 1:2 ratio that gives you an ideal noiseless 3 dB voltage gain at LNA input, that you don't have if you fix to gnd one of the inputs in single end mode.
Measurement level: yeah you can have poor grounding of the unused LNA input (if it "moves" the Gain is affected). Also a poor Common mode ground can give you wrong results that are less with differential structure.
I hope that (finally) it can help.
Mazz
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top