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.

LNA NF post sim result and measurement result different

Status
Not open for further replies.

wccheng

Full Member level 5
Joined
May 16, 2004
Messages
287
Helped
1
Reputation
2
Reaction score
0
Trophy points
1,296
Activity points
2,710
Dear all,

I have designed a LNA which is working at 2.6GHz. It is included the ESD PAD. I have done two extracted view. The first one is whole module extraction included the ESD PAD and PAD ring. The second one is extracted individually. That means LNA has one extracted view, ESD PAD has own extracted view. Then, I simulate the circuit using hierarchy mode. The simulated NF result shows that the whole module extracted post layout is higher than the individual extracted one. The simulated value of the whole module is 3.4dB. The individual extracted one is 2.5dB. The measured one is 4.8dB. Does the pad ring introduce so much noise in the circuit? Or, I should use the power cut and give a independent PAD for the RF input?

Thanks

wccheng
 

The input PAD can increase the NF of the LNA, detuning the optimum NF match, and also because is picking up more noise from the substrate.
 

Hi vfone,

How could I decrease this noise contribution? I have used the rfcmos pdk already.

Thanks
 

On-wafer test or packaged test?

If it is on-wafer test only, you should take all traces on the chip into simulation. These traces should be packed as RF elements.

If it is packaged test, you should also take bondwires, leads into simulation.
 

Dear all,

I have tried to re-do the post layout simulation of whole module. I find that the parasitic makes the noise match and conjugate match value with larger different. If I made noise match, it could give out a better NF value but worse input impedance, and vice versa. In the post simulation, if I did the noise match, the NF=1.8dB and S11=12dB. On the other hand, if I did conjugate match, the NF=3.2dB and S11=22dB. In the post layout simulation, I use RC extraction view and include the bondwire model. (the bondwire model just constructed by myself). The problem appears in the measurement. I did the measurement in package on PCB. I have tried to do the noise match. It just gives me NF=4dB and S11=6.5dB. I don't know why so high NF is gotten. Do you have any idea about this? Or, what kind of things could I improve in next tape out?

Thanks

wccheng
 

Anyone could help me?
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top