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.

Doubly balanced Gilbert Cell Mixer

Status
Not open for further replies.

shruthi08

Newbie level 5
Joined
Oct 2, 2018
Messages
8
Helped
0
Reputation
0
Reaction score
0
Trophy points
1
Activity points
69
Hello Forum,

I am working on the design of a Gilbert Cell Mixer.

My Goals are:

1. Vdd should be 1V
2. Total current Idd approximately equal to 500µA-1mA
3. Voltage Conversion gain >= 6dB

I have achieved this with the circuit attached below. Inductive degen.PNG

With inductive degeneration at the source of Gain stage, the optimum value obtained after parametric analysis is 15.5nH.

My question for this thread is,

1. Can the length of the transistors be 232nm for a 45nm technology?
2. What effects will the 15.5nH inductor value have on the layout and chip size?

I might be ignorant in a few concepts. Please feel free to correct me.
 

You're doing a mistake by using resistors as load.You loose 0.5V over those resistors and therefore chopping FETs are working in linear region.In other-words, the headroom for chopping transistors is very narrow.Why don't you use a tank circuit resonated at center frequency ( if this mixer is not wideband ) ?? This will bring an extra headroom.
15.5nH Regeneration Inductor value seems to much to me..There micght be some optimization errors.
 
  • Like
Reactions: Badoo

    Badoo

    Points: 2
    Helpful Answer Positive Rating
232nm is a bit funny value, because at 45nm the deviation can be 5-10nm from the litography as I know, I am not sure, but ok. If you can use smaller size probably it could help on a lot of things by the way.
You used 15.5nH ideal inductor. Chip inductor is absolutely not ideal, has a lots of parasitics, and takes huge area, recommended to use it over 1-2GHz, but you didn't mention your input frequencies.
If you use an LNA before this mixer maybe you don't need inductive degeneration, on silicon the matching criteria between the LNA and mixer is not so critical if the distance is small enough between them.
I think switching transistors can operate in triode region, it is not a problem, and with the 500uA current the value and area of them can be low enough. For direct conversion probably this the most suitable way... what you also forgot to mention, how is your output (intermediate frequency, bandwidth).
And you didn't share the noise figure, IP3 or P1dB value either. Very important specifications.
 
Last edited:
  • Like
Reactions: Badoo

    Badoo

    Points: 2
    Helpful Answer Positive Rating
Hello BigBoss,
Thank you for the ideas. I had a strong feeling about the inductors being too high. This is one way to increase the voltage conversion gain as seen in many IEEE papers.
As suggested I will try to use the tank circuit for the load. The reason I included resistor at the load is because the formula for gain is,

Gain formula.PNG
 

Hello Frankrose,

Sorry for not mentioning the frequency of operation. I am designing the Mixer at 2.4GHz. I will take all the points you have mentioned and work on designing a better mixer.

Thank you
Regards
 

Why you use an inductor as the degeneration?
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top