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.

Fully Differential Amplifier to drive Capacitive Load and Resistive Load

Status
Not open for further replies.

ChetanGK

Newbie level 6
Joined
Dec 17, 2018
Messages
11
Helped
0
Reputation
0
Reaction score
0
Trophy points
1
Activity points
167
Hi everyone,

In continuation of my old question:
https://www.edaboard.com/showthread.php?388562-Down-conversion-Mixer

I am planning to design a fully differential amplifier as a front end amplifier for my sensor application which later drives a passive mixer followed by a passive LPF. I am new to system-level design. Please help me out

1. I know I have to drive a resistive load as voltage mode passive mixer will have switches which imposes ON resistance (switches) which is very low. This resistance will load my opamp and reduces the DC gain. My question is should I design the mixer first and get to know the exact input impedance of the mixer?

2. I read a few papers and Textbooks (Jacob Baker and Razavi) which uses Class AB as an output buffer to drive resistive loads. Is it possible to use Source follower (only 1 CMFB loop) as the output buffer instead of Class AB (needs 2 CMFB loops and also needs floating current source)? And please do suggest output buffer topology.

Fews specifications are Open loop DC gain - 70dB, Capacitive Load (feedback cap) - 3pF, GBW - 150MHz, Closed-loop gain around is 3 to 6dB, Resistive load (unknown as of now), no constraint on power and the critical parameter is noise.

Please let me know your suggestions and thanks in advance.
 
Last edited by a moderator:

1. I would do the same, design a mixer first, and model a drive opamp to see how the OpAmp should behave.
2. I am almost sure it is not true for every class AB fully-differential OpAmp that they require 2 CMFB loops, and sure that not every class AB uses floating current source. I cannot suggest you any architecture until I saw your mixer results. If you can afford any current and/or your load impedance is high enough why wouldn't you design class A topology?
 
Thanks for your reply @Frankrose...

1. I am planning to design a 25% duty cycle passive direct down conversion sampling (NRZ) mixer for Rf range (20K to 20MHz) and LO range (20K to 20Mhz). The initial design steps which I am following is selecting a Cap load of 1pF and then F3db = 1/(2*pi*Ron*Cload) = 10 *Fb , where fb = bandwidth of RF (~ 20Mhz). Then I get value of around 800 Ohms for Ron. is this correct? assumption of 10 times Fb is from this literature: A method for reducing the variation in "on" resistance of a MOS sampling switch.

2. My load impedance of the opamp will be less and hence I cannot use Common source Stage as second stage (Class A). Please correct if i am wrong.
 

Your OpAmp load impedance will be less .... than, how much? Negative feedback around the OpAmp decreases output impedance of the OpAmp, I am not sure it is not enough for you to drive the quasi resistive mixer input impedance with a common source amplifier. You still have to charge and discharge a capacitor in the mixer, which is 1pF, I believe that a class-A OpAmp should handle that if your closed loop gain is maximum 6dB, and you require ~20MHz bandwidth. However, a design of a class-AB common-drain amplifier, or a Monticelli tybe class-AB have many advantages, obviously.
 
I mean by "My load impedance of opamp will be less" is that the impedance seen at the input of the mixer is less (which will be a series combination of Ron||Rop of the complementary switch with Load cap). I fear that low impedance from mixer will drop by DC gain of the opamp. I will try Class A opamp and do some initial output impedance calculation with close loop and let you know.

Meanwhile, I just wanted to confirm whether the above-mentioned design steps for the mixer is the correct starting point?? Or the transistor sizing of switch has to be calculated from other parameters like conversion Gain or any other parameter? I could not find any systematic approach and if there is another way around please do share.
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top