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

Basic question about pole frequency

Status
Not open for further replies.
P

palmeiras

Full Member level 6
Joined
Feb 22, 2010
Messages
375
Helped
61
Reputation
122
Reaction score
50
Trophy points
1,308
Location
South America
Activity points
4,199
Hi guys,

Please, consider the attached circuit.
Based on the Miller theorem, C3 (miller capacitor) can be divided in two capacitors connected to ground. The first one (CX) is connected from node 2 and ground, and the second (CY) one is connected from node 3 (vout) and ground.

The capacitor CY is given by C3/(1-(1/AV)). Since this circuit has large gain, CY ~ C3.
Therefore, if we consider that R2 << ro, we can calculate the pole frequency at node 3:

f = 1/(2*pi*R2*CY) ~ 8kHz.
R2 = 2k
CY = 10n

However, the bode plot of my simulation show that the high frequency corner (-3 db) frequency is about 20 kHz. (NOT 8 kHz). Why?
Why is there such discrepancy?

Thanks
 

Attachments

  • -3dB_corner_frequency.png
    -3dB_corner_frequency.png
    54.6 KB · Views: 68

The assumption CY ~ C3 is correct, but you need to look what happens on the input side, you can't ignore CX.

Increase C1 by a factor of 10 for trial.

In a real circuit, it's unlikely to have zero source impedance.
 
Thanks, FvM!!!
Why should I look what happens on the input side?
I mean, why does (CX) affect the output pole?
The output pole only depends on the output capacitor, doesn’t it?
Since the simulated -3db frequency is a about 20 kHz, it is like CY is lower than C3 (and I did not understand that). Because CY ~ C3.

- - - Updated - - -
 
Last edited:

The output pole isn't changed, but there's a lead-lag element in the overall frequency characteristic compensating it. Simple miller theorem with assumed constant gain doesn't help well to understand the circuit behavior.

Instead calculate the complete transfer function or look also at the base node in the simulation. Increasing C1 can help you as a first step.
 
FvM.
You are completely right! I have increased C1 and the -3dB corner frequency became the one I have expected before, when I have applied the Miller theorem. ( about ~ 8 kHz).
Why did the pole move to that one we expected? What did you have in mind when you gave your suggestion?
Is it related to this lead-lag element (I don’t know what it is) ? Could you, please, clarify?
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top