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.

question about to disconnect the loop of bandgap please help

Status
Not open for further replies.

tedd

Newbie level 5
Joined
Mar 30, 2007
Messages
9
Helped
2
Reputation
4
Reaction score
1
Trophy points
1,283
Activity points
1,331
AS shown in the attachment is the common used bandgap, i want to simulate ac response to gain PM, when i disconnect the input of amplifier (X OR Y). the transfer function has positive sign and the PHASE of the open loop gain is from 0 to 180 when frequency is incresing. but when disconnect the gate of the PMOS transistors (Z node), the transfer function has minus sign, and the PHASE of the open loop gain is from 180 to 0. It seems incorrect. why does this accurs???
thank you very much [/url]
 

I alredy find out the problem, because Z node is associated with dominat pole, break this node may
causes problem
 

Re: question about to disconnect the loop of bandgap please

But I think you can break Z too,and you can put a equivalent load at the broken point to see the PM,
 

fundamentally, Z can be break to run ac simulation,
but when Z is the dominant pole, break z may lead to inaccuracy in PM and bandwidth. Additionally, it is hard to determine "equvalent" load.
 

Re: question about to disconnect the loop of bandgap please

I think you can get the dc point of the three pmos, and then the dc point can be used to get the equivalent load.
 

Re: question about to disconnect the loop of bandgap please

When breaking the loop, you usually add a large L and a large C (GH/GF) to form a LP filter. So it's connected in DC and disconnected in AC. In this case, it does not hurt to break Z.

tedd said:
fundamentally, Z can be break to run ac simulation,
but when Z is the dominant pole, break z may lead to inaccuracy in PM and bandwidth. Additionally, it is hard to determine "equvalent" load.
 

you gays can refer to MiddleBrook's paper for the problem. In my design, Z is as dominant pole and is connected to a 4pf compensation capacitor ( not show in the attachment), when break this node, where the compensation capacitor can be loacted?? the opamp output side? or the PMOS gate input side?? when you put it at the opamp output side, the PMOS will introdcue difference when comparing with close loop condition.
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top