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

Wrong BGR voltage result in a bandgap design

Status
Not open for further replies.
R

rajanarender_suram

Full Member level 4
Joined
Jan 20, 2006
Messages
214
Helped
22
Reputation
44
Reaction score
10
Trophy points
1,298
Activity points
2,375
Attached is the circuit used for the design of BGR(Fig1) & the Output voltage(Fig2)


By using the circuit iam geeting the constant BGR voltage as 1.1718V

But my hand calculation shows the value of 1.22V.

Why is it falling to the value of 1.1718 & what should i have to do to improve this value to 1.2??

The folded cascode amplifier in the circuit is having gain of 70dB..
 

Re: band gap design

You figure shows 1.1718V, not 1.718V.
 

band gap design

I think you should increase Rtrim. It is a scalable resistor, isn't it ?
If so it will solve your problem without affecting the bandgap behavior.
 

band gap design

are the currents flowing in two legs of the BJT the same?
VBG=VEB2+Vtlnn(Rptat+R2+2Rtrim)/Rptat, correct?
why the out node connects back to the gate of M10, what's the vaule of VDD, will M10 turn-on?

Added after 15 minutes:

i think you can do DC analysis, and print the real vaule of each resistors, and see if the resistor ratio is the same as the ratio based on your hand caculation
 

band gap design

It is suspicious that you have a bandgap curve at the wrong voltage. Is the startup circuit completely disconnecting after startup? Try it without the startup circuit and start it using initial conditions to check.

Keith
 

Re: band gap design

banosey said:
I think you should increase Rtrim. It is a scalable resistor, isn't it ?
If so it will solve your problem without affecting the bandgap behavior.
Click to expand...

Rtrim is helping me if its type is different than Rptat R1 & R2.. i.e Rptat R1 & R2 are Ppoly than Rtrim is Npoly resistor. if all are same type than the value of constant voltage is again same. But i didnot understand why???

prcken said:
are the currents flowing in two legs of the BJT the same?
Click to expand...
yes
prcken said:
VBG=VEB2+Vtlnn(Rptat+R2+2Rtrim)/Rptat, correct?
Click to expand...
yes
prcken said:
why the out node connects back to the gate of M10, what's the vaule of VDD, will M10 turn-on?
Click to expand...
M10 wiil turn On .. its connected like this to control the start-up current

prcken said:
i think you can do DC analysis, and print the real vaule of each resistors, and see if the resistor ratio is the same as the ratio based on your hand caculation
Click to expand...
the resistor ratio is slightly lesser then the calculated value

keith1200rs said:
It is suspicious that you have a bandgap curve at the wrong voltage. Is the startup circuit completely disconnecting after startup? Try it without the startup circuit and start it using initial conditions to check.
Keith
Click to expand...

Yes the startup circuit is completely disconnected after starting
 

band gap design

So, you haven't used the same type of resistor for all resistors?

Keith.
 

Re: band gap design

keith1200rs said:
So, you haven't used the same type of resistor for all resistors?

Keith.
Click to expand...

By using Rtrim different from other resistors iam able to push up the BGR voltage to 1.2V
 

band gap design

You shouldn't need to rely on the differential temperature coefficients of resistors to make a bandgap operate at the correct voltage. Something is wrong. What is the source of the design paper?

Keith.
 

Re: band gap design

keith1200rs said:
You shouldn't need to rely on the differential temperature coefficients of resistors to make a bandgap operate at the correct voltage. Something is wrong. What is the source of the design paper?

Keith.
Click to expand...

I got it from the book "Analog Design essentials" Willy sansen Pg 468

Where he says Rtrim is used to trim the value of bandgap to 1.2V. than only possibility i saw is using Rtrim a different resistor than others.(the attached paper made me to think in that direction)
 

band gap design

PTAT current is equal to Vtlnn/Rptat, you can increase VBG with increasing the Rtrim value, i think not necessarily need to use different resistor type for Rtrim.
 

band gap design

In the paper the different temperature coefficient resistor is in a different place to Rtrim in your circuit.

I don't have the book you mention, but the voltage of the bandgap should be close to 1.22V without trimming. Normally the trimming is to correct for mismatches due to processing variations. The nominal voltage with perfect matching should be correct without the trimming.

It might be worthwhile posting your circuit with device sizes & process.

Keith.
 

band gap design

Acctually, I got a 1.18V bandgap voltage by wafer data, but in that case, the PTAT is about 180nA. I think you can check your VBE. Is it equal to your hand calucation value?
 

Re: band gap design

firry said:
Acctually, I got a 1.18V bandgap voltage by wafer data, but in that case, the PTAT is about 180nA. I think you can check your VBE. Is it equal to your hand calucation value?
Click to expand...

Is 1.18V from typical condition? It will be beneficial to us if you can provide the min,max, and typical values of your bandgap voltages.
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top