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.

How to keep Bandgap output voltage as constant ?

Status
Not open for further replies.

realtek

Member level 5
Joined
Mar 15, 2004
Messages
89
Helped
1
Reputation
2
Reaction score
1
Trophy points
1,288
Activity points
870
I encounter a big problem
In my Bandgap Circuit
The output voltage Vo will changed whne input VDD changed
I need some special Bandgep Circuit to guarantee when input Vdd change but output voltage Vo only changed in a very little range
Is there any paper or book major in discussing bandgap circuit?
 

The BandGap Output Voltage must not depends on the VDD supply Voltage. It should only depend on the Energy Gap of the Silicium (1.12eV) between Valence and Conduction Band. You should probaly check the PSRR of your amplifier, and maybe cascode some current sources to solve your problem
 

h**p://users.ece.gatech.edu/~rincon/classes/ece4430/handouts/l39_bg.pdf
 

I think is your current mirror has the problem.or your erramp is not good for the input range.
 

The BGR has bad line regulation.
To improve the line regulation one could use cascode or have a pre-regulated supply.
Line regulation is a DC phenomenon, basic aim is to increase the resistance i.e. the resistance of the PMOS device "rds" must be much greater than sum of the diode forward resistance and the resistance in series with it

Added after 2 minutes:

The BGR has bad line regulation.
To improve the line regulation one could use cascode or have a pre-regulated supply.
Line regulation is a DC phenomenon, basic aim is to increase the resistance i.e. the resistance of the PMOS device "rds" must be much greater than sum of the diode forward resistance and the resistance in series with it
I hope that your transistors are in deep saturation when the supply is varied. Also see what is happening to your startup
 

checkout the eda book download forum
there r few bg thesis and papers
 

Sorry. I don't see any relation that bandgap voltage should be with "should only depend on the Energy Gap of the Silicium (1.12eV) between Valence and Conduction Band". Bandgap voltage is called because its zero TC is just around the bandgap of silison.

Check your current mirrors, better you can use cascode current mirrors if you supply voltage is not a problem, for example VDD>3V. Also check your error amplifier, if the gain fails (for example < 50 ~ 60 dB open loop gain) to form a close feedback loop. Check the size of your current mirrors, especially the channel length. DO NOT use short channel devices since we want to avoid the channel length modulation problems. I suppose that you didn't not mean the vaired bg with vdd was tested with process corners. Or there will be other issues.
 

1.check your AMP's PSRR
2.check your AMP's gain when vdd changed
 

Bandgap voltage is called because its zero TC is just around the bandgap of silison. It's true.
 

I think, the most possibility is that the input range of your Ampilfier is not enough for your vdd change range.
 

Check the signal path between VDD and your bandgap output. Check when VDD change, does the transistors works in the right states. Also you can design some regulator if you need a low TC (e.g. below 10ppm), which can help to simplify the problem
 

I am sure ur Line Regulation of reference is not correct! Pls check ur bias circuit ,maybe the bias operation point change with VDD.
 

I agree that it's a line regulation problem, probably due to using small channel lengths that give low output resistance.
 

i have a question
my BGR output voltage followes the source puwer change. It seems that the opamp doesn't work. my all start-up circuit is normally work. what is the reason?
 

if you go over your ckt's formula for output voltage,it should be defined by band gap voltage.If the formula is right that means transistors or BJT or model have some problem
 

reading the books writen by Allen and Gray,expecially the problems in chapter 4 of Allen's
 

current mirror has the problem.or your erramp is not good
 

the output voltage of bandgap reference will change in different corner such as tt,ff,ss. The value of the resistor also affent the out voltage,but it is unimpossible if the TC&PSRR is very low.
 

pk3316 said:
Check the signal path between VDD and your bandgap output. Check when VDD change, does the transistors works in the right states. Also you can design some regulator if you need a low TC (e.g. below 10ppm), which can help to simplify the problem

please explain to me what is ppm (part per million) unit means for TC ??
does TC always use unit mV/C ???
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top