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.

[SOLVED] multiple reference voltages generation with BGR

Status
Not open for further replies.

goatmxj666

Member level 2
Joined
Mar 5, 2023
Messages
43
Helped
0
Reputation
0
Reaction score
1
Trophy points
8
Activity points
397
Hello,

I designed current-mode BGR in cadence and output voltage is 2V.

I need reference voltages: 0.5V, 0.7V, 1V, 1.5V, 2V

So I wonder how can I make those reference voltages with my BGR?

my plane is just using resistor divider at the output voltage node.

or using 4 current mirror with pmos which is located at output stage of BGR, and adjust mirroring ratio & resistor values for each voltages. (0.5V, 0.7V, 1V, 1.5V)

I would be very appreciated if you give me some tips for how to deal with BGR for actual circuits that need reference voltages, such as 0.5V, 0.7V, 1V, 1.5V, 2V!

Thanks
 

Solution
Since you are using an Current mode BGR, it automatically generates a Vbg/R current.
And by passing this current through an appropriate resistor, you can get the value of reference voltage you want as long as you have the headroom. Vbg* R2/R1

Now the problem you would face is the matching of the current mirrors and the matching of the resistors.
Adding any other things on top of this would just be adding additional things you need to match. And also there would be other things to consider in case of OpAmps such as input offset etc.

If you specifically want a buffered voltage reference, to be able to provide current/charge, then you definitely need a separate opamp /reference buffer.
The bandgap core itself wants to be about 1.2V. You must
have a buffer w/ gain and and feedback ladder to get 2V.

Construct that ladder with fine enough segments and you
can pick taps which make the right voltages. Beware
"cross-regulation" if any real current will be pulled, or
noise injected at those non-1.2V-feedback taps.

If you have those issues then a single bandgap reference and
some plurality of buffers. Might group all the "zero current and
quiet" clients on one divider ladder and give anything filthy, its
own buffer and divider / FB branch.
 

I would say that your current plan is the better plan.
Since it requires matching of fewer devices.

Although, if you plan to route the voltages to far away places, it is better to route it as currents and have a resistor at the destination.
Just ensure that you have kept the same resistor units and orientation at the source and destination for some matching.
 

I would say that your current plan is the better plan.
Since it requires matching of fewer devices.

Although, if you plan to route the voltages to far away places, it is better to route it as currents and have a resistor at the destination.
Just ensure that you have kept the same resistor units and orientation at the source and destination for some matching.
Thanks!

What about using OP AMP and resistors with BGR for higher reference voltage? (>1.2V)

just Like conventional LDO structure but opposite polarity of OPAMP.

How do IC designers use reference voltages: resistor divider, op amp, diode connected MOSFET or function generator voltage?
--- Updated ---

The bandgap core itself wants to be about 1.2V. You must
have a buffer w/ gain and and feedback ladder to get 2V.

Construct that ladder with fine enough segments and you
can pick taps which make the right voltages. Beware
"cross-regulation" if any real current will be pulled, or
noise injected at those non-1.2V-feedback taps.

If you have those issues then a single bandgap reference and
some plurality of buffers. Might group all the "zero current and
quiet" clients on one divider ladder and give anything filthy, its
own buffer and divider / FB branch.
Thanks,

I heard that the current-mode bandgap circuit can obtain the desired voltage value by adjusting the resistance value at the output stage as well as 1.2V. (not voltage-mode BGR)

Am I wrong???
 

Since you are using an Current mode BGR, it automatically generates a Vbg/R current.
And by passing this current through an appropriate resistor, you can get the value of reference voltage you want as long as you have the headroom. Vbg* R2/R1

Now the problem you would face is the matching of the current mirrors and the matching of the resistors.
Adding any other things on top of this would just be adding additional things you need to match. And also there would be other things to consider in case of OpAmps such as input offset etc.

If you specifically want a buffered voltage reference, to be able to provide current/charge, then you definitely need a separate opamp /reference buffer.
 
Solution
And by passing this current through an appropriate re

Since you are using an Current mode BGR, it automatically generates a Vbg/R current.
And by passing this current through an appropriate resistor, you can get the value of reference voltage you want as long as you have the headroom. Vbg* R2/R1

Now the problem you would face is the matching of the current mirrors and the matching of the resistors.
Adding any other things on top of this would just be adding additional things you need to match. And also there would be other things to consider in case of OpAmps such as input offset etc.

If you specifically want a buffered voltage reference, to be able to provide current/charge, then you definitely need a separate opamp /reference buffer.
Thanks for the reply it helped a lot!!!
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top