Browse Prior Art Database

Integration of a zero in a feedback loop compensation

IP.com Disclosure Number: IPCOM000168861D
Publication Date: 2008-Mar-31
Document File: 5 page(s) / 70K

Publishing Venue

The IP.com Prior Art Database

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 60% of the total text.

Page 1 of 5

Integration of a zero in a feedback loop compensation

Summary

The purpose of this invention is to present new loop compensation circuitry that introduces a zero in the feedback loop, without adding a parasitic pole.

Description of the invention

One classical approach to the problem of the stability in a regulation loop is to add a zero, as described in figure 1 here after.

Vout / Vin using Laplace transform

Iout = Gm * Vin Vout = Zout * Iout

+

      - Gm Vin

Iout

Vout

p

ZoutRoutCout *

   1 +

=

Rout

Cout =

    p Cout Rout Zout *

)

*

Zout

1 ( +

*

    p Cout

*

    p Cout Rout
GM
Vin

Vout =

+

= *

)

*

1
( *

    p Cout

Gain

Figure 1. A first classical approach

The addition of a zero (1 + Rout Cout p) in the loop has necessary the drawback of the creation of a pole (Cout * p), which limits the advantage of the zero effect.

Another classical solution is to use an operational amplifier as described in Figure 2 below.

Page 2 of 5

C2

Vout

  ) 1
1
(
1

    + +

C1

R2

Vin

  2 *
1
(
*
2
1
(
*
*
)
2
1
(
*
1

   C C
R

=

R1

    p C
C
R

   p C
R

+

+

   C C

 * )
2
1

p

+

Vin

Vout

Figure 2. A second classical solution

The same drawback appears here also: with a zero added: (1 + (R1 * C1 * p), 2 poles have been created: one at origin (R1 * (C1+C2) * p ) and another defined as
( 1 + R2 * (C1 * C2/(C1+C2)) * p), which can be considered as a local bandwidth limitation. This is generally not desired.

Figure 3 below shows the principle of the proposed new circuitry:

A voltage to current converter is used in a feed back loop.

The equation is very simple: R

     Vin Iout =

If we add a capacitor C in parallel to the resistor...