Design Article
Design Type II compensation in a systematic way
Liyu Cao, Ametek Programmable Power
4/4/2012 4:20 PM EDT
(Editor's note: This article looks at a topic which combines a pair of subjects which can strike fear into many designers: power supplies and filters. But love it, fear it, or hate it, it's an important topic.)
Power supplies and Type II compensators are widely used in the control loops for power converters. A type II compensator has two poles (one at the origin) and one zero, and the zero is placed somewhere between the poles. Designers use this type of compensator to provide a phase boost to the control loop. It is known that the compensator reaches its maximum phase boost at the geometric mean of the zero’s frequency and the second pole’s frequency.
To design this type of compensator, the popular approach is to place the desired loop crossover frequency at the geometric mean of the zero and pole. This approach can produce maximum phase margin from a given compensator. However, it does not take the other important loop parameter, the gain margin, into consideration.
In general, to make a control loop work properly, it is necessary to have both enough phase margin and gain margin, In this sense, it is important to have a design procedure that can take care of both design parameters.
In this note, a systematic design procedure is developed to design a type II compensator, which gives the designer an easy way to take both phase margin and gain margin into consideration and to adjust the compensator until the design requirement is satisfied. Due to its length and complexity, "Design type II compensation in a systematic way" is presented as a single Word document, here.
About the author
Liyu Cao is a Sr. Engineer at Ametek Programmable Power, where he is engaged with designing programmable DC and AC power supplies and control loops. Liyu holds a Ph.D. degree in electrical engineering from Tsinghua University, Beijing, China.
Editor's note: Liked this? Want more?
If you are interested in "analog" issues such as signal input/output (sensors and transducer, real-world I/O); interfacing (level shifting, drivers/receivers); the signal chain; signal processing (op amps, filters, ADCs and DACs); and signal integrity, then go to the Planet Analog home page for the latest in design, technology, trends, products, and news. Also, sign up for our weekly Planet Analog Newsletter.
If you are interested in "power" issues such as components; efficiency; thermal concerns; AC/DC and DC/DC supply topologies; batteries; supply ICs; complete supplies; single- and multi-rail management; and supply monitoring: then go to the Power Management Designline home page for the latest in design, technology, trends, products, and news. Also, sign up for our weekly Power Management Designline Newsletter.


Weatherbee
4/11/2012 11:47 AM EDT
Please post a PDF. This .doc file doesn't display correctly in MS Office for Mac. Thanks.
Sign in to Reply
Weatherbee
4/11/2012 11:50 AM EDT
PDF/A is best for these types of documents. Since it is designed for archival all of the Fonts are embedded in the document and there shouldn't be any wierd font problems with viewing on different platforms.
Sign in to Reply
Liyu
7/25/2012 1:45 AM EDT
If you can send a email to liyu.cao@ametek.com, then I would be happy to send you a copy in the PDF format.
Sign in to Reply
sharps_eng
4/17/2012 6:18 PM EDT
Doc opened OK in Pages for Mac.
I worked through a good part of it, and wasn't convinced it presented a general solution?
Not sure i want to spend more time proving that either way. Happy to dig in though if there is evidence the doc has been peer reviewed.
Well done, though, its a good area to work on, there is a real need to simplify these problems.
Sign in to Reply
louP
4/23/2012 11:49 AM EDT
Equation 1 needs to be verified? The C1 term in the denominator should be (C1+C2), where usually C2 much greater then C1 . Which simplifies to C2, not C1?
Sign in to Reply
Liyu
7/25/2012 1:36 AM EDT
Hi,
Here's the article's author. You are right on Equation 1. The C1 term in the denominator should be replaced by C1+C2. Thanks for your comment.
Sign in to Reply