Design Article

Hitchhiker's ride to the world of current mode control

Sameh Sarhan, staff application engineer, National Semiconductor Corp.

7/11/2007 11:13 AM EDT

Hitchhiker's ride to the world of current mode control
If need is the mother of innovation, then money is its father. The rapid growth of the electronics industry, with its miniaturization trend, created the need as well as the money-making opportunities. Semiconductor companies responded by unleashing their mighty intellectual power, each seeking a larger piece of the pie.

As a result, the market flooded with all these fancy chips that claim to do wonders. Although having a variety of solutions is a good thing for customers, it can be confusing. The trick is knowing how to sort through all of those options to ultimately choose the best fit for a given application. Unfortunately, acquiring this knowledge can be challenging, as it may require much time and effort, given the wide variety of these options.

In this article, we will address this challenge when it comes to the different types of current mode control schemes. We will discuss the basic concept behind current mode control, using simple analogies, then, explore the key highlights of the most famous schemes. The goal is to provide technical, but simple information that might help you sort through these different current mode flavors. So, fasten your seat belts, and come join us on this Hitchhiker's Ride to the World of Current Mode Control.

The basics of current mode control

Before hitting the road, it might be a good idea to know a little bit more about our destination. First, let's review the basics: When a constant current flows in a fixed load resistor, it generates a constant voltage. But what will happen if the load resistor changes? To maintain the same constant voltage, the current level has to change as the load resistance varies. That is exactly what current mode control of switching power converters is all about. The idea behind the current mode control is to create a voltage- controlled ideal current source. This current source is programmed to ensure a constant voltage at the output of the power converter regardless of load current changes.

This approach is implemented through two control loops. A current control loop (inner loop), which monitors the inductor current information, creates the voltage-controlled current source. The second loop is a voltage loop (outer loop), which monitors the converter's output voltage and constantly programs the controlled current source to regulate the output voltage at a given set point.

So why bother with this complicated approach to provide voltage regulation? To answer this question, let's remember some more basics about the ideal current source. The current level of the ideal current source doesn't change as its supply voltage changes. It can also be set to any value, no matter how low or how high. Furthermore, the ideal current source can change its current level to a different level instantaneously.

Now imagine what a power converter which utilizes an ideal voltage-controlled current source can do. It would be able to generate almost any output voltage out of any input voltage by setting the proper current level (i.e. no minimum or maximum conversion ratio constraints). Also, the generated output voltage would not change as the input voltage to the converter changes (Ideal line rejection). Moreover, if the converter load resistance suddenly changes, the current source would change its amplitude instantaneously to ensure that the converter output remains regulated at the desired set point (ideal load transient response). If a converter could do all of that, it should be called the ultimate power converter!

Ideal is not real!

Unfortunately, in the current mode world, an ideal voltage-controlled current source does not exist, and the ultimate converter is the converter yet to be invented. For starters, the inner current loop needs to sense the inductor current information then use it to turn-on or/and turn-off the power converter switch(es). Sensing this current information is associated with time delays that would force minimum on-time and/or off-time constraints. These constraints would limit the output voltage range that a power converter can generate at a given switching frequency and input voltage.

In addition, the inner current loop has a limited bandwidth that would relatively slow down its response to a sudden change in the converter load current. Additionally, it suffers from inherent instabilities, generally referred to as sub-harmonic oscillation. If the loop is compensated adequately for these instabilities, it would not be able to respond fast enough to sudden input voltage changes. These imperfections would change the name of the game from "seeking ideal performance" to "seeking the best trade-offs for a given application". This would create a need for different current mode schemes that can offer the best trade-off for different applications. On the next few stops, we will explore some of these schemes.

Current mode control flavors

As we discussed earlier, current mode control schemes sense the inductor current information then use it to influence the decision to turn-on and/or turn-off the power converter switch(es). Generally the scheme would be named based on the type of inductor current information being sensed and/or how the information is used to control the power switch(es). That brings us to our first stop:





solen

11/3/2007 4:46 AM EDT

I want to make a basic power generator, half bridge, zerocross, resonant? for Induction heating, witch control do you suggest.

Sign in to Reply



ilhan

7/2/2008 2:48 PM EDT

try to use zerocross

Sign in to Reply



ledman

7/25/2008 10:55 AM EDT

WHAT WOULD BE THE BEST POWER SOURCE TO POWER CONSTANT CURRENT TO A MCU WHICH WILL BE USED TO FLUCTUATE CONSTANT CURRENT AT LIGHTENING FAST INTERVALS NANOSECONDS AND MILLISECONDS TIMERS?

Sign in to Reply



MrPWM

7/29/2008 1:38 PM EDT

The author writes, "If the loop is compensated adequately for these instabilities, it would not be able to respond fast enough to sudden input voltage changes." This is incorrect. If the compensating ramp in a PCMC system is at 1/2 of the down slope, the current adjusts quickly to the new value, and the only error is a quickly-damped pertubation of the sub-harmonic at half the switching frequency.

Sign in to Reply



ledman

8/7/2008 1:35 PM EDT

is there a device or demo kit that will let me pulse for 200 nanoseconds on 200 off cycles providing constant current? this will power high power leds. if not, what is the fastest possible rate of pulse?

Sign in to Reply



HaloMike

8/26/2011 10:14 AM EDT

Figure 4 has the wrong title.

Sign in to Reply



Please sign in to post comment

Navigate to related information

Datasheets.com Parts Search

185 million searchable parts
(please enter a part number or hit search to begin)

Feedback Form