Electronics Weekly Magazine
Loading

Sign-up for newsletters:

Electronics Weekly newsletters - Sign up for Made By Monkeys, Mannerisms, Gadget Master and Daily and Monthly newsletters

A current mode to tackle large step down ratios

Thursday 30 August 2007 10:11

Many older monolithic buck type switching regulators have incorporated on-chip feedback loop compensation. While this makes for a very simple design procedure, it does not typically allow for optimisation of loop dynamics. This forces the need to select power-path components to suit the feedback requirements, which is not an ideal situation.

Current mode control

Current mode control, a control scheme which reduces the power path to a first order system, drastically simplifies the loop compensation task. However, current mode control is not the answer either because noise sensitivity is often an issue.

A recently developed version of current mode control, known as emulated current mode (ECM) control, dramatically improves the ability to operate with very high step down ratios, while maintaining good noise immunity.

With input voltages up to 75V, it is possible to have significant design margins while still being able to produce the low output voltages used by today's digital circuitry. Loop compensation becomes an almost trivial exercise.

So how does current mode control work in the first place?

The fundamental concept is to turn the power stage into a programmable current source which in turn gets its commanded current level from an error amplifier. The error amplifier looks at the output voltage and varies the current command based on the output voltage's variation from the ideal value.

The usual way to control the inductor current is to measure it, frequently by looking at the switch current, and turn the high-side Fet (control Fet) off when this current reaches the requested level.

Output filters

From the perspective of the output filter - which consists of a capacitor and the load resistance connected in parallel - the inductor is made to look like a programmable current source. Any small signal variations in the error amplifier's output will result in a small signal current variation through the inductor. These small signal current variations flow through the impedance of the output filter network, resulting in small signal voltage variations at the output.

Since the output RC filter is a single order system, the control to output small signal response is also single order, and the result is a system that is very easy to stabilise.
ECM, on the other hand, measures the current in the catch diode just prior to turning the control Fet back on. This information is then captured by a sample and hold circuit that gets gated by the regulator's on-board clock.

The diode current information is held and then the control Fet is turned on. A small current source then begins to charge a ramp capacitor, the value of which has been chosen proportional to the value of the inductor. The charging current is programmed proportionate to the input-output differential voltage.

As such, the resulting ramp voltage on this capacitor has a slope that is proportional to the inductor current slope.

When the ramp voltage is then added to the previously sampled current measurement, the result is a trapezoidal waveform that looks exactly like the control Fet's current waveform, without all the usual non-idealities. This gives ECM the ability to accurately command very narrow on-pulses which is a very desirable trait for a regulator intended for large step-down ratios.

Loop compensation

The latest simple switcher buck regulators take advantage of this ease of compensation and return a degree of control to the user by making the loop compensation accessible. This is a change from previous versions, which although hugely popular, relied completely on internal, factory programmed gain characteristics.

Of course, to truly capitalise on the loop gain flexibility, the operating frequency should be flexible as well so the user can make performance tradeoffs between efficiency, size, and dynamic performance.

For instance, if dynamic performance is desired and efficiency is a lesser concern, then the designer would choose to run at a faster clock frequency, thereby minimising the stored energy in the LC filter and allowing for better transient response.

Craig Varga is applications engineer manager at National Semiconductor

 

Comments powered by Disqus

Share the content

Most Viewed

Products

Related Jobs

Resources