# Signal Chain Basics #96: Active Loop Filter Designs

Editor’s note : Dean Banerjee, Application Engineer, Texas Instruments, is our guest blogger this month. He is an applications engineer for TI’s Signal and Data Path Solutions Business Unit. He has been involved with phase-locked loop (PLL) frequency synthesizers for over 17 years. Dean has also authored two books: PLL Performance, Simulation, and Design, and From Continuous to Discrete. He holds a master’s degree in applied mathematics from the University of Illinois and an MSEE degree from Southern Illinois University.

Introduction
At the heart of a clocking device is the phased-locked loop (PLL). A key part of the PLL is the loop filter, which converts correction currents from the PLL charge pump into a control voltage for the voltage-controlled oscillator (VCO). For reasons of cost, noise, and size, the most effective implementation of the loop filter is usually with passive components. However, some situations warrant the use of active devices in the loop filter. This is to get more tuning range for the VCO, isolate the VCO input capacitance from the loop filter, monitor the tuning voltage, or isolate the loop filter from excessive VCO leakage currents. For these situations, using the approach in Figure 1 allows the charge pump voltage to be centered for optimal performance. It also puts fewer demands on the input speed of the operational amplifier (op amp), and has a pole after the op amp to attenuate its noise.

Figure 1

Active loop filter.

Choosing a good op amp
Table 1 shows some important op amp characteristics and their impacts on active filter designs.

Table 1

Non-ideal op amp characteristics.

Table 2 shows some op amps from TI with different characteristics that might be good to consider for active filters.

Table 2

Some possible TI op amps to consider.

Active filter design principles
Calculating the filter zero (T2) and poles (T1 and T3) for the filter in Figure 1 is beyond the scope of this article, but is provided in PLL Performance, Simulation, and Design . For the pole T3, it is good to choose this to be larger than T3 as this pole rolls off the op amp noise. Once these three-time constants are calculated, there are still two more degrees of freedom, since there are five components. This allows the designer to specify components R1 and R3 (Table 3) .

Table 3

Choosing components R1 and R3.

Figure 2

Phase noise for active and passive filters of the same bandwidth.

Results
Figure 2 compares an active and passive filter of the same bandwidth. The active filter has some degradation near the loop bandwidth due the voltage noise of the op amp. At lower offsets, there is also some noise degradation, probably because the op amp does not react fast enough to the fast correction pulses of the charge pump.

Acknowledgement
A special thank you goes to Hooman Hashemi of Texas Instruments for his help and insights on op amps portion of this article.

Join us next time when we will discuss why faster is better: GSPS ADCs enable wide bandwidth RF digitizers.

## 2 comments on “Signal Chain Basics #96: Active Loop Filter Designs”

1. xiaonanhai
December 12, 2014
 This series laser driver is designed for driving diode lasers with low noise current at high efficiency, > 76%. The output current of laser drivers can be set either by an analog voltage ranges from 0 to 2.5V or by a potentiometer to from 0 to 20A, 30A, 40,45A.  analogtechnologies.com/High_Efficiency_AC_Input_Laser_Drivers
2. samicksha
December 12, 2014

When we discuss on driving of Laser Diodes ,I believe it is little tough to design circuit for producing constant regulated voltage.

This site uses Akismet to reduce spam. Learn how your comment data is processed.