Application of DSP

Introduction
Digital Signal Processors (DSPs) designed for closed loop control implementations are
extensively used in areas of motor control, uninterruptible power supplies (UPS), and
motion control applications. With the availability of low cost, high performance DSP
controllers featuring high CPU bandwidth and integrated power electronics peripherals such
as, analog-to-digital (A/D) converters, pulse width modulator (PWM) with built-in dead-
time and asynchronous power stage protection, power supply designers have started to
consider this technology as a suitable option for their real-time power conversion and control
applications.
Compared to traditional analog control, DSP controllers provide many distinctive
advantages:
• standard control hardware design for multiple platforms
• less susceptibility to aging and environmental variations
• better noise immunity
• ease of implementations of sophisticated control algorithms
• flexible design modifications to meet a specific customer need
• single chip solution for both control and communication functions
The use of DSPs in power supply applications brings new challenges to many analog
designers in their effort to change the design from the existing analog space to its new digital
environment.
==============================================================
Power Factor Correction:
PFC Stage Hardware Interface to TMS320LF2407A

Figure 1 shows a power factor correction (PFC) stage interfaced to a TMS320LF2407A

DSP. This is an ac-dc boost converter stage, which converts the ac input voltage to a high
voltage dc bus and maintains sinusoidal input current at high input power factor. As
indicated in Figure 1, three signals are required to implement the control algorithm. These
are, the rectified input voltage Vin, the inductor current Iin, and the dc bus capacitor voltage
Vo. The converter is controlled by two feedback loops. The average output dc voltage is
regulated by a slow response ‘outer loop’; whereas, the inner loop that shapes the input
current is a much faster loop.
The instantaneous signals Vin, Vo and Iin, are all sensed and conditioned by the respective
voltage and current sense circuits. The sensed signals are then fed back to the DSP via three
ADC channels ADCIN0, ADCIN1, and ADCIN2 respectively. The rate at which these signals
are sensed and converted by the ADC is called the control loop sampling frequency fs. The
digitialized sensed bus voltage Vo is compared to the desired reference bus voltage Vref. The
difference signal (Vref - Vo) is then fed into the voltage loop controller Gvea. The digitized
output of the controller Gvea, indicated as ‘B’, is multiplied by two other components, ‘A’
and ‘C’, to generate the reference current command for the inner current loop. In Figure 1,
the component ‘A’ represents the digitized instantaneous sensed signal Vin. The component
‘C’ is calculated as,

where, Vdc is the calculated average component of the sensed digitized signal Vin.
In Figure 1, Iref is the reference current command for the inner current loop. Iref has
the shape of a rectified sinewave and it’s amplitude is such that it maintains the output dc
voltage at a reference level Vref, against variation in load and fluctuation in line voltage. The
sensed digitized inductor current Iin is compared with the reference current Iref. The
difference between Iref and Iin is passed into the current controller Gca. The output of this
controller is finally used to generate the PWM duty ratio command for the PFC switch.
==============================================================
Vibration Analysis:
One application of digital signal processing (DSP) is to the study of vibration signals from
mechanical processes. These signals can come from different types of transducers that
commonly have a voltage or current output proportional to the corresponding vibration. The
main characteristics of a vibration signal are its frequency, amplitude and phase angle with
respect to a reference signal. These parameters generally change with time, as occurs during
a frequency sweep test as the excitation frequency is varied across an interval. A signal can
present important phase variations and distortion even in small frequency intervals.
However, with adequate analysis of the signal the characteristics of its different frequency
components can be determined. Commercial tools exist for the estimation of frequency,
phase and amplitude by means of synchronous filtering. This technique consists in analyzing
a vibration signal along with a frequency reference signal. These tools can be overly
sophisticated for the study of simple vibration signals, and can impose strict limitations to
the development of signal processing schemes. For example, they normally do not allow the
customizing of processing functions to meet the user's needs.

Representation of vibration signals

There are different ways to represent vibration signals, one of them is in polar vector form
(Figure 1). In this diagram a phasor of magnitude and phase angle is shown, representing the
amplitude and the phase angle of vibration. When is replaced by the phase angle between
two vectors of the same frequency, these polar plots can be used to represent both parts of
the Bode plots, although frequency cannot be read directly and has to be written separately.

A method based on synchronous filtering is proposed for the analysis of vibration signals
coming from rotating machinery. The block diagram of figure 2 shows the proposed system
to consist of the following stages: sensing, conditioning, acquisition, processing and PC
analysis. In the diagram of figure 2 an electric motor is shown as the system being analyzed.
The validation tests for the developed system are carried out on this motor

The sensing stage is employed to convert mechanical vibrations into electric signals, called
vibration signals in this work. The conditioning stage is formed by an overvoltage protection
and a filter to eliminate vibration signal noise. The processing and acquisition stages are
based on a commercial Bittware Gamma 20 board (which uses an Analog Devices
ADSP21020 signal processor) and the Bittware V2 audio, respectively. These algorithms are
based on detection of: zero crossings of the vibration signal, maximum vibration amplitude
and maximum slope on the reference signal. Detection of zero crossings is accomplished
with a linear interpolation algorithm, capable of determining points between samples. This
algorithm is used with vibration signals with mean value equal to zero. Zero crossings are
used to determine the vibration phase and frequency. The vibration frequency is computed
as the inverse of the period between the two zero crossings that limit an oscillation period.
For the reference signal, the frequency is computed as the inverse of the period between two
maximum slope occurrences. Phase angle is computed using the zero crossings of the
vibration signal together with the points for which a maximum slope of the reference signal
occurs. The period between two maximum slopes is considered as a full vibration cycle. To
help the user in the interpretation of the output from the processing stage, an interface was
developed for the graphical and tabular display of the results. The interface that was
developed as part of the algorithm allows the user to get quickly familiarized with its use.
Also, the graphical display of results makes possible a better understanding of the
algorithm's output.
 ==============================================================

Induction Motor Control:

Traditionally motor control was designed with analog components, they are easy to
design and can be implemented with relatively inexpensive components. However, there are
several drawbacks with analog systems. Aging and temperature can bring about component
variation causing the system to need regular adjustment, as the parts count increase the
reliability of the system decreases. Analog components raise tolerance issues and upgrades
are difficult as the design is hardwired. Digital systems offer improvements over analog
designs. Drift is eliminated since most functions are performed digitally, upgrades can easily
be made in software and part count is also reduced since digital systems can handle several
functions on chip.
Digital Signal Processors go on further to provide high speed, high resolution and
sensorless algorithms in order to reduce system costs. Providing a more precise control to
achieve better consumption or radiation performances often means performing more
calculations, the use of some 1-cycle multiplication & addition instructions included in a DSP
speeds-up calculations. Generally fixed point DSPs are preferred for motor control for two
reasons. Firstly, fixed point DSPs cost much less than the floating point DSPs. Secondly, for
most application a dynamic range of 16 bits is enough. If and when needed, the dynamic
range can be increased in a fixed-point processor by doing floating-point calculations in
software.
Benefits of the DSP Controllers
The performances of an AC induction motor are strongly dependent on its control. DSP
controllers enable enhanced real time algorithms as well as sensorless control. The
combination of both allows to reduce the number of components and to optimize the design
of silicon, to achieve a system cost reduction. A powerful processor such as a DSP controller
does the following:
• favours system cost reduction by an efficient control in all speed range implying
right dimensioning of power device circuits
• performs high level algorithms due to reduced torque ripple, resulting in lower
vibration and longer life time
• enables a reduction of harmonics using enhanced algorithms, to meet easier
requirements and to reduce filters cost
• removes speed or position sensors by the implementation of sensorless
algorithms
• decreases the number of look-up tables which reduces the amount of memory
required
• real-time generation of smooth near-optimal reference profiles and move
trajectories, resulting in better-performing
• controls power switching inverters and generates high-resolution PWM outputs
• provides single chip control system
In the IGBT (Insulated Gate Bipolar Transistor) Module shown in fig 2, has the three diode
bridge rectifier with capacitors which gives the rectified DC voltage to IGBT. The IGBT
Module of rating 25A, 1200V is used which consist of three phase inverter or a four quadrant
chopper which requires six switching devices, six anti- parallel diodes and a switching
device for braking. In addition all six switching devices (IGBT) require gate drive circuitries.

All these require many external interconnections with additional inductance, causing an
additional over voltages across the IGBT. IGBT power module consists of six IGBT, six anti-
parallel diodes and the breaking switch, and all are in single unit, on insulated
semiconductor substrate. Circuitries for detecting over current rise in temperature are also
built-in. Such modules are called intelligent Power Module. A Digital Signal Processor (DSP)
approach implementation of advanced motor drive systems requires the following features
from a typical motor controller. Capability of generating multiple high frequency, high-
resolution PWM waveforms, fast processing to implement advanced algorithms to minimize
torque ripple, on line parameter adaptation, precise speed control etc, Implementing
multiple features using the same controller (motor control, power factor correction,
communication, etc.), making the complete implementation as simple as possible (reduced
component count, simple board layout and manufacturing etc.), Implementing a flexible
solution so that future modification can be realized by changing software instead of
redesigning a separate hardware platform. A new class of DSC controllers has addressed
these issues effectively. These controllers provide the computational capability of a DSC core
and integrate useful peripherals on chip to reduce the total chip count. The DSC family
controller is becoming a viable option for even the most cost sensitive applications like
appliances, HVAC (high voltage alternate current) systems etc. In addition to traditional
mathematical functions like digital filter, FFT (Fast Fourier Transform) implementations, this
new class of DSP’s integrates all the important power electronics peripherals to simplify the
overall system implementation. This integration lowers over all part count of the system and
reduces the board size. The DSC controller DSPIC30F4011 is for motor control applications.
To put DSPIC30F4011 DSC Processor chip into application for motor control, the gate driver
must be used to amplify the logic signals to 15V and powerful enough to turn on and turn off
the thyristors. Because they have different sources and gates at different voltage levels, the
gate drivers (amplifiers) cannot use a common ground. Therefore, the dc power supply for
those gate drivers should be isolated. PWM is widely used in power electronics to
“digitalize” the power so that a sequence of voltage pulses can be generated by the on and
off of the power transistors. The fundamental component has variable magnitude and
variable frequency. Pulse Width Modulation technique is used to generate the required
voltage or current to feed the motor or phase signals. This method is increasingly used for
AC drives with the condition that the harmonic current is as small as possible and the 16
maximum output voltage is as large as possible. Generally, the PWM schemes generate the
switching position patterns by comparing three-phase sinusoidal waveforms with a
triangular carrier. The maximum output voltage based on the space vector theory is double
time as large as the conventional sinusoidal modulation. It enables to feed the motor with a
higher voltage than the easier sub-oscillation modulation method. This modulator allows
having a higher torque at high speeds, and a higher efficiency. For a better understanding of
the space vector process and to represent the switching state of the inverter we define a
switching function Sa for phase A as follows: Sa = 1 when the upper transistor of phase A is
on, and Sa = 0 when the lower transistor of phase A is on. Similar definitions can be made for
phase B and C [6]. The signals Sa, Sb, Sc, controlling the lower transistors, are the opposite of
Sa, Sb, Sc with an addition of dead-bands. Dead-band is the name given to the time
difference between the commutations of the upper and lower transistor of one phase. The
two transistors of each phase are then never conducting at the same time. The aim of the
dead-band is to protect the power devices during commutation by avoiding conduction
overlap and then high transient current. [1] The speed sensor has most common ways to
sense motor speed on the shaft is the use of an incremental encoder and a tachogenerator. In
the case of an encoder, the DSPIC30F4011 includes a module, the quadrature encoder which
perfectly handles the situation and calculates the speed and the direction of the rotation
using only two digital inputs and a 16 or 32 bit internal timer register. There are several types
of tachogenerators, some build a dc voltage proportional to the motor speed, and others
generate a number of pulses per rotor revolution. In the first case, one of the 16 A/D
converters channel of the DSPIC30F4011 is connected to the tachogenerator output. For a
Hall Effect sensor, the pulses enter a capture and a software driver allows the frequency
measurement. The implemented software is called at fixed time intervals no longer than the
minimum period of the measurable frequency. As there is only one signal, it is not possible
to measure the motor speed sign. An interface is inserted to add the sign to this speed
measurement. The speed sign is memorized in a variable and only when the motor speed
goes under a predetermined speed this variable is updated with the current sign. This also
allows the user to execute fast speed reversing cycles of the motor.

In this system, control kernel DSC accept velocity and which is fed by photoelectric encoder,
associating with the space vector algorithm to drive module IGBT, and use keyboard to set
parameter and use numeral tube to display electric-mechanic velocity real time.
==============================================================
Harmonic Measurement:

Electric power is an important and indispensable energy in modern society. However, a large
number of non-linear power and electronic devices that occurred in modern power systems with the
technology development results in high-order harmonic in power-net getting worse. It directly
distorts current and voltage waveform periodically, among which a series of component sine waves
(high harmonics) with frequency multiples of the fundamental frequency. The high harmonics is
called power harmonics. Harmonic detection system is mainly responsible for on-site power data
collection, processing and analysis and uploads them to upper computer based on real needs. It has
to handle a wide variety of tasks: periodic tasks (data acquisition, various calculation etc.) and
unexpected tasks (communication, human-computer interaction, etc.) and at the same time ensure the
instantaneity of the system, so the processing equipment is required to have high data sampling rate
and processing capabilities, but have real-time task scheduling ability, as well as cost. Its design
should include hardware and software.

The Integral Design of the System

At present, the main approaches of harmonic detection system are: acquisition card and IPC-based
implementations, microcontroller-based design method, DSP-based design implementation,
microcontroller and DSP-based design implementation and ARM and DSP-based design realization
and so on. This design uses DSP program. It fully gives play to DSP chip’s digital signal processing
advantage to complete the data processing, storage and communication with the host computer of
the signal, significantly improving the system’s processing efficiency. In this way, it can
appropriately simplify the hardware circuit, the allocation and design of software resource is
relatively independent and the program’s modification and transplantation is easier. The system uses
a DSP chip for data sampling and harmonic detection algorithm, and upload the data to the host
computer that achieves the goal of real-time, comprehensive and high precision monitoring and
management harmonic parameters.

After completing the multi-channel signal detection and through signal pre-processing circuit, DSP is
to conduct analog digital conversion. It controls high-speed A/D to realize high-speed collection for
three voltage and three current channels, then to finish analysis and calculation of the harmonic
power indicators, and finally via RS232 and CAN communication interface to transfer the data to the
PC machine and with the harmonics analysis software to long-term monitor the power network to
identify various problems affecting harmonics. Its monitoring and processing speed is even higher
when adopting the advanced digital signal processors and high-speed A/D, better meeting users’
real-time test needs.

FOURIER BASED METHODS

When harmonic analysis is performed in energy metering or power quality monitoring
systems, phase currents and voltages are sampled simultaneously. Next, they are processed
to compute power quality on the fundamental and harmonic components, including: active,
reactive and apparent powers, rms values, power factors, and harmonic distortion. Fast
Fourier transform (FFT) analysis comes immediately to mind. The procedure, as shown in
Figure 1 , is as follows:
• Determine the period of the fundamental component. This time consuming process is
typically realized by low-pass filtering the phase voltages to isolate the fundamental, and
then measuring the time between consecutive zero crossings. Any error in determining the
period propagates to the error in amplitude and phase of the harmonics.
• Modify the sampling frequency to obtain 2N samples per period. This implies using
analog-to-digital convertors that allow variable sampling frequencies.
• Acquire 2Nsamples corresponding to one or more periods.
• Execute FFT algorithm. Samples taken across multiple periods increase the accuracy of the
computations, but impose a heavier burden on the DSP and a slower overall response.
 ==============================================================
Spectrum Analysis:
Over the past decade, many new digital communications systems have been developed that
are far more bandwidth efficient with respect to the amount of information that is packed
into a given channel bandwidth. These new systems achieve this efficiency by using
advanced digital modulation techniques. These advanced modulation techniques produce
signals that are similar to random noise when viewed in both the time and frequency
domains. The noise-like characteristics of the signal present challenges to the engineers and
technicians working with them when trying to measure the actual signal parameters. The
increased deployment of today’s RF data communications systems using these advanced
digital modulation techniques increases the likelihood that the engineer or technician will
run into situations where power measurement of these digitally modulated signals will
occur. New equipment is being developed to measure and characterize the RF signals
generated by these new digital communications systems. The vector signal analyzer (VSA) is
a new tool that is replacing the spectrum analyzer (SA) as the tool of choice for RF designers
working on these systems. The VSA is a powerful tool that can perform many of the same
measurement and characterization tasks that the SA can, but it can also perform many more
useful digital demodulation. The SA and VSA operate in different manners. These
operational differences can result in measurement errors if these differences aren't properly
considered while making a measurement. One of the important measurements in RF
communications is the spectral mask measurement. This article discusses the equivalent
setup on a VSA to make a spectral mask measurement as traditionally performed on a SA.
The spectrum analyzer measures the power (or amplitude) of the signal vs. the
frequency of the signal. SAs are best suited for measuring tonal signals such as continuous
wave (CW) where the time-varying signal characteristics do not alter the detected power
levels. Two important architectures are used in an SA: the classical
superheterodyne/detector architecture and the FFT architecture. A simplified block diagram
of superheterodyne/detector SA is shown in Figure 1. The signal of interest is presented to
the RF input of a wideband RF mixer where it mixes with a signal from a local oscillator
(LO). The LO input of the mixer is swept over the frequency range of interest. The output of
the mixer is the signal of interest downconverted to a low intermediate frequency (IF) that is
typically 70 MHz. The mixer, being a non-linear device, also produces the harmonic signals
and signals at the sums and differences of the original frequencies and their harmonics. The
output of the mixer is fed to a switched bandpass filter, called the resolution bandwidth filter
(RBW). This filter bandwidth is selectable so that the SA can display different resolution
levels on the frequency axis. This allows the SA the ability to display a small signal next to a
relatively large signal. If any of the mixed signals fall in the passband of the RBW filter, they
are further processed and rectified by an envelope detector. This voltage is amplified by a
log-amp to produce a voltage proportional to logarithm of the RF power for better dynamic
range.
The log-amp output voltage is then fed to a low-pass filter to help eliminate some of
the signal variations over time. This filter is selectable and called a video bandwidth filter
(VBW). (The video bandwidth name results from the fact that this signal was used to deflect
the electron beam in the original analog spectrum analyzer’s video display.)