Source Characterization for Power Supply Specification

Last month we introduced this series with the article titled, “Specifying Electronic Power Supplies from a System Point of View: Introduction”.  In this article we explore importance of load characteristics when specifying a power supply.

The figure 1 shows a power supply in simplified form, connected to a generalized load. The feedback and control circuit samples the output voltage, compares it to a reference (not shown), and adjusts the source voltage to maintain the load voltage constant. This process is not perfect, and the power supply specifications describe the deviation from perfection. The deviation from perfection must be within the requirements demanded by the load for the load to operate satisfactorily.

Figure 1: Simplified representation of power supply and load.

The Importance of Knowing the Load Characteristics

The load is the reason for being for a power supply and imposes a major portion of the performance requirements for the power supply. The power supply is never a perfect black box and it is extremely important to treat it as a vital and integral part of your system. It must meet the demands placed upon it by the load in several key performance measures. The most common measures are discussed below:

Static Requirements

Typically, the load requires one of the following parameters to be provided and controlled to within a certain tolerance band: voltage, current, or power. For example, a subassembly might be designed to operate with a controlled input voltage of 5 volts. When excited with a controlled input voltage of 5 volts, the subassembly responds by drawing up to 10 amperes, and consuming up to 50 watts of power. This example illustrates that the power supply must maintain the output voltage at 5 volts and be able to provide up to 10 amperes of current since up to 10 amperes is what the load draws when excited by 5 volts. In this case, power is an alternate way of expressing compliance since power is the same as voltage times current.

Some loads may require different controlled parameters at different times. Such an example is a battery charger which might require constant current for battery charge mode and constant voltage for battery maintenance mode.

Loads will require the controlled parameter to be within a certain tolerance band for proper operation. The power supply must maintain the controlled parameter within the tolerance band.

These parameters may be expressed in terms of average, RMS, or a peak value with a duration qualifier.

Dynamic Requirements

Loads also exhibit dynamic characteristics which change over time.

Time Transients

Many types of loads frequently change their effective impedance. An example might be a computer printer which exhibits rapid step changes in effective impedance. For such a device to function properly, the power supply must be able to rapidly source spurts of output current while maintaining the output voltage within a specified band. This means that the power supply must have enough output capacitance and high enough control loop bandwidth to maintain the output voltage within the prescribed limits. Loads which have this type of behavior must have power supplies specified to limit the droop on the leading edge of the pulse, and recover to within a certain band of the steady state output in a prescribed time interval.

Dependence on Voltage

Non-linear loads change impedance as voltage is increased. One example is a typical solid state circuit which might draw very little current at low voltages, and then begins to draw current with a very rapid and nonlinear increase as voltage is increased.

A configuration which causes greater problems is cascading a power supply with a second power supply of a switching converter design. A switching converter has a nonlinear negative resistance characteristic. At very low input voltages, below the turn-on threshold, the current may be miniscule. When the input voltage is increased to the turn-on threshold, the input current suddenly draws very high current. As the input voltage is increased, the input current decreases, following a constant power characteristic. If care is not taken in the first power supply design and cable length, the load (the switching converter) will cycle on and off because of the voltage drop in the cable length and/or the output impedance of the power supply.

Dependence on Frequency

Loads can create a frequency dependence which is not obvious to the untrained user. This frequency dependence is of at least two forms:

1. Resonant behavior can occur due to inductance and capacitance in both the power supply and the load. Power supplies can resonate with load capacitance or inductance if the power supply is not designed well for the load. This resonance will usually take place at frequencies determined by the reactive elements in the system. This effect is usually undesirable unless the system is designed to be resonant.
2. The power supply control loop behavior can be adversely influenced due to load capacitance and inductance. The presence of substantial capacitance or inductance can move the control loop poles and zeros, substantially changing the transient response and ripple rejection capability of the power supply by decreasing or possible increasing the bandwidth of the power supply.

Dependence on Temperature

Loads must operate in their intended environment. More often than not, the power supplies for these loads must operate in the same environment. These environments may be benign, such as a test laboratory, or severe, as in down-hole oil and natural gas exploration. The power supply must be able to work in the environment of the load, or the power supply environment must be separated from the load to allow satisfactory operation.

Upcoming Topics

This series of articles about specifying power supplies from a system perspective will cover in more detail various aspects of system level concerns. Some of these aspects to be covered are:

• Source Characterization
• Efficiency
• Thermal Environment
• Packaging

Watch for the upcoming articles on “Specifying Electronic Power Supplies from a System Point of View “.

If you need assistance with power electronics design, call or email us today for help with your requirements. You can also go to our power electronics consultant website for more information about our services for business clients. Thank you for reading this tutorial article entitled “Specifying Electronic Power Supplies from a System Point of View: Load Characterization”

The post Specifying Power Supplies for a System: Load Characterization appeared first on POWER ELECTRONICS CONSULTANT.

Power Supplies Serve to Match Load to Source

There are many commonly available sources of electrical power as well as endless types of sources yet to be discovered or developed. We also have countless applications for this power.

However, most applications can not take power directly from the source. There must be a matching process that changes the  source power to useable power for applications. This matching process is the function of the electronic power supply.

Notice that the phrase “electronic power supply” is a misnomer. An electronic power supply does not create power — rather, it simply transforms the available source power to a useful form for your application.

Figure 1 illustrates the relationships between the source and the power supply and the power supply and the load.

Figure 1: Power supply relationsip with the source and the load.

A System Level Approach Needed at the Right Time

Approaching power supply application, specification, and design with a system level perspective is of critical importance for the success of projects. The system includes the source, the load, and even the environment around the power supply. Additionally the system power supply must be treated with the same importance as other aspects of the system design.

The source and load characteristics must be understood before the power supply requirements are determined. The time to evaluate for power supply specifications is right after the initial load and source characteristics are known. It is quite common that a suitable power supply design is not reasonably possible given the initial load and source characteristic determination. If this is the case, either the load or source characteristics need to be changed before proceeding further with the system design. Otherwise, the result will be a failed project or a project with extremely high cost overruns. This problem is a very common mistake that is easily preventable.

A Major Cause of Project Catastrophes

Perhaps a reason for project failures is a belief that electronic power supplies are perfect black boxes, capable of perfectly supplying power to the load. They are not perfect black boxes and failure to consider their limits and adverse behavior can be catastrophic to a project program.

The cost to make make design changes increases exponentially as time in the development cycle progresses. This exponential characteristic happens due to the work that must be done to re-document, re-test, and re-work designs and existing product. This is called the waterfall effect. Problems not found until late in the development cycle can prove to have catastrophic cost consequences. Don’t think that the power supply design can wait till the end! Design power supplies into your system at the right time.

Power Supplies are Multi-Disciplinary and Complex

Electronic power supplies, especially switch-mode type, have very complex analog energy transfer processes and require a diverse skill set to design. The skill set includes not only knowledge of high frequency with high power analog electronics, but also sufficient knowledge of mechanical packaging and heat transfer concepts to prevent the unit from overheating.

Many vendors have simplified switching power supplies for general purpose applications. However, a system knowledge of the source and the load are still important as well as understanding the characteristics and limitation of the power supply itself.

Upcoming Topics

This series of articles about specifying power supplies with a system perspective will cover in more detail the following system level concerns. Some of these concerns are:

• Load Characterization
• Source Characterization
• Efficiency
• Thermal Environment
• Packaging

Watch for the upcoming articles on “Specifying Electronic Power Supplies from a System Point of View”.

If you need assistance with power electronics design, call or email us today for help with your requirements. You can also go to our power electronics consultant website for more information about our services for business clients. Thank you for reading this tutorial article entitled “Specifying Electronic Power Supplies from a System Point of View: Introduction”

The post Specifying Power Supplies for a System: Introduction appeared first on POWER ELECTRONICS CONSULTANT.

Coaxial Power Cable Introduction

A coaxial power cable for DC power transmission has at least the following advantages over other forms of conductors:

• Information content can be combined with DC power from a power supply on a single coax cable saving cost and encouraging simplicity.
• The controlled geometry of a coax cable confines high frequency power supply switching noise components to the space between the center conductor and shield, preventing radiation from the cable and resulting pickup by unintended receivers. The coax cable eliminates undesired electromagnetic radiation.

Since coax cables are not usually thought of for DC power transmission, this article will discuss some important cable specifications pertinent to DC power handling capability.

Coaxial Power Cable Specifications

Dielectric Type – There are two main types of coaxial cable dielectrics: polyethylene (PE) and polytetrafluoroethylene (PTFE) also known by the DuPont trade name of Teflon. The pertinent properties of each are summarized in the following table:

Generally speaking, PTFE is superior in each of the listed properties.

Center Conductor Ampacity – The ampacity of the center conductor is not specified in coax cable data tables for center conductors. However, some tables give the diameter of the center conductor or the center conductor gauge size, from which the ampacity can be derived for a given ambient temperature and coax cable dielectric type. An authoritative table for extensive coax cable data is found in each of the following publications:

• Reference Data for Radio Engineers
• The ARRL Handbook for Radio Amateurs

A useful relationship to remember for coaxial power cable applications is that as conductor diameter doubles, the conductor cross-sectional area quadruples and therefore the DC ampacity quadruples. Derated ampacity can be found from the Wire Parameter Calculator which is based on MIL-STD-975 and using 200C insulation and with the ambient temperature equal to 70C. You must make an adjustment to find the ampacity for use with either PE or PTFE dielectrics or at other ambient temperatures. In determining the relationship for the ampacity adjustment factors, an understanding of the following relationship is needed: conductor temperature rise is proportional to power dissipated in the conductor, and power dissipated in the conductor is proportional to the square of the current in the conductor. Therefore,

the ampacity adjustment factor accounting for the melting temperature of the chosen dielectric is:

and the ampacity adjustment factor accounting for the actual ambient temperature is

Multiplying the above factors together and simplifying gives a consolidated ampacity adjustment factor of:

Therefore, the derated ampacity for a chosen coaxial dielectric and operating ambient temperature is determined by multiplying the ampacity found from the Wire Parameter Calculator for a given conductor size by the consolidated ampacity adjustment factor.

For example, coaxial cable RG-58, Belden part number 8240 has a AWG20 center conductor and a PE dielectric with a melting temperature of 120C per the table above. The Wire Parameter Calculator reports a single wire ampacity for a 200C rated wire at 70C ambient temperature of 6.15A. If the maximum operating ambient temperature is 40C, then the consolidated ampacity adjustment factor calculates to 0.78. The ampacity of RG-58 at 40C operating ambient temperature is 0.78 * 6.15A = 4.82A.

Center Conductor Resistance – Another factor to consider in coaxial power cable applications, especially when the cable run is very long is the resistance of the conductor. Knowing the center conductor gauge size, any wire table or calculator can be entered to find the resistance per length of the conductor gage size. Multiply the resistance per length by the cable length and multiply be two again to account for the outbound and return paths to get the total cable resistance. The current squared times resistance losses must be small compared to the transmitted power for the system to be practical.

Dielectric Strength – If the coaxial power cable application involves high voltage transmission, we should estimate our margin to dielectric breakdown. It is important to realize that generally speaking, dielectric strength figures are given for perfect conditions. Dielectric materials are typically stretched, compressed, dinged, nicked, abrased, etc., leading to degraded dielectric strength. Knowing the dielectric thickness and dielectric strength, the safety margin between operating voltage stress and dielectric breakdown stress can be determined. Safety factors of 10X are not unreasonable.

For example, RG-58, which has a PE dielectric, has a dielectric outer diameter of 0.119 inches and a center conductor outer diameter of 0.0285 inches. Therefore, the dielectric thickness is found to be (0.119 – 0.0285) / 2 = 0.04525 inches = 1.78 mm. With a published dielectric strength of 20kV/mm, the breakdown voltage of the insulator is about 35.6kV. However, prudent design requires a very large safety factor. Using a 10x safety factor results in an maximum operating stress level of 3.56kV. This DC level put us in the same ballpark as the published RMS voltage rating of 1900V for AC applications.

If you need assistance with power electronics design, call or email us today for help with your requirements. You can also go to our power electronics consultant website for more information about our services for business clients. Thank you for reading this tutorial article entitled “Coaxial Power Cable For DC Power Transmission”

The post Coaxial Power Cable For DC Power Transmission appeared first on POWER ELECTRONICS CONSULTANT.