INTRINSICALLY SAFE (IS) ACTIVE POWER SUPPLIES by Mark Edward Walpole Assoc. Dip. Elec. Eng., B Eng. (Hons.) Submitted for the Degree of Master of Engineering (research) Queensland University of Technology Faculty of Built Environment and Engineering School of Electrical and Electronic Systems Engineering Brisbane March, 2003
128
Embed
INTRINSICALLY SAFE (IS) ACTIVE POWER SUPPLIESeprints.qut.edu.au/15896/1/Mark_Walpole_Thesis.pdf · intrinsic safety, intrinsically safe ... equivalent circuit, intrinsic safety Standards,
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
INTRINSICALLY SAFE (IS)
ACTIVE POWER SUPPLIES
by
Mark Edward Walpole
Assoc. Dip. Elec. Eng., B Eng. (Hons.)
Submitted for the Degree of
Master of Engineering (research)
Queensland University of Technology
Faculty of Built Environment and Engineering
School of Electrical and Electronic Systems Engineering
Brisbane March, 2003
Keywords (ii)
Mark Walpole, Intrinsically Safe (IS) Active Power Supplies, M.Eng. Thesis, QUT 2003
intrinsic safety, intrinsically safe, active power supply, modelling, equivalent circuit,
Chapter 2 Review of IS Power Supplies and Intrinsic Safety..6
2.1 Evolution of Intrinsic Safety .........................................................................6
2.1.1 Mechanism of Electrical Arcs .....................................................................7 2.1.2 Mechanisms of Ignition...............................................................................7 2.1.3 Energy Transferred from the Electric Arc ...................................................9 2.1.4 Development of the Principles of Intrinsic Safety .....................................11
2.2 IS Power Supplies .....................................................................................13
2.2.1 Evolution of IS Power Supplies ................................................................13 2.2.2 Modern IS Power Supplies .......................................................................14 2.2.3 Design Methodologies of IS Power Supplies............................................15
2.3 Types and Terminology of IS Power Supplies ...........................................17
2.3.1 Three Types of IS Power Supplies ...........................................................17 2.3.2 Definition of IS Power Supplies Terminology ...........................................19
2.4 IS Active Power Supplies...........................................................................21
2.5.1 Current Australian and International Standards .......................................25 2.5.2 Comparison of AS 2380.7 and AS/NZS 60079.11 ...................................27 2.5.3 Participants in Ensuring Intrinsic Safety ...................................................27 2.5.4 Accredited Intrinsic Safety Testing and Certification Bodies ....................29
2.6 Certification, Assessment and Testing of IS Power Supplies ....................31
2.6.1 Certification – Determining Conformance to a Standard ..........................31 2.6.2 Assessment of IS Active Power Supplies.................................................32 2.6.3 Testing IS Active Power Supplies using the STA .....................................34
4.3.1 Measuring Transient Characteristics using the STA ................................57 4.3.2 Measuring Transient Output Characteristics using a Relay .....................60 4.3.3 Limitations in Measuring Transient Output Characteristics ......................62
4.4 Transient Characteristics of Sample IS Active Power Supplies................. 64
Group I capacitive circuits ...........................................................114
Group I inductive circuits .............................................................115
List of Tables (viii)
Mark Walpole, Intrinsically Safe (IS) Active Power Supplies, M.Eng. Thesis, QUT 2003
Tables
Table 2-1: UK. National Coal Board DC IS power supplies [3] ................................13 Table 2-2: Summary of SIMTARS recommended design methodology [9]..............16 Table 2-3: Definition of active and passive power supplies......................................19 Table 2-4: Definition of linear and non-linear power supplies ..................................19 Table 2-5: Defining the types of IS power supplies ..................................................20 Table 2-6: Maximum values of V and I for Group I active power supplies [11] ........22 Table 2-7: Relevant Acts and Regulations [17] ........................................................25 Table 2-8: Summary of SIMTARS intrinsic safety assessment procedure [9]..........33 Table 2-9: Summary of SIMTARS intrinsic safety testing procedure [9] ..................34 Table 4-1: Measured steady-state parameters – sample active power supplies .....55 Table 4-2: Measured transient parameters – test circuit with STA...........................59 Table 4-3: Measured transient parameters – test circuit with a relay.......................61 Table 4-4: Instantaneous voltage and current for inductors and capacitors.............63 Table 5-1: Component equations for the RLC equivalent circuit model ...................73 Table 5-2: Experimental RLC equivalent circuit model – component values ...........74 Table 5-3: Component equations for the RC equivalent circuit model .....................80 Table 5-4: Experimental RC equivalent circuit model – component values .............81 Table 6-1: Measured transient parameters of sample active power supplies ..........93 Table 6-2: PAAM calculating component values (RC equiv. cct. model) - PS 1 ......95 Table 6-3: PAAM component values (RC equiv. cct. model) for PS 1, 2 and 3 .......96 Table 6-4: Comparison of results - PAAM vs. STA testing.....................................100
List of Figures (ix)
Mark Walpole, Intrinsically Safe (IS) Active Power Supplies, M.Eng. Thesis, QUT 2003
Figures
Figure 2-1: Energy system of an electrical arc ...........................................................8 Figure 2-2: Ignition kernel growth vs. ignition energy and quenching distance ........10 Figure 2-3: Power supply circuit topologies and their V-I characteristic [12] ............18 Figure 3-1: Plan and elevation views of STA wire holder and cadmium disk [24] ....38 Figure 3-2: Oblique view of the STA wire holder and the cadmium disk ..................39 Figure 3-3: STA wire and cadmium disk making and breaking contact....................40 Figure 3-4: STA making contact - discharging a capacitive circuit ...........................42 Figure 3-5: Test circuit with STA and wire path for a single traverse .......................43 Figure 3-6: Measured output current (IO) and voltage (UO) for a single traverse......44 Figure 3-7: Periodic make and break of wires on the cadmium disk ........................45 Figure 3-8: Measured periodic make and break waveform ......................................45 Figure 3-9: Geometry of arc scribed by the wire on cadmium disk ..........................46 Figure 3-10: STA electrical circuit.............................................................................48 Figure 3-11: STA calibration circuit with current measuring resistance....................50 Figure 3-12: Measured V and I waveforms for the STA calibration circuit ...............51 Figure 4-1: Block diagram of sample IS active power supply DC stage...................53 Figure 4-2: Steady-state test circuit..........................................................................54 Figure 4-3: Steady-state output characteristics ........................................................55 Figure 4-4: Transient characteristics test circuit with STA........................................57 Figure 4-5: Measured transient output characteristics (STA) ...................................58 Figure 4-6: Transient characteristics test circuit with a relay....................................60 Figure 4-7: Measured transient output characteristics (relay) ..................................61 Figure 4-8: Power supply output capacitance – external discharge path .................63 Figure 4-9: Active power supply NL to FL transient characteristics..........................64 Figure 4-10: Active power supply FL to SC transient characteristics .......................65 Figure 4-11: Active power supply NL to SC transient characteristics.......................66 Figure 4-12: Active power supply NL to SC transient characteristics.......................68 Figure 5-1: PAAM - RLC equivalent circuit model topology .....................................72 Figure 5-2: Experimental RLC equiv. cct. and steady-state characteristic ...............75 Figure 5-3: Experimental RLC equivalent circuit – transient tests............................75 Figure 5-4: Over damped RLC equiv. cct. NL to SC transient characteristics..........76 Figure 5-5: Over damped RLC equiv. cct. SC to NL transient characteristics..........77 Figure 5-6: Under damped RLC equiv. cct. NL to SC transient characteristics........78 Figure 5-7: Under damped RLC equiv. cct. SC to NL transient characteristics........79 Figure 5-8: PAAM - RC equivalent circuit model topology .......................................80
List of Figures (x)
Mark Walpole, Intrinsically Safe (IS) Active Power Supplies, M.Eng. Thesis, QUT 2003
Figure 5-9: Experimental RC equiv. cct. and steady-state characteristic.................81 Figure 5-10: Experimental RC equivalent circuit - transient test circuit ....................82 Figure 5-11: Measured RC equiv. cct. NL to SC transient characteristics ...............82 Figure 5-12: Measured RC equiv. cct. SC to NL transient characteristics ...............83 Figure 5-13: Illustration of ignition curve safe and unsafe areas..............................86 Figure 6-1: Measured transient output current response for PS 1 ...........................95 Figure 6-2: PAAM RC equivalent circuit model for PS 1 ..........................................96 Figure 6-3: PAAM ignition curve plots for PS 1, 2 and 3 [24] ...................................97
Abbreviations (xi)
Mark Walpole, Intrinsically Safe (IS) Active Power Supplies, M.Eng. Thesis, QUT 2003
Selection, installation and maintenance Part 1: General requirements
• AS 2381.7 – 1989 Electrical equipment for explosive atmospheres -
Selection, installation and maintenance Part 7: Intrinsic safety i
• AS/NZS 60079.0:2000 Electrical apparatus for explosive atmospheres
Part 0: General Requirements
• AS/NZS 60079.11:2000 Electrical apparatus for explosive atmospheres
Part 11: Intrinsic safety i
• Standards Australia, HB13 - 2000 Handbook Electrical equipment for
hazardous areas
The relationships between the various international bodies and committees that
govern the International Standards are quite complex. A number of authors [7, 20]
have questioned this complexity and referred to the many vested commercial and
political interests involved. In brief the IEC Standards (IEC 60079-x series) are used
as a basis for the European Committee for Electrotechnical Standardisation
(CENELEC) Standards (EN 50 0xx series). Each CENELEC Standard is adopted
and renumbered to a British Standard (BS 5501.x series).
With the exception of the United States of America (USA) nearly all other nations
are progressing towards the adoption of the International Standard [7]. The adoption
by Australia and other nations of the IEC 60079-x series of Standards is a significant
step toward the development of harmonised International Standards.
Dill [10] highlighted a number of deficiencies in the intrinsic safety Standards and in
the certification documents. In his preamble, Dill subtly criticised the committee’s
responsible for the Standards for their lack of contact with researchers in the field
and failure to incorporate the latest knowledge in the Standards.
Chapter 2 Review of IS Power Supplies and Intrinsic Safety - 27 -
Mark Walpole, Intrinsically Safe (IS) Active Power Supplies, M.Eng. Thesis, QUT 2003
During the course of this research program, the new Queensland Coal Mining
Safety and Health Act 1999 (see Table 2-7) and associated regulations were
invoked. The main change relevant to IS active power supplies is that Mines
Department approvals are no longer required in Queensland. Mine managers now
have the responsibility of ensuring certified IS active power supplies are fit for their
intended purpose.
2.5.2 Comparison of AS 2380.7 and AS/NZS 60079.11
The differences between the two Australian Standards for IS power supplies are:-
• minimum value of voltage for simple circuits has increased from 1.2 V (AS
2380) to 1.5 V (AS/NZS 60079)
• reduction in the number of assessment curves from ten curves catering for
construction materials (AS 2380) to six curves (AS/NZS 60079)
• minor variations of the values in the ignition curves
Both Standards still fail to sufficiently clarify the measurement of let through energy
when testing crowbar (over-voltage) protection circuitry in IS power supplies. Both of
the Standards prescribe an upper limit but do not define how the measurement is to
be performed.
The nameplate information for IS apparatus requires improvement. The method
used by the German testing bodies (of quoting the limitations of the ranges for
external inductance and capacitance together) would reduce the potential for the
unwary to inadvertently connect an IS device to an unsafe cable or load. For IS
power supplies additional parameters need to be included which define the V-I
characteristics of the power supply as well as internal resistance, inductance and
capacitance [12].
2.5.3 Participants in Ensuring Intrinsic Safety
The legal roles and responsibilities of the parties involved in ensuring intrinsic safety
are defined within the Statutory Acts and associated Standards. In this section,
these roles are discussed at length and the costs associated with the intrinsic safety
process are highlighted in order to clarify the participation of the various stake
holders.
Chapter 2 Review of IS Power Supplies and Intrinsic Safety - 28 -
Mark Walpole, Intrinsically Safe (IS) Active Power Supplies, M.Eng. Thesis, QUT 2003
Designers and/or suppliers of intrinsic safety equipment have a responsibility to
ensure that their equipment is both functional and safe. The design strategy of a
commercial product traditionally involves compromise between design and
manufacture cost, device performance, and market pricing structures while still
satisfying the requirements of the intrinsic safety Standards. Market demands, new
technologies and competition all act to influence designers in their quest to produce
a saleable item. The certification process also imposes a considerable cost burden,
which must be considered. These costs are all ultimately passed on to the product
purchaser.
The factors that determine the cost and/or duration of the certification process falling
within the responsibility of the party seeking certification are [15]:-
• type of certification requested
• quality of the design and manufacture of the equipment
• nature and complexity of the equipment
• level of pre-compliance review
• quality, completeness and accuracy of the associated documentation
• time taken to modify and resubmit the equipment if required
• quality and responsiveness of the communications between the party
seeking certification and the accreditation body
The role of the third party certification body is to assess and test where necessary to
determine conformance to an Australian and/or International Standard. The services
provided by certification bodies are utilised by designers, suppliers, and users of
intrinsic safety equipment. The assessment, testing and certification process are
themselves covered by relevant Standards to which the certifying body must
conform to ensure that it retains its accreditation, i.e. its authority to certify
equipment.
In Australia, any testing of explosion protected equipment must be covered by the
National Association of Testing Authorities, Australia (NATA) laboratory
accreditation and the certification activities accredited by the Joint Accreditation
System of Australia and New Zealand (JASANZ). The main factors that determine
the cost and/or duration of the certification process, which are the responsibilities of
the certification body are [15]:-
Chapter 2 Review of IS Power Supplies and Intrinsic Safety - 29 -
Mark Walpole, Intrinsically Safe (IS) Active Power Supplies, M.Eng. Thesis, QUT 2003
• assessment, testing and certification processes, which are the methods used
to determine conformance to the requested Standard
• quality and responsiveness of the communications between the certification
body and the party seeking certification
• the current work volume of the certification body
The users of the intrinsic safety equipment have a responsibility to ensure that their
equipment is functional and safe throughout the serviceable life of the equipment.
This responsibility includes the application and usage, maintenance and repair of
the equipment, establishment and maintenance of documentation, and other
statutory and inspectorate requirements. It also includes timely response to
addressing any issues arising from publication of safety alerts, product recalls, and
requests for re-certification.
Australian and International Standards bodies set the requirements by which
certification is determined. They have a responsibility to maintain these Standards,
while responding to industry trends, and advances in technology. They also need to
ensure that the Standards remain relevant with acceptable levels of risk associated
with the use of the equipment in specified hazardous locations.
Australian has a number of industry associations such as the Australian Coal
Association (ACA) and the Association of Electrical and Electronic Manufacturers
Australia (AEEMA), which promote, lobby and influence matters that impact upon
industries using IS equipment.
2.5.4 Accredited Intrinsic Safety Testing and Certification Bodies
Third party testing bodies are used to establish that a particular apparatus or system
complies with the specified Standard. There have been examples of differences in
the interpretation of the Standards between the third party testing bodies, both on an
international and national level [21]. The impact on the NSW coal mining industry,
because of the Safety Alerts issued in 1998, also raised questions by industry
observers on the assessment, testing and certification process [21], [2].
In Australia the two main accredited laboratories capable of certifying IS equipment
are Safety In Mines Testing And Research Station (SIMTARS) and TestSafe
Chapter 2 Review of IS Power Supplies and Intrinsic Safety - 30 -
Mark Walpole, Intrinsically Safe (IS) Active Power Supplies, M.Eng. Thesis, QUT 2003
Australia. Both SIMTARS and TestSafe Australia issue certificates and approvals for
conformance with Australian Standards and legislative requirements for [22]:-
• Certificates of Conformity for Groups I and II explosion protected electrical
equipment
• Certificates of Conformity for electrical equipment used in NSW or
Queensland coal mines, and others
Safety In Mines Testing And Research Station (SIMTARS) acts as a semi-
autonomous, professionally independent division of the Queensland Government's
Department of Natural Resources and Mines. The testing, calibration, certification
and other specialised services for electrical equipment used in hazardous locations
is carried out by the Engineering Testing and Certification Centre (ETCC). Evidence
of conformity issued by SIMTARS include [23]:-
• NATA reports for Australian and equivalent International Standards
• Certificates of conformity to intrinsic safety Standards to AS 2380.7, AS/NZS
60079.11, and others
Some of the major International accredited testing bodies include:-
• Health and Safety Executive – mining (HSE (M)) in Britain
• Berggewerkschaftliche Versuchsstrecke (BVS) in West Germany
• Underwriters’ Laboratories Inc (UL) in America
These testing bodies are authorised to certify to the International intrinsic safety
Standard and their own national Standard [7].
Chapter 2 Review of IS Power Supplies and Intrinsic Safety - 31 -
Mark Walpole, Intrinsically Safe (IS) Active Power Supplies, M.Eng. Thesis, QUT 2003
2.6 Certification, Assessment and Testing of IS Power Supplies
The certification process in Australia confirms compliance to one of the two
Australian Standards AS 2380.7 or AS/NZS 60079.11. A testing and certification
body determines conformance to the Standards by circuit analysis, spark ignition
testing, or a combination of both. Conformance to the thermal ignition requirements
of the Standards can be determined by temperature rise tests. In practice, a
significant part of the certification process involves assessment and testing.
Generally, the various testing and certification bodies regard their procedures and
assessment methods as proprietary information. These are therefore not generally
available to the public. As part of this investigation, the author has summarised
SIMTARS intrinsic safety assessment and testing procedures and they are
presented in Sections 2.6.2 and 2.6.3.
The certification process requires a number of reviews to be performed at critical
points within the assessment and testing process. During the latter stages of the
process, a final review takes place and if compliance is confirmed an appropriate
certificate is issued.
2.6.1 Certification – Determining Conformance to a Standard
AS 2380.7 or AS/NZS 60079.11 categorises IS electrical apparatus initially by their
location relative to the hazardous area. IS electrical apparatus are able to be located
within a hazardous area. Associated equipment must be located in a safe area but
the interconnecting wiring may enter the hazardous area.
The Standards then further categorise IS electrical apparatus by whether the
equipment is self-contained, part of a system, or entity concept equipment [24].
Generally IS power supplies are categorised for accreditation as associated
electrical apparatus and are certified as entity concept equipment or as part of an
integrated system. Associated electrical equipment require the following output
parameters to be defined: Maximum output voltage (UO), Maximum output current
(IO), Maximum external capacitance (CO), Maximum external inductance (LO), and
Maximum external inductance to resistance ratio (L/R).
Chapter 2 Review of IS Power Supplies and Intrinsic Safety - 32 -
Mark Walpole, Intrinsically Safe (IS) Active Power Supplies, M.Eng. Thesis, QUT 2003
2.6.2 Assessment of IS Active Power Supplies
As the complexity of the electrical/electronic circuitry increases it becomes more
difficult and less reliable to determine the equipment’s conformance to the relevant
Standards by analysis alone. The interaction between the load and the active
components within the active power supply cannot be easily analysed. It is for this
reason that use of the ‘ignition curves’ is not applicable for IS active power supplies
[24].
In most situations IS active power supplies are subjected to a combination of both
circuit analysis and spark ignition testing using the STA. A summary of SIMTARS
general intrinsic safety assessment procedure is shown in Table 2-8 [9]. IS active
power supplies are subjected to this assessment procedure with the exception of the
initial part of Step 7. At this point IS active power supplies assessed by SIMTARS
are subjected to spark testing using the STA as described in Section 2.6.3.
There is scope to make a number of improvements in the assessment process by
increasing the pre-submission work and documentation. This would be performed by
suitably qualified designers. During the assessment procedure outlined in Table 2-8
a number of steps repeat the work performed earlier by the designer. These steps
are listed as follows:-
1. Identify all sources of energy
2. Identify components on which intrinsic safety depends
4. Segregation of components by creepage and clearance distances
5. Circuit Calculations - ratings for all components
6. Circuit Parameters - maximum voltages and currents determined
7. Identification of potential ignition sources - sparking and heating
The additional work outlined above, if performed and documented by designers,
would provide a self-review. A second benefit is a reduction in the assessment time
and cost. Magison [6] identified this in his design methodology Task 3 as presented
in Section 2.2.3.
Chapter 2 Review of IS Power Supplies and Intrinsic Safety - 33 -
Mark Walpole, Intrinsically Safe (IS) Active Power Supplies, M.Eng. Thesis, QUT 2003
Table 2-8: Summary of SIMTARS intrinsic safety assessment procedure [9]
halla
This table is not available online. Please consult the hardcopy thesis available from the QUT Library
Chapter 2 Review of IS Power Supplies and Intrinsic Safety - 34 -
Mark Walpole, Intrinsically Safe (IS) Active Power Supplies, M.Eng. Thesis, QUT 2003
2.6.3 Testing IS Active Power Supplies using the STA
It is essential to ensure the intrinsic safety of the active power supply is tested using
the STA under numerous variations of the load parameters and over the full range of
its output characteristics to ensure that incendive sparking is not possible. These are
time consuming tests and are not required for linear power supplies. The sparking
potential of linear power supplies with ‘well defined’ circuits can be determined by
assessment alone using ignition curves.
A summary of the SIMTARS general intrinsic safety testing procedure is presented
in Table 2-9 [9].
It is an exhaustive process to establish whether a circuit is IS at all possible circuit
configurations and values under both normal and fault conditions. If the apparatus is
being certified under the Entity Concept, then in addition to establishing the IS
status, Entity Concept parameters (LO, CO and L/R ratio) must also be determined
using a trial and error method.
Table 2-9: Summary of SIMTARS intrinsic safety testing procedure [9]
halla
This table is not available online. Please consult the hardcopy thesis available from the QUT Library
Chapter 2 Review of IS Power Supplies and Intrinsic Safety - 35 -
Mark Walpole, Intrinsically Safe (IS) Active Power Supplies, M.Eng. Thesis, QUT 2003
Of the research performed on the STA the main areas of interest relate to the
mechanics and materials used for the electrode wire and contact disk. In addition,
some research has been done on the application of the STA to testing high current
IS apparatus [25]. In Chapter 3 properties of the STA are investigated further.
There is scope for possible improvement in the testing phase of the certification
process. The testing procedure outlined in Table 2-9 is a lengthy and tedious
exercise, contributing significantly to the costs involved. The STA uses up to four
wires located in the wire holder. These tungsten wires are subjected to flexing and
after a period they either bend or break off. The wire holder rotates at 80 rpm and,
therefore, it is difficult to monitor the state of the wires during a test. Wires bend or
break frequently and, if unobserved, this may require the test to be repeated. By
sensing the current flow using additional circuitry, broken wires may be easily
identified. This would allow a test to halt immediately and the wire to be replaced.
The test could then be resumed with minimum lost time.
There are a number of requirements stipulated in the Standards that must be
maintained throughout the test for the results to be valid. These items include:-
• concentration of the explosive testing gas
• flow rate of explosive test gas through the testing chamber
• nominated voltage and current to the device under test
• number of revolutions of the wire holder
• occurrence of an explosive ignition
• results of the pre and post STA sensitivity check
Partial automation, data acquisition, and recording would improve the operation and
efficiency of the testing phase of the certification process.
Chapter 2 Review of IS Power Supplies and Intrinsic Safety - 36 -
Mark Walpole, Intrinsically Safe (IS) Active Power Supplies, M.Eng. Thesis, QUT 2003
2.7 Summary
Hazardous areas with explosive atmospheres are common in industries such as
processing, manufacturing, and in underground coal mines. Statutory requirements
may stipulate that IS devices must be used. IS power supplies have been designed,
manufactured and certified to meet specific criteria in accordance with intrinsic
safety Standards. These Standards specify the amount of energy that the IS power
supply is permitted to deliver to the IS circuit.
The intrinsic safety accreditation process may involve both assessment and testing
to determine conformance to the intrinsic safety Standards. Increase in the
complexity of modern IS active power supplies has complicated the assessment and
testing process and extended the time taken to determine conformance.
The sparking potential of linear and trapezoidal type power supplies can be
determined using ignition curves included in the intrinsic safety Standard.
Researchers have observed significant levels of output energy when subjecting
rectangular type (active) power supplies to dynamic load conditions. The Standard
specifies that the sparking potential of active power supplies must be determined
using the STA. The STA is used to determine whether this amount of energy can
cause an explosive ignition.
The dynamic behaviour of active power supplies has been investigated by a number
of researchers and this literature was reviewed. The determination of the output
energy when subjecting the active power supply to dynamic loads is further
investigated in this thesis.
In this chapter, two of the research goals of Section 1.2 have been fulfilled. -(a) the
different types of IS power supplies and the relevant terminology have been clearly
defined, and -(b) the existing practices for assessment and testing of active IS
power supplies have been reviewed and opportunities for improvement identified.
The remaining research goals require determining the limits of intrinsic safety for
active power supplies and to this end the principal instrument for determining the
sparking potential of IS circuits, the STA is investigated in the following chapter.
- 37 -
Mark Walpole, Intrinsically Safe (IS) Active Power Supplies, M.Eng. Thesis, QUT 2003
Chapter 3 Electrical Investigation of the STA The spark test apparatus (STA) is used to determine the sparking potential of the
intrinsically safe (IS) power supply by simulating fault conditions likely to occur in the
field. The main concern for spark ignition is the presence of exposed conductors that
can touch (make) and then separate (break). The STA is connected to the circuit
under test, produces a variety of ‘makes’, and ‘breaks’ at different velocities and
intervals. In this chapter, the STA is investigated to determine how the sparking
potential of a device under test is established.
Sections 3.1 and 3.2 give an introduction to the STA and identify how the sparking
potential of an IS apparatus is determined. In Section 3.3 the periodic nature and
randomness of the STA is discussed. This is followed by the measurements of the
electrical circuit of the STA in Section 3.4. In Section 3.5 the sensitivity of the STA is
discussed.
3.1 Introduction to the STA
The STA consists of a small gas chamber to which a flammable test gas of known
concentration is applied at a low flow rate. The chamber contains an insulated wire
holder and an insulated cadmium disk as shown in Figure 3-1.
The wires on the wire holder and cadmium disk simulate the electrical contacts in a
switch that makes and breaks the circuit under test [24]. The wire holder is able to
secure up to four wires and is positioned above the cadmium disk so that their
circumferences overlap, as shown in Figure 3-1 and Figure 3-2. The wires are
located equidistant around the edge of the wire holder and extend down so that they
can make contact with the surface of the cadmium disk. Only one wire is able to be
on the cadmium disk at any one time. The wire holder is driven at 80 rpm in a
clockwise (CW) direction and the cadmium at 19.2 rpm in a counter-clockwise
(CCW) direction. As the wire holder rotates, one of the four wires makes contact
with the cadmium disk, traverses the surface of the cadmium disk, and then
disconnects as illustrated in Figure 3-2. A short period lapses before the next wire
makes contact with the cadmium disk [24].
Chapter 3 Electrical Investigation of the STA - 38 -
Mark Walpole, Intrinsically Safe (IS) Active Power Supplies, M.Eng. Thesis, QUT 2003
Figure 3-1: Plan and elevation views of STA wire holder and cadmium disk [24]
The cadmium disk has two parallel grooves on its surface which cause the wire and
cadmium disk surface to intermittently break contact as the wire end leaves the
edge of the groove and then re-make contact when it reaches the other side of the
groove. The angle at which the wire departs from the edge of a groove and the
tension of the wire determines the speed of departure at the break, the speed of
arrival of the make, and the time period the wire is located in the groove [24].
halla
This figure is not available online. Please consult the hardcopy thesis available from the QUT Library
Chapter 3 Electrical Investigation of the STA - 39 -
Mark Walpole, Intrinsically Safe (IS) Active Power Supplies, M.Eng. Thesis, QUT 2003
Figure 3-2: Oblique view of the STA wire holder and the cadmium disk
The ability to accurately replicate test results using the STA is affected by the
statistical probability that an explosive ignition may occur. Ignition probability is
based upon the occurrence of a potentially explosive ignition event every 1600
revolutions of the wire holder.
Wire path acrosscadmium disk
Wire holder
Wire
Chordalgrooves19.2 rpm
80 rpm
Cadmium disk
Chapter 3 Electrical Investigation of the STA - 40 -
Mark Walpole, Intrinsically Safe (IS) Active Power Supplies, M.Eng. Thesis, QUT 2003
3.2 Low Voltage Electric Arcs and the STA
The STA facilitates the discharge of the energy storage components within the
electrical circuit under test by repetitively making and breaking the electrical circuit.
The sparks produced by the STA are located in the spark gap between the end of
the wire and the cadmium disk. The dielectric in the spark gap is the prescribed
flammable test gas at a known concentration. The conditions under which the end of
the wire and the cadmium disk make and break the circuit during the STA operation
are illustrated in Figure 3-3 (a) to (g).
Figure 3-3: STA wire and cadmium disk making and breaking contact
KEY: v = the relative velocity between the wire and the cadmium disk v1 = 208 mm/s v2 = 250 - 2000 mm/s dependent on angle to chordal groove v3 = 250 mm/s
Wire holder
Wire
Cadmium disk
Wire holder
Wire
Cadmium disk
Magnified view of wire across disk surface
disk surface
wire
Scoured disk surface
(b) Wire makes with side of disk
Wire holder
Wire
Cadmium disk
(a) Wire approaches side of disk
(c) Wire traverses surface of disk
v1 v1 v1
Wire holder
Wire
Cadmium disk
Cadmium disk groove
Wire holder
Wire
(d) Wire breaks with edge of groove
(e) Wire makes with edge of groove
Cadmium disk
v2 v2
Wire holder
Wire
Cadmium disk
(f) Wire approaches edge of disk
Cadmium disk
(g) Wire breaks with edge of disk
Wire holder
Wirev3 v1
Chapter 3 Electrical Investigation of the STA - 41 -
Mark Walpole, Intrinsically Safe (IS) Active Power Supplies, M.Eng. Thesis, QUT 2003
When the voltage applied across the terminals of the STA is greater than the
breakdown voltage of the dielectric an arc is formed. The arc will persist only while
the applied voltage is greater than the breakdown voltage of the dielectric and
sufficient current is available.
It is assumed that the wire and cadmium disk have been separated for a period
sufficiently long for the capacitor to be charged. The discharge of the capacitor will
generally occur when the wire is making contact with the cadmium disk as in Figure
3-3 (a) and (b), and in greater detail in Figure 3-4.
The discharge of an inductive energy storage component will generally occur when
the wire breaks contact with the cadmium disk. This is after a period of time where
the wire and cadmium disk have been in contact for a sufficient period to allow the
inductor to be energised as in Figure 3-3 (d) and (f). Resistive circuits may also
cause spark ignition where there is an intermittent making and breaking of a high
current circuit as in Figure 3-3 (c).
The energy available from the apparatus under test at the terminals of the STA can
significantly exceed the minimum ignition energy (MIE) of the explosive test gas in
the STA chamber without causing an explosive ignition.
As discussed in Section 2.1.3 the energy available at the spark gap typically
exceeds the MIE as the amount of energy transferred from the arc to gas is highly
dependent upon the physical shape, arrangement and materials of the electrodes.
The exact amount of energy transferred from the arc to the surrounding flammable
test gas in the STA is difficult to determine exactly and is dependent upon:-
• distance between the wire and cadmium disk
• geometry of the wire end, disk surface or edge at the spark gap
• duration of the time spent near the quenching distance
• relative velocity between the wire and cadmium disk
• the instantaneous values of arc voltage and current
Chapter 3 Electrical Investigation of the STA - 42 -
Mark Walpole, Intrinsically Safe (IS) Active Power Supplies, M.Eng. Thesis, QUT 2003
Figure 3-4: STA making contact - discharging a capacitive circuit
The only indication that sufficient energy has been transferred from the arc to the
surrounding explosive gas mixture is the occurrence of an explosive ignition. This
confirms that an amount of energy equal to or greater than the MIE of the explosive
gas mixture has been transferred from the electric arc to the test gas in the spark
gap between the wire end and the cadmium disk.
KEY: velocity v, separation distance d, and time t Rotational speed of wire holder = 80 rpm and wire path radius r = 24.84 mm Angular velocity ω = (2 * π * 80 rpm) / 60 = 8.38 rad/s Linear velocity v = ω * r = 8.38 * 24.84 = 208.10 mm/s
vd = d0
t = 0
Edge of Cadmium disk viewed from above.The wire is approaching the edge of the cadmium disk. The voltage across the wire and cadmium disk is insufficient to breakdown the spark gap distance d0.
Wire
(a)
(b) vd = d1
t = t1The voltage across the wire and cadmium disk is sufficient to breakdown the spark gap distance d1 and an arc is formed.
Arc
(c) vd = d2
t = t2The voltage across the wire and cadmium disk is sufficient to breakdown the spark gap distance d2 and the arc continues.
(d) vd = dq
t = t3 The separation distance d = dq and the arc is quenched.
(e) vd = 0
t = t4 The wire is now in contact with the disk and the separation distance d = 0.
Chapter 3 Electrical Investigation of the STA - 43 -
Mark Walpole, Intrinsically Safe (IS) Active Power Supplies, M.Eng. Thesis, QUT 2003
3.3 Periodic and Randomness of the STA
When the STA is connected to the direct current (DC) resistive circuit as shown in
Figure 3-5 (a) and the wire on the cadmium disk encounters both of the chordal
grooves as shown in Figure 3-5 (b), the current (IO) and voltage (UO) waveforms are
depicted in Figure 3-6. The spikes in the waveforms indicate when the wire breaks
contact with the cadmium disk.
Figure 3-5: Test circuit with STA and wire path for a single traverse
The measured waveforms in Figure 3-6 for the output voltage and output current
show small variations whilst the wire is over the surface of the disk. These variations
are due to physical irregularities of both the wire and the disk. Irregularities may
include: scratches on the disk surface, loose particles on the disk surface, worn
edges on the side of the disk and grooves, wire bending and wire splitting.
Steady state measurements:-
STA contacts open UO = 10V, IO = 0 mA STA contacts closed UO = 42 mV, IO = 42 mA
(a) STA connected to resistive circuit.
Wire positions on path:- a – wire makes with disk b – wire breaks with edge of groove c – wire makes with edge of groove d – wire breaks with edge of groove e – wire makes with edge of groove f – wire breaks with disk
(b) Wire path traversing cadmium disk
Uo
STA
Io
Rm = 1
AB
Chnl. A - UoChnl. B - Io = Um / Rm
Uo
Um
OSCILLOSCOPE
Ω
R = 237 Ω
10 V
Wire path
abcdef
Cadmium disk rotation CCW
Chapter 3 Electrical Investigation of the STA - 44 -
Mark Walpole, Intrinsically Safe (IS) Active Power Supplies, M.Eng. Thesis, QUT 2003
The period that the wire is over the surface of the disk is dependent on the rotational
speed of the wire holder and the free length of the wire. The waveforms in Figure
3-6 were measured with the wire holder rotating at 80 rpm. The end of a single wire
is on the disk for a period of 130.4 ms followed by a 57.1 ms period with the wire off
the disk before the next wire makes contact. The total period for the end of a wire is
187.5 ms. As the four wires are located equidistant on the periphery of the wire
holder then the 187.5 ms period corresponds to 90o CW rotation of the wire holder.
A 90o rotation of the wire holder through the 50:12 gearbox ratio results in a 21.6o
CCW rotation of the cadmium disk.
Figure 3-6: Measured output current (IO) and voltage (UO) for a single traverse
The STA is periodic based on 12.5 revolutions of the wire holder, corresponding to 3
revolutions of the cadmium disk. The periodic nature of the STA is illustrated in
Figure 3-7.
Note: Wire positions a,b,c,d,e and f correspond to Figure 3-5.
a b c d e f wire positions on path
Output current Io
0
10
20
30
40
50
0 30 60 90 120 150 ms
mA
Output voltage Uo
0
2
4
6
8
10
0 30 60 90 120 150ms
V
Chapter 3 Electrical Investigation of the STA - 45 -
Mark Walpole, Intrinsically Safe (IS) Active Power Supplies, M.Eng. Thesis, QUT 2003
Figure 3-7: Periodic make and break of wires on the cadmium disk
The measured current waveform in Figure 3-8 depicts two revolutions of the wire
holder. The waveform shows eight periods corresponding to each of the four wires
in the wire holder traversing the cadmium disk. The first period (0 to 188 ms) shows
the wire intersecting the two chordal grooves on the cadmium disk. In the following
period (188 to 375 ms), the cadmium disk has rotated and the wire intersects only
one of the chordal grooves. During the fifth period at the 800 ms point, the cadmium
disk has rotated to an angle where the groove is tangential to the wire path. The
wire scrapes along one of the edges of the groove rapidly making and breaking the
circuit under test.
Figure 3-8: Measured periodic make and break waveform
0 1 2 3 4 5 6 7 8 9 10
Wire ON disk
Wire OFF disk
Period of wire holder 0.75 sec [12.5 revs per STA period]
Period of Cadmium disk 3.125 sec [3 revs per STA period]
Period of wire 0.1875 sec [50 wires across disk to STA period]
Period of STA 9.375 sec [wire holder at 80 rpm]
Time sec
O u tp u t c u rre n t Io
0
10
20
30
40
50
0 300 600 900 1200 1500 m s
m A
1 2 3 4 5 6 7 8 Period
Chapter 3 Electrical Investigation of the STA - 46 -
Mark Walpole, Intrinsically Safe (IS) Active Power Supplies, M.Eng. Thesis, QUT 2003
The tungsten wire is harder than the cadmium disk and therefore, it scratches the
surface of the cadmium disk. As the STA is periodic, the scouring forms a pattern as
illustrated in Figure 3-9 (b). The wire path forms an arc on the cadmium disk and the
geometry of the arc has been determined graphically in Figure 3-9 (a). The pattern
in Figure 3-9 (a) matches that observed in the picture of the cadmium disk in Figure
3-9 (b).
When a new cadmium disk with a smooth surface is used, the STA has poor
sensitivity. A conditioning process is described in the intrinsic safety Standard. The
purpose of the conditioning process is to roughen the surface of the cadmium disk.
At the end of the conditioning process, a distinct pattern is observed on the surface
of the cadmium disk as observed in Figure 3-9 (b). This process gradually improves
the sensitivity of the STA to the point where it will successfully have an explosive
ignition using the calibration circuit.
Figure 3-9: Geometry of arc scribed by the wire on cadmium disk
The sensitivity of the STA varies and is dependent upon both its physical properties
and environmental conditions. The wire condition and humidity appear to be the
main factors contributing to variations in STA sensitivity. These are minimised by air
conditioning (i.e. controlling the humidity) of the atmosphere within the testing
laboratory and ensuring that the test gas is at constant temperature with a low
moisture content. Wires can be prepared to minimise splitting and require regular
cleaning and straightening. Wire replacement is recommended if any deterioration of
the wire is noticeable.
(a) Graphically determined using (b) photo of scoured
QuickCAD release 7 cadmium disk
Wire path on disk (arc)
Arc angle88.33 deg
Cadmium disk
Arc radius
21.34 mm
3.36
mm
15.35 mm
Chapter 3 Electrical Investigation of the STA - 47 -
Mark Walpole, Intrinsically Safe (IS) Active Power Supplies, M.Eng. Thesis, QUT 2003
3.4 Electrical Circuit of the STA
The individual circuit loops within the STA were measured using an LRC meter to
derive the electrical circuit presented in Figure 3-10 (a). The distributed components
in Figure 3-10 (a) have been lumped together and are presented in Figure 3-10 (b)
which illustrates the equivalent circuit of the STA.
During the measurement, it was observed that the value of RWH and RCD varied as
the wire holder was rotated. This indicates that the brush contact resistance with the
rotating shaft varies. Average values for RWH and RCD are presented in Figure 3-10.
The Standard specifies maximum allowable values for the STA as self-capacitance
30 pF (contacts open), self-inductance 3 µH (contacts closed), and resistance of
0.15 Ω (contacts closed, measured at 1 A DC) [24]. The measured value for the
resistance is considerably higher than the permissible values specified in the
intrinsic safety Standard.
The small values of series inductance, resistance, and shunt capacitance do not
significantly load the circuit under test. The voltage available at the terminal of the
STA is present across the wire holder and cadmium disk before the wire making
contact with the side of the cadmium disk. The low internal resistance of the STA
ensures the maximum short-circuit current is present during the time that the wire is
in contact with the cadmium disk.
Chapter 3 Electrical Investigation of the STA - 48 -
Mark Walpole, Intrinsically Safe (IS) Active Power Supplies, M.Eng. Thesis, QUT 2003
Figure 3-10: STA electrical circuit
WH – Wire holder CD – Cadmium disk av. – average value av. R WH = 1064 mΩ L WH = 0.5 µH R CONTACT = 34 mΩ C OPEN = 7 pF L CD = 1.1 µH av. R CD = 489 mΩ
(a) Electrical circuit (distributed component) of the STA R CLOSED = R WH + R CD = 1553 mΩ L CLOSED = L WH + L CD = 1.6 µH R CONTACT = 34 mΩ C OPEN = 7 pF
(b) Electrical circuit (lumped component) of the STA
WH
CD
External terminals
R WH L WH
C OPEN
R CD L CD
WH
CD
R CONTACT
R CLOSED L CLOSED
WH
CD
External terminals
C OPENWH
CD
R CONTACT
Chapter 3 Electrical Investigation of the STA - 49 -
Mark Walpole, Intrinsically Safe (IS) Active Power Supplies, M.Eng. Thesis, QUT 2003
3.5 Sensitivity of the STA
The intrinsic safety Standard prescribes a test to determine the sensitivity of the
STA. The calibration circuit is a DC inductive circuit as shown in Figure 3-11 and is
intended to cause an explosive ignition in the STA test chamber. The sensitivity test
is performed before and after all spark ignition tests. If the post sensitivity test fails
then the spark ignition test is invalid.
Widginton states “… ignitions can arise simply because of variations which are
known to occur in the sensitivity of the spark test apparatus, or as a consequence of
the probabilistic behaviour of the spark test apparatus” [16]. The probabilistic
behaviour of the STA is termed the ‘probability of ignition’ which, for the ignition
curves included in the intrinsic safety Standard, represents a probability of 1 ignition
in approximately 400 revolutions of the wire holder (4 wires) resulting in
approximately 1000 sparks [16]. This equates to a probability of < 1% [5].
The transient conditions that occur during the break of the inductive calibration
circuit is the period when the stored energy in the inductor is delivered to the STA
terminals and ultimately to the wire end and cadmium disk. It is at these times that
peak energy and potential for ignition occur. Generally an explosive ignition occurs
near an instance of a rapid rise in the available energy and not during steady-state
periods. A rapid rise in available energy coincides with peak output current as the
circuit is opened and has a short duration related to the time constant of the circuit.
The steady-state circuit measurements are presented in Figure 3-11 and typical
waveforms for voltage and current for the calibration circuit during a non-explosive
ignition are presented in Figure 3-12 (a). As the STA calibration circuit is opened, an
arc was formed. The energy transferred during this period was less than the
minimum ignition energy (MIE) of the surrounding explosive test gas.
In the case presented in Figure 3-12 (b), an explosive ignition did occur as the circuit
was opened. The total amount of energy transferred to the test gas caused ignition
kernel growth to exceed the quenching distance and hence become a self-
propagating flame front. When an explosive ignition occurs the energy value
associated with the MIE of the test gas must have been transferred from the arc to
the test gas. During the explosive ignition the measured peak values of the output
Chapter 3 Electrical Investigation of the STA - 50 -
Mark Walpole, Intrinsically Safe (IS) Active Power Supplies, M.Eng. Thesis, QUT 2003
voltage 197 V and output current exceeded 3.2 A resulting in the output power
exceeding 630 W for a short duration.
Figure 3-11: STA calibration circuit with current measuring resistance
Steady state measurements:-
STA contacts open UO = 24V, IO = 0 A STA contacts closed UO = 101 mV, IO = 100.8 mA
InductorUo
STA
Io
Rm = 1
AB
Chnl. A - UoChnl. B - Io = Um / Rm
Uo
Um
OSCILLOSCOPE
Ω
24 V
L
214 Ω
RLRS
24 Ω 92.9 mH
Chapter 3 Electrical Investigation of the STA - 51 -
Mark Walpole, Intrinsically Safe (IS) Active Power Supplies, M.Eng. Thesis, QUT 2003
Figure 3-12: Measured V and I waveforms for the STA calibration circuit
Upper trace: Output voltage UO Lower trace: Output current IO (a) Typical output voltage and current - Wire is traversing chordal groove and as
the circuit opens a non-explosive ignition occurs
Upper trace: Output voltage UO Lower trace: Output current IO (b) Output voltage and current during explosive ignition - As the circuit opens a
explosive ignition occurs
Explosive ignition wire and disk separating
Non-explosive ignition wire and disk separating
Chapter 3 Electrical Investigation of the STA - 52 -
Mark Walpole, Intrinsically Safe (IS) Active Power Supplies, M.Eng. Thesis, QUT 2003
3.6 Summary
In this chapter the research goal in Section 1.2 of investigating the intrinsic safety
testing practices for IS active power supplies has been fulfilled. The principal
instrument for testing and determining the sparking potential of active power
supplies and other electrical circuits is the STA. The STA replicates conditions that
are likely to occur in the field and hence is able to determine if the output energy
from the electric circuit is low enough for it to be regarded as conforming to IS
requirements.
The investigations carried out as part of this thesis have determined that the STA
has a periodic make and break sequence between the wires and cadmium disk.
However, the roughened surface of the cadmium disk results in variable conditions
throughout the duration of contact between the wire and the disk surface. The
chordal grooves on the cadmium disk provide a range of separation and approach
velocities between the wire and cadmium disk.
The STA is sensitive to the physical condition of the wire and cadmium disk. For
optimal performance straight wires where the wire end is without splits or deformity,
and a conditioned cadmium disk with a roughened surface is required. The STA is
also sensitive to environmental conditions, in particular to humidity.
The STA wire and cadmium disk apply an intermittent transient short-circuit and
open-circuit loads to the circuit under test. It is the transient response of the circuit
under test that determines the output power during this transient period and thus the
available output energy. In the case of active power supplies, the energy stored in
components within the active power supply is transferred to the STA during these
transient periods potentially creating a low energy electric arc. The amount of
energy available at the output of an active power supply is dependent on the
transient characteristics of the active power supply and this is investigated in
Chapter 4.
- 53 -
Mark Walpole, Intrinsically Safe (IS) Active Power Supplies, M.Eng. Thesis, QUT 2003
Chapter 4 Characteristics of IS Active Power Supplies In order to carry out the investigations described within this chapter a number of
direct current (DC) intrinsically safe (IS) active power supply samples were provided
by a local underground coal mining company. The IS active power supplies had
similar electrical circuit topology and varied only in their nominal DC output ratings.
The three sample IS active power supplies are identified as PS 1, PS 2 and PS 3,
and their nameplate ratings are presented in Table 4-1
In this chapter an analysis of the output steady-state and transient characteristics of
this type of IS active power supply is undertaken. The sparking potential of these
power supplies is identified and defined by parameters that are measured from their
steady-state and transient output characteristics.
4.1 Sample IS Active Power Supplies
Due to proprietary privilege, no documentation or circuit diagrams were available for
the sample IS active power supplies. The circuits were traced and analysed and a
generic block diagram is included in Appendix A 1. The functional block diagram of
the output stage of a sample IS active power supply circuit is presented in Figure
4-1. The IS active power supply includes active components in the voltage regulator,
current limiter, and over-voltage crowbar protection circuitry. In this particular IS
active power supply the current limiter includes the intrinsic safety control circuitry.
Figure 4-1: Block diagram of sample IS active power supply DC stage
Voltage Regulator
V reg
I Limit
Low passfilter
C C
Bridge rectifier
Low passfilter
AC Line supply side of circuit not shown
+
-CurrentLimiter
Crowbarprotection
IS DC Output
IS Cntrl
Output current sensing circuitry
Over voltage sensing circuitry
IS Control circuit
Chapter 4 Characteristics of IS Active Power Supplies - 54 -
Mark Walpole, Intrinsically Safe (IS) Active Power Supplies, M.Eng. Thesis, QUT 2003
4.2 Steady-state Output Characteristics
The steady-state output characteristics of three sample IS active power supplies
were measured and are presented in Figure 4-3 with the measured values of all
three sample power supplies presented in Table 4-1. The load resistance RLOAD in
the test circuit of Figure 4-2 (a) is slowly reduced from infinity (open-circuit) to short-
circuit. At each measuring point, the output voltage and current are recorded once
they have stabilised. The two linear sections of the steady-state output
characteristics in Figure 4-3 are emphasised to illustrate the differences between
no-load, full-load and short-circuit values.
The two linear sections in Figure 4-3 correspond to the two operating modes of the
sample IS active power supply. The normal mode of operation is from no-load to full-
load where the voltage regulator in Figure 4-1 maintains a constant output voltage
as the output current and load resistance varies.
The fault mode of operation is where the current demand exceeds the rated full-load
current. In this mode of operation the current limiter in Figure 4-1 limits the output
current to approximately the full-load value and reduces the output voltage for
further demands of output current as load resistance is reduced. The current limiter
effectively reduces the output power available.
Figure 4-2: Steady-state test circuit
(a) Steady-state test circuit (b) Steady-state measurement
DCActive PowerSupply
+
-
Uo
Io
R LOAD
FAULT MODE
Amps
Volts
IO FLIO SC
UO SC UO FL UO NL
NORMAL MODE
Full load (FL)Short circuit (SC)
No load (NL)
[Current limiting]
[Voltage regulation]
Chapter 4 Characteristics of IS Active Power Supplies - 55 -
Mark Walpole, Intrinsically Safe (IS) Active Power Supplies, M.Eng. Thesis, QUT 2003
IS active power supplies with predominantly capacitive energy storage components
pose a spark ignition risk during the contact closure transient period. The
parameters that define this transient period are; no-load output voltage, period from
non-zero output current to peak output current, peak output current and
corresponding output voltage values, duration of the output current decay from the
peak value to steady-state value, and the steady-state output current and
corresponding output voltage values. These parameters are measured from the
transient no-load to short-circuit load applied by the STA when the maximum peak
output current occurs.
The maximum peak output current sample is identified after recording a number of
sample waveforms. Transient waveforms are included in the sample if the energy
storage components are fully charged by a suitable open circuit time, and there is no
STA wire bounce on initial contact with the cadmium disk. This excludes contact
closures that occurred as the wire traversed the chordal grooves on the cadmium
7.61A IO 1.76A
0A 12.61V
9.44V UO
2.41V 0 V t0 t Peak=160µs tSS=7.12ms
Upper trace: Output current (IO) Lower trace: Output voltage (UO)
Chapter 4 Characteristics of IS Active Power Supplies - 59 -
Mark Walpole, Intrinsically Safe (IS) Active Power Supplies, M.Eng. Thesis, QUT 2003
disk. The transient with the maximum peak output current is selected from the
recorded samples and the parameters measured. A sample size of 20 was found to
be statistically significant and typically included a sample with the maximum peak
output current.
This sample size was determined by repetitive experiments and obtaining a
population of measurements. Due to the inherent variations in the STA, numerous
measurements were required to derive a population. It was found that within any 20
consecutive samples of the population, a sample occurred where the peak output
current was equal to the maximum peak output current of the total population.
The measured parameter values of maximum peak output current and
corresponding voltage for the transient no-load to short-circuit load are presented in
Table 4-2. Data values of the sample transient voltage and current waveform were
tabulated in an Excel spreadsheet (see Appendix A 2) where the value for output
power was determined using equation (4.1). The output energy was determined
using equation (4.2) where the integration was approximated using the trapezoidal
method. Plots of output current, voltage, power and energy are also included in
Appendix A 2.
Initial contact of wire and cadmium disk Time Time IO(t) (A) UO(t) (V) PO(t) (W) EO(t) (mJ) t 1 0 0 12.61 0 0 t Peak 160 µs 7.61 9.44 71.84 5.5 t SS 7.12 ms 1.76 2.41 4.26 122.6
t 1 Contact makes and circuit closes at time t 1 t Peak Peak output current occurs at time t Peak
Times of interest
t SS Steady-state conditions occur at time t SS
The value of output energy when the peak output current occurs is approximately
5.53 mJ. This is considerably higher than the MIE of Methane 0.28 mJ [5]. This
amount of energy is a spark ignition risk, though as discussed in Sections 2.1.3 only
some of this energy is transferred to the ignition process.
Table 4-2: Measured transient parameters – test circuit with STA
Chapter 4 Characteristics of IS Active Power Supplies - 60 -
Mark Walpole, Intrinsically Safe (IS) Active Power Supplies, M.Eng. Thesis, QUT 2003
4.3.2 Measuring Transient Output Characteristics using a Relay
To overcome variations inherent in the STA a relay contact was substituted for the
STA wire and cadmium disk as shown in Figure 4-6. A signal generator was used to
provide an independent square wave voltage to drive the relay coil. By varying the
signal generator frequency and voltage the relay closing speed could be adjusted to
a value between 0 to 200 mm/s. However, although contact bounce was a problem
at higher speeds. The relay contact applied a short-circuit load and a storage
oscilloscope was used to measure the output voltage and current during the
transition.
Figure 4-6: Transient characteristics test circuit with a relay
The measured results as shown in Figure 4-7 shows a transition where a short-
circuit was applied. During the experiment it was observed that the highest values of
peak output current occurred when the relay was operated manually with the relay
test button. This produced a slow closure of the relay contact with no contact
bounce. This led us to believe that a wetted contact may provide an alternate
solution to the problem of contact bounce.
RM – Current measuring resistor (1Ω)
Contactcloses at t1opens at t2
time
time
t1 t2
t1 t2
Uo NL
Uo SC
Io Peak
Io SCIo NL
Uo(t)
Io(t)
Io AmpsRM
DCActive PowerSupply
+
-
AB
Chnl. A - UoChnl. B - Io = UM / RM
Uo
UM
OSCILLOSCOPE
IO
Uo Volts
Chnl. B
Chnl. A
Chapter 4 Characteristics of IS Active Power Supplies - 61 -
Mark Walpole, Intrinsically Safe (IS) Active Power Supplies, M.Eng. Thesis, QUT 2003
resistance, and a current measuring resistor (1 Ω). This resistance value was
selected as a conservative approximation to ensure that the proposed alternative
assessment method (PAAM) was more sensitive to ensure pass margins with high
levels of confidence. Parasitic inductance and capacitance are minimised by
separated short wire lengths.
When measuring the transient response of a circuit with either capacitance or
inductance, the instantaneous values of current and voltage are dependent upon the
rate at which the voltage or current is changing, as shown in Table 4-4.
RC = Effective series resistance (ESR) of C RM = Current measuring resistor
Inductor L eL = L di / dt vR = i * R i = 1/L ∫ eL dt i = vR / R E = vR + eL = i * R + L di / dt
Capacitor C i = C dvC / dt i = vR / R vC = 1/C ∫ i dt vR = i * R E = vR + vC = i * R + 1/C ∫ i dt = RC dvC / dt + vC
where E – applied source voltage i – instantaneous circuit current vR – instantaneous voltage across resistor R eL – instantaneous voltage across inductor L vC – instantaneous voltage across capacitor C
Table 4-4: Instantaneous voltage and current for inductors and capacitors
EC
i
vC
vR RL
i
e L
v R R E
RM
DCActive PowerSupply
+
-
ContactAB
Chnl. A - UoChnl. B - Io = UM / RM
Uo
UM
OSCILLOSCOPE
RC
C IO
Chapter 4 Characteristics of IS Active Power Supplies - 64 -
Mark Walpole, Intrinsically Safe (IS) Active Power Supplies, M.Eng. Thesis, QUT 2003
4.4 Transient Characteristics of Sample IS Active Power Supplies
Three transient output characteristics were obtained from the sample IS active
power supplies. Transient output characteristics were obtained for both the normal
and fault modes of operation to separately determine the transient behaviour of the
relevant blocks in Figure 4-1. The third transient output characteristics
encompassed both modes of operation and determined the transient behaviour
during the transition between operational modes. For each of the three transient
output characteristics a transient load is applied, the output voltage and current then
stabilise. This is followed by the removal of transient load so that there are two
transitions during a single test.
The transient output characteristics for the normal mode of operation are determined
by rapidly changing the load from no-load to full-load at time t1 and, after a period to
stabilise, rapidly removal of the load from full-load to no-load at time t2. The test
circuit and transient output characteristics for UO and IO are illustrated in Figure 4-9.
During the transition from full-load to no-load at time t2 in Figure 4-9 the output
voltage over shoots and has a damped oscillation as it stabilises at the steady-state
no-load output voltage as illustrated in the detail inset. Output current reduces
rapidly from the steady-state rated full-load value to zero. The oscillation near time t2
has a short period and decays quickly. During this transition, there is an increase in
the output energy which is a potential source for spark ignition. The parameters
defining the period near time t2 are the values of the first two peaks of output voltage
and the corresponding output currents.
Figure 4-9: Active power supply NL to FL transient characteristics
RFL – full-load resistance UO – output voltage Rm – current measuring resistor Um – voltage across Rm NL – No-load FL – Full-load IO - output current
Detail
Rm
Contactcloses at t1opens at t2
DCActive PowerSupply
+
-
R FL
A
BChnl. A - UoChnl. B - Io = Um / Rm
Uo
Um
OSCILLO-SCOPE
Uo(t)
time
timeIo(t)
Uo NLUo FL
Io FLIo NL
t1 t2
Uo Peak
Io Amps
t1 t2
Uo Volts
Chapter 4 Characteristics of IS Active Power Supplies - 65 -
Mark Walpole, Intrinsically Safe (IS) Active Power Supplies, M.Eng. Thesis, QUT 2003
The transient output characteristics for the fault mode of operation are determined
by the transition from full-load to a short-circuit at time t1 and, after a period to
stabilise, a rapid reduction in the load from a short-circuit to full-load at time t2. The
test circuit and transient output characteristics for UO versus time and IO versus time
are illustrated in Figure 4-10.
During the transition from full-load to short-circuit at time t1 in Figure 4-10 the output
current rises rapidly to a peak followed by a non-linear decay to the steady-state full-
load current. The output voltage reduces rapidly from the steady-state full-load value
to the steady-state short-circuit value. During this transition, there is an increase in
output energy, which is a potential source for spark ignition. The parameters that
define the period near time t1 are the value of peak output current and the output
voltage.
Figure 4-10: Active power supply FL to SC transient characteristics
The test circuit and transient output characteristics of a no-load to short-circuit
transition are illustrated in Figure 4-11. During the transition from no-load to short-
circuit at time t1 in Figure 4-11 the output current rises rapidly to a peak followed by
a non-linear decay to the steady-state full-load current. The output voltage reduces
rapidly from the steady-state no-load value to the steady-state short-circuit value.
During this transition, there is an increase in the output energy, which is a potential
source for spark ignition. The parameters that define this period are the value of the
peak output current, the time constant of the exponential decay, and the output
voltages at these points.
RFL – full-load resistance UO – output voltage Rm – current measuring resistor Um – voltage across Rm FL – Full-load SC – Short-circuit IO - output current
Rm
DCActive PowerSupply
+
-
R FL
A
BChnl. A - UoChnl. B - Io = Um / Rm
Uo
Um
OSCILLO-SCOPE
time
time
Contactcloses at t1opens at t2 Uo SC
Uo FL
t1 t2
t1 t2
Io AmpsIo PeakIo SCIo FL
Uo(t)
Io(t)
Uo Volts
Chapter 4 Characteristics of IS Active Power Supplies - 66 -
Mark Walpole, Intrinsically Safe (IS) Active Power Supplies, M.Eng. Thesis, QUT 2003
Figure 4-11: Active power supply NL to SC transient characteristics
Considering the transient output current responses in Figure 4-10 and Figure 4-11
the initial current rise at time t1 is attributed to the capacitive energy storage
components in the circuit of Figure 4-1 discharging into the short-circuit load. As the
contacts in the test circuit are closing at time t1 an arc is formed and output energy is
transferred to the arc.
The peak output current and the initial part of the decay near time t1 are caused by
the rapid discharge of the energy storage capacitors. The peak of the initial output
current rise is dependent upon the voltage across the energy storage capacitors and
the circuit resistance between the energy storage capacitors and the short-circuit.
The later part of the output current decay is due to the non-linear components in the
intrinsic safety control and current limiter circuit of Figure 4-1. As the output current
demand exceeds the rated value the IS control circuit drives the current limiter. This
increases its resistance to limit the output current, reduces the output voltage, and
limits the output power.
The time between time t1 and when the peak output current is reached is the
response time of the current sensing and intrinsic safety control circuit of Figure 4-1.
After the initial peak the output current returns to the steady-state short-circuit value.
The period between the peak output current and when steady-state short-circuit
values are reached is the response time of the intrinsic safety current limiter
circuitry.
On removal of the transient short-circuit at time t2 there is no evidence of output
voltage oscillation. The oscillation observed in the full-load to no-load transient
Rm – current measuring resistor UO – output voltage FL – Full-load Um – voltage across Rm SC – Short-circuit NL – No-load IO - output current
Rm
DCActive PowerSupply
+
-
Contactcloses at t1opens at t2
A
BChnl. A - UoChnl. B - Io = Um / Rm
Uo
Um
OSCILLO-SCOPE time
time
t1 t2
t1 t2
Uo NL
Uo SC
Io Peak
Io SCIo NL
Uo(t)
Io(t)
Io Amps
Chapter 4 Characteristics of IS Active Power Supplies - 67 -
Mark Walpole, Intrinsically Safe (IS) Active Power Supplies, M.Eng. Thesis, QUT 2003
characteristics of Figure 4-9 has been damped by intrinsic safety control and current
limiting circuits which are active during the initial part of the short-circuit to no-load
transition. The output current drops rapidly from the steady-state short-circuit value
to zero. This indicates that there are minimal inductive energy storage components
in the output stage of this type of IS active power supply.
Values of the no-load to short-circuit transient characteristics voltage UO(t) and
current IO(t) were tabulated in an Excel spreadsheet (refer Appendix A 4) where the
value for output power as shown in Figure 4-12 (b) was determined using equation
(4.1). The output energy as shown in Figure 4-12 (b) was determined using equation
(4.2) where the integration was approximated using the trapezoidal method to
calculate the area under the output power versus time curve.
The peak output power occurs with the peak output current shortly after time t1. The
output energy rises rapidly from time t1 to a knee and then continues to slowly
increase due to the steady-state output power. The value of the output energy at the
knee is the transient output energy rise and is the available energy in the arc that
potentially can be transferred to the surrounding explosive test gas. The time
between t1 and the knee is the duration of the arc. On the removal of the transient
short-circuit at time t2 the output power drops rapidly to zero and the output energy
ceases to rise.
Chapter 4 Characteristics of IS Active Power Supplies - 68 -
Mark Walpole, Intrinsically Safe (IS) Active Power Supplies, M.Eng. Thesis, QUT 2003
Figure 4-12: Active power supply NL to SC transient characteristics
(a) Output current IO(t) and voltage UO(t)
(b) Output power PO(t) and energy EO(t)
12V 1A PS No-load to Short-circuit output characteristic
0
2
4
6
8
10
12
14
0 0.02 0.04 0.06 0.08 0.1
time (ms)
Ou
tpu
t vo
lts (U
o)
0.00
1.00
2.00
3.00
4.00
5.00
6.00
7.00
8.00
Ou
tpu
t cu
rren
t (Io
) Am
ps
Uo Io
12 V 1A PS No-load to Short-circuit output characteristic
0.00
20.00
40.00
60.00
80.00
0 0.02 0.04 0.06 0.08 0.1
time (ms)
Out
put p
ower
(Po)
W
0.00
200.00
400.00
600.00
800.00
1000.00
1200.00
Out
put e
nerg
y (E
o) u
JPo Eo
Chapter 4 Characteristics of IS Active Power Supplies - 69 -
Mark Walpole, Intrinsically Safe (IS) Active Power Supplies, M.Eng. Thesis, QUT 2003
4.5 Summary
The steady-state output characteristics of the sample IS active power supplies
identified a normal mode where the output voltage is regulated and a fault mode
where the output current is limited. Transient output characteristics can be
determined by measuring instantaneous output voltages and currents using a
storage oscilloscope and a relay contact to switch between two load conditions.
Output voltage and output current values can be measured during the transient
period.
The transient output characteristics of the sample IS active power supplies identified
a number of transient load conditions where the output power is significantly higher
than the maximum steady-state output power. During these transient load
conditions, there is a rise in the available output energy. This is a potential source of
spark ignition. During the no-load to short-circuit load transient period there was a
change over between the modes of operation of the sample IS active power supply,
from normal mode to fault mode and during this time the highest simultaneous
values of voltages and current were measured.
The sample IS active power supplies analysed in this section have transient output
characteristics consistent with circuits containing predominantly capacitive energy
storage components. The parameters that define the transient output current and
voltage in this chapter are used in Chapter 5 to develop a proposed alternative
assessment method (PAAM).
- 70 -
Mark Walpole, Intrinsically Safe (IS) Active Power Supplies, M.Eng. Thesis, QUT 2003
Chapter 5 Development of the PAAM The research performed by Dill and Kanty [11] established a way of determining the
sparking potential of a circuit utilising a comparative method. If the static and
transient output characteristics of an intrinsically safe (IS) active power supply were
recorded then any time later the sparking potential of that same power supply could
be determined by comparing its present static and transient output characteristics
with the recorded characteristics. The implications are that the steady-state and
transient output characteristics contain sufficient information to determine the
sparking potential of a circuit.
Three alternative assessment methods are discussed in this chapter. In Section 5.1
the first two methods are briefly described followed by a third method based on the
development of an equivalent circuit. The third method is the proposed alternative
assessment method (PAAM). In Sections 5.2 to 5.4 the equivalent circuits models
used in the PAAM are developed. The PAAM and its limitations are discussed in
Sections 5.6 and 5.7 respectively followed by the conclusions in Section 5.8.
5.1 Assessment Methods for IS Active Power Supplies
The first method is based on the determination of a finite value for the output energy
derived from the transient output characteristics of an active power supply. This
value of output energy could then be used to determine whether the active power
supply’s sparking potential is low enough to be regarded as intrinsically safe.
Whilst this appears to be a simple technique, consideration should be given to the
test conditions under which the transient output characteristics of the active power
supply are produced. The test conditions should be such that there is an optimal
transfer of energy from the electric arc to the surrounding test gas. The effective
amount of energy transferred to the ignition process needs to be determined and a
relationship established between the energy transferred to the test gas and the
sparking potential in terms of intrinsic safety limitations.
A determination of the effective amount of energy transferred to the ignition process
and the development of a relationship between this energy and the sparking
potential is beyond the scope of this thesis.
Chapter 5 Development of the PAAM - 71 -
Mark Walpole, Intrinsically Safe (IS) Active Power Supplies, M.Eng. Thesis, QUT 2003
The second method uses circuit analysis of the active power supply to determine the
maximum power transfer under transient short-circuit conditions. This would require
impedance matching between the internal impedance of the active power supply
and the impedance of the transient short-circuit. Under transient short-circuit
conditions the internal impedance of the active power supply can vary significantly.
An assessment based on analysis of dynamic impedance was deemed too complex
for consideration as a practical assessment method to determine sparking potential.
The third method entitled ‘proposed alternate assessment method‘ (PAAM) is
developed throughout the remainder of this thesis and features the modelling of an
IS active power supply via the use of an equivalent circuit. Ideally the equivalent
circuit would simplify the IS active power circuit, containing fewer components while
still producing the same output characteristics as the IS active power supply.
According to Dill and Kanty [11] the equivalent circuit can be used to establish the
sparking potential if it has the same steady-state and transient output characteristics
as the IS active power supply.
The existing assessment method using the ignition curves included in the intrinsic
safety Standard (refer Appendix A 5) is applicable to ‘well defined’ circuits. A ‘well
defined’ circuit is a circuit such as a direct current (DC) voltage source and
comprises of one of the following component combinations: a series resistor, or
resistor and inductor, or resistor and capacitor. If the equivalent circuit is one of
these ‘well defined’ circuits then its sparking potential and that of the IS active power
supply can be determined by using existing assessment techniques.
Two equivalent circuit models are presented in this chapter. The first equivalent
circuit discussed is entitled ‘RLC equivalent circuit model’ where the circuit topology
includes resistance, inductance and capacitance. The second equivalent circuit
discussed is simplified ‘RC equivalent circuit model’ as the circuit topology includes
only a resistance and a capacitance. The RC equivalent circuit model is a ‘well
defined’ circuit.
Chapter 5 Development of the PAAM - 72 -
Mark Walpole, Intrinsically Safe (IS) Active Power Supplies, M.Eng. Thesis, QUT 2003
5.2 The RLC Equivalent Circuit Model
The RLC equivalent circuit model attempts to model both conditions that occur
where a transient output energy rise is observed during both the transient
application of a short-circuit and the transient removal of a full-load.
The circuit topology of the RLC equivalent circuit as presented in Figure 5-1 is
determined by analysing the output characteristics of the sample IS active power
supply. The steady-state output characteristics illustrated in Figure 4-2 show that the
full-load voltage is slightly less than the no-load voltage, indicating the existence of a
series resistance RS.
The first transient considered is the ‘no-load to short-circuit’ transition as described
in Figure 4-11. At time t1, when the short-circuit is applied, the current rapidly
increases from zero to a peak value, followed by a non-linear decay to the steady-
state short-circuit value. The voltage during this period decays from the steady-state
no-load voltage to the steady-state short-circuit voltage. This indicates a shunt
capacitive energy storage component C with a corresponding series resistance RC
which includes the effective series resistance (ESR) of the capacitor.
The second transient considered is the ‘full-load to no-load’ transition described in
Figure 4-9. At time t2, when the full-load resistance is removed, the current decays
from steady-state full-load value to steady-state no-load value, indicating a series
inductive energy storage component L with a corresponding series resistance RL.
The output voltage exhibits an overshoot followed by an oscillation that decays to
the steady-state no-load voltage, indicating a damped oscillatory circuit.
Figure 5-1: PAAM - RLC equivalent circuit model topology
US - DC voltage source RS - Source resistance L - Inductor RL - Inductor resistance C - Capacitor RC - Capacitor ESR resistance UO - Output voltage IO - Output current
+
-
RS + RL L
RC
CUS
Uo
Io
Chapter 5 Development of the PAAM - 73 -
Mark Walpole, Intrinsically Safe (IS) Active Power Supplies, M.Eng. Thesis, QUT 2003
The damping factor (ξ) of a series RLC circuit exhibiting an under damped
oscillation can be estimated from the ratio of the magnitude of the first two
overshoots of the oscillation. The natural frequency of the oscillation (ωn) can be
estimated using the damped oscillation frequency (ωd) and the damping factor [27].
The characteristic equation for the damped second-order response can be solved so
that the damping factor and natural frequency are related to the series circuit
component values.
The component values for the RLC equivalent circuit model are determined using
the equations presented in Table 5-1. The parameters measured in Table 5-1 are
determined from the steady-state and transient characteristics of an IS active power
supply.
Component Equations Description US = UO NL UO NL = SS no-load circuit voltage
RS + RL = UO NLIO SC
- RLOAD RLOAD known (external component) Note(i) IO SC = SS short-circuit current
RC = UO NLIO Peak
Note (i) IO Peak = TS(i) peak output current Note(i)
Underdamped case ξ < 1 , ωd < ωn Critically damped case ξ = 1 Over damped case ξ > 1
L = (RS + RL + RC)
(2*ωn*ξ)
C = 1
(L*ωn2)
UO exhibits a damped oscillation UO exhibits an exponential like behaviour UO exhibits an exponential like behaviour
Damping factor ξ = log e
x1 x2
√(π2 - (log e x1 x2
) 2) [27]
Natural frequency ωn = ωd
√(1 - ξ2) [27]
where x1 = amplitude of first overshoot of TS(v) x2 = amplitude of first undershoot of TS(v) T = period of TS(v) oscillation ωd (damped frequency) = 1/T
SS – Steady-state characteristics in Figure 4-2 TS(i) NL – SC - Current transient characteristics in Figure 4-11 TS(v) FL – NL - Voltage transient characteristics in Figure 4-9 Note (i) - In some cases where RC or RL are calculated as low ohm values, special component types are selected such as a capacitor type with low ESR or manufactured such as an inductor with low internal resistance.
Table 5-1: Component equations for the RLC equivalent circuit model
Chapter 5 Development of the PAAM - 74 -
Mark Walpole, Intrinsically Safe (IS) Active Power Supplies, M.Eng. Thesis, QUT 2003
5.3 Experimental Verification of the RLC Equivalent Circuit Model
Once the RLC equivalent circuit model was defined, component values were
determined for an over damped and under damped circuit. Experimental RLC
equivalent circuits were constructed using the component values listed in Table 5-2,
and tested to measure the steady-state and transient output characteristics.
The value for RS is higher than typically found in power supplies. A high value for RS
was used to ensure that the time constant involving the inductor was significantly
different from the time constant related to the capacitor. This would allow
identification of their respective affects on the circuit.
Value Component Over damped Under damped
US – DC voltage source 10 V 10 V RS – Series resistance 216 Ω 216 Ω L – Inductor (air cored) 92.8 mH 92.8 mH RL – Inductor resistance 24 Ω 24 Ω C – Capacitor 10.29 µF 972 nF RC – Capacitor (ESR) 0.91 Ω 5.2 Ω RM – Current measuring resistor 1.526 Ω 1.526 Ω ξ - Damping factor 1.27 0.4
The under and over damped experimental RLC test circuits presented in Figure 5-2
(a) produce the same steady-state characteristic for both the under and over
damped cases as shown in Figure 5-2 (b). The DC voltage source US has a current
limit that is activated as the current demand exceeds full-load value. The steady-
state characteristics for the under and over damped experimental RLC equivalent
circuits formed a rectangular shape consistent with an active power supply.
Table 5-2: Experimental RLC equivalent circuit model – component values
Chapter 5 Development of the PAAM - 75 -
Mark Walpole, Intrinsically Safe (IS) Active Power Supplies, M.Eng. Thesis, QUT 2003
Figure 5-2: Experimental RLC equiv. cct. and steady-state characteristic
The under and over damped experimental RLC equivalent circuit output transient
characteristics are measured using the circuit shown in Figure 5-3. The DC voltage
source US had its current limiter de-activated for the measurement of the transient
output characteristics. In the case of the over damped experimental RLC equivalent
circuit the ‘no-load to short-circuit’ output transient is presented in Figure 5-4 and the
‘short-circuit to no-load’ output transient presented in Figure 5-5.
Upper trace: Output current (IO), Lower trace: Output voltage (UO)
Chapter 5 Development of the PAAM - 80 -
Mark Walpole, Intrinsically Safe (IS) Active Power Supplies, M.Eng. Thesis, QUT 2003
5.4 The RC Equivalent Circuit Model
The RC equivalent circuit model as shown in Figure 5-8 is the same as the RLC
equivalent circuit model with the exception that the inductance component has been
removed. The RC equivalent circuit component values can be determined from the
steady-state and transient characteristics of an IS active power supply.
Figure 5-8: PAAM - RC equivalent circuit model topology
Only the steady-state and transient ‘no-load to short-circuit’ characteristics of the
sample IS active power supply are required to determine the component values.
Table 5-3 shows the equations required.
Component Equations Description US = UO NL UO NL = SS output voltage
RS = UO NLIO SC
- RL RL known (external component) ISC = SS short-circuit current
RC = RS
(RS + RL)
IO SC.RS
IO Peak + IO SC - RL Note(i)
IO Peak = TS(i) peak output current
C = τ.(RS + RL)
(RS.RC + RS.RL + RL.RC) τ = time constant of TS(i) peak current decay
SS – Steady-state characteristics in Figure 4-2 TS(i) – Current transient characteristics in Figure 4-12 (a) Note (i) - In some cases where RC is calculated as low ohm values, special component types are selected such as a capacitor type with low ESR.
US - DC voltage source RS - Source resistance C - Capacitor RC - Capacitor ESR UO - Output voltage IO - Output current
Table 5-3: Component equations for the RC equivalent circuit model
+
-
RS
RC
CUS
Uo
Io
Chapter 5 Development of the PAAM - 81 -
Mark Walpole, Intrinsically Safe (IS) Active Power Supplies, M.Eng. Thesis, QUT 2003
5.5 Experimental Verification of the RC Equivalent Circuit Model
A experimental RC equivalent circuit using component values as shown in Table
5-2. was constructed as shown in Figure 5-9 (a) and subsequently tested to
measure the steady-state output characteristics presented in Figure 5-9 (b).
Component Value US – DC voltage source 10 V RS – Series resistance 99.3 Ω C – Capacitor 10.29 µF RC – Capacitor (ESR) 0.91 Ω RM – Current measuring resistor 1.526 Ω
Figure 5-9: Experimental RC equiv. cct. and steady-state characteristic
The experimental RC equivalent circuit steady-state characteristics in Figure 5-9 (b)
is a rectangular shape consistent with an active power supply. The experimental RC
equivalent circuit output transient characteristics were measured using the circuit
shown in Figure 5-10. The ‘no-load to short-circuit’ output transient is presented in
Figure 5-11 and the ‘short-circuit to no-load’ output transient presented in Figure
5-12.
Table 5-4: Experimental RC equivalent circuit model – component values
[23] SIMTARS, "SIMTARS - Safety in Mines Testing and Research Station",
[Online], 2000, Last update 28 August 1998, Available:
www.dme.qld.gov.au/simtars/index.htm.
[24] Standards Australia and Standards New Zealand, AS/NZS 60079.11:2000
Electrical apparatus for explosive gas atmospheres Part 11: Intrinsic safety i:
Standards Australia International Ltd. and Standards New Zealand, 2000.
[25] S. Halama, J. Cerri, and J. Bigourd, "Intrinsic safety of high intensity
sources", presented at 22nd International conference of safety in mines
research institutes, Beijing, China, 1987.
[26] E. Hughes, Hughes Electrical Technology, Sixth ed, England: Longman
Scientific & Technical, 1987.
[27] J. Schwarzenbach and K. F. Gill, System Modelling and Control, 2nd ed,
Britain: Edward Arnold Ltd, 1984.
Appendices - 107 -
Mark Walpole, Intrinsically Safe (IS) Active Power Supplies, M.Eng. Thesis, QUT 2003
A 1. Generic Block Diagram of IS Active Power Supply
C
Brid
ge
rect
ifier
Low
pass
filterIsol
atio
nPr
otec
tion
Filte
rTr
ansf
orm
er
ON/
OFF
AC Inpu
tVo
ltage
Sour
ce
Volta
ge
Regu
lato
r
V re
g
I reg
+ -Cu
rrent
Regu
latio
n&
IS C
ontro
l
Crow
bar
Prot
ectio
n
IS D
C O
utpu
tIS
Cnt
rl
Volta
ge
Sens
eCu
rrent
Sens
e
RR
VI
OUT
PUT
STAG
E O
F IS
ACT
IVE
DC P
OW
ER S
UPPL
Y
Appendices - 108 -
Mark Walpole, Intrinsically Safe (IS) Active Power Supplies, M.Eng. Thesis, QUT 2003
A 2. Measured Output Characteristic using STA ORM(YOKOGAWA) Data for transient output characteristic measurement using STA Number of data 700 Trigger point 23040 Trigger time 01-07-20 15:04 Sample rate 50 kHz Io =B17/$B$11 Tag name C01 C02 Uo =C17 Unit V V Po =G17*H17 No. Eo inst =(I17+I18)*0.5*$B$10*1000000
Mark Walpole, Intrinsically Safe (IS) Active Power Supplies, M.Eng. Thesis, QUT 2003
Measured output characteristic using STA
0
2
4
6
8
10
12
14
0 2 4 6
time (ms)
Out
put V
olts
(Uo)
0
2
4
6
8
Out
put C
urre
nt (I
o) A
mps
Uo Io
Measured output characteristic using STA
0
10
20
30
40
50
60
70
80
0 1 2 3 4 5 6 7
time (ms)
Out
put p
ower
(Po)
W
0
20000
40000
60000
80000
100000
120000
140000O
utpu
t ene
rgy
(Eo)
uJ
Po Eo
Appendices - 110 -
Mark Walpole, Intrinsically Safe (IS) Active Power Supplies, M.Eng. Thesis, QUT 2003
A 3. Measured Output Characteristic using Relay ORM(YOKOGAWA) Data for transient output characteristic measurement using a relay Number of data 2001 Trigger point 5760 Trigger time 01-07-17 11:22 Sample rate 100 kHz Io =B17/$B$11 Tag name C01 C02 Uo =C17 Unit V V Po =G17*H17 No. Eo inst =(I17+I18)*0.5*$B$10*1000000
Mark Walpole, Intrinsically Safe (IS) Active Power Supplies, M.Eng. Thesis, QUT 2003
Measured output characteristic using a relay
0
2
4
6
8
10
12
14
0 1 2 3 4 5 6time (ms)
Out
put v
olts
(Uo)
-1012345678910
Out
put c
urre
nt (I
o)
Am
ps
Uo Io
Measured output characteristics using a relay
0102030405060708090
0 2 4 6 8time (ms)
Out
put p
ower
(Po)
W
-20000
0
20000
40000
60000
80000
100000O
utpu
t ene
rgy
(Eo)
uJ
Po Eo
Appendices - 112 -
Mark Walpole, Intrinsically Safe (IS) Active Power Supplies, M.Eng. Thesis, QUT 2003
A 4. No-load to Short-circuit Output Characteristic ORM(YOKOGAWA) Number of data 100 Trigger point 1920 Trigger time 01-06-26 15:12 Sample rate 100 kHz Io =B17/$B$11 Tag name C01 C02 Uo =C17 Unit V V Po =G17*H17 No. Eo inst =(I17+I18)*0.5*$B$10*1000000