Fullscreen HTML5Med Res (??MB)HTML4

Helping to keep the skies safe . . . by Dr David Maddison Airborne Weather Radar Airborne Weather Radar enables flight routing to avoid extreme weather in order to keep passengers and crew safe and more comfortable – and to avoid damage to the aircraft. Huge advances have been made in this technology in recent years, including Rockwell Collins new “MultiScan” ThreatTrack Radar, released only last year at the Singapore Airshow. D espite the fact that flying is the safest way to travel, the year 2014 was perceived by many as a bad year for aviation – although that really depends upon how you analyse the statistics. According to the Geneva-based Bureau of Aircraft Accidents Archives (BAAA) there were 111 aircraft accidents in 2014, the lowest number of accidents since 1927. The BAAA counts any aircraft crash in which the aircraft is certified for at least six people plus the crew. It also counts shoot-downs but does not count military aircraft except troop carriers and other aircraft that can carry more than six passengers. Deaths are a different matter, however and 2014 saw 1,328 people die in aircraft crashes and shoot-downs, the most since 2005, according to BAAA statistical methods. Cause 1950s 1960s 1970s 1980s Pilot Error 42 36 25 29 Pilot Error weather related 10 18 14 16 Pilot Error mechanical related 6 9 5 2 Total Pilot Error 58 63 44 57 Other Human Error 3 8 9 5 Weather 16 9 14 14 Mechanical Failure 21 19 20 21 Sabotage 3 5 11 12 Other Cause 0 2 2 1 1990s 2000s Average 29 21 5 55 8 8 18 10 1 34 18 5 57 6 6 22 9 0 32 16 5 53 6 12 20 8 1 Fatal accident causes for commercial aircraft with 19 or more passengers on board from 1950 to 2010. Over that period weather-related fatalities, both involving and not involving weather-related pilot error have been a factor in 28% of accidents. (www.planecrashinfo.com/cause.htm). 16  Silicon Chip However, according to the Aviation Safety Network (which counts only civilian planes which are certified for 14 passengers or more and does not count corporate jets, shootdowns or sabotage) in 2014 there were 692 people killed in aircraft incidents making it the safest year since 1945. This would obviously exclude the 298 killed when Malaysian Airlines flight MH17 was shot down over the Ukraine. According to the BAAA there were 163 weather-related fatalities in 2014 and 162 of those were on Air Asia flight QZ8501 that crashed into the Java Sea off Indonesia. The aircraft was an Airbus 320-200. Historically, weather-related aircraft fatalities due to crashes are a factor in 28% of cases. Avoiding weather of sufficient severity to put an aircraft Adverse weather effect on an aircraft: a lightning strike. siliconchip.com.au World’s first airborne weather radar – the ECKO airborne “cloud and collision warning search radar” from 1950. at risk is of particular importance. Planning to avoid potentially dangerous weather starts at the flight planning stage but after take-off weather continues to be monitored both from reports radioed to the aircraft and on-board weather monitoring systems, the most important of which are the pilot’s Mark I eyeballs! Some aircraft operating in some areas also transmit weather data to meteorological authorities where it is fed into weather models to supplement data from weather balloons and other sensors. Apart from the possibility of severe weather causing fatal aircraft crashes, a much more common occurrence is injury to passengers and damage to aircraft caused by turbulence. In order to assist aircraft operators avoid bad weather once in flight they use aircraft mounted weather radar systems (radar is an acronym for RAdio Detection And Ranging). Airborne weather radar detects bad weather in the aircraft’s flight path and allows the pilot(s) to avoid the worst of it. Another primary purpose of airborne weather radar is to ensure that course deviations to avoid bad weather are kept to the minimum that is necessary to avoid the adverse weather, without adding excessively to the distance to be flown which increases the time taken and adds to operating costs. A problem? The ability for radar to detect weather conditions was first noted during World War II where it was seen as a problem as radar returns from certain weather systems containing rain, snow and sleet could mask enemy activity. Ways were then developed to filter out such undesirable returns but scientists and engineers started studying the phenomenon after the war as a means to monitor weather and it has been extensively developed ever since. The first airborne weather radar was from the UK company ECKO who, in 1950, developed the “cloud and collision warning search radar”. In later developments in 1953 a researcher with the Illinois State Water Survey produced the first radar image of a “hook echo”, a particular type of weather radar signature associated with tornadoes. This demonstrated the viability of using radar to detect severe weather conditions and even siliconchip.com.au The first weather radar image of a “hook echo” which is associated with tornadoes, taken in 1953. provide early warning of developing severe conditions. Early ground-based and airborne weather radars provided information on the reflectivity of whatever targets they illuminated but could give no information on the speed, of say, water droplets in a storm which would be indicative of wind speed. Initial research on weather radar systems focused on observations of the precipitation within a weather system and its development, movement and structure, as well as making observations of the relationship between the characteristics of the radar echo and precipitation rate. When there was a greater precipitation rate there were more water droplets for the radar beam to reflect from and therefore the radar return was greater. Doppler radar for weather In 1950s research began on Doppler radar for weather applications although the earliest Doppler radar systems were developed during World War II. The Doppler effect is the familiar property of a moving noise source such as a siren changing in frequency as it approaches an observer and then moves away. The same phenomenon applies to radar signals whose return echo is influenced by the velocity of the target they are bouncing off, such as rain drops. Early Doppler radars used large and sensitive analog filters and were not practical for airborne operation except under special circumstances. It required the development Feet, Nautical Miles & Flight Levels While Australia (and indeed most countries) have adopted the metric system, in aviation Imperial units are still used: heights are generally expressed in feet, distances in nautical miles and speed in knots (which is of course nautical miles per hour). You may also come across the term “flight level” with values between zero and perhaps 500. While a flight level strictly speaking is a barometric pressure (based on an international standard air pressure at sea level), it is conveniently used to express a height above sea level expressed in thousands of feet. Therefore an aircraft said to be flying at flight level 360 means it is 36,000 feet above sea level. April 2015  17 VISUAL TOP RADAR TOP A primary threat to en-route weather avoidance is the fact that thunderstorm cell tops are non-reflective because they contain ice – a poor radar reflector. of fast computers and digital signal processing in the 1970s and the development of digital Doppler radar to enable useful and easy to visualise weather information to be interpreted from such Doppler shifts in the return radar echoes. As an aside, it is interesting to note that an unexpected Representative values of radar reflectivity as a function of height for equatorial oceanic and continental geographical areas and mid-latitude areas. For a given cruise altitude of 35,000 feet note the very large variation of reflectivity between the equatorial oceanic (black vertical bar) regions and the mid-latitude continental (yellow vertical bar) regions, corresponding to almost a 20dB range or 100 times power ratio. Note also the dramatic loss of radar reflectivity above the typical altitude for freezing of water at 16,000 feet. dBZ is a a logarithmic measure used for radar systems representing the radar echo intensity. (Diagram courtesy Rockwell Collins.) 18  Silicon Chip problem during the development of early radar systems was that the Doppler shift induced by the reflection of a radar pulse from a fast moving object effected a phase shift in the returning signal, causing the signal to be cancelled and thus reverse phase-shift compensation had to be built into the radar set. The development of Doppler radar enabled not only the shape and location of a weather pattern to be determined but also the velocity of precipitation within that weather pattern, and by inference, wind speed. Doppler radar also allows the elimination of returns travelling at a particular velocity. For example, with airborne radar, ground returns can be eliminated. Airborne weather radar can be classified as either the more conventional and familiar two dimensional radar, or the more recently developed three dimensional radar. It might also come as a surprise to some that modern commercial aircraft do not have general purpose radars that indicate the presence of other aircraft or terrain. Avoidance of these is effected by pilot observation, flight planning, transponders on aircraft and automated aircraft systems. What you see is not what you get! Monitoring weather systems with radar might seem straightforward but there are many complicating factors. For a start, what is visible to the naked eye may not be visible to radar. For example, the cloud tops of thunderstorms contain mainly ice and that is a very poor radar reflector. Typically, above 16,000 feet the temperature will be below zero centigrade and so water will be in the frozen state. The cloud top will be visible to the naked eye but not to the radar, or there will be very poor radar visibility so the flight crew have to correlate in their mind what they see with their eyes and how that relates to the radar information being received. As a general rule, the lower two thirds of a cloud are visible to radar and the top one third is invisible, due to the presence of non-radar-reflective ice crystals. Of course, even though the cloud top may be invisible to radar it does not mean it is not of concern and there can be turbulence within that area of the cloud which can affect flight operations. The presence of certain weather patterns that are visible to the radar below 16,000 feet can be used to infer that there will be certain formations above them and what their properties may be. This important point will be discussed later. By convention, a display for weather radar is coded by three different colours according to precipitation activity. Green or Level 1 refers to light precipitation activity, little or no visibility and possible reduced turbulence; yellow or Level 2 corresponds to moderate precipitation, very low visibility, moderate turbulence and passenger discomfort; while red or Level 3 refers to heavy precipitation, possible thunderstorms, severe turbulence and the possibility of aircraft damage. Black corresponds to no return. A typical cloud will have heaviest precipitation at the bottom, with less higher up in the cloud. It should be noted that the radar reflectivity varies enormously for different types of weather and is dependent on several factors. Mid-latitude continental thunderstorms have a much greater radar reflectivity than, say, equatorial oceanic thunderstorm clouds. This has lead to problems in the past as a weather radar might be optimised for typical weather conditions in, say, the United States where it is siliconchip.com.au manufactured and where most of its planes fly (mid-latitude continental area) but it would not work so well for an Australian operator in areas where many of its planes fly (equatorial oceanic). In fact the variation in radar return from these two types of conditions may vary by a factor of 20dB or 100 times (see graph). Some examples of the different radar characteristics of thunderstorms are as follows: continental land-based thunderstorms (eg, USA) typically have high moisture content at high altitude and are more radar reflective than other types. Oceanic thunderstorms (eg, Bay of Bengal) have low radar reflectivity as their moisture is located at low altitudes and the cloud tops are invisible to radar. Mid-latitude land based thunderstorms (eg, Brazil) have an intermediate radar reflectivity between that of continental land-based thunderstorms and oceanic thunderstorms. In addition to geographical variation in the radar reflectivity of storms, there is also a seasonal variation. An additional problem is how to determine the severity of a thunderstorm cell. They may look the same to the eye and on the radar but one might be much more risk for hail and lightning than the other. Thus it is clear that a weather radar should ideally take all these factors into account. When monitoring weather patterns from aircraft it is important to get a complete view of meteorological activity. With conventional 2D airborne weather radar the image provided is in one field of view like a slice and so the flight crew have to manually tilt the radar beam up and down to get a full picture of the weather. There is a fairly significant flight crew workload associated with obtaining comprehensive weather information with 2D radar. For an instruction guide on the operation of a typical modern 2D airborne weather radar you may wish to see the video “EJETS WEATHER RADAR OPERATION” http:// youtu.be/VusX0V2zvU8 Ground-based weather radar For those interested, there are numerous weather-related websites along with radar Apps for smart phones. We featured the Australian Bureau of Meterology Doppler Weather Radar in the January 2010 issue (also see www. bom.gov.au/australia/radar/ 3D Radar or www.weatherzone.com.au/ I n r e c e n t radar/, among others). times two companies have developed airborne 3D weather radar. One company is Honeywell with their IntuVue RDR-4000 system and the other company is Rockwell Collins with their WXR-2100 MultiScan ThreatTrack system. The objective with 3D radar is to reduce flight crew work load and to provide a more comprehensive picture of weather activity. This leads to greater safety and airline efficiency. The Honeywell system is currently used on the Boeing 737NG, 777, C-17, and Airbus A380 aircraft and has been A typical MultiScan radar display showing various weather threats. siliconchip.com.au April 2015  19 A typical MultiScan display and how it correlates with what is seen out of the cockpit windows. selected for the Airbus A350, Gulfstream G650 and KHI CX aircraft platforms. Rockwell Collins system The Rockwell Collins system has been installed on all Qantas aircraft, and is standard on all new Boeing 787 Dreamliners, Boeing 747-800s and Boeing Business Jets and is an option for the Airbus A320s, A330s and A340s, and Boeing 777s and Next-Generation 737s. Qantas is a pioneering operator of the Rockwell Collins Milestones in radar development 1865 James Clerk Maxwell publishes “A Dynamical Theory of the Electromagnetic Field” with the original four Maxwell’s Equations which describe how electric and magnetic fields are generated and relate to each other. 1887 Starting in November of that year, Heinrich Rudolf Hertz discovers electromagnetic waves, proves Maxwell’s Equations and publishes a series of papers, the first being “On Electromagnetic Effects Produced by Electrical Disturbances in Insulators”. 1899 Guglielmo Marconi recalls his 1899 work in 1922 and says a “ship could radiate or project a divergent beam of these [electromagnetic] rays in any desired direction, which rays, if coming across a metallic object, such as another steamer or ship, would be reflected back to a receiver screened from the local transmitter on the sending ship, and thereby immediately reveal the presence and bearing of the other ship in fog or thick weather.” 1900 Nikola Tesla in Century magazine wrote “by their [standing electromagnetic waves] use we may produce at will, from a sending station, an electrical effect in any particular region of the globe; [with which] we may determine the relative position or course of a moving object, such as a vessel at sea, the distance traversed by the same, or its speed.” 1904 Christian Hülsmeyer demonstrates detection of a ship at a distance with his “Telemobiloskop” and is sometimes credited with the invention of radar but it does not give the range of 20  Silicon Chip an object direct. It is the first patented device that can detect objects at a distance. 1917 Lucien Lévy invents the superheterodyne receiver. 1921 The magnetron is invented by Albert Wallace Hull. 1922 US Naval Research Laboratory engineers Albert H. Taylor and Leo C. Young detect a wooden ship in the Potomac River by accident when conducting communications experiments and later in 1937 develop a practical ship-based radar. 1930 Lawrence A. Hyland at US Naval Research Laboratory demonstrates the reflection of radio waves from an aircraft. 1936 The development of the klystron at General Electric by George F. Metcalf and William C. Hahn (the invention has also been attributed to the brothers Russell and Sigurd Varian of Stanford University in 1937). From this time on radar was rapidly developed, especially as the Second World War loomed and was soon started. For more details on the history of radar you may wish to look at http://en.wikipedia.org/wiki/History_of_radar Some YouTube videos of interest are: “Radar: Technical Principles: Mechanics” pt1-2 1946 US Army Training Film” http://youtu.be/64LUeQ4DAqg and “Heroes and Weapons of WWII : 01. The Men Who Invented Radar” http://youtu.be/5x37BVCvFRk siliconchip.com.au Rockwell Collins WXR-2100 as installed in aircraft cockpit and displayed on the central monitor. system and has been using it since it was first released to the market in its original version in 2002. In fact, Qantas played a major role in its development. Qantas flies across the Pacific Ocean frequently, often at night and thus had a particular incentive to want better weather radar than Air Turbulence In Australia, there are about 25 in-flight turbulence related injuries every year according to the Australian Transport Safety Bureau (ATSB) with many more unreported. Some injuries are serious with broken bones and head injuries. In a typical severe turbulence event, 99 percent of passengers will not be injured. Since Australia has some of the best flying conditions in the world, it is expected that flights outside of Australia will encounter more serious problems than flights within Australia. From 2009 to 2013 there were 677 turbulence related instances reported to the ATSB on flights to and from Australia with 197 minor injuries and 2 major injuries. Australia’s Civil Aviation Safety Authority (CASA) classifies several types of turbulence and their causes as follows. Types of turbulence Light turbulence - briefly causes slight, erratic changes in altitude and/or attitude. Light chop - slight, rapid and somewhat rhythmic bumpiness without noticeable changes in altitude or attitude. Moderate turbulence - similar to light turbulence, but greater intensity. Changes in altitude/attitude occur. Aircraft remains in control at all times. Variations in indicated air speed. Moderate chop - similar to light chop, but greater intensity. Rapid bumps or jolts without obvious changes in altitude or attitude. Severe turbulence - large, abrupt changes in altitude/atsiliconchip.com.au conventional 2D weather radar which requires a lot of flight crew interpretation and represents a high work load. While both Honeywell and Rockwell Collins make superb radar systems, the Rockwell Collins system has an Australian connection and it is the focus of the remainder titude. Large variation in indicated airspeed. Aircraft may be temporarily out of control. Extreme turbulence - aircraft is violently tossed about and is impossible to control; may cause structural damage. The causes Thermals - Heat from the sun makes warm air masses rise and cold ones sink. Jet streams - Fast, high-altitude air currents shift, disturbing the air nearby. Mountains - Air passes over mountains and causes turbulence as it flows above the air on the other side. Wake turbulence - Near the ground a passing plane or helicopter sets up small, chaotic air currents. Microbursts - A storm or a passing aircraft stirs up a strong downdraft close to the ground. Preventing injury from air turbulence Occasional injuries are sustained by passengers due to air turbulence, mainly by being thrown about the cabin or by having items fall on them from open overhead lockers. Almost all air turbulence related injuries can be avoided by ensuring objects are securely placed in overhead lockers and the lockers are kept closed and that seat belts are worn at all times, not just during take off and landing. Also, crew instructions should be followed at all times and you should familiarise yourself with the safety information card in the seat back pocket. April 2015  21 Rockwell Collins MultiScan ThreatTrack Radar – features and development milestones 2002 Release of first model and it is put into immediate service by Qantas. This version was called MultiScan. 2003 On delivery flights of Qantas aircraft from the USA to Australia the flight time was used to test and develop the next version of the system. The aircraft flew with both a certified version of the radar and also the test unit. Data obtained were eventually incorporated into the 2008 model. 2006-2007 Rockwell Collins rented a Boeing business jet and flew around the world for three months to verify what was learned during the Qantas flights and this information was incorporated into the 2008 model. 2008 A major upgrade was made from the 2002 model. Storm top prediction was possible due to the addition of a geographic database of storm models in different areas at different times of the year. This was called MultiScan VI. 2014 Hail and lightning detection was added (predictive overflight). The ability to track 48 different thunderstorm cells and vertical analysis of thunderstorm cells were added. This version was called MultiScan ThreatTrak. Note that the essence of the radar system is in its smart software, it represents a revolution in software rather than a revolution in hardware. Earlier model systems can be upgraded to the current specifications by software upgrades and some minor hardware changes in some cases. Windshear is a hazard pilots dread because in the vast majority of cases they receive no visual warning of the phenomenon. Here the red and black stripes represent the actual windshear location along with a warning reminder at upper right. 22  Silicon Chip of this article but the same general principles apply to the Honeywell system. A video of the Honeywell IntuVue radar is available at “IntuVue® 3-D Weather Radar” http://youtu.be/w8IYyFmJcF0 and a training video for its use “RDR-4000 IntuVue™ Weather Radar Pilot Training for Boeing Aircraft | Avionics | Honeywell Aviation” http://youtu.be/WNVtJeccNSM A video of the MultiScan radar can be seen on YouTube at “MultiScan ThreatTrack weather radar -- The worst weather is the one you can’t see coming” http://youtu. be/zJDduGPvOEA and Boeing crew training videos can be seen at “MultiScan Weather Radar Module 1 Boeing” http://youtu.be/EUjxFVRTdtw and “MultiScan Weather Radar Module 2” http://youtu.be/Ai_P-MwlrOw The Rockwell Collins MultiScan ThreatTrack has the following technologies: Geographic Weather Correlation: Recall from earlier in this article that there is significant regional variation in the radar reflectivity of thunderstorm cells as well as variation according to the time of year. This is a relatively recent discovery that occurred from 1997 onwards after the launch of the TRMM satellite (Tropical Rain Forest Measuring Mission). This satellite has amassed a vast database of thunderstorm reflectivity information which, along with the work of leading climatologist Dr Ed Zipser, has enabled Rockwell Collins to embed thunderstorm models into the radar which are specific to particular geographic locations and time of year. As previously noted, the tops of storm clouds are invisible to radar but the radar model is able to make predictions of the altitude of the true top of the storm cell and its level of hazard by knowing the location and time of year. Core Threat Analysis: The radar can track up to 48 thunderstorm cores at once and also predict their severity. Automatic Temperature-Based Gain: As the outside air temperature decreases, the cloud tops become less radar reflective so this feature increases the radar energy used to This particular windshear occurred during taxiing. In these pictures from the cockpit the windshear can be seen in the form of a line squall approaching the aircraft down the taxiway. The pilot delayed his takeoff for 30 minutes until the thunderstorm had passed the airport and took off safely. siliconchip.com.au illuminate the cloud, effectively decreasing the proportion of the cloud that is invisible to radar. OverFlight Protection: Traditional manual operation of 2D weather radars involves pointing the radar beam at the lower radar reflective portion of storm clouds. As a matter of simple geometry, if no adjustment to the beam angle is made, as the aircraft approaches the cloud the beam moves higher and higher until it is in the non-reflective part of the cloud and the storm cell disappears from view. The overflight protection feature keeps the beam pointed 6,000 feet beneath the flight path to keep the reflective portion of the cloud in view. Predictive OverFlight Protection: Storm cells can grow at up to 6,000 feet per minute and when this happens a “bubble” of turbulent air is pushed above the cloud top. This feature tracks the rate of growth of storm cells and warns if there is a fast-growing cell in the vicinity, which is to be avoided. SmartScan: As an aircraft turns with traditional radar there is a black wedge indicated a lack of data in the direction of the turn. This feature ensures that data is immediately acquired in the direction of the turn. Two Level Enhanced Turbulence: USA FAA regulations require turbulence with a ±0.3G RMS severity to be displayed on weather radars but in addition to displaying that, areas with a less severe but still uncomfortable level of turbulence are also displayed. Flight Path Hazard Assessment: The radar system looks for different hazards according to those relevant to the phase of flight. For example during take off and landing the main concern is storm cells with convective activity and the Core Threat Analysis feature is used to evaluate the threat; during cruise the main threat is accidental penetration of thunderstorm cloud tops, so the Geographic Weather Correlation, Automatic Temperature Based Gain and the Predictive Overflight Protection features are invoked to prevent flying through the cloud tops. In addition to these features there is also a wind-shear alert and an attenuation alert to warn when a storm cloud has absorbed so much radar energy that nothing behind it will be visible (radar shadow). Quiet, dark cockpit In keeping with modern aircraft flight philosophy of minimising the pilot workload (the “quiet, dark cockpit”) the MultiScan radar ensures that only relevant information is displayed. Threatening weather can be detected out to a maximum range of 320 nautical miles while non-threatening weather 6,000 feet beneath the aircraft is not displayed. In order to minimise the display of unnecessary information ground clutter is removed by the use of a global terrain model so that returns from the ground can be ignored. With so many aircraft flying around the world, many fitted with the same model of radar, one might wonder if there is potential for the radars to interfere with each other. This is not a problem as each radar pulse is sent out with a slightly different frequency and the radar will reject any pulse it receives that is not at the frequency that was sent out. The radar dish for this system is around 70cm in diameter and sweeps side to side and up and down every four seconds. siliconchip.com.au Radar and the pulse repetition frequency and Doppler compromise Traditional radar works by sending out a short signal pulse and then turning off the transmitter and listening for any signals reflected back to the radar antenna. Knowing that the radar signal travels at the speed of light, it is possible to determine the distance to an object, by dividing the total time for the pulse to return by two. The required length of the period between pulses has to be enough time for the signal to travel out from the radar set to the target object and return. A long period between pulses allows objects to be seen at a long distance compared to a short inter-pulse period which will only allow objects to be seen a short distance away. A compromise short inter-pulse period, corresponding to a high pulse repetition frequency (PRF) allows a potential target to be illuminated with more radio energy. This makes an object easier to see than if illuminated with a low pulse repetition frequency. However, a low pulse repetition frequency is needed if distant objects are to be observed and they are illuminated with less radio energy due to a lower number of pulses. In fact, the energy reflected from the target back to the transmitter is also subject to the inverse square law so the energy received back at the radar set has a fourth root dependence, not a square root dependence. A consequence of this is that to double the effective range of a radar system the power has to be increased by a factor of 16. With Doppler radar there is a further compromise which is that there is an inverse relationship between the distance that the radar can see to and the velocity that can be measured. When the PRF is low a long distance can be measured but only a low range of velocities. When the PRF is high, a much higher range of velocities can be measured but the range is reduced. A more recent development in radar, at least as far as commercial augmentation is concerned, is Frequency-Modulated Continuous Wave, or “Broadband” Radar, which unlike traditional radar doesn’t use a high-energy pulse but is “always on”. Contradictory though it may sound, FMCW radar uses a lot less energy, is a lot safer to operate close to and is particularly applicable to marine use (see the feature in November 2010 SILICON CHIP). Conclusion Great advances have been made in radar since it was first invented. The most recent advances are being made not so much in radar hardware but in the software used to interpret and make use of the radar data. As applied to airborne weather-radar, recent developments in 3D radar which serve to both reduce pilot work load and greatly increase the analysis of weather systems and the possible threats posed will make our skies even SC safer than they are now. April 2015  23