Corrugation Analysis Trolley (CAT)

CAT 3 in use on a tram system

The CAT is an extremely accurate and robust instrument that can be used to measure both rail corrugation and so-called acoustic roughness.  It can be used and carried by a single person: the equipment, when packed in its flight case, weighs about 20kg. There are few, if any, instruments that can measure metres of rail with similar accuracy and none, to our knowledge, which can be used conveniently by a single person.

The CAT has resulted from decades of work on measurement and understanding of rail corrugation.  A trolley profilometer for corrugation measurement was made at Cambridge University in the 1970s, and was later used also by British Rail Research to measure acoustic roughness for prediction of wheel/rail noise.  A further development (christened “The Possum”) was made in the late 1980s to measure long wave corrugation for Australian National, who have since replaced it with a CAT. The first prototype of the present CAT was produced in 1997. Since then, examples of the present equipment have been used reliably for corrugation surveys, Quality Assurance work following reprofiling on all types of railway line, and for acoustics work.

CAT 3 in its robust, wheeled carrying case

During 2009 RML introduced the Mark 3 CAT, which is designed specifically to measure both conventional flat-bottomed (Vignole) rail and also grooved rail which is severely worn and deformed, and possibly also deeply embedded in the street. The lateral position of the measuring head can be adjusted in less than a minute right up to within 10mm of the gauge and field faces of the rail. The CAT is probably the least expensive equipment of its type available to measure acoustic roughness or for accurate rail corrugation measurements, yet in many ways also (as a client told us), "The CAT is unique".

Repeatability, calibration and conformance with accepted standards

The CAT can undertake measurements routinely to the accuracy required to demonstrate compliance with the very demanding requirements of the European Standard for reprofiling rails, EN13231-3:2006.  CAT measurements of ground track in Europe, made in the mid-90s, were indeed used to propose acceptable levels of residual rail corrugation after reprofiling that were the basis for some of the criteria in EN13231-3:2006, as is apparent from this paper.  The European Standard has since been revised to incorporate, amongst other changes, significantly relaxed levels of residual corrugation for reprofiled track. 

A method for objectively determining the accuracy of corrugation measuring equipment is also stated in this paper.  This technique, which we developed when no other supplier had an objective method for determining the accuracy of their equipment, is the basis of the test to validate datum equipment for corrugation measurement in EN13231-3:2006.  A typical comparison of CAT measurements of a calibration beam with datum measurements from a Coordinate Measuring Machine (CMM) is shown below.  Measurements such as this are included in the calibration certificate with each CAT that we supply.

Typical comparison of CAT and CMM measurements of a calibration beam, 30-100mm filter

The consistency of output from the CAT is illustrated below.  This shows some standard output in which measurements from two CATs are compared on a section of quite smooth rail.  The graph on the upper left hand side compares the moving average of RMS amplitudes in the 30-100mm wavelength range, to the right of this is a graph showing percentage exceedence while the graph below shows the one-third octave spectra.  There is excellent correlation of these two different instruments, with a difference of 0.2 micron in RMS amplitude over the 190m site and correlation of the spectra within a couple of dB.  The spectra are plotted relative to the reference curve given in EN ISO 3095:2005 and the acoustic TSIs.

Comparison of measurements over 190m from two different CATs, illustrating close correlation of measurements and many standard features of the software

The CAT was used in a so-called “road test” that was undertaken in 2007 under CEN auspices to assess the proposed measurement protocol that is now incorporated in EN15610:2009.  One-third octave spectra from the main test site are shown below.  The CAT is instrument H, other instruments are based on straight-edges of about 1.2m length, except for instrument G, which is the original trolley profilometer developed in Cambridge in the 1970s.

Results of “road test” for protocol in EN15610 made under CEN auspices in 2007

Using CAT for rail reprofiling work

Many CATs are used to assess where corrugation of rails is excessive and whether this has been removed satisfactorily by the reprofiling operation.  An example is shown below of measurements taken a couple of days before grinding and the day after grinding of a heavily corrugated site on a metro.  Residual irregularities are well inside the level of EN13231-3:2011 (which allows a level of +/-10 microns to be exceeded for 5% of the site for irregularities in the 30-100mm range), although only marginally inside the stricter requirements of EN13231-3:2006.  It is nevertheless clear from the one-third octave spectrum that there is some residual corrugation at 50mm, from which corrugation would rapidly develop.  Sometimes a simple acceptance criterion is insufficient.

CAT measurements of a heavily corrugated site before and after grinding

Using CAT to measure “acoustic roughness”

Although many CATs are used routinely for corrugation surveys and for Quality Assurance of rail grinding, the instrument is also sufficiently accurate to be used for the even more demanding requirements of measuring acoustic roughness.  The requirements for this are laid down in standards EN ISO 3095:2005 and EN 15610:2009.  The CAT far exceeds the requirements of these Standards insofar as hundreds of metres of rail can be measured continuously, accurately and quickly, rather than simply isolated lengths of about 1m. Measurements are shown below that were made with a CAT and with a conventional straight-edge instrument on both rails of a 100m test section with a very low level of short wave roughness.  Correlation is excellent up to the 200mm wavelength limit of the 1.2m straight-edge instrument.  Spectra for the CAT, based on continuous records of 100m length, are calculated up to 2m wavelength.

The capabilities of the CAT for measurement of corrugation and acoustic roughness are illustrated in an article that appeared in the April 2010 edition of International Railway Journal (IRJ) and in a couple of presentations that Dr Grassie of RailMeasurement Ltd has made on rail corrugation and wheel/rail noise in 2011 (here and here). Measurements are shown from a variety of railway systems worldwide, which have been supplied to RML by users.  Interesting differences between different types of railway system are apparent that have not (to our knowledge) been noted before.  Probably this is at least in part because no instrument of sufficient accuracy has previously been available and used as widely for measurements of this type.  The IRJ article also shows the requirements for residual corrugation in the European reprofiling standard, EN13231-3:2006, as a one-third octave spectrum and compares this to the spectrum in the acoustics standard EN ISO 3095:2005.  Since the requirements are very similar, it follows that an instrument to demonstrate compliance with EN13231-3:2006 should be of similar accuracy to one that is used to demonstrate compliance with EN ISO 3095:2005.  Fortunately we can supply you with such equipment, even to work at 10km/h or more on a reprofiling train or hi-rail.  This helps the supplier of reprofiling services to know that they are reprofiling track to a high standard, and the railway system to know that the contractor is doing what they require.

RailMeasurement Ltd do not currently make equipment to measure acoustic roughness of railway wheels.  However, we have colleagues who undertake such measurements: please contact us if this is of interest to you. 

Using CAT for surveys

CATs are much used for surveying track to propose where reprofiling is required and also to measure reprofiled track to ensure it has been reprofiled to standard.  If surveys are being undertaken of a complete line or small railway system, it is often inconvenient or impossible to measure the complete line continuously. RailMeasurement have produced a software option that enables the user to concatenate files, producing a single file of several kilometres length.  An example is shown below for a 10.5km stretch of rail.  This example also shows how the standard CAT software can be used to tabulate areas where reprofiling is required, according to criteria selected by the user.  You can hire a CAT from RailMeasurement to undertake your own survey or ask us to quote for this service.

Measurements from several concatenated files over 10.5km, showing also a table of potential reprofiling sites

Using CAT to measure welds

In 2009 we received a request from a good client who was trying to expedite measurement of a plague of badly misaligned welds in track, and who wanted to know if the CAT could be used to help.  We modified the CAT software to mimic the manual measurements that were being undertaken of weld peaks.  There was encouraging correlation of results (see below) but the project was not completed.  If you are interested in this application please contact us.  We are confident that the CAT, or perhaps even more our RCA or MSRCA, could reliably reproduce the irregularities at rail welds.  Our experience is that the irregularity presented by a rail joint depends significantly (and unsurprisingly) on loading.

Comparison of manual and CAT measurements of weld peaks:

Please contact us if you would like to discuss a novel application of the equipment.

Further information on the accuracy and repeatability of our corrugation measurement products is given here, whilst a comparison of our product specifications and use is available here.