FIG. 3 shows the conceptual operation of an illustrative embodiment of the impact load detector described herein. As illustrated, an analog to digital converter (ADC) 40, which can be implemented on the first circuit board 38A shown in FIG. 2, can be connected to and receive data from various sensors, including one or more strain gauges 30, one or more accelerometers 34, one or more temperature sensors 36, and/or one or more other sensors 37. The analog to digital converter 40 can perform analog to digital conversion of the raw sensor data acquired by each of the sensors. However, it is understood that one or more of the sensors can comprise a digital sensor. To this extent, an embodiment of the analog to digital converter 40 can be configured to acquire data from one or more digital sensors.

The analog to digital converter 40 can perform further processing of the data, such as performing simple noise filtering, and/or other preprocessing of the data. In an embodiment, the sensors can have significantly different sampling rates. For example, the temperature sensor 36 can be sampled at a very low rate (e.g., once per minute), while the accelerometer 34 may be sampled at several kHz while a rail vehicle is passing over the rail. As a result, the temperature sensor 36 can use a low speed converter, while the accelerometer 34 can use a high speed converter. To this extent, an embodiment of the analog to digital converter 40 can be configured to accommodate the different sample rates on different channels. Furthermore, an embodiment of the analog to digital converter 40 can comprise multiple converters, each of which can be configured to accommodate a different sample rate.

In any event, the signals from the analog to digital converter 40 can be passed to a filter bank 42 (e.g., implemented on the second circuit board 38B shown in FIG. 2), which can extract various characteristic signatures for analysis. A digital signal processor (DSP) 44 can further process the data and/or prepare some or all of the data for transmission to a remote system 50 for analysis. In an embodiment, once the filter bank 42 and digital signal processor 44 have prepared the data of interest, the data is placed in an output queue 46. Data in the output queue 46 can be sent to a TCP server 48, which can transmit the data for processing on a remote system 50, e.g., through an ethernet connection 49.

As illustrated, data and/or updates also can be received by the impact load detector from a remote system 50. Such a two way connection can enable a remote system 50 to provide software updates, such as modifications to the filter bank 42, new algorithms for data analysis in the digital signal processor 44, and/or the like. Furthermore, a remote system 50 can provide other data for use during onboard data processing by the impact load detector. For example, a remote system 50 can provide wheel sensor data which can permit measurement of the speed of a passing vehicle, a direction of travel of a passing vehicle, and/or the like. This information can assist in rail vehicle and/or rail wheel identification, verification of speed, determination that an unsafe speed is reached, incorrect track usage by a rail vehicle, etc.

FIG. 4 shows an illustrative rail wheel evaluation system 100 according to an embodiment. As illustrated, a rail 2 can include a sensing device 50A and an impact load detector 10 located adjacent thereto. As a train or one or more rail vehicles travel along the rail 2, the sensing device 50A and the impact load detector 10 can acquire and communicate data to a remote processing system 50B via a shared communications connection, such as an ethernet communications link 49. In an illustrative embodiment, the sensing device 50A comprises a wheel sensor, which can acquire data regarding a speed and/or direction of travel of the rail vehicle(s) and trigger the impact load detector 10 to commence acquiring data in response to a rail wheel being detected. However, it is understood that embodiments can include any of various different sensing devices 50A configured to acquire data regarding any of various aspects of the train operation and/or physical characteristics of the rail vehicles.

The remote processing system 50B can be capable of analyzing the data of the sensing device 50A or the impact load detector 10 in isolation. Furthermore, the remote processing system 50B can combine this data in manners known to those skilled in the art to determine whether a fault is present on a passing rail vehicle. For example, a flat spot on a rail wheel must produce impacts in exact timing with one revolution of its wheel. As a result, the remote processing system 50B can compare the timing of impacts detected by the impact load detector 10 with the speed and spacing of the wheels detected by the sensing device 50A to confirm that the detected impacts correspond with the timing for the wheel revolutions.

It is understood that in addition to detecting a flat spot, the remote processing system 50B can detect one or more other faults that may be present on a rail vehicle. Other illustrative faults include an out-of-round wheel, hunting oscillations, and/or the like. Non-ideal wheel conditions, such as out-of-round and hunting oscillations result in certain specific signatures resulting from impacts to the rail. For example, while a flat spot has a characteristic signature of an abrupt spike in the signal, the out-of-round wheel generates a broader signal with different periodicity ratios to the speed. Hunting oscillations have a different periodicity to speed ratio (less than one signal per rotation of wheel). Using the data described herein, the impact load detector 10 and/or the remote processing system 50B can identify such conditions using any of various known solutions.

In any event, the remote processing system 50B also can evaluate the signals from the impact load detector 10, the sensing device 50A, and/or any other sensors, to verify the functionality of the sensors, or detect any failures. The remote processing system 50B can transmit alerts, raw data, service requests, and/or the like, for processing on a central server 50C in response to the evaluation. As illustrated, the remote processing system 50B can use any of varying communications routes to transmit such alerts to the central server 50C. For example, the remote processing system 50B can use a satellite phone uplink 52A, e.g., when located in a remote area, to contact a communications satellite and establish a downlink to the central server 50C. Similarly, a wireless link 52B, such as a cell phone connection, can be made through a cell tower and provide a communications connection with the central server 50C. Also, a direct land-line connection 52C also can provide connectivity to the central server 158.

While only a single sensing device 50A and impact load detector 10 located adjacent to a single rail 2 are illustrated, it is understood that embodiments of the system 100 can include any number of such devices located adjacent to one or both rails 2, all of which can communicate with the remote processing system 50B and/or each other via a suitable communications connection, such as the ethernet communications link