Factors that Could Influence the Accuracy of a Residual Speed Reading on a Crash-Involved Motorcycle
Introduction
Sometimes after a collision, a motorcycle’s speedometer or tachometer will exhibit a residual reading other than zero. In some instances, physical evidence will make it evident that the residual speed reading is not a reliable indicator of the motorcycle’s impact speed. However, if the motorcycle is equipped with a speedometer with a stepper motor, then there is the potential for the residual reading to be accurate and useful in the reconstruction.
Past Research
The authors of Reference 1 explain: “A stepper motor converts electrical current into mechanical movements. The feature of these motors that serves as the critical factor…is that they require continuous electrical power to adjust the speedometer indications either upward or downward – this includes returning to a 0 mph indication. In the case of instantaneous loss of electrical power to the stepper motor, there is no way to generate the stepper motor movement needed to return the needle to its zero position. In the absence of any other applied force, a needle attached to an unpowered stepper motor will remain in its position indefinitely. As a result of this feature, speedometers equipped with electrical stepper motor driven needles will maintain the speed indication that was displayed at the moment of the power loss, colloquially described as a ‘frozen’ speedometer.”
Based on the research by these authors, the process of establishing the reliability of a residual speedometer or tachometer reading on a motorcycle could involve the following steps:
1. Determine if the speedometer or tachometer on the subject motorcycle employed a stepper motor.
2. If a stepper motor was present, determine the stepper motor manufacturer, model, and/or type.
3. Quantify the torque resistance of the speedometer or tachometer needle to rule out that the inertial forces of the collision or post-collision handling did not change the position of the speedometer needle.
This reference notes that, “for motorcycle speedometers with stepper-motors of high torque resistance, the frozen needle indication should be a reliable indication of the motorcycle’s impact speed in situations where there is no pre-impact lock-up of the speed-sensing wheel.” On the other hand, Reference 2 observes that “residual readings on speedometers with low resistance to needle motion are not reliable indicators of that vehicle’s speed at impact, even with a sudden power loss coincident with the time of the collision.”
The authors of Reference 1 also noted that the presence of a stepper motor can sometimes be detected with testing of an exemplar motorcycle of the same year, make, and model as the accident-involved motorcycle and having the same original manufacturer speedometer or tachometer. They state that “initial observations can be made simply by starting the vehicle. In our testing, all confirmed stepper motor-driven speedometers performed a visible calibration procedure at the onset of power. When the motorcycle gauge cluster was initially provided power (such as by turning the ignition on), the stepper motor-driven needles, initially at their zero values, rotated to their maximum gauge position, paused briefly at their maximum gauge reading, and then returned back to their zero-value. This start up movement is a re-calibration procedure that allows the stepper motor to know its exact location by driving to its internal dead stop, or when the coils in the motor stop generating a feedback pulse.”
The presence of a stepper motor can be further confirmed by instantaneously removing “power to the exemplar vehicle’s gauge cluster while the indicator needle(s) is displaying a non-zero value. In order to provide speedometer needle readings while safely stationary, the motorcycle’s rear drive-axle should be safely and securely hoisted where the drive-wheel is completely off of the ground, such as by using a manual motorcycle hydraulic lift, or the motorcycle’s center-stand…The simplest method of achieving this instantaneous power loss is to remove the motorcycle’s primary fuse, or if it can be located, pulling the motorcycles gauge cluster fuse, while the gauge needle is displaying a non-zero value. However, due to the complexity of some motorcycle electronic systems, identifying the correct fuse for ensuring instantaneous loss of power to the vehicle’s gauge cluster can be challenging. If the analyst were to be unsuccessful at locating a fuse directly linked to the instrument cluster, the next step would be to remove the wiring harness directly from the speedometer body. This typically requires the removal of some body panels and the removal of the speedometer from its mount. Once the gauge cluster has been removed, and while the drive-wheel is rotating, the primary harness into the gauge cluster should be quickly disconnected. If the needle freezes at the pre-power loss value, then it can be concluded that the subject vehicle’s gauge is driven by a stepper motor, and the analyst can continue to dismantling. The subject speedometer, or a replacement part from the original manufacturer with the same part number as the subject gauge, can be cut open in the back for visual inspection of the component parts and structure. Because manufacturers parallel source some components such as stepper motors, it is important to identify, when possible, the specific make and model of stepper motor inside the subject instrument gauge. This process renders the speedometer unusable, but it is also the most conclusive procedure to determine the stepper motor type.”
Reference 1 identifies the following motorcycles that had stepper motor speedometers and tachometers:
2008 Harley Davidson Heritage Softail Classic FLSTCI
2014 Harley Davidson Heritage Softail FLSTC103
2015 Harley Davidson Street Glide Special FLHXS
2014 Triumph Bonneville
2015 BMW R1200RT
2015 Indian Chieftain
2014 Triumph Tiger Explorer (tachometer only)
Reference 3 reported testing of six electronic needle-display speedometers. The authors elevated the rear wheels of the motorcycles and accelerated them to pre-determined speeds. They then disconnected the speedometer wiring harnesses. They found that the “dial indicator would move slightly up, down, or remain in place depending on the model of the speedometer. The observed change of indicated speed was within ±10 mph upon power loss.” In addition, these authors subjected speedometers to impact testing on a linear drop rail and quantified the minimum acceleration to cause needle movement. This type of testing constitutes an alternate method to establishing high torque resistance in the stepper motor and has the advantage of being more directly related to the type of loading that would be expected during a collision. The results obtained by these authors are summarized below.
1999 Harley Davidson FXDX [gear driven (spur)] – “During the power interruption testing, the FXDX model speedometer stayed within ±1 mph of the indicated speed prior to the power interruption…During the drop testing, the FXDX speedometer dial…indicator rotated approximately 2-3 mph due to the impact, in the direction of pre-impact momentum…the deceleration reached nearly 350 g, over an impact duration of roughly 7 ms.”
2013 Harley Davidson FLTRU [gear driven (worm)] – “During the power interruption testing, the maximum drop in speed of the FLTRU model speedometer was approximately 8 mph…At no point during the power interruption testing did the dial indicator observably increase in speed…During the drop testing, the FLTRU was very resistant to rotation at relatively high deceleration rates. After a six foot drop test, the dial indicator on the FLTRU remained within a 1 mph deviation. The six foot drop test again produced deceleration values at nearly 350 g over 7 ms.”
2015 Honda GL 1800 Gold Wing [direct drive] – “During the power interruption testing, the GL 1800 model speedometer returned to zero after every test, regardless of dial indicator position.”
2015 BMW R1200GS [gear driven (worm)] – “During the power interruption testing, the R1200GS speedometer stayed at the indicated speed. There was no visible increase or decrease in speed at any dial indicator position. During the drop testing, the R1200GS speedometer dial was very resistant to motion at relatively high deceleration rates. The dial indicator remained at the pre-test position below roughly 200g. At impact levels reaching 230 g, the dial indicator rotated in the direction of pre-impact momentum roughly 1-2 mph. At impact levels of 310 g, the dial indicator rotated approximately 2-3 mph.”
2014 Triumph T100 Bonneville [gear driven (worm)] – “During the power interruption testing, the T100 speedometer stayed at the indicated speed, very similar to the R1200GS speedometer. There was no visible increase or decrease in speed at any dial indicator position. During the drop testing, the T100 was similar to the FLTRU speedometers and very resistant to moving at relatively high deceleration rates. After a six foot drop test, the dial indicator on the T100 rotated in the direction of pre-impact momentum approximately 1-2 mph. The six foot drop test again produced deceleration values at nearly 350 g over 7 ms.”
2012 Kawasaki ZX-14R [direct drive] – “During the power interruption testing, the ZX-14R speedometer moved ±10 mph. The dial indicator would either increase or decrease in indicated speed, or not move at all depending on the position of the stepper motor before power loss…During the drop testing, the ZX-14R speedometer was more likely to rotate than the other speedometers tested.” The speedometer rotated approximately 7 mph due to an impact that reached 50g over 13 ms.
The drop testing reported in Reference 3 subjected the speedometers to significant impact forces. The significance of forces applied to the speedometer during a crash would need to be evaluated on a case-by-case basis. These forces would depend on a number of factors, including the relative speed of impact between with motorcycle and the struck vehicle, the distance from the front of the motorcycle to the speedometer, the stiffness of the motorcycle forks, the stiffness of the struck vehicle structure, and whether or not the speedometer itself became a part of the crushing region of the motorcycle during the collision. As an illustration, consider a scenario in which an upright motorcycle strikes a non-moving, non-deforming barrier at a speed of 40 mph. Assume an impact duration of 55 ms and a haversine collision pulse. With this collision pulse shape, a peak acceleration of approximately 72.9 g gives the motorcycle a of 44 mph, and thus, yields a collision with 10% restitution. Under this scenario, the motorcycle will experience a maximum dynamic crush of approximately 1.5 feet at approximately 40 ms into the collision. If the speedometer is 1 foot back from the front of the motorcycle, then it will be traveling approximately 31.8 mph when it reaches the barrier.
Now, assume that portion of the motorcycle comes to a stop against the barrier in another 10 ms, that will produce an average acceleration on the speedometer of approximately 145 g, and perhaps a peak acceleration of 290 g. This would be likely to be on the high-end of the accelerations that would be experienced by a speedometer that remains attached to the motorcycle, since this illustration assumes a non-moving, non-deforming barrier. On the other hand, consider a scenario in which the motorcycle impacts a soft region of a passenger car, the motorcycle’s speedometer is outside of the crushing region of the motorcycle, and the impact duration is longer. Assume an impact duration of 155 ms and a haversine pulse. In this scenario, the speedometer would experience a peak acceleration of around 26 g, significantly below the acceleration levels tested in Reference 3.
Based on these studies, a frozen speedometer reading would reliably report the reading that was on the motorcycle speedometer at impact when:
1. The speedometer is driven by a stepper motor.
2. A power loss to the speedometer occurs during the collision.
3. The speedometer needle does not move significantly when a power loss occurs.
4. The speedometer needle has high torque resistance and does not move significantly under impact loading.
Factors to Consider in Practice
A remaining question is whether the speedometer reading that was present on the motorcycle speedometer when impact occurred is a reliable indicator of the actual speed of the motorcycle at impact. In instances where a motorcyclist employs emergency level braking prior to impact, there may be some discrepancy between the actual speed of the motorcycle and the reading on the speedometer. One reason for this could be latency in the speedometer (i.e., how quickly the speedometer responds to actual changes in speed). Another reason could be wheel-slip due to the heavy braking. This is similar to the issue that comes up with the pre-crash speeds reported by a passenger car event data recorder, where when heavy braking is present, there can be a discrepancy between the vehicle-indicated speed and the actual speed of the car. Neither of these issues has been addressed in prior literature related to motorcycle speedometers, and testing is needed to address them.
Another issue that could cause a discrepancy between the speed on the speedometer and the actual over-the-ground speed of the motorcycle is lean of the motorcycle. To see this, consider a motorcycle traveling straight down the road without any lean. If this motorcycle uses a rear wheel speed sensor to feed the speedometer speed, then the motorcycle would need to have a rolling radius for the rear tire listed within its programming. Now, if the motorcycle is traveling and continues to travel at a steady speed as the rider leans the motorcycle, the motorcycle will roll onto a portion of the rear tire that has a lower effective rolling radius. In order for the actual over-the-ground speed of the motorcycle to stay constant, the wheel will have to roll at a higher rate. However, the rolling radius value within the programming will not have changed, and so, the motorcycle will calculate and report a speed that is higher than actual because it will use a higher rolling radius than actual. Thus, if a motorcyclist leans to swerve prior to a collision, the reading on the speedometer at the time of impact could be slightly higher than the actual speed. This issue also has not been addressed by prior literature and testing is needed to address this.
References
Montalbano, P., Melcher, D., Keller, R., Rush, T. et al., “Testing Methodology to Evaluate Reliability of a “Frozen” Speedometer Reading in Motorcycle / Scooter Impacts with Pre-Impact Braking,” SAE Technical Paper 2016-01-1482, 2016, doi:10.4271/2016-01-1482.
Anderson, Robert D., “Post-Collision Speedometer Readings and Vehicle Impact Speeds,” Collision: The International Compendium for Crash Research, Volume 5, Issue 2, Fall 2010, pp. 33-41.
Fatzinger, E., Shaw, T., and Landerville, J.,
“The Effects of Power Interruption on Electronic Needle-Display Motorcycle Speedometers,” SAE Technical Paper 2016-01-1474, 2016, doi:10.4271/2016-01-1474.