Imagine you are a race engineer on a high-level motorsport team. You are responsible for the development of the vehicle on the track and you must give your blood and sweat to make sure your driver can squeeze the last bit of speed out of the car, and if both of you are not able to win, your job is not being made correctly. Now, if there is some problem that makes your car go slower than you predicted, how would you know what was wrong in order to solve the problem?
This is where data acquisition (also called data logging) comes into play. An engineer who worked with me during an internship once told me that few things are more valuable in motorsport than information. If you have a racing team, accurate information is one of the things you have to invest from the beginning. It is good information that can give you a reasonable judgement when making key decisions. Data acquisition provides the engineers with the information they need in order to improve the performance of the race vehicle/driver system. To study that, let us begin with the definitions.
What is Data Acquisition?
Well, let’s say we need to gather information on your car in order to improve its performance. You can count just on your instruments and gauges to that. The race driver would have to look at the instruments (a lot of them to look) every time anything worth of attention happened (and inevitably take his eyes from the track), then remember what the readings on the gauges indicated (all of that while suffering massive accelerations) and later after the session ends, pass the information to his engineer. It sounds like impossible to the human nature, right? Data logging, as the name implies, is the recording of the information measured during the time the car is running, for later in-depth analysis, and a permanent log can be made for future reference. It will also help to define what it is not.
What Data Acquisition is not?
Once and for all, data acquisition is not telemetry. The confusion arises from the fact that they both are similar, but they are not the same. Data acquisition is the recording of information for later analysis, which requires the data engineer (or any other person in charge of doing the job) to the download the data from the Data Acquisition System (DAS) while the car is on the pits, and analyse it later. Telemetry, on the other hand, is the recording of information for instantaneous analysis. This means that data is transmitted by radio or similar means to a remote receiver, generally installed in the boxes or to a team base located near. Basically it is an extension to data logging, but it is not the same thing. Now, to get you a little bit more excited, I will show you two videos of telemetry in action, the first one on a Formula One car, from Sauber F1 Team, in a filming day in Fiorano, with their C33 from 2014 season. The second video is of a lap from the Audi R18 e-tron Quattro, during the 2012 24 Hours of Le Mans.
WEC 24 LeMans 2012 – Audi R18 e-tron quatro – telemetry
Now, before going on with the post, I would like just to thank all the people responsible for producing both videos. Seriously, they are so awesome that I cannot even describe it. As I write more posts on data acquisition, you will be able to “read” them, and start identifying patterns and know what to expect. Now that we have put the record straight, let us move on to the main reasons why data acquisition is used in motorsport.
6 Reasons to use Data Logging
The use of data is so common in modern days motorsport that is hard to find where it is not applied. From factory to track, almost every part of the vehicle development rely on some data collected on track. The main reasons why racing teams use data acquisition are listed below.
#1 To Analyse Vehicle Performance
It is impressive the amount of conclusions that can be drawn about the vehicle performance from a small amount of parameters. As you will see in later posts, just by looking at the engine RPM and vehicle speed data, it is possible to get a huge amount of information, and it can be used to support the information given by the driver. The race engineer can identify any handling problems the car might have and the locations on the track where it occurred. This is then used to decide which setup changes should be made for the next driving session. This is the main reason why data logging is used on the track, followed by the reason below.
#2 To Analyse Driver Performance
Again, by looking at only a few parameters (RPM and speed are the most important, but also steering angle and throttle position), a lot of conclusions can be made about how the driver is performing. The comparison between different laps can help to find out why the driver is quicker on a lap than on the other, and show how much room for improvement there is to work on. Also this kind of data helps to understand the driver’s style and to compare it with different drivers performances. The latter type of analysis is particular useful in championships where more than one driver uses the same car.
#3 To Help with Vehicle Development
On motorsport categories where the teams design their own cars (e.g. Formula One and World Endurance Championship), there is a continuous development cycle race after race. The process is generally like this: the design team design the car, which is taken to the track where it is tested and races. After the race weekend, the race engineer will gather a lot of information on what went wrong and what should improve, and then will brief it to the design team, when he gets to the factory. The design team will them take the information passed from the race engineer and them work on the improvements. The importance of data acquisition here is quite obvious. Also, specific goal-driven measurements will identify which solutions work and which don’t, helping to determine in what direction to focus.
#4 To Monitor Reliability and Safety
The parameters that represent vital functions can be logged in order to ensure reliability. For example, the analysis of oil pressure and temperature, along with water temperature can indicate the presence of possible failures on the engine or gearbox. Battery voltages and other electrical parameters can indicate malfunctioning of electronic circuits or hybrid power units. Moreover, safety can also be monitored through parameters like tyre pressures.
#5 To Determine Vehicle Parameters
On categories where teams make use of race car simulations, the models can be developed by the team itself, or a third party software can be used. In both cases, all relevant parameters should be known by the programmer or the responsible for running the simulations, to guarantee enough accuracy of the model. Data logging comes into play here, by helping to record parameters that need to be measured under racing conditions.
#6 To Keep Maintenance Logs
The DAS records massive amounts of data from the car. This data can show how long the car has run and what happened during this time. This, coupled with important parts of the car, can be used to create a lifetime log of critical components, and determine which parts should be replaced, and when rebuilds should be performed.
What Kind of Data?
There is a huge amount of parameters to look at when logging data, and these parameters can be related to any part of the car. However, they are generally categorized in three types, namely powertrain and vital functions, driver and chassis. Of course these parameters are interrelated, as a race car/driver system is very complex in nature, and the parameters from one input of the system reflects in the output of other systems.
The main powertrain-related parameters to keep track are engine RPM (among engine parameters, this is the main one in terms of vehicle performance), oil pressure and temperature, water temperature, airbox pressure, inlet air temperature, fuel pressure, fuel mass flow, gearbox and differential temperature, turbocharger boost pressure, exhaust gas temperature, battery voltage, and so on.
Chassis-related parameters may be wheel speed (of one or the four wheels, we will get on that on later posts), lateral and longitudinal G-forces, steering angle, damper displacement, suspension loads, ride height, tyre pressures and temperatures, brake disc temperatures, and aerodynamics parameters like air speed and local pressures.
Driver-related parameters are those from powertrain or chassis, which the driver has some control on. Examples include steering angle, throttle position, gear, brake position and brake line pressure.
Some data acquisition specialists agree that the most important parameters to log are speed, engine RPM, steering angle and throttle position. All in all, speed is what really matters when it comes to racing, and throttle and RPM are directly connected to the development of speed. Also, the use of steering angle together with throttle angle can provide a lot of info on a race car’s handling.
It is also a good idea to add a lateral G-force measurement, and that is for a real good reason: most of the data analysis softwares use the measurement of lateral accelerations together with the speed information to calculate and display a map of the circuit where the race goes on. This is actually the path taken by the car during a lap. It is useful because the software shows a dot on the track map showing the current position of the car at the instant analysed. This is displayed together with a cursor (a vertical line on the data graphs) to show the exact value of the measured parameters at the instant. This way it is easier to recognize the patterns that emerge on the graphs associated with specific points of the track (for example, the crescent speed from corner exits to the end of straights).
Measuring the Data
The measurements recorded by a DAS are actually taken by sensors installed all over the car (as you saw on Sauber F1 video, there are about 140 built into a Formula One Car!). The sensors are components that transform an actual physical property into an electric signal (more on how sensors work on the next post). These signals are then compared by the DAS with its stored calibrations, and a value for the property is calculated and recorded. Different types of sensors are required to measure different parameters, but a single parameter can be measured by different types of sensors. For example, the vehicle velocity can be measured by a hall effect magnetic sensor installed on the wheel, by a Correvit optical sensor, by a pitot tube (A typical Formula One car uses the three at the same time), or even by GPS measuring. Specific sensors can measure temperatures, angular and linear speed, angular and linear displacements, pressures, strains in materials, accelerations, changing magnetic fields and so on.
Recently, GPS-based (Global Positioning System) DAS have become greatly available. The system can measure successive instantaneous positions of the car, and from them, calculate the vehicle speed, its heading, longitudinal and lateral accelerations and provide a track map. These systems are a relatively cheap solution to log speed and accelerations.
One sensor connected to the DAS is called a channel. DASs are always limited in what concerns the maximum number of channels they can log, and that brings a problem to the data engineer to solve (how many and which channels should he log?). There are options available from all needs and budgets, from a single channel rpm only device to professional devices that offer more than 120 channels. The point is, each channel can only record one parameter at a time. Of course you may have more sensors than channels, and then you will have to choose which sensors you are going to use in each driving session. For example, you can log performance related parameters (RPM, speed, steering wheel angle, G-forces and throttle position for example) during free practice and qualifying sessions and switch to durability-related parameters (e. g. oil pressure and temperatures, water temperature, gearbox and differential temperatures, battery voltage) during the actual race.
Recording the Data
In terms of recording your data, you should have two concerns about your DAS: sampling rates and storage memory. Sampling rates, or sampling frequencies, are the quantity of measures (sampling) recorded per second. Depending on the level of the DAS, it can vary from 10 Hz (10 samples per second) to 1000 Hz. In case of more advanced systems, it is possible to select different channels to be logged at different rates, which makes a much more efficient way to use memory. This can be used for parameters that do not change abruptly, and hence, do not need a high sampling rate (e.g., water temperature which could be easily be measured once every 10 seconds). On the other hand, some parameters need to be measured at a much higher frequency, as it is the case for damper displacement, which can be really fast on some race cars while hitting a kerb.
The number of channels logged combined with the sampling rates, determines how much memory will be needed in order to store all the data logged. Simply put, more channels or higher frequencies means more data, and thus, more memory needed. The good thing is, some DAS systems are “modular”, allowing more channels and more memory to be added later.
Downloading and Analysing
So far, I have talked about two main aspects of data acquisition (measuring and recording). But there are two other major parts of the whole thing, which are downloading the data and analysing it. Downloading is the process of collecting the data from the car, after the driving session has ended, or the car has entered into the pits. This is generally done by connecting a cable from the DAS to a computer (usually a laptop) and sending the appropriate commands to the computer. Some DAS have a memory card instead of a cable to retrieve data, and some use wireless connections to do that.
Once transferred, the data can be “read” and analysed in the most convenient way possible (generally through graphs and charts). The data is generally plotted in the form of a graph of the parameter of interest in the y-axis versus time or distance in the x-axis.
The data starts to be recorded when the DAS is powered up, but some systems start recording when some speed sensor detects forward motion. This is useful to get rid of useless data from periods like engine warm up.
At first, the graphs may look like a random plot on the screen, without any meaningful shape, but after examining it for a brief period of time, you realize the patterns that emerge from different places of the track. With some practice, it is possible to see how the plot of a performance related parameter during an entire lap will look like for a particular track.
Future posts on this section of Racing Car Dynamics will show you how data from these parameters will look like, and how you should interpret them. If you want to read more about this, please make sure to subscribe to our newsletter below. You can also follow us on Twitter, Facebook or Google+. See you!
KEEP UP TO DATE