Thursday, March 28, 2013

F1 Engine Maps

20 articles later, hello again and welcome to the next installment of F1 Framework topics. This time we have former Formula 1 engineer as guest author on exclusive insight into F1 engine maps - a term often cited around with not much explanation.
Maurizio Bollini is the owner of MET Milano (www.met.it), a consultancy firm involved in motorsport. In the past he worked as engine engineer for Michael Schumacher when he was Ferrari F1 driver, during 1996 to 2006. Maurizo can be contacted at maurizio.bollini@met.it.

F1 Engine Maps Explained

With the generic term “Engine Maps” we refer to a wide set of one-dimension or two-dimension parameter tables loaded into the ECU to control all the engine parameters, that are, at least, throttle opening, injection and ignition timings (duration, phase, etc...). The main two-dimensions maps take the throttle opening and the engine speed as inputs and return a proper parameter as output. All maps are filled with proper numbers during hours of engine calibration done at the test rig and/or at track. Let’s consider now two important maps for the driver: the Driver Pedal Map and the Driver Torque Demand Map.

Driver Maps

The Pedal Map and the Torque Demand Map are two important maps that allow the driver to manage the engine drivability through the redefinition of torque delivery at wheel and in turn, to improve the laptime performance.
The F1 Powertrain Control Unit (PCU)  has a Torque Control strategy properly implemented to control the engine. The torque demand, as requested from the driver by the acceleration pedal, is calculated by the chain composed by the Pedal Map and the Torque Demand Map. The input variable is the accelerator pedal position, given by the Drive By Wire potentiometer and the output variable will be the throttle position, actuated from the PCU by hydraulic or electric actuators on the engine intake butterflies or barrels.
The required maps are the following:
  1. The engine torque map. This is a 2-dimension table with engine speed and throttle as inputs and torque as output. This map is defined point-by-point or by ramps at the test rig with the fired engine and the torque meter. Sometimes is trimmed on track if the car is equipped with torque meters on the transmission.
    Short example of why this is so important - we are supplemented by radio message to his team by Charles Pic in Free Practice 1, Monaco Gran Prix:
    "Engine map with retarded ignition timing helps throttle response and low speed rear grip, but still needs to improve a bit".
  2. The inverse engine torque map. It is calculated by the engine torque map. It is a 2-dimensions table with torque and engine speed as inputs and throttle as output. It is computer calculated.
  3. The driver demand torque map. This is the 2-dimension table with engine speed and driver normalized torque as input and engine torque as output. This is a reshaping of the engine torque. This is the core of the drivability because allow the engineers to completely reshape the engine torque in function of the engine speed, within the boundaries of maximum and minimum torque available (which is the torque at off-throttle – engine brake - and at wide-open-throttle).
  4. The pedal map. This is a 1-dimension map where the input is the normalized accelerator pedal position (0-100%) and the output is the normalized torque (0-100%). It can be considered as a “gain” on the driver demand torque map.
Let see a typical engine map. This map is the torque behavior of the engine. Notice the fluid dynamic effect which makes changing of slopes of the torque at constant throttle – this is the source of drivability issues especially when condition are not optimal (worn tires, low grip, etc).

The driver demand torque map could be like the following. Notice the torque at constant throttle now is linear, unregarding of the original torque. It can be shaped freely. It is like to have a “different engine”.

The negative slope is useful to compensate the wheel spin. In fact when the driver hold on the pedal,  if the wheel spin that develops is causing the engine speed to increase, the torque is actually reduced, minimizing the wheel spin. Different drivers are comfortable with different shapes, so the map is driver defined.
With different pedal maps, the torque driver torque map change as follows:

Legend for each column: 
  1. Soft pickup – driveability for low speed corner or wet conditions
  2. Standard linear map
  3. Smooth top end – for high speed corner
Changing the pedal map is an easy way to change the sensitivity of the pedal while conserving the characteristic shape of the driver demand torque map. You notice that the negative slope is conserved while the distance of constant pedal torques change.

How the driver uses the Engine Maps

On the practical way, the engineers provide a selection of pedal maps and torque maps, driver selectable by the driver using rotary selectors on the steering wheel. In addition to a manual selection, if the regulation permits, pedal maps and torque maps can be automatically selected by the space information on the ECU or selected by gears. In this way, the driver could have the optimal pedal map in function of the corner (different between low speed corners or high speed corners) or have negative slope torque map for low gears and standard map for high gears - all selected automatically.
The rotary selector for the pedal map is labeled somewhat like PEDAL and the driver demand torque map is labeled TRQ. Sometimes the two functions can be overlaid on single rotary selector, to save one rotary selector for other uses. It is recommended to always have a WET configuration on the rotary selectors, which can be used when grip is low. Pedal map for wet is usually soft-pickup shaped.
Everyday work at track is to modify the pedal maps to give the best support to the driver. Driver demand torque maps are usually developed on telemetry analysis overnight at track.

The programming part

Programming tables and thus engine maps are usually structured text files. Some are of proprietary format, other are standard formats like XML.
The data manipulation is done using Excel and VBA with custom macros, Matlab or custom Windows applications.
Engine maps are organized into a database with version support in order to quickly update to previous revision in case of issues.
The System Monitor application included in the MES Standard ECU package include a tool to modify engine maps as well other parameters but usually these tools are on the PC connected to the car or to the dyno, to reprogram the ECU.
Engineers usually prefers to work with shared corporate power tools to manipulate the maps offline.
The tools to manipulate the engine maps required specific algorithm for reshaping as well to merge different data sets with specific criteria, not always found on the support applications more dedicated for data uploading/downloading to ECU.


I hope you liked the words from the inside of Formula 1, dear readers. Maurizio can be further contacted via his Twitter handle. There will be more coming from him in the near future.