Mk1 was UBCO Motorsports' first competition legal vehicle and existed in 2 forms. Mk1.1 and Mk1.2, Mk1.1 was centred around making it to competition and was kept as simple as possible. Mk1.2 was significantly more sophisticated.
Mk 1.1
Insert Summary Here
Mk 1.2
Competition summary:
Electrical
A primary objective of our electrical system was to create a modular system which could easily be adapted to suit our needs as the car evolved through it’s development. For this reason a Haltech Elite 1500 ECU was selected as a central module. An expansion module located in the nose of the car reads all the sensors in the front half of the car then relays data over CAN to the main ECU. Our data logging requirements are met by a suite of AIM products including an MXL dash, a SmartyCAM and several expansion modules. The beauty of a modular electronics system is how flexible it allows us to be when testing new hypotheses and developing new systems.
Driver Aids
Traction Control
After our first season of development we quickly realized the speeds possible were beyond the driving skills of many of our team members. If we were to want to be competitive we had two possible options, one being allocating insane amounts of seat time to each of the drivers, the other being develop a system to help manage the car’s power. The hardest part of driving the car at pace was to keep the car in line on corner exit as the power came on extremely quickly and with our current chassis setup this often lead to spinning. Traditional traction control systems are often frustrating to use as they intervene extremely aggressively and in some cases are solely based on wheel slip. We wanted to avoid an overly intrusive system as in many cases it can be beneficial to induce a minor slide to help neutralize the car’s handling. To accomplish our goal, we decided to use a yaw sensor and a lateral accelerometer to map our slip table; this lead to what could be defined as a hybrid traction control and stability control system. The lateral accelerometer is used to determine the magnitude of the corner speed and the yaw sensor is used to determine how quickly the car is changing direction. An example being a high speed corner, a very low yaw rate will be present yet high lateral acceleration may be present. This scenario is an example of where intervention is unlikely and a higher wheel slip percentage is allowed. In a moderate speed tight corner, a high yaw rate and moderate acceleration will be present; this is where the system is most likely to intervene. In summary the system is used to help reduce the probability of spinning the car while maintaining the ability to neutralize the car with throttle inputs.
Drive by Wire
In our first season we had significant issues with throttle response as the CBR’s power curve is extremely abrupt in the mid range. This coupled with a relatively large throttle body made drivability and precise control an issue. To address this issue for the new season it was decided to convert to drive by wire. A drive by wire system allows for the mapping between the throttle pedal and the throttle pedal to be non-linear. By creating a gentle initial mapping for the first section of the pedal it will enable precise throttle control for low speed maneuvers while retaining full throttle capabilities. Drive by wire also allows the ECU to have control over the throttle body for uses such as traction control. In many cases drive by wire systems are safer than a cable throttle as they allow for a wide variety of diagnostic functions. An example being when the throttle plate is mechanically jammed, in this scenario the ECU is able to recognize a discrepancy between the requested and actual throttle positions. Once a discrepancy as been found spark and fuel may be removed to return the engine to idle and introduce limp flags. This type of failure is much harder to detect with a mechanical throttle system.
Launch Control
To help get the car off the line as quickly and consistently as possible we have implemented a launch control system. The system is based off wheel slip as well as pre-set targets for ignition, fuelling and speed. The system is always active once certain criteria are satisfied, namely: wheel speed below 15 km/h, coolant temperature over 70ºC and a requested throttle over 85%. The launch control system is configured from parameters of actual vehicle performance to allow it to perform well on various surfaces as well as adverse weather.
Clutch Assist
Many of our drivers were having a hard time getting used to the clutch in the car over the past season. To help reduce this issue a clutch assist system as is present on many production vehicles was introduced. Our system operates off a clutch pedal position sensor which is then mapped through the ECU to both idle the car up and advance the timing while the clutch is engaging. The combination of these two changes should help by firstly giving the driver a sense of the clutch’s state and secondly making the car harder to stall. To ensure the system is safe it is only active up to 5km/h, is only able to request a 10% idle up and is not active once the clutch pedal is above 90%.
Flatshift
A flatshift system was introduced to help reduce shift times. The critical inputs to arm the system are vehicle speed, throttle position and clutch pedal position. If a criteria of a throttle position over 80%, vehicle speed over 15kph and a clutch pedal position under 90% engagement are met the system interrupts the timing to allow for a gear change without lifting out of the throttle. The primary benefit of flatshift is that no engine momentum is lost, ie the crankshaft speed is retained allowing for an amount of energy to be available immediately upon clutch re-engagement.
Engine Management
Fuel Strategy
A volumetric efficiency strategy was selected to control our fuelling for several reasons. Primarily, the need to compensate for changing atmospheric conditions, volumetric efficiency allows for these corrections to be included with very little additional work. Haltech’s volumetric efficiency strategy can automatically correct for air temperature and to an extent for barometric pressure. A volumetric efficiency tune is also very simple to modify for hardware changes as the car evolves. Fuel load is calculated based on a manifold pressure sensor. This was chosen primarily due to the restrictor which we have noticed causes undesirable fueling conditions during transient throttle. A manifold pressure sensor and RPM combination has been more representative of actual air flow through the engine than a throttle position and RPM based map.
Fuel Correction Factors
Barometric Pressure
Barometric pressure should be accounted for due to the locations where the car is being run. Altitude change can have adverse affects on naturally aspirated engines, especially on low displacement high rpm applications. To ensure our fuel mixtures remain safe and relatively optimal the Haltech’s on-board pressure sensor is used to measure atmospheric pressure and correct the fuelling map accordingly. The car was tuned in Kelowna, British Columbia, as a result this is where a correction factor of zero will be present. At any other altitudes the map will be modified to suit the conditions.
Air Temperature
The fuelling map is corrected based on the air temperature, this is critical because of the wide temperature spread the car is operated through. Some test days are run in temperatures below 10ºC as we are located in a cold climate. For this reason, the car must be able to function in both hot and cold climates. Much like a production vehicle we require it to start and operate normally in all conditions. In cold weather more fuel is required as the air is denser with a higher oxygen content reaching the combustion chambers. In hot weather the opposite is true, less fuel is required. The change between our daily lows and highs can be in the range of 20ºC throughout a test day, this makes our temperature correction factors extremely important.
Coolant Temperature
Coolant temperature correction factors are critical to starting behaviors as well as to the response of the vehicle. When an engine is cold it often is not very well sealed, essentially operating with reduced compression until oil has a chance to circulate, pistons expanding to their engineered sizes and valves expanding to their operating sizes. The culmination of these small factors make a substantial difference to the mixture. Mixtures are kept rich to ensure a quick warm up and to avoid internal engine damage. In our team’s case we are often pulling the car out of an unheated trailer into cold conditions, cold start is extremely important as we are often operating in near zero conditions through the spring and fall. As a means of engine protection the mixture is also richened when coolant temperatures exceed 110ºC to help avoid any engine damage.
Wideband O2 and Fuel Trim
Our Haltech ECU has been configured to use both a short term and a long term fuel trim learning strategy. This means we are able to read the true air-fuel ratios and correct the amount of fuel delivered when they are compared to the desired AFRs. The short term fuel trim is allowed to act in small corrections, for example during the warm up cycle. The long term fuel trims are active over the life of the ECU and are adjusted over many data points for each load site in the fuel table. Long term fuel trims are an extremely effective way of optimizing a fuel map as they are able to improve the fuelling map as the car is driven. In essence the more the car is driven the closer to the target AFRs the engine management system will get. Haltech has also included a “quick tune” feature which uses the logged data from the fuel trims to touch up the base fuel map. If this feature is used iteratively in conjunction with a good base dyno tune the base map can be extremely well optimized. Fuel trims are also important for diagnostic purposes as they can be used to detect intake and exhaust leaks electronically. If the fuel trims go extremely rich, over 10%, it can be an indicator of either an intake or exhaust leak. Desired thresholds can be set for check engine lights and diagnostic faults being reported from these trims.
Exhaust Gas Temperature
We integrated an exhaust gas temperature correction factor into our tune as we wanted to be able to run very close to the stoichiometric values for gasoline. An ideal air-fuel ratio for gasoline is 14.7:1, this is where the optimal burn takes place and the most energy is released from the gas. Most production vehicles are tuned to run at 14.7:1 or higher in cruising situations for the sake of fuel consumption and to richen down into the low 13:1 range under high load. This enrichment degrades the efficiency and has very little effect on power output. The reasoning for richening the mixture under load is to use gasoline as a coolant for the combustion chamber and exhaust. Many modern engines, including our CBR600RR, use aluminum pistons which are vulnerable to heat damage under extreme conditions. In an effort to optimize fuel consumption we are running a very lean mixture even under load. Our justification for this is the CBR600 is able to produce far more power in it’s original form than it is with our 20mm restrictor. Heat and power are directly proportional to each other, and thus how we have deemed it appropriate to run lean while we are making a lower power output than the original designed output. Our target is to maintain AFRs of between 13.8:1 and 14.7:1 at all times, these mixtures have been deemed safe when combined with our exhaust gas temperature correction system. A danger zone of above 500ºC was decided upon as a starting threshold for enrichment based on Aluminum’s 660ºC melting temperature. Above 500ºC an enrichment factor starts with a small amount of 4% and ramps up exponentially to 50% by 650ºC. If 645ºC is reached an engine limp mode is operational until the temperatures return to a safe range.
Timing Strategy
A throttle position based strategy was selected as it was found to have the fastest response to any throttle inputs. A manifold pressure based system would provide a very similar outcome however would be subject to the mechanical lag of the throttle body response and the change in air flow (pressure). When combined with our drive by wire system it means the ECU now has the upper hand on desired timing angle; it is now commanding the throttle blade rather than catching up to it. Having this advantage although in the order of milliseconds makes a notable difference to the feel of the car.
Timing Correction Factors
Coolant Temperature
During a cold start and through warm up timing is retarded to avoid any harmful knock while the engine is going through a vulnerable state. Once operating temperatures are reached normal timing maps are resumed. If 110ºC is exceeded timing is retarded to help reduce coolant temperatures in an effort to protect engine internals.
Knock Sensor
We equipped our car with a knock sensor as ignition angle has been proven to be a very important factor to engine efficiency. In order to be able to safely obtain the optimal ignition angles some form of knock detection is essential; persistent knock at 12 000rpm can quickly damage engine internals. Our knock control strategy works on the principal of innocent until proven guilty. The base ignition table is setup as the best case scenario then the knock control system retards the timing as required when knock is detected. If no knock events are detected for a period of time any timing removed will be restored until another knock event is detected. Event detection is based on the amplitude of a signal of a known frequency of knock in a CBR. If the signal is above the threshold it is recorded as an event, if is below the signal is ignored. Knock is especially important for our team as we have no ability to store fuel or easily obtain race gas, this means we are often running pump gas of unknown quality.
Engine Protection Systems
Limp Modes
We have configured our Haltech ECU with 3 stages of limp mode, levels 1,2 and 3. Level 1 is the least severe with a rev limiter being imposed at 11 000 rpm, a fuel enrichment of 5% and timing being retarded by 2 degrees. An example of a level 1 fault could be a coolant temperature or an exhaust gas temperature being exceeded by a limited margin. A level 1 limp will be reset once the conditions are lifted or the ignition is cycled. Level 2 is more severe and results in a rev limiter at 6000 rpm, a fuel increase of 7% and timing being retarded by 5 degrees. An example of a level 2 fault would be a sustained over-temperature for any critical temperatures such as coolant temperature. If an Inlet Air Temperature sensor failure were detected it too would be treated as a level 2 fault. Level 2 may be reset by cycling the ignition. Level 3 is the most severe limp mode which limits rpm to 2000, reduces timing by 6 degrees and fuels an additional 8%. A level 3 failure would be any oil-related failure or an extreme lean mixture detected by the wideband. Level 3 faults may be reset by cycling the ignition. All limp modes will return if the conditions are still present after they are reset.
Driver Warnings
Any induced fault will trigger a malfunction light informing the driver through the AIM MXL dash as well as the introduction of a limp mode if the circumstances require. Any event will be marked by the data logger as well as the Haltech’s fault logging functions. There are additional conditions which have been deemed dangerous to engine health yet a slow enough occurrence to be dealt with by a driver alarm. An example of this being oil temperature which is only available to the AIM dash and not to the ECU.
Diagnostic Systems
Drive by wire
As mandated by FSAE rules our car is equipped with the required redundant sensors for throttle blade position and accelerator pedal position. Any difference between the sensor readings is treated as sensor implausibility, if detected the throttle is retuned to idle with a fault flag set. A brake pedal sensor is present which if the accelerator present is over 20% and the brake pedal is depressed the car is shut down. This is because it is likely due to a stuck accelerator pedal. A standalone brake system encoder is also present which shuts the car down in the event of an ECU failure. The plausibility device also makes use of the brake pedal sensor and compares it to the throttle blade’s position, if it is detected over 20% with the brakes firmly depressed it will shut the car down.
Sensor diagnostics
Our current configuration has minimum and maximum operating voltages configured for all on-board sensors. This allows for easy detection of total sensor failure or wiring failure. Any sensor failure will set a fault code. Certain sensors also have advanced diagnostic functions for operational pressure ranges or for expected resistances. If any expected parameter is exceeded a detailed fault code is reported along with engine parameters at the time of the fault.
Engine Performance
If engine performance is not as expected fault codes will be reported for the suspected cause. An example being an extremely rich long term fuel trim caused by an exhaust or intake leak. An oxygen sensor fault may also be detected by slow response times or lack of response. Misfires can be detected electronically using a combination of the knock sensor, wideband and the behavior of the coil pack. This allows us to have advanced troubleshooting capabilities and near production vehicle capabilities.
OBD2 Compliance
We elected to make our vehicle OBD2 compliant as per to allow us to use a variety of diagnostic tools if need be. Being OBD2 compliant also makes connecting to a dyno much simpler as we are now using the same standard as every production vehicle since 1998.
Data Logging
AIM MXL
For our primary logging system, we are using an AIM MXL Pista logging dash. This serves two purposes, the first being displaying multiple parameters to the driver. The second is the high speed logging capabilities for both engine parameters as well as external sensors which can be used for primary sensor validation as well as vehicle setup. Our AIM system has secondary coolant temperature sensors, additional exhaust gas temperature sensors, additional transmission data as well as every sensor available to the Haltech being relayed over CAN. Future expansion plans include suspension travel sensors, steering angle sensors, brake pressure sensors and many many more. An example of the purpose of logging the available data is evident while configuring our Drexler differential. Individual rear wheel speeds may be used in conjunction with calculated channels to determine what effect a change has had. Data logging is also extremely valuable when developing vehicle controls systems such as our traction control. Data has been the foundation for our strategy, we began by looking for indicators of a slide from any of the on board sensors in confirmed slide events. Slide events were confirmed using both GPS data and the video from the SmartyCAM(with data overlay and timestamp). When we began to see a correlation between the yaw rate sensor, the lateral accelerometer and various types of vehicle behavior they were the chosen sensors. From there base maps were created of yaw rate, lateral acceleration and wheel slip based upon logged data. The tables will be fine tuned based on driver input and the logged data. Data is the most critical aspect to our development program, our goal is to quantify every change and to eliminate any human element of our progression. This is very important when we have multiple drivers and opinions.
Haltech ECU
Several calculated channels internal to the Haltech ECU may be logged to a USB key for advanced troubleshooting. Some data is not available to the AIM logger and must be logged directly by the Haltech, for this reason Haltech’s own logging software has proven extremely valuable. For high speed issues it is also preferential to use the ECU’s logger as it has virtually no delay. The biggest issue we have experienced with the Halech logger is the limited number of channels and the short periods of time it is able to log for.