Load and Boost Control

The MG1 ECU uses a model based on a combination of airflow and torque. The model is set up so that a specific airflow is requested in order to hit a requested load %. The requested load is a percent of the total load, not engine load, so the percentage scaling only goes from 0-100%.

Torque/Airflow Model Differences

There are two different generations of cars within the 2019-2022 range and both operate using a different strategy.

Torque Strategy 1 (~2019-2021 MG1 ECU) - Target Cylinder Filling

While the model year designation is not concrete, it’s a fairly easy way to get a good idea. The best way to determine if it’s using this strategy is the existance of a Target Cylinder Filling table.

Target Cylinder Filling - Load Request, RPM, with an output of Cylinder Filling as z-data
Optimum Engine Torque (moment) - Inverse with airflow as x-axis with torque output as the z-data
Optimum Torque for Function Monitoring - The parallel system for monitoring the torque model and cylinder filling (aka airflow/relative load %) and RPM with load % (aka torque) as the z-data

All of these tables control the torque together, and must be working in harmony or you’ll hit limp modes and run into errors.

Torque Strategy 2 (~2021-2023 MG1 ECU) - Optimum Engine Torque Approach

Again, model year is not the hard and fast for identifying this strategy. If it doesn’t have a Target Cylinder Filling table, it should just be using the Optimum Engine Torque table instead.

Optimum Engine Torque - Airflow and RPM as axes with torque output as the z-data
Optimum Torque for Function Monitoring - The parallel system for monitoring the torque model and cylinder filling (aka airflow/relative load %) and RPM with load % (aka torque) as the z-data

Regardless of which version you have, as you increase or change airflow the the entire Target Cylinder Filling / Optimum Torque maps must be rebuilt with the new airflow data assigned across a range of load % (torque) targets and their related airflow at target and airflow at 100% torque. By doing this complex task, you are able to target higher airflow while the vehicle creates the same calculated torque. When you get this correct, boost is well controlled, the throttle stays open, and there is little to no ignition-based torque management.

After you rebuild the Target cylinder filling and optimum engine torque tables (or in the case of torque model approach 2, the Optimum engine torque table alone), you must rebuild the Optimum torque for function monitoring table so that the monitoring system and the torque model report comparable airflow and torque relationships to the ECU. Failure to do so will result in a hard limp mode.

Maximum Relative Load tables set the highest percentage of the available load in that area the computer is allowed to request.

Meanwhile, Target Cylinder Filling / Optimum Engine Torque determines the target airflow for the given load request, essentially behaving as an airflow model for the car, so this table requires some careful tuning.

Load Control, however, is not that simple

On this vehicle the torque and airflow model are very interconnected and limited by other ECU functions.

The Optimum Engine Torque table provides the percent of maximum allowed torque that is provided for the relative load, and torque request from the driver. Thus everything becomes interrelated.
There is an intrinsic limit to target load (torque) allowed by the ECU
- The ECU will allow higher airflow but will not allow an unlimited torque request.
In order to avoid throwing errors, the ECU must understand that the target airflow is always fairly close to the calculated limits.
The Torque to Airflow model is always checked for accuracy by a parallel airflow to torque model. If these are not mathematically identical, the ECU will present a torque error and the car will go into limp mode.

As a result, in order to raise what you’re asking for, you’ll need to increase torque targets, driver requests/limits, as well as the drivetrain and wheel torque limits in order to roughly match the increase in torque gained by the increased filling.

Maximum Filling Allowed from Intake Temperature provides a hard limit to airflow and torque so you’ll need to raise these to be slightly above what you’re attempting to hit.

Boost Target

Managed by the ecu to achieve the airflow request (Cylinder filling). The ECU will simply run the boost necessary to achieve an airflow target (after any temperature or fill limits).

Maximum Boost and Maximum Intake Manifold Pressure provide a simple ceiling for manifold pressure. It is a very good idea to start these low and gradually raise them as your calibration progresses as they can provide a safety limiter to prevent damaging boost levels. Your final values should be slightly above what you’re attempting to target so that the driver won’t experience them unless something is wrong.

Load Limits

Each vehicle has an intrinsic torque limit. This means that while you could place all the load request values at 100%, the car will never actually achieve that target. An easy way to see which torque limit you’re running into will involve taking the time to increase all of the torque limiters and targets one at a time so that the limiter you’re hitting has the least interference from other tables. This should cause the car to hit an obvious number that can easily be found in the tables.

For all of these cars this highest available load % is sufficient to increase power to desired levels.

Once the limiters are raised, any higher load request % will increase desired airflow. The Target Cylinder Filling table sets the relationship between load % and Airflow. After rescaling the range of the table it will be able to facilitate higher airflow and more power.

Higher Load Request % (as long as other limiters are also not hit) will increase desired airflow. Target Cylinder Filling describes the relationship between load % and airflow. Stock, the table doesn’t require an appropriate range in order to allow much of an increase in power. In order to make more airflow and power, you’re going to need to scale the table differently to cover the affected areas.

Once that has been rescaled you’ll need to adjust the Optimum Engine Torque map. Similar changes need to be made to that table so that the engine torque model matches the airflow model. If these values are not similar enough to one another in function, the ECU will not properly target load and will cause errors or limp modes to occur.

“Approach 2” (Late Model Non-Cylinder Filling Cars)

With these vehicles it is a little bit more tricky in order to create an airflow model given the absence of the Target Cylinder Filling table. In order to increase power you will first need to increase the load target so that the car can achieve it’s intrinsic torque (relative load/request %/load %). Once you have found this maximum load request there is no more effective way to make more power until you increase airflow.

In the picture above, we have a load request table with the approximate values that equate to maximum load request for a 2022 S5.

The stock Optimum Engine Torque (Moment) doesn’t reference a high enough airflow in order to support the new load request. To rectify this we’ll need to extend the torque map with a higher value. In regions with higher airflow values, the optimum torque map must make sure to maintain the relationship with airflow and torque requests in order to be happy.

The first part of this is going to be to increase the relative load % and the request.

From the tables above you can see that the request has been increased to match the newer load % appropriately.

The second part of this task is to create a mathematically similar Optimum Engine Torque for Function Monitoring map. Since the changes are only in the upper load regions, it’s fairly easy. However these changes MUST reflect the same torque and airflow relationships as the Optimum Engine Torque map otherwise you’ll see errors or limp modes occur

The third and last part is increasing the power. Now that you have a balanced airflow and torque model, you will likely be hitting the intrinsic torque limitations. If your torque model already extends to 100% of the load request, you’ll need to increase the airflow request for a given torque at a given RPM. From there you’ll also need to make sure that the airflow at the request is raised so that the new higher airflow maintains a similar reported torque to the ECU.

As a simple example

Torque request of 50% Load - this is the intrinsic max at a specific RPM. 738*.5 = 369 ft lbs.
The airflow at this load is 120 (relative load %, target cylinder filling).
You desire to raise the power, so you want to increase airflow by 10% at this point.
The new airflow is 132% relative load.
The new related 100%, load % airflow is now 264 compared to 240.
120 / 240 = 0.5, = 369
132 / 264 = 0.5 = 369

So, there is 10% higher airflow and the same reported torque. The car makes more power and with the same relative load limitations.

Ignition Timing

While fundamentally simple, there are a lot of moving parts to the ignition timing maps for these cars. The Optimum Ignition Timing maps are very carefully modeled stock, and represent the ignition timing under ideal conditions (no knock or heat soak etc.) where the car will run the highest output torque. As such you’ll want to leave those completely alone unless you’ve drastically changed physical characteristics of the engine internally. Keep in mind as well, that any changes you make to that table will break the calculations for the torque model, so you would need to change a large number of tables. In almost every instance, it’s easier to leave these tables alone, and simply use the other tables to calibrate.

The main ignition timing maps you will use are

Ignition Timing Normal
Ignition Timing Normal 2
Timing Map
Timing Map Variant 2

For 93 octane fuels, there is very little reason to change timing if you aren’t drastically increasing load and boost. Typically we’ll only remove a degree of timing or so under high load/RPM conditions.

This ECU uses an Optimum Ignition Timing map, which is under ideal conditions with no variables, that timing equals the amount of torque in the model the car will produce. Because this is under ideal conditions running 100% of this timing is NOT recommended under any circumstances. This also means that it looks at the timing you request under normal conditions and looks at what percentage it is of the ideal timing to understand how much torque the engine will produce. This means that when you reduce timing the car will be aware that the reduction in timing will reduce torque output (and vice versa). In conditions when the torque target is left the same and you reduce ignition timing, the car will change the manifold pressure and airflow request in order to hit your target torque.

This means that if you change timing without changing your torque request, the car will still run the same amount of power, and just make up for it in other areas.

On some vehicles there are per cylinder ignition increase or decrease. This allows you to compensate for particularly knock-prone cylinders with the least amount of impact on torque output. Under light use, these values do not impact the torque calculations as much (or related airflow/boost etc.)

Controlling Torque with Ignition TIming

In order to control the amount of torque produced, the ECU will manage airflow through boost, throttle closures, as well as decreasing ignition timing. A properly tuned vehicle will use minimal throttle closures and will run the average timing from the 4 main timing tables. If the torque is reduced within your power curves despite reasonable manifold pressure and throttle openings, it is very likely that a timing reduction is the root of the issue. Ignition timing changes are highly effective when used to control torque. If this is happening you may need to reexamine your airflow and torque models, as they may not be in line with some of the intrinsic limitations of the torque system or the vehicle.

Knock Control

Like most other VAG cars, these MG1 ECU vehicles use per cylinder knock corrections. These systems, when working properly, are highly active and adaptive. Knock thresholds, gains, increment, increment and decrement rates are all adjustable. With few exceptions we prefer to keep these systems fairly conservative and mostly stock in stock applications.

While some active knock response is expected and normal, constant corrections of around 4 degrees retard or higher is outside of the norm and usually means some calibration changes are needed. Reducing cylinder pressure, charge temperature, desired timing or enriching the fuel side is recommended to get these back down to a lower level of correction.

These motors and their proper operation is fairly dependent on running the correct timing. Even a small change like 1-2 degrees below where it wants to be can be worth 25-30 ft-lbs of torque. And because most of these engines are limited by relatively small turbochargers, the power created by timing is not going to be replaced by airflow.

Fuel Control

Unlike older vehicles, fuel is largely less controllable than most people have come to expect. Many of the early MG1 6 & 8 cylinder vehicles use a fueling approach that counts on the vehicle hitting component protection at wide open throttle. Typically this system swings very rich to cool things down in the engine and then back to lean for clean burn and then possibly back to component protection again later in the pull. This tends to keep the Exhaust Gas Temperatures (EGT) lower overall due to the rich mixture.

This cycle of lean to rich and back is unfortunately without much of a mechanism to edit as these cars do not have a traditional high load requested lambda target. It is possible to minimize some of the extreme rich conditions within the component protection by going no more rich than you normally would at WOT, however you’ll want to make sure the car still runs within an appropriate safety margin.

In some of the newer MG1 vehicles, (2021+) fueling is not controllable at all. These engines always target 1 lambda and will only enrich for short periods of time. There is very little you can do to change this and, shockingly, it hasn’t required much adjustment beyond the factory tables in a stage1 configuration. While the fueling seems unreasonable at higher boost pressures, Audi/Porsche have made it work for these cars.

Contact Us:

COBB Customer Support

Web Support and Tech Articles: COBB Tuning Customer Support Center

Email: support@cobbtuning.com

Phone support available 9am to 6pm Monday-Thursday. 9am to 4pm Friday (CST)

866.922.3059