LNF GDI Torque and Load Information

[405HP_Z06]

I have been doing some research on Bosch GDI in an effort to better understand what, where, and why certain things happen when tuning the LNF PCM. This is what I have assembled from Bosch literature and v2.22 of HPTuners. I have been struggling with the different types of loads and torques used in the PCM and I think I have it straight now.

This first diagram illustrates how engine torque demand from the driver is represented as internal engine torque.



Description from Bosch literature:

Torque Demand (TD) - Driver (TDD)
Vehicle Operator Command Interpretation
The position of the accelerator pedal is interpreted as torque demand from the driver. It is measured by using two independent angle sensors and is converted into a standardized accelerator pedal angle (0 – 100%).
The characteristic of the throttle device is stored in a map. This map, depending on the standardized pedal angle and engine speed, contains a relative torque that is used to determine the desired internal engine torque level. This desired torque level is obtained by an interpolation between the minimum and maximum adjustable torque. The minimum internal engine torque represents the pre-control level for idle control and is derived from and engine drag torque and the torque demand of the accessories, dependent of engine speed. The maximum engine torque is calculated, in consideration of the engine characteristic, from the maximum possible fresh air charge at the respective operating point. The engine speed and the density of the aspirated air first determine the maximum possible fresh air charge.

To increase driving comfort at positive and negative load changes, overrun cut-off, and fuel feed restart the calculated desired level then runs through a change limitation.

This second diagram illustrates how internal engine torque is calculated based on engine efficiency.



Description from Bosch literature:

Torque Structure (TS) – Torque Modeling (TMO)
Modeling of Actual Torque via Engine Efficiency
The main function calculates the induced actual torque of the engine. To implement this, the following actual levels are used: charge level, engine speed, Lambda, ignition angle, and reduction stage. The function also calculates the induced torque adjusted without transmission or surge damping intervention in the event of a transmission intervention.
A theoretically optimum, induced torque is derived from the variables engine speed and charge, describing the actual operating point. The following efficiency chain is used to calculate an induced, real, actual torque:

1. The cut-off efficiency is determined proportionally to the number of fired cylinders. If all the cylinders are fired, the efficiency equals 100%; if none are fired, the efficiency is 0%.
2. The ignition angle efficiency is a function of the deviation of the actual ignition angle from the optimum ignition angle. This efficiency corresponds to the thermodynamic efficiency of the engine.
3. The Lambda efficiency is obtained from the efficiency characteristic as a function of the air/fuel ratio. This efficiency and the ignition angle efficiency reflect actual consumption.

Input, corrections, and insight welcome; I am trying to understand this like most others. The GDI engine is a real complex relationship of engine modes and torque structure that are not used in PFI PCM's.

[Chris@HPTuners]

Nice summary

this torque based engine control is common for many controllers. Most switched to that control strategy when electronic throttle became popular.

we won't be adding the pedal to torque mapping unless it becomes necessary, it is not a "limiter" as such. I've never seen a calibration where 100% pedal wasn't 100% torque.

most of the other torque calulations aren't used for anything useful (ie. limiting performance in an way). The other input tables such as cruise control etc. aren't really calibratable in any useful way.

i'm not a real big fan of the Bosch code, it seems very simple in concept (maybe too simple) but very complex in implementation, almost unecessarily so. Maybe the German design or maybe i'm just too used to the GM code

[405HP_Z06]

Nice summary

this torque based engine control is common for many controllers. Most switched to that control strategy when electronic throttle became popular.

we won't be adding the pedal to torque mapping unless it becomes necessary, it is not a "limiter" as such. I've never seen a calibration where 100% pedal wasn't 100% torque.

most of the other torque calulations aren't used for anything useful (ie. limiting performance in an way). The other input tables such as cruise control etc. aren't really calibratable in any useful way.

i'm not a real big fan of the Bosch code, it seems very simple in concept (maybe too simple) but very complex in implementation, almost unecessarily so. Maybe the German design or maybe i'm just too used to the GM code

Thanks Chris, it seems not many like Bosch code. What about the optimum timing tables? Is there a PID for ignition angle efficiency?

Also, the individual cylinder fuel cut-off piece is still cryptic. I understand the PCM will cut fuel to individual cylinders to control torque, but which table controls this? Is fuel cut-off limited to certain engine modes? Fuel cut bit, maybe?

[Chris@HPTuners]

Thanks Chris, it seems not many like Bosch code. What about the optimum timing tables? Is there a PID for ignition angle efficiency?

Also, the individual cylinder fuel cut-off piece is still cryptic. I understand the PCM will cut fuel to individual cylinders to control torque, but which table controls this? Is fuel cut-off limited to certain engine modes? Fuel cut bit, maybe?

yeah there is calibration for the ignition effciency or "optimum spark". All pcm's use this kind of table and you don't need to mod it. In fact we had this table in for LS1's and most people try to zero it becuase they confuse it for a limiter of some sort, which really screws things up.

The cylinder fuel cut is used in two ways, 1) for DFCO when the PCM is cutting fuel under decel 2) for torque management limiting, which is covered by the table there under ETC Limits. Bascially the PCM just mutliplies the torque by the ratio of cylinders firing/total. Drop 1 cyl lose 25% of your torque.

A lot of people don't like the Bosch code cos its very difficult to disassemble most of the time. It uses a compiler convention that most would associate with Apple Mac's rather than embedded controllers.

[405HP_Z06]

Okay, so the Tq Mgt Cylinder Disable table. What are the X and Y axis? If I input 0 in the whole table the PCM will not use fuel cut-off to control torque. What is the methodology to adjust this table? Also, I don't think this is an ETC limiter since the throttle blade is controlled independently of Pedal angle and fueling is based on load and engine mode.

[Chris@HPTuners]

it's complicated and i haven't been bothered working them out exactly. The axes values are various modes and severity types, you can see as the numbers get closer to 3 it is more likely you will have more cyls disabled (the higher the number the more cylinders are likely to be disabled).

setting it to zero prevents cylinders being dropped.

[405HP_Z06]

it's complicated and i haven't been bothered working them out exactly. The axes values are various modes and severity types, you can see as the numbers get closer to 3 it is more likely you will have more cyls disabled (the higher the number the more cylinders are likely to be disabled).

setting it to zero prevents cylinders being dropped.

Okay, so let me take a stab at the row axis:

0 - Stratified Mode
1 - Homogeneous Mode
2 - Homogeneous Stratified Mode
3 - Homogeneous Lean Mode

Column axis: I have no idea and didn't find anything specific in Bosch literature concerning this table.

Concerning the use of cylinder cut-off to control torque. The PCM uses ignition, crankshaft, air, and throttle to manage torque output. Crankshaft (cylinder cut-off) is the most effective system the PCM uses for fast torque reduction. What's the impact to the other control systems if the PCM can't use the most effective system under extreme condtions?

The PCM would have to use ignition and throttle blade control to manage torque because the air system is too slow. I'm not sure how well the PCM could manage large torque reductions with ignition and throttle without causing very adverse effects.

Thanks for all the input and support!

[Chris@HPTuners]

the impact is it will not cut fuel for torque reduction

on other vehicles it commonly uses this as a last resort, spark is generally used first (such as trans shifting, idle and other things that require very fast torque control), then throttle is used such as RPM limit, speed limit etc. fuel cut is rarely used becuase it is very erratic and introduces large transients into the drivetrain.

I suspect the column axis is some kind of "severity mode".

[405HP_Z06]

the impact is it will not cut fuel for torque reduction

on other vehicles it commonly uses this as a last resort, spark is generally used first (such as trans shifting, idle and other things that require very fast torque control), then throttle is used such as RPM limit, speed limit etc. fuel cut is rarely used becuase it is very erratic and introduces large transients into the drivetrain.

I suspect the column axis is some kind of "severity mode".

I set myself up for that answer!

Cylinder cut-off is used as the primary method of torque reduction due to the quick impact to torque reduction commands. Different modes are used to make transients transparent to the driver during these events. I'll have to do some testing on this to see how this works out.

Is the data point in this table 'time' in ms or 'degrees' of crankshaft rotation?

[Chris@HPTuners]

the data value is actually a binary mask denoting which cylinders will be cut.

eg. 15 = 1111b = all cylinders

also the row axis is the current cylinder number eg. so that the next cyl cut is in sync with the current firing cylinder. ie. its a timing thing.

the column axis is the desired number of cylinders to cut for torque reduction. Notice the values are binary combinations of 1,2,3,4 cylinders cut.

Chris...

[405HP_Z06]

the data value is actually a binary mask denoting which cylinders will be cut.

eg. 15 = 1111b = all cylinders

also the row axis is the current cylinder number eg. so that the next cyl cut is in sync with the current firing cylinder. ie. its a timing thing.

the column axis is the desired number of cylinders to cut for torque reduction. Notice the values are binary combinations of 1,2,3,4 cylinders cut.

Chris...

Thanks Chris, so this table isn't mode and severity based but rather a look up of which cylinders to cut IF cylinder cut-off is invoked.

To adjust this table, it should be done in a column fashion? I.e. If I wanted to limit cylinder cut-off to 2 cylinders, I would zero out columns labeled 2 and 3?

Thanks again for the information!

[Chris@HPTuners]

if you wanted to limit it to 2 cyls you would copy the 2nd column to the 3rd and 4th col. So that if it asks for 2,3,4 cyls cut it only gets 2.

[405HP_Z06]

if you wanted to limit it to 2 cyls you would copy the 2nd column to the 3rd and 4th col. So that if it asks for 2,3,4 cyls cut it only gets 2.

Thanks, so to disable this altogether zero out the whole table. To reduce the number of cylinders used when fuel cut-off is invoked to manage torque, choose the number of cylinders the system structure can use and copy that column to the remaining columns from left to right?

[Chris@HPTuners]

yes

[mexico]

Is there a new beta version for the Bosch Stuff? The latest one I have doesn't have the parameters that you are listing.

[Bill@HPTuners]

Redownload & install the newest PUBLIC version 2.22 which can be found on your customer login page. It has the latest parameters available for these applications.

[405HP_Z06]

Is there a new beta version for the Bosch Stuff? The latest one I have doesn't have the parameters that you are listing.

All of the parameters we are discussing are currently available in the latest build of v2.22. Download the latest build from your customer portal and install to ensure you have the latest build.

**Edit: Sorry Bill, we were responding at the same time.**

[405HP_Z06]

Chris,
A question concerning this table:



Looking further into cylinder cut-off, it appears there are three types of cylinder cut-off. See the following:



From this I read that the Tq Mgt Cylinder Disable table is for Fixed Suppression cylinder interventions where cylinder timing is required.

The second type, Suppressions for Rapid Torque Reduction, looks like it uses the Max Torque table as the the reference for percentage of max torque allowed vs. RPM. Is the row axis on this table cylinders like the Tq Mgt Cylinder Disable table? e.g. 1,2,3,4 from top to bottom?

The third type cuts all cylinders and uses engine speed limiters, etc. as the reference.

Am I close?

[405HP_Z06]

TTT for Chris.

[SSpdDmon]

Regardless of which type of cut it makes and based on the descriptors, I don't know if I would want to change this table....even more so if it was related to the 1st or 3rd.

When I first saw it the day I got HPT, I thought it was related to the 3rd type as I had my car go into limp home mode a month or so ago. Based on the way it performed, it felt a lot like the way this table looks.

It might be that the left side of the table is as follows:

Normal operation
Fixed Suppressions
Suppressions for Rapid...
Disabling of Complete...

...instead of cylinder 1, 2, 3, 4. Why make one cylinder do all of the work instead of just doing some sort of torque reduction across the board?