Boost by Gear with OEM Logic

[rolfe.oliver]

It is torque based. When you push the throttle it has no direct effect on the throttle blade. It is a torque request. So if your tires will spin at 100 Lb/ft and you limit power to 90 the blade will open far enough to give that amount of power.

You will accelerate as fast as is possible with your current set up.

If you have ever watched Street Outlaws one guy said he has a $50,000 box that makes sure he can get down a dirt road. This is a lots cheaper form of that device.

Those Davis profilers are no joke.

Sponsored

 

[mejohn50]

This is really interesting stuff, nice work! My questions are truly for understanding, not a critique.

If I understand this strategy, you are essentially employing an electronic throttle stop: the pedal is pushed to the floor, but the angle of the blade in the throttle body is limited to various angles under specified conditions. Are your torque values referenced from previous dyno data, and then correlated to a specific CFM that results in that approximate torque, or are you using some sort of strain gauge to actually measure torque live?

Is the throttle response and rate of RPM increase affected by this strategy? I ask because in effect, this is partial throttle acceleration, and I want to wrap my head around it. I'm planning on a twin turbo system and was thinking more along the lines of boost by gear, but am uncertain how difficult/possible it would be to reliably control boost at very low levels (IE only a pound or two in 1st gear)

Thanks.

@K4fxd gave a really good explanation of what's going. I'll give a little more detail on what it's doing behind the scenes.

The shortest answer is it is capping the driver demand torque request value based on gear and RPM.

The longer answer...

Like he said, the pedal position creates a torque request, which then gets turned into a desired air mass value (load) through a series of tables and calculations. That desired air mass value gets turned into a throttle opening value. I am simplifying the process, but these are the basic parts that matter for this.

My car has never been on a dyno, I got my torque model values from existing calibrations (Whipple, Roush), personal testing and calculations, and from working with other enthusiast tuners.

As you can see in the below screenshots, the ETC torque request (light blue line) matches the value in the attached torque by gear and RPM table screenshot even though I am requesting way more torque according to the driver demand table. Then you can see the engine torque value (yellow line) attempt to follow that value. The values on the left correspond to the gear the car is in at that moment.

1st gear, 300 ft lbs:

2nd gear, 400 ft lbs:

3rd gear, 500 ft lbs:

Torque by gear table:

The rate of RPM increase does change because I am asking the car to accelerate slower by asking for a lower torque value. The whole premise, like k4fxd said, is to make the car have traction when it normally wouldn't. Throttle response is unchanged below the torque request value, but once my throttle input exceeds the torque limit value, the throttle pedal is dead and does nothing else for power. That's the point. I want to be able to go WOT in any gear at any speed on the street and not spin.

I do not know how well this would work on a turbocharged car. Turbochargers do some wonky stuff to the torque request and actual torque values by nature of how they work. I can assume the logic would try to work, but having never tried it I can't say if it would be as successful as this.

Hope that helps.

 

[gorditas]

Thanks.

@K4fxd gave a really good explanation of what's going. I'll give a little more detail on what it's doing behind the scenes.

The shortest answer is it is capping the driver demand torque request value based on gear and RPM.

The longer answer...

Like he said, the pedal position creates a torque request, which then gets turned into a desired air mass value (load) through a series of tables and calculations. That desired air mass value gets turned into a throttle opening value. I am simplifying the process, but these are the basic parts that matter for this.

My car has never been on a dyno, I got my torque model values from existing calibrations (Whipple, Roush), personal testing and calculations, and from working with other enthusiast tuners.

As you can see in the below screenshots, the ETC torque request (light blue line) matches the value in the attached torque by gear and RPM table screenshot even though I am requesting way more torque according to the driver demand table. Then you can see the engine torque value (yellow line) attempt to follow that value. The values on the left correspond to the gear the car is in at that moment.

1st gear, 300 ft lbs:

2nd gear, 400 ft lbs:

3rd gear, 500 ft lbs:

Torque by gear table:

The rate of RPM increase does change because I am asking the car to accelerate slower by asking for a lower torque value. The whole premise, like k4fxd said, is to make the car have traction when it normally wouldn't. Throttle response is unchanged below the torque request value, but once my throttle input exceeds the torque limit value, the throttle pedal is dead and does nothing else for power. That's the point. I want to be able to go WOT in any gear at any speed on the street and not spin.

I do not know how well this would work on a turbocharged car. Turbochargers do some wonky stuff to the torque request and actual torque values by nature of how they work. I can assume the logic would try to work, but having never tried it I can't say if it would be as successful as this.

Hope that helps.

All very interesting, thank you for taking the time to explain it further. I fully understand that the pedal position is only one input not only to the angle of the throttle blade, but as to which the rate it moves - via servo, commanded by the PCM. Your explanation of where you are getting the torque values/tables really covers most of what I didn't understand. When it comes to performance car tuning, most of my experience is with carburetors, cams, and timing. I have a lot of repair experience with EFI cars, but most of that was focused on things that referenced tables for fuel, not torque. When I was growing up, 225hp was a "fast" car. The last Mustang I owned was a 2001 with "Performance Improved!" heads that made 260 HP, if I recall.

Keep up the posts. The aircraft I fly is all fly by wire, not only for the flight controls, but also the thrust levers (throttles). The funny thing is, the closest correlation to this topic though is the old turboprop I used to fly. We had limits on torque, not for fear of the equivalent of breaking traction, but because you could actually break a gearbox or motor mounts. As we climbed into thinner air, full rated torque could only be made to a certain altitude where the engine became thermodynamically limited: you would exceed the maximum temperature of the exhaust section of the engine. But instead of computer controls, it was up to us, with two separate engines to manage.

 

[mejohn50]

All very interesting, thank you for taking the time to explain it further. I fully understand that the pedal position is only one input not only to the angle of the throttle blade, but as to which the rate it moves - via servo, commanded by the PCM. Your explanation of where you are getting the torque values/tables really covers most of what I didn't understand. When it comes to performance car tuning, most of my experience is with carburetors, cams, and timing. I have a lot of repair experience with EFI cars, but most of that was focused on things that referenced tables for fuel, not torque. When I was growing up, 225hp was a "fast" car. The last Mustang I owned was a 2001 with "Performance Improved!" heads that made 260 HP, if I recall.

Keep up the posts. The aircraft I fly is all fly by wire, not only for the flight controls, but also the thrust levers (throttles). The funny thing is, the closest correlation to this topic though is the old turboprop I used to fly. We had limits on torque, not for fear of the equivalent of breaking traction, but because you could actually break a gearbox or motor mounts. As we climbed into thinner air, full rated torque could only be made to a certain altitude where the engine became thermodynamically limited: you would exceed the maximum temperature of the exhaust section of the engine. But instead of computer controls, it was up to us, with two separate engines to manage.

Thanks. The game has changed dramatically with drive by wire throttle and adjustable cams.

The gen 3 coyote PCM and operating system logic is ridiculously capable from the factory. We just don’t get a manual like a stand alone buyer would, plus we have to deal with OEM programming for economy/efficiency and warranty considerations.

 

[WildHorse]

Ok, this is NOT a critique, but in it's common denominator form your limiting torque to prevent wheel spin, in the lower gears, and easing up in the higher gears. Whereas the factory tune has a set limit no matter what gear it's in, correct ? Would using say a 3.15 or a 2.76 axle ratio accomplish the same thing ?

 

[mejohn50]

Ok, this is NOT a critique, but in it's common denominator form your limiting torque to prevent wheel spin, in the lower gears, and easing up in the higher gears. Whereas the factory tune has a set limit no matter what gear it's in, correct ? Would using say a 3.15 or a 2.76 axle ratio accomplish the same thing ?

My stock 2019 Mustang GT tune does not have a torque limit that is realistic anywhere in the calibration I have seen. The lowest value I have seen in any forward gear is 479 ft lbs of torque in first gear, which is lower than the 420 ft lbs the engine is rated from the factory. Perhaps there is some torque multiplication going on or a series of conditions that would activate a specific limiter that I am not aware of, but I have not seen any data that verifies that.

If you can show me how to change differential gears with a laptop or on the fly with my cruise control buttons, I am all ears. I already have 3.15 rear gears. I also have 28" tall drag radials, and I can't hook those on the street in lower gears, but I can hook them at the drag strip.

With this PCM logic, I can tell it to make any amount of torque I want, up to the max possible engine torque, with either a quick PCM flash or by hitting a couple cruise control buttons in about two seconds. I can have multiple torque limit values stored in the various tune slots. I can have a street tires slot, a drag tires on the street slot, and a no limits slot for drag radials at the track. Additionally, as you can see by the screenshots, I can make the limit a curve based on RPM for each gear (up to sixth). There’s plenty of granularity to work with.

Changing differential gears wouldn't accomplish the same thing and isn't the answer.

 

[Rolls]

I bought the PCMTech single car edition, but I have NO idea how to use it. The car is currently running on a 93 tune with HPTuners but engineermike is going to tune it once I get the new clutch and engine broken in. Then we are going to do the flex fuel tune too.

He will get you sorted. He knows all the tricks and we keep in contact so if you get stuck we will get you sorted out.

 

[Rolls]

My next plan is to dial in the torque values more accurately for my setup, and then I am going to integrate this into one of my PCMTEC Custom OS slots so I can switch between a torque limited slot and a slot that has full power in all gears.

Wow, you are probably one of the first people to bother doing the torque tables properly like an ecoboost. If you get it spot on it gives you a lot more flexibility in the tune. I always thought the WOP == WOT strange, it was like they were too lazy to calibrate it properly and just went if you ask for 80% throttle give it full send.

 

[mejohn50]

Wow, you are probably one of the first people to bother doing the torque tables properly like an ecoboost. If you get it spot on it gives you a lot more flexibility in the tune. I always thought the WOP == WOT strange, it was like they were too lazy to calibrate it properly and just went if you ask for 80% throttle give it full send.

I do like the flexibility it gives setting it up like the Ecoboost, but my torque tables need some work. I had to do some brute force work to the driver demand table to get it to open the throttle body 100% while at WOT, but it’s currently working (for testing purposes).

And I agree. It seems lazy, but there are many things that seem like they were calibrated kind of lazy from the factory. Some of the spark values in unused regions of some of the mapped points don’t make a ton of sense for example.

 

[SolarFlare]

I see you saying you can “finally go wot in any gear”……but are you really?

 

[mejohn50]

I see you saying you can “finally go wot in any gear”……but are you really?

My foot is to the floor as you can see in the log screenshots I’ve posted. Not sure what the confusion is.

Is the throttle *blade* wide open? No. Of course not, at least not while the PCM is modulating torque for me. That’s the point. It’s seamless for me because the PCM is better at modulating torque than I am.

 

[SolarFlare]

Ok just making sure I didn’t miss somekthing that’s all….like a reverse pedal commander.

 

[illtal]

Ok, this is NOT a critique, but in it's common denominator form your limiting torque to prevent wheel spin, in the lower gears, and easing up in the higher gears. Whereas the factory tune has a set limit no matter what gear it's in, correct ? Would using say a 3.15 or a 2.76 axle ratio accomplish the same thing ?

No, because if you make enough engine brake torque to overcome traction in a gear, lowering the final drive will only make you accelerate slower in the upper gears where you will have less mechanical assistance through the final drive.

In essence it's a more sophisticated TC, open loop, with no feedback from the wheel speed sensors. If you know what limit is needed to not break traction you can hard code your WOT (in this case airflow/airload) to not break that number.

This method is the way that aftermarket ECUs deal with wheelspin. WOT simply will not allow more power than the tires can handle, so there's almost always full power. There are instances when you will get wheelspin on a road course without a yaw/inertial sensors, because there is less available traction when turning than going straight.

 

[mejohn50]

I’ve made a somewhat crude video showing this in action.

The graphs from the data log are at the bottom, the boost gauge is on the right, and the tach and gear indicator on the left.

You can see my foot is to the floor in the log the whole time (light blue line). You can see the torque request (orange line) change with each gear shift, and you can see the engine torque (white line) follow the torque request.

The boost gauge also shows an increase in boost with each gear change.

This is loads of fun. My car is actually usable on the street at low speeds now. It still needs a ton of work, I’ve spent no time (yet) trying to dial it in, but the basic concept is there and it works using 100% OEM functionality in the PCM.

 

[Midwestracer]

What all do I need to employ this and how will it work with my current tune will it overlay with it or will the engine need to be tuned for my setup first then play with this and my car to find the sweet spot

Sponsored