Dyno tuning?

[nickg07]

it doesn't show what the after effects are going to be so when u do a engine stage u have no idea what your new power band is going to be

That's true but it's more a case of it being an issue because you can't remove it and put it back in the parts setup screen. Dynoing the car after doing the upgrade it won't have any difference. They just need to make it a swappable mod or something.

 

[Ills1999]

That's not how it works in real life either, or at least, not universally.

I'm not sure how much the average person here knows about how engines work and are tuned, so I'm going to go through the 101 level of engines and tuning.

A car engine depends on two things to make the best power, all other things being equal: having the right amount of fuel in the combustion chamber when it fires, and triggering the spark at the right time.

The right amount of fuel in the combustion chamber for best power is generally considered to be ~12.5 x the amount of air by weight. This means that petrol must be precisely metered for the amount of air flowing into the engine. The original scheme for accomplishing this metering is the carburetor, which meters the fuel through mechanical means. The modern scheme is electronic fuel injection, which uses some kind of sensor suite coupled to a computer (called by various TLAs, an ECU in GT parlance) to deduce the amount of air going into the engine and direct the injectors to fire to provide a controlled amount of fuel. (In between there was mechanical fuel injection, which I know little about and shall not discuss further.)

The right time to trigger the spark varies with both the speed and the load on the engine. In general, triggering too soon will fire while the piston is still going up (not just bad for power, but likely to break something), while triggering too late will fire after the piston has already started going down (bad for power). As the speed of the engine increases, the spark needs to be triggered earlier to allow time for the combustion to progress through the cylinder. An example of this would be a late-80's Ford Mustang, which has base timing of ~12* before top dead center (BTDC) which is advanced to ~35* BTDC over ~3500 RPM (exact numbers may be different, but all are in the right ballpark).

The original scheme for altering timing was purely mechanical in the distributor, with a system of weights to handle advancing the timing and points producing the spark. By the mid-'80's, there were semi-mechanical systems where the weights and points were replaced by an electronic system controlled by the ECU but there was still a rotor for delivering spark to the plug wires (which would allow base timing to be changed by clocking the distributor body).

With electronic fuel injection, there are two major sensor suites and schemes for deducing the amount of air going into the engine. One of these is known as "speed-density" and the other is known as "mass air flow". With the speed-density scheme, the physical and flow characteristics of the engine are pre-determined, with the density of the charge air measured by a pressure sensor and temperature sensor on the intake. With a mass air flow scheme, the actual mass of air is determined by measuring atmospheric pressure, air temperature, and the volume of air passing through a sensor.

So, how does all of this relate to tuning?

While timing is somewhat important, it's usually set pretty well from the factory. With a mechanical or semi-mechanical system, you can do some things with timing, with a purely electronic system (coil on plug, etc), the only way to fool with the timing is to mess with the ECU.

The big issue is the fuel metering. Remember how I said a speed density system has the flow characteristics of the engine pre-determined? That means that pretty much anything that fools around with the engine is going to screw up its behavior---and offhand, I think it's more likely to end up lean (with more air going through than is supposed to be) than rich, which is very bad for the health of an engine when you're driving it hard. In this case, your statement, "Something as simple as an intake and exhaust on an old Civic would ruin the engine's tune" is correct.

But you remember how I said that speed density isn't the only possible system for metering fuel in an engine? In a mass air scheme, the computer may have a fairly broad ability to deal with an engine that will move more air. It will depend on whether the air meter has the capacity to signal more air and whether the fuel injectors can move enough fuel (and even that can be partly addressed by changing the fuel pressure regulator to get higher pressure on the fuel rail). In a mass air car, you can do a significant amount of tuning and see the engine take it in stride. An example of this that I'm relatively familiar with is the Ford 2.3L turbo four-cylinder engine from the mid-late '80's. As it shipped stock in a Thunderbird Turbo Coupe, this engine was spec'ed at 190 bhp at the crank in a manual transmission car. The stock air meter, fuel injectors, and ECU can handle ~250 rwhp---that's with things like a cone air filter, replacement exhaust, port-matching in the intake, maybe bumping the amount of boost, but no tuning of the fuel or spark whatsoever, just the stock provided by the ECU.

Of course, that's also not getting into the issue that you don't want 12.5 AFR all the time, you will want higher for part-throttle and off-throttle for better fuel economy when you don't need max power, and you maybe even want higher for full throttle depending on whether you're normally-aspirated or boosted, and so on...

(And how do carbs fit into it? I'm not sure how well they typically deal with moving more air through the engine, but I do know that they need more fine-tuning than an ECU due to less capability to adjust to different atmospheric conditions, and they have a lot of latitude for aiming for a particular operating envelope by tweaking settings or internals.)



tl;dr: What kind of tuning is required for maximum performance depends on how extreme the modifications are you're making to the car, and how well-equipped the car was to handle modifications in stock form. Some can take a lot more power just with simple bolt-ons, some require modifying the ECU for even relatively minor changes, and carbs require different kinds of tuning altogether.

A lot of good info there AND you don't even get into cam profiles and cam timing which does indeed shift the powerband around, contrary to the pronouncement of the member to whom you were responding.

 

[R34EVO9]

That's true but it's more a case of it being an issue because you can't remove it and put it back in the parts setup screen. Dynoing the car after doing the upgrade it won't have any difference. They just need to make it a swappable mod or something.

that's true too I kinda forgot that lol. They need to fix that too

 

[HBR-Roadhog]

that's true too I kinda forgot that lol. They need to fix that too

It should be changed and broken down into more optional parts, cams, displacement, pistons, valves and such along with porting and balancing and stuff. There are however some changes one would do to an engine that can not be easily undone. For example once you bore the cylinders you can not unbore them and if the goal is realism should not be an option in the game, buy a new engine and start over would be the option there.

 

[MrWednesday]

Yes I know that as should have been obvious in my post. My point was that it should be the rear diff either in the form of an adjustable final drive or options as to what ratio you install when you buy a new diff, at least some of the more common ratios for high and low gearing. This combined with the close ratio 5 speed or 6 speed and in some cases even with the stock tranny would be good for performance and add a more realistic element to the tuning and upgrading of the cars.

I just find it hard to believe that so many games have a fully adjustable tranny [un realistic] but none seem to have optional rear ratios for the diff where as in the real world changing the rear ratio is a very common thing when building a hotrod

Agreed that some step-changes in rear-end ratio would be an appropriate allowable mod short of a fully customizable and infinitely variable ratio as with the present fully customizable transmission.

 

[niky]

That's not how it works in real life. Before cars were smart enough to tune themselves the engine would get all out of whack as you added different parts. Something as simple as an intake and exhaust on an old Civic would ruin the engine's tune, resulting in rich smoke in the exhaust. That's what we're talking about. As you add things you generally increase the engine's performance, but to get the maximum effect you need to retune the engine to maximize the available benefit. That's exactly how the system should work.

tl;dr: What kind of tuning is required for maximum performance depends on how extreme the modifications are you're making to the car, and how well-equipped the car was to handle modifications in stock form. Some can take a lot more power just with simple bolt-ons, some require modifying the ECU for even relatively minor changes, and carbs require different kinds of tuning altogether.

The real problem with dyno-tuning for mods is it's a big: "It depends"

-

With older cars with terrible stock intake and exhaust systems, relatively simple intake and exhaust modifications could have drastic effects on both power and fuel economy. I'm not calling the 2.3 old... well... it is... but there is that.

With older EFI cars and some modern cars with Mass Airflow sensors, changing an intake or exhaust, altering the speed of the airflow over the MAF sensors, or the quality of the air going past the O2 sensor, causes changes to the sensor inputs reaching the ECU. Ergo: your intake makes power not because it flows more air (or not solely because of this), but because it fools the engine into running leaner.

Sometimes, an exhaust will throw a CEL or MIL, as the O2 sensor starts reading too much O2 and the ECU enters a feedback loop it can't recover from, resulting in an error (cheap fix... lengthen the O2 sensor bung to keep it out of the exhaust airflow). Sometimes an intake will do worse (witness the Mazdaspeed CAI, or certain Subaru intakes with improperly sized MAF adapters, which make the car go boom-boom rather than zoom-zoom).

Then you get to cars equipped with Mass Air Pressure sensors and/or ECU algorithms that CAN adapt to all these things, and which typically neither gain nor lose power from most intake and exhaust modifications... or which gain a negligible amount that will go away after a thousand kilometers of running down the road.

-----

I don't personally design aftermarket intake and exhaust systems, but I have friends who do, and I've sat in on many tuning and development sessions. And I've had experience developing these parts for my car alongside them. With a modern, adaptive ECU, these sessions are a challenge at best, and are hair-pullingly frustrating at worst. Even WITH ECU tuning. I'm not surprised that some aftermarket intakes now come with their own chips, to recalibrate the MAF to work with the new tube.

-----

With Gran Turismo, the game works under the shortcut assumption that the intake and exhaust parts are properly designed for your application and will make a certain percentage of power. And it works under the assumption that your ECU tune is optimal. While it would be fun to have to retune for each and every mod, the insane variety of effects each mod would have on the tuning parameters of each car means that any game simulating this would have to adapt a "one-size fits all" tuning strategy that would be as unrealistic and unrewarding as the current power tuning system once you got deep into it.

(as an aside... just wait... GT6 will still give you the choice of fitting race headers to the SkyActiv engines, even though they come with long-tube scavenging exhaust manifolds as stock!)

I'd like ECU programming, simply to be able to do the other things I've mentioned. Anti-lag (via throttle programming), launch control, gear-specific power maps, multiple power maps for endurance racing economy and etcetera.

-----

Now cams. Cam tuning would be lovely. Complex. mind-boggling for those of us who still can't wrap their heads around proper overlap, but lovely. Better yet, aggressive cams (like mine) will make even the most anodyne engines gurgle at idle.

 

[Master__Shake_]

...Now cams. Cam tuning would be lovely. Complex. mind-boggling for those of us who still can't wrap their heads around proper overlap, but lovely. Better yet, aggressive cams (like mine) will make even the most anodyne engines gurgle at idle. ...

I'd be happy just letting us run custom grind cams for modifications, but dyno tuning would still be fantastic. Every car in my garage would sound like the '66 Cobra at idle.