skale7
07-31-2007, 09:01 PM
Automatic Transmissions and Efficiency
Alright, first post. Everything here comes from a day of reading and about an hour on the road today, in a 1999 Toyota Sienna with a 3.0L V6, if you want to know. I didn't do any of the fun stuff like coasting on neutral or PG or whatever other risky stuff you guys do. I am a relatively new driver, and really don't feel I have close to the experience needed for the fancier tricks.
First of all, I guess it would help if I explain the torque converter, albeit in a very stylized (and, therefore, inaccurate) way. Think of a rotating fan in front of a really small windmill, about the same size. When power rotates the fan, it blows air to the windmill. The air then turns the windmill. Essentially, a torque converter works the same way. The flywheel rotates with the engine, pushing fluid across the blades connected to the axle, and turning that. Not the most direct method, but it works with only a 4%-8% drop in efficiency, and it allows the engine to idle without stalling. Because of this "fluid coupling" the engine is never directly connected to the transmission (except at higher speeds, but then you're not using the TC). Engineers design the TC to deliver no power at idle (or very little, i.e. if you put the car in park and let go of the brake on a flat road, the car will move forward very slowly, that's the TC at work) but to deliver as much power as possible once you press the accelerator. In reality, it is a sort of curve. At idle, the T is very inefficient at delivering power to the wheels, and it gradually increases in efficiency as the power increases. That's intentional, they are designed that way. Giving the wheels too much power at idle is bad for the brakes, if they slip. The bup-bup-bup of the car sometimes when you are waiting at a traffic light is the brakes slipping due to the engine delivering too much power to the wheels. When you let the brakes and accelerate, the TC applies more torque to the wheels, so that it can move. Unlike manual cars, the engine RPM is is not indicative of the speed, because there is no hard connection between the engine and wheels.
Regulation of shifts is much more complicated than this, but here's what happens. There are three important parts to know. The pump (connected to the TC) spins with the engine, and the resulting fluid pressure indicates the speed of the engine. The governor does the same thing, connected to the wheels. The throttle valve indicates how much you push the throttle. All three push a piston, called the shift valve, in different directions. The position of the piston determines what gear the tranny should be in.
When feathering the pedal, the car shifts very late, and therefore is not the most efficient. The AT is not going to shift unless it has enough power to do so. At the same time, because of the TC, not all of the power is going to the wheels anyway, making it even more inefficient. If you slam the pedal, the auto engineers assumed you wanted to accelerate hard, and again the tranny delays the shift to give you more power and higher RPMs, and you can take a guess as to how inefficient that is. If you do a little experimenting, you can find the "sweet spot" acceleration which is the most efficient. I would call it brisk, but I don't want to give a quanitiative description, because it is likely different for every car.
At freeway speeds, the TC power drain becomes negligible (or it locks, whatever), so here RPMs do make a difference. The van runs 55-60 at 1900 RPM, which seems to me to be great. I really have no idea what mileage I got, but I was pushing the pedal only a hair and still keeping a good speed. I read that once at speed you only need 10-15 hp to keep moving, and it sure felt like it. At 1900, the van also had enough power to go up and down the hills without any trouble.
I really noticed the effect of big trucks on mileage. There was a tanker in front of me, going at about the same speed, and I could take my foot completely off the pedal and still continue at speed. I was easily three seconds behind, which is safe. I was there for about six minutes or so, when the tanker took an exit. A chevy 4x4 on the side, however, really screws up my speed because of its wake, which makes the van sway, and I lose speed somehow.
That's all I found out yesterday. If anyone wants to disagree or give me tips or something I would really like it.
Sameer
Alright, first post. Everything here comes from a day of reading and about an hour on the road today, in a 1999 Toyota Sienna with a 3.0L V6, if you want to know. I didn't do any of the fun stuff like coasting on neutral or PG or whatever other risky stuff you guys do. I am a relatively new driver, and really don't feel I have close to the experience needed for the fancier tricks.
First of all, I guess it would help if I explain the torque converter, albeit in a very stylized (and, therefore, inaccurate) way. Think of a rotating fan in front of a really small windmill, about the same size. When power rotates the fan, it blows air to the windmill. The air then turns the windmill. Essentially, a torque converter works the same way. The flywheel rotates with the engine, pushing fluid across the blades connected to the axle, and turning that. Not the most direct method, but it works with only a 4%-8% drop in efficiency, and it allows the engine to idle without stalling. Because of this "fluid coupling" the engine is never directly connected to the transmission (except at higher speeds, but then you're not using the TC). Engineers design the TC to deliver no power at idle (or very little, i.e. if you put the car in park and let go of the brake on a flat road, the car will move forward very slowly, that's the TC at work) but to deliver as much power as possible once you press the accelerator. In reality, it is a sort of curve. At idle, the T is very inefficient at delivering power to the wheels, and it gradually increases in efficiency as the power increases. That's intentional, they are designed that way. Giving the wheels too much power at idle is bad for the brakes, if they slip. The bup-bup-bup of the car sometimes when you are waiting at a traffic light is the brakes slipping due to the engine delivering too much power to the wheels. When you let the brakes and accelerate, the TC applies more torque to the wheels, so that it can move. Unlike manual cars, the engine RPM is is not indicative of the speed, because there is no hard connection between the engine and wheels.
Regulation of shifts is much more complicated than this, but here's what happens. There are three important parts to know. The pump (connected to the TC) spins with the engine, and the resulting fluid pressure indicates the speed of the engine. The governor does the same thing, connected to the wheels. The throttle valve indicates how much you push the throttle. All three push a piston, called the shift valve, in different directions. The position of the piston determines what gear the tranny should be in.
When feathering the pedal, the car shifts very late, and therefore is not the most efficient. The AT is not going to shift unless it has enough power to do so. At the same time, because of the TC, not all of the power is going to the wheels anyway, making it even more inefficient. If you slam the pedal, the auto engineers assumed you wanted to accelerate hard, and again the tranny delays the shift to give you more power and higher RPMs, and you can take a guess as to how inefficient that is. If you do a little experimenting, you can find the "sweet spot" acceleration which is the most efficient. I would call it brisk, but I don't want to give a quanitiative description, because it is likely different for every car.
At freeway speeds, the TC power drain becomes negligible (or it locks, whatever), so here RPMs do make a difference. The van runs 55-60 at 1900 RPM, which seems to me to be great. I really have no idea what mileage I got, but I was pushing the pedal only a hair and still keeping a good speed. I read that once at speed you only need 10-15 hp to keep moving, and it sure felt like it. At 1900, the van also had enough power to go up and down the hills without any trouble.
I really noticed the effect of big trucks on mileage. There was a tanker in front of me, going at about the same speed, and I could take my foot completely off the pedal and still continue at speed. I was easily three seconds behind, which is safe. I was there for about six minutes or so, when the tanker took an exit. A chevy 4x4 on the side, however, really screws up my speed because of its wake, which makes the van sway, and I lose speed somehow.
That's all I found out yesterday. If anyone wants to disagree or give me tips or something I would really like it.
Sameer