Tested: The Most Efficient Way To Accelerate With An Automatic (Toyota Matrix)

[mechman600]

Originally Posted by euromodder

And if you briefly lift the throttle , get it to shift up, then get on the gas again (staying short of kick down) ?

With my wife's Toyota Matrix, I accelerate briskly until 60 km/h, where I release the throttle just enough for the trans to shift into high gear and TC lockup and then give it enough throttle to reach 75% "LOD" (SG2) - the max before it unlocks the TC.

I find the "LOD" display the most helpful. "TPS" is not really useful to me at all because it is not RPM dependent.

[E.Roy]

Originally Posted by euromodder

And if you briefly lift the throttle , get it to shift up, then get on the gas again (staying short of kick down) ?

I use this technique, but its not always as effective as one would like, plus I find it annoying. I would rather have the car do it automatically , even if it means I have to tune the car to do it.

My wife's past 2007 Yaris auto 4spd was quick to upshift, and would hold the gear under much larger throttle openings than this 2000 Corolla, even though the Corolla makes more power/torque

[mechman600]

Originally Posted by E.Roy

^^^ The small vacuum leak would raise your idle though, which should consume more fuel.

Unless you hooked your controlled vacuum leak up through a 12V vacuum switch that you could turn on and off with a switch.
Just saying....

[Regenerit]

Something to take note of if you have a dyno sheet available for your vehicle that tells you the air to fuel ratio at various rpms and horsepower. You may be able to find a sweet spot in your acceleration and also know what rpms to avoid.

For example in this dyno sheet for my 2011 Genesis coupe, past 2500rpms my air fuel ratio begins steadily dropping before plummeting down below 10:1 at 4600 rpms. I know for sure I should avoid going past 4600 rpms when accelerating, to avoid dumping gas. And for the best A:F ratio i should go no higher than 2500rpms in general when accelerating. Oddly enough, My engine spins at 2500 rpms at 55mph. Not bad.

[UltArc]

Originally Posted by Regenerit

Something to take note of if you have a dyno sheet available for your vehicle that tells you the air to fuel ratio at various rpms and horsepower. You may be able to find a sweet spot in your acceleration and also know what rpms to avoid.

For example in this dyno sheet for my 2011 Genesis coupe, past 2500rpms my air fuel ratio begins steadily dropping before plummeting down below 10:1 at 4600 rpms. I know for sure I should avoid going past 4600 rpms when accelerating, to avoid dumping gas. And for the best A:F ratio i should go no higher than 2500rpms in general when accelerating. Oddly enough, My engine spins at 2500 rpms at 55mph. Not bad.

Attachment 13893

You just gave me an excuse to get the Mustang and Insight on a dyno!

Err, reason...

[2003protege]

Wow, this is something I've slowly been getting a feel for even in my standard transmission vehicle, that tank fuel efficiency has way more to do with cruising speed and driving without brakes than it does acceleration method - even though I think everyone has some sort of indoctrinated sense of inefficiency under heavy acceleration.

[oldbeaver]

Originally Posted by mechman600

The most efficient speed to cruise at with an automatic transmission (at least the kind with a torque converter) is the minimum speed that the torque converter will lock up in high gear. For my wife's Toyota 2007 Matrix (1.8L, four speed auto), this speed is 60 km/h (37 mph). The most efficient way to accelerate above this speed is at max load without the torque converter unlocking. For the Matrix, this is 75% "LOD" on the ScanGauge II.

The question that has plagued me for a while is, what is the most efficient way to accelerate up to 60 km/h? Slowly? Quickly? Today I finally got around to executing a test using my SGII.

I first determined a fixed distance to test with by accelerating very slowly from a stop to 60 km/h. As soon as 60 km/h was reached, this was the fixed ending point, using signs and trees beside the road as markers. This distance ended up being 740m (2400ft), going by Google Earth.

For every [quicker] run after that, 60 km/h was reached before the ending point, so the remaining distance was travelled with the torque converter locked at a steady 60 km/h, with the final L/100km figure taken at the fixed ending point.

The two constants were the test distance and 60 km/h vehicle speed at the ending point.
The variable was the acceleration rate up to 60 km/h.

What I found was quite surprising. Different acceleration rates have little effect on efficiency with this car, from ulta-slow to very fast. And the most efficient acceleration rate is much quicker than I ever thought it would be.

Test results chart:


Test results graph:

This is a good example of a well done investigation. You experimented what happen in real world and didn´t follow other people's opinions.

For me, opinions will never substitute experimentation.

Congratulations.

Oldbeaver

[BambuFlute]

Very cool info thank you. I had the opportunity to drive my 2001 Yaris 1500 automatic (called Echo in Australia) on the same 500km trip twice this year. The first time for a week with my 80 year old mum & my wife, the second time with just my wife and I. Aside from ~60kg less (no mum, but more of other things) in the car on the second trip everything else was the same (tyre pressure 35psi, fuel grade 98, molybond in the oil). Of course I drove a lot more conservatively with a fragile old lady in the car on the first trip, and I used the cruise control as much as possible. The result was L/100Km = 5.44 (odometer 157,971km at start).
The second trip was on a weekend, so I drove at the speed limit and accelerated hard in situations such as mountain climbs, & passing trucks, and I used the cruise control less often. Surprisingly, L/100Km = 5.64 on the second trip (odometer 161,328km at start). If you had told me this was going to happen I would not have believed you. No one was more surprised than me after I entered the numbers in my spreadsheet. On both occasions I bought fuel in the same garages, and I do my best to use the same pump when I fill up (pumps vary). Clearly, the variable valve inlet and engine management, even on my old car, works well at higher engine revs. Strike for modern engine management systems.
I've also noticed that my wife's 2005 Yaris auto (odometer 65,000km) is very efficient in peak hour traffic (7.12 L/100km over 10,000km). She knows nothing about car engines or transmission so this is a good test of how a car would normally behave with the aircon on and being driven hard in traffic.

[BLSTIC]

Originally Posted by mechman600

In theory all you would need to do is create a small (metered) vacuum leak. Your throttle would be more closed at the same load to maintain the same MAF/o2 sensor readings. Not sure what the trans goes by as far as downshift logic goes, but if it goes by throttle plate angle, that would do it. If it goes by algorithms based off multiple sensor readings, it would't work.

That wouldn't work, as someone stated previously. It would give a high idle and stuff... What you want is a throttle that flows more air at low angles than normal, but still closes. There are two options:

a: Bigger throttle body. Expensive and/or hard, idle mechanisms may be changed. More air per degree of opening, higher max flow.

b: modify the one you have. You can carve a channel around where the throttle sits at (say) 15%. If you carve it wide enough and smooth enough, the effective throttle opening will be larger from roughly 8-20%. With the right shape you could probably increase low angle flow by 30%. More air per degree of opening at light loads. More DIY, modifies exactly what you want. No shiny bits or potential horsepower gain though.

Alternatively if your car is older and has a single TPS (most anything with a cable) you can use a voltage interceptor kit (jaycar electronics do a pretty trick one) to make the computer think that the throttle is less open than it is, but still reads correct at idle and full throttle. That way you can haul along at 2/3rds throttle and the computer will keep you in closed loop and change up at half redline...

[E.Roy]

That is a cool method BLSTIC with grinding a channel, thanks for sharing that idea. Might be useful especially to someone whose TPS is non adjustable like me. Could you also place a resistor inline in the voltage signal wire to the ECU, would that reduce voltage?