Throttle % while accelerating?
 

Hi I have just recently started on my quest for fuel economy in my 1995 saturn SL1 5spd. My best tank mpg to date is 38.4. Anyway to my question I was wondering what would be best for fuel economy - more throttle in a higher gear but less rpm or less throttle in a lower gear but higher rpm? Is it best to accelerate quickly say 75% throttle and shift at a higher rpm up to 40-50mph and then cruise or use about 25% throttle and shift at a lower rpm up 40-50mph then cruise? i know its based on the car and the engine. I know that the SOHC in my saturn makes peak torque at 2450rpm. Should i take the rpms higher than that while accelerating or shift around 2500rpm? Im just in search of the best way to get up to a certain speed for cruising. Thanks in advance

Between 70% to 100% throttle and shift between 1,800 - 2,500 rpm.

There is an article on www.autospeed.com that tests this I just can't place my finger on it right now. I'll edit it in if I can find it.


Edit: Found it!

Brake Specific Fuel Consumption
http://www.autospeed.com/cms/A_110216/article.html

About 2/3 to 3/4 throttle (not quite enough to throw it into WOT enrichment), and shift early (low RPM) is most efficient.

this has actually been a burning question of mine for a while now. make me wish I had a 5 speed again

Now, what about an auto trans?

For an auto, accelerate with as much throttle as you can while not driving the RPMs too high. After a while, you find the sweet spot, which is different on every car.

Best rate of acceleration is lowest manifold vacuum in highest gear that will produce the desired rate of acceleration. If acceleration is too slow use next lower gear, while maintaining low manifold vacuum.

Any throttle position below the point of full load enrichment is fine and it can range quite a bit from 20%to 80%. This is because throttle position does not determine load. Load is a function of manifold vacuum. Once manifold vacuum reaches 100%, additional throttle will not increase power unless you go to full load enrichment, or in the case of my VX you loose lean burn, which is another stage of enrichment.

In an auto transmission you want the lowest manifold vacuum you can maintain without causing the transmission to hold the lower gears any longer than its earliest shift point.

Every engine is more efficient when it has the maximum amount of air inducted into the cylinders at the bottom of the intake stroke. RPM too low you loose the harmonics of induction resonance and that benefit. RPM too high and you are loosing energy to friction and reciprocation losses, which are due to the accelerating and decelerating masses in all reciprocating engines. Good intake harmonics can actually increase the volume of air slightly on the intake stroke with the inertia of the incoming air contributing a sight "supercharging" effect.

I think it is not well understood generally, that very low throttle positions can virtually eliminate manifold vacuum in almost any throttled engine as long as the load is sufficient, which is accomplished by choosing higher gears than most drivers use for acceleration.

Think of it this way. I call it pulse accounting. Every combustion pulse is much more energy efficient when it uses all of the engines available compression, which requires the manifold restriction to be the lowest possible amount. When manifold vacuum is 0 the maximum amount of air is in the cylinder to be compressed and that will produce the maximum amount of power per unit of fuel consumed.

Without wide open throttle enrichment.

If you produce that amount of manifold vacuum at the RPM where your engines BSFC is highest, you have duplicated that BSFC, and that is as good as it can get.

regards
Mech

I'll try that. I figured out how to short-shift this auto under 3K RPM, but I can't push the pedal far for a little after the shift.

Though if I barely touch the throttle (1-2%), it'll shift even sooner, but accelerate like a turtle. It will act like it's in high idle at that throttle level.

Some times you can "force" a shift on an auto. Accelerate heavily 70-80% throttle when the rpm (the plural is on the r of revolutions) is about right for the next gear let up on the accelerator then push back down till just before it shifts down again.

On my Maxima it turns out there is an adjustment to the throttle positioning sensor. Basically I make the compuet think the throttle is closed more than it is making the tranny want to stay in a higher gear. It may be the same on other vehicles. On my Max it took loosening 2 phillips screws and twisting the senor in the same direction as the throttle plate twists, which made the tranny shift early.

I found this out my accident when I installesd a new TPS. I adjusted it reverse of how I described, and it downshifted sooner and refused to upshift without higer RPMs.

Hmmmm, wonder what happens if i go the other way....

Just don't go too far with that adjustment. Setting it to shift sooner tends to lead to lower line pressure in the tranny (in some cars), which can cause soft shifts and/or slippage, leading to shorter tranny life.

I'll need to look at that now. I'm wondering about the kickdown rod, though, and if pulling that off can help it not force downshifts when I don't want it to.

I've been trying to figure out how to get some feedback to the driver on acceleration efficiency for a while. With a stickshift and a bsfc map it is pretty easy to make a guess. But if you have an auto it will have varying degrees of slippage, and if you have no bsfc map then what?

I think a reasonable approach is to use something mpguinoish that reports the change in speed over the fuel used as an instant value, and as something that displays an interval, say until the rpm drops indicating another shift, then perhaps a third accumulator that reset at 0mph so that you can relatively easily:

hunt for a peak value while accelerating with pedal position.
hunt for a peak average within a gear with different strategies.
hunt for a peak average from zero to whatever speed you want.

Of course hills and wind will throw off any hope of absolute readings, but might be useful enough with relative readings.

I've asked a few times about this, hope it isn't beating a dead horse. But does it make sense? Sound like a useful tool? Any serious suggestions?

Can you adjust the kick down rod to bring on up shifts earlier?

Pull rod=lengthen
Push rod=shorten

ATs that have kick down cables, rods, or vacuum servos, in most cases will shift very early if those components are not adjusted properly. At least it was that way some years ago.

regards
Mech

I've had a car (and worked on a car) where the kickdown cable was inoperative, ie it always wanted to start in 3rd (both 3 speed autos) and you had to manually downshift to function correctly. I wonder if a old school tranny couldn't be adjusted with higher line pressure and disable the kickdown to make for a manual shift auto? My cars have electronic controlled trannys so I'm sure that's a whole new ball of wax, but older trans seem to be a bit simpler.

My TPS adjustment I mentioned earlier was to the extreme allowed, I didn't force it to go further than engineered. And if power is needed, sometimes a manual downshift is in order. But I can now cruise at 35 in locked 4th without the trans trying to downshift constantly.

Some will start in 3rd, others will still start in 1st and shift very early with the cable disconnected.

For most non-electronic autos (such as the 46RE in my Jeep), you can get a modified valve body that can make them full manual shift.

Has anyone done any testing with heavy acceleration vs. light? I can see good and bad points either way.

Dan-o

I was under the impression that the slowest you accelerated, the less you used gas. (SGII is on the way, so I used my butt-o-meter on this one). Maybe that's why I can get past 31MPG, even tho I drive slower than anyone in town.:p

thanks for the head-up !

I've done all jackrabbit starts on this tank, so we'll see how it goes. Though there are multiple variables at work, so I'll have to compare it to a few tanks with similar weather and no changes to the car.

But accelerating faster gets you up to cruise speed sooner. Like I said, I would like to see/do some ABA tests on this.

Dan-o

with the scanguage I've noticed accelerating fast halves the MPG figure i get when accelerating conservatively. Mind you that MPG figure is >10 MPG in a car capable of 30 mpg, so IDK. Might play with the thing sometime when i have the time.

No love from "goose-ing" the throttle
 

I don't want to shoot down anything yet, but I've tried a tank of "jackrabbit starts" (pre scangauge) and I'll tell you that it burns a heck of a lot more fuel in the long run than normal acceleration. Maybe if I was going on the freeway for an extended period of time that might work, but usually I am only going a mile or two before having to stop for something. I can verify this from my scangauge, from about 20-60 MPH I can see anything from 3-5 MPG (75% throttle) or 12-22 MPG (15% throttle). Without directly measuring the volume that has flowed, I can't say conclusively, but given the chance (and a fat foot), this Jeep can eat a gastank quickly. Hope this helps more than confounds anybody

- Andrew

with an automatic trans, the jackrabbit approach seems less likely to work.

Reason being, the greater the difference between the engine speed and the transmission input shaft speed, the more losses are occuring. When you gun it in an auto, the engine immediately revs up and starts sucking more air and gas, and the transmission and wheels have not increased speed at the same rate.


The trick with an auto is finding a balance between bsfc peak and increasing losses in torque converter slippage. I think that takes some feedback.

In an auto transmission of the older general types (not CVTs, 8 speeds, transmissions that learn the way you drive, etc.), you want to find the minimum speed for an up shift first. Then find the greatest amount of throttle position you can give it without delaying the up shift.

I find that normal acceleration of traffic (at least around here) is pretty good. Your driving tactic in any auto transmission is to get into the highest gear the quickest, as a general rule.

Also try to minimise downshifting if you find you need to accelerate.

If you live where there are some grades, that are not too steep, you can pulse slightly up the grade and then semi coast down the grade and possibly improve your mileage.

The wife's Rogue and my Insight both have CVTs and they are different. The Rogue will accelerate up to 45 MPH at 1400 RPM without the RPM changing. The Insight can go from one stop to another with very little average MPG loss if you accelerate gradually and use regenerate about twice as long as it took you to get to speed.

DFCO is also a good tactic in an auto, if you have to stop. I find much of the art of getting good mileage consists of taking a lot of time to come to a stop. Of course avoiding stopping altogether is better, but with 46 traffic lights on my daily route you will always get nailed by a few of them.

One stretch of 4 miles on my route is about a light every .3 mile. If you do not average 47 MPH the lights will get you, even though the speed limit is 45. Go 45 and you will hit 8of the twelve lights yellow, until you catch one of them and then you start off at the beginning of the light timing window.

I don't use EOC in my cars. It could get better mileage possibly, but in the VX you are better to just use lean burn and stay in 5th gear at any speed over 30 MPH. I will coast down any grade that allows me to idle at 450 RPM and about .125 gallons fuel per hour. Coasting in the VX means your are getting 8 times the speedo reading in fuel mileage. EOC kills the lean burn feature for about 30 seconds after a restart.

I could probably get better mileage with EOC and no lean burn, but my average speeds are usually close to 40 MPH or higher.

regards
Mech

That's exactly what I did this tank, and I only managed upper 27s. Doesn't seem like it does much good for my car.

problem is that an upshift means more torque needed to turn the transmission input shaft, and more converter slippage.

when I come off the dirt road I live on, I generally have to climb a roughly 5% grade that lasts nearly a mile. It's a conundrum for sure. Do I accelerate gently and never make it to 4th gear or do i hurry up and get into 4th before the hill gets too terribly steep? Once I get to 45 or so I can hold 4th, but the whole time I'm accelerating I'm climbing for the most part. I alway balanced jackrabbit with granny and chocked up the extra fuel spent on getting the tranny up to temp so that the torque convertor can lock in sooner.

unfortunately climbing a hill is one of those activities that should put you closer to bsfc peak for a typical vehicle (allowing you to store energy in altitude for a nice glide later while running near peak efficiency). I think the only solution (short of a manual swap) is to override gear selection and lockup so you can climb in the highest gear that works without spinning the converter.

its called ecoMODDER for a reason ;)

On my Jeep, I tightened the TV cable on the tranny up a little about a month ago. With the same amount of throttle off the line, it now shifts about 150 - 200 rpm later than before. However, due to reduced torque converter slippage, mpg remained the same, while firming up the shifts a little.

The smaller your engine is, relative to the vehicle, the slower you can accelerate efficiently. With my huge displacement in a 4200 lb vehicle, I can basically idle up to speed barely touching the gas, but it's inefficient. I'm better off getting on it a little harder and letting it shift around 2100 - 2300. If I crawl around letting it shift around 1500 - 1600 (such as being behind slow people downtown), I get significantly worse MPG.

Wow. I thought I was doing pretty good getting 50-60 mpg in my civic and I'll tell you what, I accelerate like a turtle and use the least throttle I can to get where I'm going!

I'm assuming this is a complicated issue, where there will always be a wide range of results in different situations and especially with different vehicles.