Shifting and Fuel Economy...
 

Okay I know it's been ages since I've been on here last, but now that I've got le small efficient car again, and there is no longer 5 feet of snow around, its time to make er more efficient...

Now a question, since my car is a 5speed manual, I'll post some numbers, at 2000rpm in each gear and the speed I'm traveling...

1st - 20kmph
2nd - 32kmph
3rd - 43kmph
4th - 62kmph
5th - 84kmph

These are rough numbers, but fairly accurate, now in first gear, I push the gas harder than a feather touch and with the gravel and everything on the road it spins the tires, so my question is as follows...

Is it more or less fuel efficient to start in second gear (skip first) and when is the best rpm range to shift (if I shift and the rpm are near or below 1000 it bogs and I can't imagine it is good for fuel economy etc)...

...but, *if* you're really interested in fuel economy, you'll *start* out in the highest gear that you can use (2nd or 3rd or even 4th, if possible) so as to get "up to" your intended speed and into the highest gear possible as quickly as possible, because two things "eat" fuel: acceleration (time & velocity) and speed (velocity).

...of course, exactly *which* gear you actually start in will depend upon traffic, vehicle load, engine warm-up, etc.

I often start in 2nd gear.

I shift when I reach 2000/2100 rpm, or when the next gear will be >1400 rpm when I land there.

Really, though, I do 2nd and 3rd so fast, get into 4th by 20 mph (32kph). The shift into 5th at ~35 mph (55 kph).

1400-2100 rpm is the best range on my car, and likely many others. The lower end of that is best for cruising, the higher end for accelerating.

I have to disagree with the crowd on this one, do not start in 2nd/3rd.

Why ?

You have to slip the clutch more to build up speed before you can fully engage (i.e. lift off) it. A slipping clutch is a wearing clutch and a worn clutch is spendy, nasty, and prone to all sorts of other issues when "Mr Mechanic" starts to work on it. I have £2K (about $3K) invested in my clutch so I know :mad:

Try the following.

Start in 1st, gentle throttle - maybe even only idle to get moving. Rev gently to 2.5K to avoid wheelspin - getting this right feels nice. :D This means you can fully lift off the clutch peddle asap.

Then press down fully and shift to 3rd, let the engine go down to 1K and lift off the clutch peddle again fully. There will be less jerk because the speed difference (and also clutch wear) is minimised.

Then Gently accelerate. Then repeat and change to 4th/5th depending on your chosen cruise speed.

With this approach you get out of the high-rev scenario quickly and also your clutch is full in / out as much as possible, so less slip and less wear.

The key here is clutch out at as low RPM as your engine can take given the circumstances.

So you go from 1st at 2500 rmp -> to 3rd at 1000 rpm?

Why do you let the clutch out all the way, and then have to wait?

I've found shifting from 1st to 2nd is very smooth by letting the clutch only part way out, because I can shift sooner, more quickly and at much lower rpm.

Letting the clutch all the way out from 1st to 2nd would result in jerking, as you say, because that takes more time and the rpm's get too high.

I'm not convinced that starting a car in anything but first gear (unless you're on a downslope and can get some movement before using the clutch) is a good idea. Here's why:

Starting the car in 2nd or 3rd will make the clutch wear out exponentially faster.

Let's say, for the sake of argument that 2nd gear will spin the clutch pad twice as fast as 1st gear. One might think, "So my clutch might wear out twice as fast, no big deal, I can still make it last 75k miles..." However, when you factor in heat and amount of time needed to slip the clutch, we're talking about an exponential factor. A hot clutch wears its material out much faster. Also, if you need to slip it for twice as long, at twice the speed; we're looking at clutch wear of (2*2)x= y where x equals some sort of heat factor and y equals the amount of clutch material used relative to using first. So, we're using four times as much clutch material before multiplying it by however much the added heat affects it.

I'm sure my equation is not quite right, and there are some other factors. I'm a musician, not a physicist (I'm sure I'm ignoring weight and a few other variables).

So, how much is a clutch worth? I will admit one miiiight be able to squeeze a bit more FE, but I'm not convinced of that, especially since all that slippage is wasted energy, and that energy is being converted to heat on the clutch which increases wear.

That's just my $0.02.

JMac
http://ecomodder.com/forum/fe-graphs/sig4777a.png

I'm not meaning like count 1 to 10, I mean a quick pause - a second or so. Its hard to explain these techniques, we need a video :D

The delay is to allow the engine (Petrol in this case so light flywheel) to settle down to 1K. When you lift the clutch peddle out the speeds usually mesh quite well and you're ready to go.

Maybe that was in the wrong order.

Start in 1st, go to 2.5
Clutch in, throttle off
wait for revs to drop
Into 3rd
Clutch out.
Away.

Jobs a gud'un. :thumbup:

I can totally see how the lower RPM shifting would be a bit smoother.

However, I wonder if this method would prolong the amount of time it takes to get into optimum FE range.

I've been accelerating a little harder lately in an attempt to get up to high gear more quickly when possible. No ScanGauge (my results are by the tank), so I can't attest to the effectiveness of this brisk acceleration. However, I got the idea from a thread I read on here.

Skipping second gear would get you to third gear more quickly, but the poor acceleration at 1K would mean a good deal of time before getting into fourth. Maybe there would be a net gain by using second gear, but reducing the amount of time spent in third.

Perhaps this is a scenario in which tank-to-tank data is not specific enough to declare a clear winner.

It depends on your gear ratios. My first gear is really short. I know some cars with a taller first than my second gear.

185,000 miles and no slipping on my clutch.

I find that with my pickup truck -- over 7,000-lbs -- that First gear starts are a very easy way to get all that mass underway with minimal engine work. The time in gear is short, granted, but it beats slipping the clutch to start in Second. (Which itself is relative as Second is used by most with this spec truck). As the Gross Combind Weight Rating is 20,000-lbs, the use of First is a bit overkill when solo/empty, but all shifts proceed in the same manner afterwards which is like finding a rhythm and sticking with it.

.

...FWIW, a lot of 2011 Cruze 1.4LT owners are starting out in 2nd gear by using the "manual" gear-selection mode of their 6-speed automatic 6T40 transmission in order to get into higher gears quicker for better fuel economy.

...YMMV.

Not to be a pain in the drain, but isn't using the torque converter (TC) in an automatic transmission about the least economical way to accelerate? I don't know the full mechanics behind it other than I've replaced one, but from personal experience I get easily 15% better FE after I feel the TC lockup.

- hill on my daily drive to work scangauge readout:
if left in overdrive, the TC doesn't lockupin 4th and at 39 MPH instant MPG is 10.6-11
If I put the lever down one position, it will lockup in 4th, at 39 MPH, instant MPG is 12.1-13

Just my 2¢

There's no contradiction here. TC slippage is a part of accelerating with an AT. If you can avoid one more gearshift and avoid lower gearing by starting in 2nd gear, you should get better FE.

...in addition to a "slush-o-matic" converter, the 6T40 also uses "dual-clutch" engagement control.

...it actually slips the tranny into NEUTRAL when you're stopped, then re-engages when you take your foot off the brake pedal and depress the accelerator.

...which produces two interesting idiosynchracies: (1) the car tends to roll foreward or backward if you stopped on a hill and (2) there's a noticeable hesitation/then slight lurch as the clutch re-engages upon take-off.

...and, YES, it's unnerving to new owners at first!

I'm with Arragonis on this one. Save the clutch.

And I say "manual TC lockup switch" to the rest of y'all. I'm such a pimp.

I tend to give just a short blip of power in first and second, and a slightly longer blip in third, with the goal of getting to fourth and above as soon as possible.

I also try to roll through stops as much as possible to start out in second gear.

Be aware that while acceleration has an effect on FE, it is the coasting and stopping phase of driving where you can make the most difference in saving fuel.

I totally agree.

#1 -it makes more sense to accelerate with the engine at peak bsfc.
#2 -it makes more sense to accelerate smoothly in the highest gear without lugging the engine, under as much load as possible.Ee
#3 -remaining aware of your situation and using good judgement will tell you which method to use and when.

Most generally, I use first gear to get moving, nothing else. Once the speedo needle has moved, I shift into 2nd gear and begin accelerating. When I reach a speed in a gear that puts me at idle in the next gear, I shift. I do this audibly because I dont have a tachometer in my car.

In the smooth shifting arena, I'm a king. Often, people riding in the vehicle with me don't know that I've changed gears at all, except by seeing the movements inside the vehicle cabin. Up or down, I match vehicle speed to engine speed before engaging the clutch, IF I actually disengage the clutch to begin with...I often do not.

EDITED FOR CLARITY...

Christ,

Wouldn't engine efficiency while accelerating follow a parabolic curve?

Obviously lugging is bad, but even at a speed where the vehicle is moving just fast enough to not lug the engine, say, 35mph in 5th gear on your average car. Let's say you want to accelerate from there to 50mph. In 5th gear, and assuming a flat road, one would have to press pretty far on the gas pedal to get decent acceleration (even for a hypermiler) at those speeds. Where as fourth gear, I believe, would get the car up to 50 much more efficiently.

So if #1 and #2 are true, then 5th gear would be the best choice if it doesn't lug. However if 4th gear is more efficient, then #1 and #2 are false.

Thoughts?
JMac

#1 and #2 are never accurate at the same time, hence the caveat, #3.

Depending the situation, it may be more efficient to use the highest gear and load the engine as much as possible, whole another situation might warrant leaving the car in a lower gear, allowing the engine to rev to it's peak bsfc area (usually where max tq occurs, also) before shifting.

Specifically, if you need to accelerate quickly, shifting at peak torque, same with uphill.
However, it's generally considered more efficient to ignore the bsfc curve and keep the engine loaded at let rpm, as fuel consumption is based on time(rpm) and power. The idea is to use the least power you can, at full throttle, so that LOD is as close to 100% as possible all the time. Technically, the closer you are to 100%LOD, the more efficient you're being with the fuel being burned.

Now that I went back and read them again, I've mis typed that...

One method is high load, high gear w/o lugging, the other is peak bsfc...

Further clarification - the peak bsfc number always occurs at 100%load, and because of the amount of power that most engines produce, they can't actually get to 100% load, which means wasted movement, wasted efficiency.

Either method, the idea is to keep the engine loaded. The bsfc curve for a given engine changes dynamically with varying load... How it changes depends entirely on the engine and the duty.

My thoughts (and actions): I do a 5th gear 35-50 P&G routine every day. I think it works well. ;)

I use ~75-80 LOD, up to 90 at times. Above that and it goes into open loop mode, dumping extra fuel in to keep things cool.

CVT's are so much easier to understand. Here's my attempt at manual tranny shifting for best efficiency. Assuming accelerating to top gear and then cruising at X speed. Corrections welcome.

The goal is to create all your energy most efficiently. Never mind the dogma about shift as low as possible. That's the correct method once you are at steady speed. Gears mostly just change the power/time ratio. The limitation of gears is really just a function of the RPM at which engines operate.

Accelerating is the act of creating mechanical energy (HP/time) which is "stored" as kinetic (speed) and potential (top of hills). RR and Drag eat away at this but we will leave that out for now. Accelerating faster in and of itself is not more or less efficient than accelerating slowly.

Power is inversely proportional to acceleration:
100hp 1000lb car:0-60 ~20s
200hp 1000lb car:0-60 ~10s
400hp 1000lb car:0-60 ~5s
Theoretically, anyhow.


In a lower gear the "lever" the engine is acting on is longer so it simply applies it's work faster. Therefore, you should shift only when you will end up in the best bsfc range of the next gear. If you shift too early and end up "off the island" then you will create that energy more slowly at lower BSFC until your RPM gets into the sweet spot.

Is that about it?

I'm just now getting my head around this. I have always thought that accelerating faster to X speed requires more energy but it really only requires more HP which is energy over time. In the end it's still the same amount of energy to get to X speed.

Marcus, my issue with that is: the bsfc map is only totally valid when the engine is fully loaded, under theoretically ideal conditions. That never happens in real life, so the effective peak bsfc changes with throttle and load, making it very difficult to accurately judge one's island status.

"bsfc changes with load and throttle"

Isn't that what the vertical axis is? Isn't half throttle the point halfway up the chart? That's how I've always read it. In the following chart, the best bsfc looks like 3,000 rpm (that high?!) and 55 Nm, or about 2/3 of max. Wouldn't that be (about) 2/3 throttle?

http://ecomodder.com/imgs/geo-1L-bsf...nstruction.gif

That's the idea. It's a fixed amount of energy to accelerate a car to the same point. The difference is how efficiently the engine converts fuel into that energy. That's where bsfc comes in.

You're right about "falling off the island" of bsfc. It varies by engine, but in most cases you don't want to be below 1400-1500 rpm for accelerating. Steady cruise is different - lower is better.

As so often happens, I get lost in the math and physics. So did we decide that its better for FE but worse on the clutch to start in second?

I found this thread by searching before asking the same question. I have much smoother starts in second and it seems to temper my driving through the rest of the range. I do about 300 miles a week, half hwy half city. I think I'd be interested in saving wear and tear on the clutch.

Your EX compared to my DX has a shorter first gear. You'll be just fine starting in 2nd going downhill. Use first for uphill. Flat - your choice. I use 2nd but a (very) quick moment in first isn't terrible. I'm talking about 10 feet or so and then into 2nd - just enough to get rolling.

And that's why I start out in first. Who the hell would think that ever having to replace a clutch would somehow be efficient? How much time does anyone actually spend in first gear anyway? The only possible payoff would be if you're going to ditch the car or it isn't yours in the first place.

Once I'm rolling I'll skip gears as appropriate, but first doesn't get skipped.

Again, it depends on the car and the gearing. Some have a tall first, some have a short first. My 2nd gear is only slightly taller than a 1st gen Insight's first.

I have 186,000 miles and my clutch is just fine. Basjoos took his Civic's clutch past 500k when it was replaced along with the engine.

It depends how much of a "stop" you came to also :)

Being able to change a clutch *should* be within the scope of most modders skill sets, parts wise it is maybe $60, but it is still annoying.

There should not be a noticeable amount of clutch wear shifting gears, or even when bump starting. The majority of wear is when you get your heavy car rolling (assuming you are not pretending to be Don Garlits)

But using all the gears means you can better approximate bsfc peak conditions for longer, so with the exception of skipping first under the right conditions, I wouldn't skip gears (unless you have a REALLY close ratio trans).

Clutches - It depends on the car. Clutches for some cars are very spendy - I was hit for £800 for my TDI although that was an uprated one including a new SMF flywheel with a view to performance improvements (not done, but at least it won't wear out :D). That was for parts alone, and a standard one including the DMF is not much less.

DMFs are not always replaced but sometimes they are on the basis of "whilst we have it in pieces may as well do that as well" - like renewing the TDI water pump when doing a belt change. Some minicabs (Taxis using normal cars) local to me have them replaced when new with SMFs.

On the other hand the clutch for a Toyota Aygo costs only around £170 fitted. I'd be tempted to track day or stage rally one every weekend at that price :D

Yah, I'm going with smf on the tdi golf when I get to it. Since my DMF failed I am not a fan of DMF, and since I don't idle very much it shouldn't matter much if I go lighter. I put simple,affordable,durable and efficient ahead of smooth personally.

It will be a couple hundred $ to convert (still less than a new DMF+etc), and a clutch disk after that will be $22?!?

VW TDI Clutch Jetta Golf Passat Beetle 1Z AHU ALH 95-07 | eBay

I totally agree.

Off Topic :

Mine was spendy due to being uprated too - I was aiming for ~180hp - in something smaller than a new Fiesta :rolleyes:.

If you do swap make sure you get the 5/6sp one depending on your gearbox (I think you converted some time back ?) otherwise you get the rattle - like mine does - which I have learned to live with :D


On Topic :

I don't think there is an issue missing gears in terms of clutch wear, its the speed difference between the input and output sides and how long it slips which causes wear. In some cars you could hold say 2nd a little longer and then slip into 4th for cruising - that way you spend less time accelerating without lengthening the time of clutch slip and the speed difference is still quite low.

But this would need measuring car by car to see if it gained an advantage.

The clutch was replaced not because it had worn out (it still had about 1/4 of its lining left), but because the engine and the 5th gear of the transmission had worn out, so I had the entire drivetrain replaced.

That's what I was trying to say but it didn't quite come out right. Thanks.

Maybe its my ratio/turbo, but I have tried may combinations and the most fuel efficent one is slow and steady. :) At least reading this helped me understand that.