Transmission Activity

yes, absolutely. it does it on "kickdown" acceleration /and/ it does it on braking. not gentle braking, but harder braking.

pointless exercise on an automatic. but even then, on the modern autos, in conjunction with electronic throttle, yes, the engine revs /are/ meshed to the gear on shift.

there's no slipping unless the lock-up clutch is released. see below.

yes, that's what a torque converter does.

that's because the lockup clutch is released to allow more torque. more flexibility than a stick where you'd have to shift.

that's different - it's not a torque multiplier. if is however a great way of achieving absolute optimum gear for all conditions.

absolutely! i drove a "real" cvt one when i was in europe years ago, and yes it is /real/ weird at first. but it's amazing how much you can get out of a small 2-cylinder engine when it's got perfect gearing. quite fun! this particular model had 2 independent drives too, so not only did you have optimum gearing, you had limited slip diff benefits in snow & ice too.

if you're not used to traditional automatics, the transition is easy. if you're used to traditional autos, its weird for a few minutes because it doesn't "shift", but beyond that, they're actually very impressive.

no, but it'll start in 2nd. mine will anyway. useful in snow.

regarding shifting, it'll shift down any time on command, providing doing so doesn't over-rev the engine - it won't let you do that. regarding up-shifting, you can hold it back until you're ready, and even then, it'll wait until it's certain you mean it - if you have your foot down. if you're not driving hard, it's academic.

i used to think that. then i had a knee injury that prevented me driving a stick for some months, so i bought an auto. and every time i've driven a stick since, it's been a real chore. that was nearly 20 years ago. it may be that there's some bad autos out there, but the way i have my civic set up [the shift pattern is adjustable], the shift points are pretty much dead on where i'd have them manually, and for other stuff, i over-ride.

correct - they're heavy.

maybe a problem with the sequencing mechanicals? who knows.

i think in due course, semi autos will replace sticks. at least in sports cars. they offer faster, more accurate shifting, and computer control knows more about the potentials of the system than the driver does a lot of the time. a friend has a tiptronic carrera - hold the shift lever and put your foot down, and it'll select the lowest gear for the speed to give fastest acceleration, and you can revert to auto from there so it shifts up through in the fastest possible time. trust me - it's fun!

It is not a bad thing, I just don't care for it. If I want to downshift, I can do that. If I don't want to, I don't like the car doing it on its own.

That is what I mean. I figured they had fixed that aspect which is the worst part of the older ATs.

But that is the slippage. The engine speeds up races up ahead of any change in vehicle speed. It is like a slipping clutch. As for more torque (horsepower really) a lot of that is eaten up by the inefficiency of the torque converter. On cars where you can get a MT or AT with the same engine, the MT is almost always faster and gets better mileage.

Neither is the conventional AT, it just has a clever design to let the engine speed up ahead of the vehicle speed without shifting. It is basically like a limited range CVT.

I agree that it has a big theoretical advantage, especially compared to a conventional AT.

That was the original CVT. Do you remember who offered the first CVT in the US. (I don't think they ever sold the Daffodil here.)

Again though, I am not sure they are any faster or more efficient than a good MT. Consumer Reports tested the Versa with MT and with CVT. The MT version was 0.6 seconds faster to 60 and got one more mpg. (And CR panned that MT.) They also tested MT and (conventional) AT versions of Fit, Rio, Accent and Yaris. In each case the MT was 2 - 3 seconds faster and got 2 more mpg, so the CVT was clearly better than an AT but not as good as even a mediocre MT, at last on raw numbers.

And yet, there is hardly a proliferation of CVTs on the market. Most of them seem to be on hybrids in fact.

I appreciate that many people prefer AT, and if you have a bad knee, there isn't much choice. But I really enjoy driving the MT. I haven't encountered an AT yet that I could actually say I enjoyed, but I certainly haven't driven many new ones.

It was a computer glitch. As they say, to error is human. To really foul things up requires a computer.

I am open to that, but it has to come down to Civic level before it will have any relevance to me. I am sure that a Porsche would be fun no matter what kind of transmission it has.

BTW, you might want to review my post in this thread from 2005:

I had no idea they would work that way, it just seemed like the logical way to do it.

why not? it's just like you'd have on a stick.

but i don't understand the problem - what's wrong with it? engines are not perfect across all rev ranges - why not let a computer manage the efficiency curves - for that's what's happening.

how is a slipping clutch more efficient? [it's not.]

not so with the modern autos. and that's one of the big things about honda autos - it's basically a standard transmission with clutch packs instead of synchros. inherently more efficient than planetary gears.

no dude, they're totally different. "torque multiplier" is something a torque converter can do - hence its name. everything else is ratio control, be it continuously variable or discrete.

compared to /any/ transmission. there are mechanical efficiency issues with the friction interface, but that is more than outweighed by ratio flexibility and ability to keep the engine at its most efficient.

subaru?

well, the daf was only 650cc iirc, and 0-30, that wasn't much to touch it.

that depends on the management system. the modern cvt's "simulate" gear shifts which is the dumbest damned thing since it's not utilizing the inherent benefit of the system! on that basis, i'm not surprised.

that's consumer and mechanic inertia - nothing to do with benefits or reliability. trust me on that one - i've driven the daf for an extended period and it's a great system.

indeed.

Interesting discussion I started here, I guess....!

I've driven the CVT in an Audi A4 loaner and thought it was weird at first (as someone else noted). But it also had 'sport' settings where it would 'shift' thru seven 'gears'. There are more CVT's out there than you think - off the top of my head, I can think of the Audi (A4 non-quattro auto models), Ford Freestyle cross-over, and Nissan Altima, Murano, Maxima and Versa.* I think some version of the Ford Five Hundred has it as well. So, they are becoming more popular.

The best compromise seems to be the automated manual transmissions - they are a true manual trans with an automated clutch. No torque converter. (

)Audi's DSG is commonly considered the best example, although BMW hasone (whch regularly gets panned for poor auto shifting). Porsche mayhave one as well. I've driven the DSG and it is excellent, very fastshifting and a decent auto mode as well. But I still prefer aconventional manual transmission. For the Ody, I'm prefectly happy with an automatic.

*Oh, Wiki has a list of CVT equipped autos world-wide.

• Audi A4 2.0/1.8T/2.4/3.0/2.5 TDI * Audi A6 2.0/1.8T/2.4/3.0/2.5 TDI * Dodge Caliber * Fiat Punto 1.2 L * Ford Escape Hybrid 2.3 L 4 cyl * Ford Five Hundred 3.0 L 6 cyl * Ford Focus C-MAX 1.6 L TDCi 110 PS * Ford Freestyle 3.0 L 6 cyl * Honda Civic HX 1.7 L 4 cyl * Honda Civic Hybrid 1.3 L 4 cyl * Honda City 1.5 L * Honda HR-V 1.6 L * Honda Insight 1.0 L 3 cyl * Honda Jazz 1.4L / Honda Fit 1.3 L/1.5 L * Hyundai Azera 3.8 Lambda * Hyundai Sonata 3.3 Lambda * Jeep Compass 2.4 L * Lexus GS450h 3.5 L 6 cyl * Lexus RX400h 3.3 L 6 cyl * Mercedes-Benz A-Class

• Mercedes-Benz B-Class * Mercury Montego 3.0 L 6 cyl * Microcar MC1/MC2 505cc 2 cyl diesel or petrol * Microcar Virgo 505cc 2 cyl diesel or petrol * Mitsubishi Colt 1.5 L MIVEC 4 cyl with INVECS-III CVT (Asian-Oceanian version only, 72 kW) * Mitsubishi Lancer 1.6 L/1.8 L MIVEC 4 cyl with INVECS-III CVT (Asian version only) * MG F/MG TF 1.8L * BMW MINI One and Cooper. * Nissan Altima (from 2007) * Nissan Cube * Nissan Maxima (from 2007) * Nissan Micra 1.0 L/1.3 L * Nissan Murano 3.5 L * Nissan Primera 2.0 L * Nissan Sentra (from 2007) * Nissan Serena 2.0 L * Nissan Skyline 350GT-8 * Nissan Tiida / Versa * Opel Vectra 1.8 L * Rover 25 * Rover 45 * Rover Streetwise * Saturn ION Quad Coupe (2003-2004) * Saturn VUE 2.2 L AWD (2002-2005), 2.2 FWD (2002-2004) * Subaru R1 * Subaru R2 * Subaru Stella * Toyota Highlander Hybrid 3.3 L 6 cyl * Toyota Camry Hybrid 2.4L 4 cyl * Toyota Prius 1.5 L 4 cyl

Dan D '04 A4 1.8Tq 6-speed

Yeah, my friend's Prelude of mid -80s did same thing.

I would really prefer that Honda would make a Volvo 240 wagon with performance suspension and an MT, but I have pretty much given up hope.

See my comments below. Few of these are currently available in the US,. If you eliminate the hybrids, I think there are only about three. Partly this is due to most CVTs being designed for small engines.

Most CVT designs seem to be reliable and the efficiency improvement is significant compared to conventional ATs. I really think that the test drive turns off a lot of buyers because it is so unconventional. Honda is apparently selling a "7-speed CVT" - talk about an oxymoron. This is a sure sign that buyers are turned off by normal CVT operation. Hybrids may be the thing that brings CVTs out of the closet.

See comments below.

The only US Civic currently offered with a CVT is the Hybrid.

I haven't heard of a Honda City in a long, long time. Are they still sold?

Never sold in US.

Discontinued - will probably be replaced.

No CVT in US market.

I am a little skeptical of those dates. I didn't think they were on the market that long before it was discovered that every single one of them breaks. And GM wonders why it is going out of business. LOL

Subaru Justy (probably equals one of above models) was sold with a CVT in the US for a few years in the 90's. Subaru hasn't sold a CVT in the US since.

Not sure I understand. My GS-R never downshifts on its own.

It is just a personal preference. I like the engine to be positively coupled to the wheels.

No, I understand that (unlike a slipping clutch) there is a benefit to the slip designed into the AT. I just don't like the feel of it and the benefit is more than eaten up by the inefficiencies that come with it.

Looking at the differences between the ATs in the econo cars tested by CR, Honda looks about as good as the other ATs (except for the CVT.)

OK, explain it to me. My understanding is that it is just a trick to get the engine running at a slightly higher rpm to produce more power. Kind of like a mini downshift. All car transmissions are torque multipliers. They take high rpm/low torque and turn it into low rpm high torque. If it were perfectly efficient, the power output would be equal to input but of course it is always less. No way to get more power out unless you put more power in, i.e. run the engine at higher rpm.

The why doesn't the Versa with a CVT get better mileage or accelerate faster than the Versa with an MT? Unfortunately, there are few cars which allow you to directly compare CVT vs. MT.

That is what I recall. The car was called the Justy.

I think the Honda 600 would have blown its petals off. ;-)

Does the Versa do that? CR didn't mention it.

Maybe. I would gladly trade the AT in my Ody for one if it was proven reliable. Not all are. The on in the Saturn Vue was a disaster, but I guess you have to expect that from GM. The Japanese units don't seem to have any problems.

if it doesn't, then there's something wrong. how old is it?

that's what the lockup clutch in the torque converter is for.

you'd hate cvt. there's no "relationship" between revs and engine speed at all.

you're looking at fuel economy, right?

ok, yes, but we're talking about different things. torque converters can increase torque output from a little to a lot in a very limited rev range. ratio change is something different and that's what the gears are for.

see above.

civic hx was significantly more fuel efficient than the stick.

don't know. given "consumer demand", i expect so.

i've never heard of problems with the civic hx.

It is a '94 with a 5-speed manual transmission. What kind of MT cars have you driven that shift themselves???

But it isn't engaged all the time. When it disengages, that is when it "slips" and I do not find that satisfying as a driver. Like I said a the outside, it is a preference for MT. I like to drive cars and the MT is more enjoyable than the ATs I have driven. I don't like it shifting when I don't want it to and I don't like the slippy feel of the torque converter. The fact that the MT is usually faster and more fuel efficient is a bonus.

I think you are probably right. That is why I mentioned the fact that the CVT isn't very popular in the US (if anywhere) and I think it is because lots of people hate it. Why else would Honda sell a 7-speed CVT?

OTOH, it probably appeals to - or at least doesn't repulse - the hybrid buyer because it befits the unconventional nature of the car. I realize there are technical benefits to the combination of hybrid and CVT, but I am saying that the unconventional nature of the CVT is less of a negative when you are already committed to buying an unconventional vehicle. If you are attracted to the hybrid because it is unconventional, the CVT is a plus.

Fuel economy was significantly better for the MTs but the biggest difference was acceleration. The MTs blew the doors off the ATs. In terms of 0 - 60 time differences, the Yaris AT was the best - "only"

2.1 seconds slower than the MT version. Even the slowest MT car, the Kia Rio was faster than the Versa CVT. The fastest AT car, the Yaris, was 1.4 seconds slower than the Versa CVT.

You understand that the ATs torque output at a certain rpm (i.e. power transmitted) is higher when the torque multiplication is active. There are only two way this can happen. The first is to increase the efficiency of the transmission. I think we can dismiss that. The other way is to increase the power input. The only way to do that is to increase the throttle opening or increase the rpm. The throttle opening is determined by your foot (and it wouldn't be much of a trick for the torque multiplier to be just an extra jerk on the throttle and it wouldn't do much good if the throttle were already wide open.) However, rpm is largely controlled by the transmission. The transmission allows the engine to run a little faster and therefore produce more power which is transmitted to the wheels. This is what I meant by a mini downshift.

I gather that was the CVT-equipped model. They don't sell it anymore. I guess people hated it more than high fuel costs.

Like I said, they hate it just like you think I would. (Actually I would love to try one, I do think it is a neat idea and it would be fun for a while at least. Only after the novelty wore off could I tell you if I like it or hate it. I actually suspect that I would like it more than a conventional AT.)

Here is an interesting review I found. (It indicates that the Versa CVT does not have "gears," so the fact that it was slower and less fuel efficient than the "clunky" MT is significant.) I made one editorial comment in brackets.

--quote-- But the CVT is the one that shines. Its proprietary design, benefiting from Nissan's global cooperation with French carmaker Renault, is uncanny in the way it maintains optimal engine rpm within the most fuel-efficient torque range. It senses, for example, when the Versa is proceeding downhill and glides effortlessly into a lower gear range to slow the car with engine braking. [What if you don't want engine braking? It is wasting kinetic energy = fuel. What it needs in a powerful computer with optical sensors to look ahead, evaluate the situation and decide whether engine braking is advantageous.] Conversely, under pedal-to-the-floor acceleration, the CVT instantly launches the engine to its max-torque rpm, then keeps it there unchanged until a desired highway speed is reached.

This goes against every traditional sensation of driving, wherein gear changes trigger a momentary drop in rpm as the higher gear ratio is engaged. There's a lot of mechanical inefficiency in that traditional gear-change syncopation; and Versa's CVT eliminates it. When first experiencing the CVT's behavior, it feels wrong, sounds noisy. In truth, however, it's mostly the lack of noticeable gear changes that's merely thwarting a driver's subconscious expectation of rising-falling rpms as gears change.

--end quote--

Neither have I, but then there aren't a lot out there.

i misunderstood you.

originally, you said "If I want to downshift, I can do that." well, you can on an auto. on my car, the shifter is in the same place as a stick would be too. and the auto replicates engine braking when you need it - just like you'd do yourself. really, it's a good system.

you'd hate cvt. as i said before, there's zero relation between engine speed and vehicle speed - if you drive on engine revs, and it sounds like you do, you'll be suffering total loss of feedback.

check the modern civics in that department.

because they rely on dealer feedback, and dealers are morons? cvt was pretty popular in europe iirc. volvo sold them as well as daf, and they did quite well.

nah, it's a pure engineering logic decision. cvt allows extremely good engine efficiency. if you don't want that, you don't want a hybrid.

dude, compare like with like - not different car to different car if you're trying to compare trnasmissions, i.e. yaris with cvt, yaris with stick, etc. yaris stick to versa cvt doesn't work.

no, it depends on input/output speed differential. within certain rev ranges, torque transmission is very high, even with a rev differential. if it gets outside of that band, it drops right off.

that's the lockup clutch releasing.

my money's on dealer prejudice, not consumer. by the same token, the hatchback has been all but dropped in the u.s. afaikt, that's more to do with vehicles with the same utility selling for $30k rather than $15k for a hatch, not consumer demand. try buying a used hatchback civic here in the bay area - good luck! people just keep them - they never sell.

not correct. when engine braking, fuel delivery is completely stopped.

why? i don't want my car making /that/ kind of decision. seriously, auto engine braking only happens when you're foot braking - just like on a stick. you can't criticize what you've not used dude.

/all/ the 96-2000 hx autos are cvt.

The AT Civic gets slightly better highway mileage than the MT - I am guessing it may have a higher final ratio. The Accord 4 and V6 and the Fit all get better mileage with the MT. I don't have any test results, but I bet the MTs are universally faster (see below.)

Why is that in the past tense? We agree they are more efficient than conventional ATs and if they were popular, why didn't they drive conventional ATs off the market? Reliability could have been an issue but shouldn't be now (as long as you stay away form GM.)

Exactly right. The converse of that statement is that, if you want increased efficiency, you will tolerate or even embrace the CVT. And if you like the hybrid because it is odd or technologically advanced, then you will love the CVT.

It doesn't matter whether you compare like to like or unlike to like. All of the MTs are faster than all of the ATs and the CVT. Let me tabulate it for you:

0-60 45-65 1/4 mi

Fit AT 12.4 8.4 19.0 Fit MT 9.9 6.5 17.4

Versa CVT 10.1 6.4 17.8 Versa MT 9.5 5.9 17.2

Rio AT 12.8 8.1 19.3 Rio MT 10.0 7.1 17.5

Accent AT 12.5 7.7 19.1 Accent MT 9.5 6.5 17.2

Yaris AT 11.4 6.9 18.6 Yaris MT 9.3 6.0 17.3

As I said, "The MTs blew the doors off the ATs."

--quotes-- [my comments in brackets]

Unlike a fluid coupling, however, a torque converter is able to multiply torque when there is a substantial difference between input and output rotational speed, thus providing the equivalent of a reduction gear. [IOW, a mini-downshift, a slightly lower gear that allows the engine to run faster and produce more power.]

The principal difference is that whereas a fluid coupling is a two element drive that is incapable of multiplying torque [IOW, it has a

1:1 ratio of input to output] , a torque converter has at least one extra element - the stator - which alters the drive's characteristics during periods of high slippage, producing an increase in output torque. [IOW, it allows the engine to run faster and produce more power, just like a lower gear.]

The Buick Dynaflow automatic transmission was a non-shifting design and, under normal conditions, relied solely upon the converter to multiply torque. [IOW, there was no gear transmission, all ratio change was due to slippage of the TC - an early CVT!] The Dynaflow used a five element converter to produce the wide range of torque multiplication [i.e. wide range of drive ratios] needed to propel a heavy vehicle.

• Acceleration. The load is accelerating but there still is a relatively large difference between pump and turbine speed. [i.e. low gear] ...The amount of multiplication will depend upon the actual difference between pump and turbine speed, [i.e. the effective drive ratio] as well as various other design factors. [efficiency]

• Coupling. The turbine has reached approximately 90 percent of the speed of the pump. Torque multiplication has ceased [i.e. gear ratio is slightly less than 1:1 but the slight power increase from higher engine rpm is lost to inefficiency] and the torque converter is behaving in a manner similar to a fluid coupling. In modern automotive applications, it is usually at this stage of operation where the lock-up clutch is applied, a procedure that tends to improve fuel efficiency.

--end quotes--

When they say "torque multiplication" what they really mean is a drive ratio less than 1:1. IOW, every transmission is a torque multiplier, at least in the lower gears.

Right, there is no torque multiplication in the TC unless it is slipping.

I personally like hatchbacks and I know they have a following. I too am surprised and disappointed that they have all but disappeared. I don't know I am ready to sign on to the conspiracy theory though. If there were a strong market, Someone like Mazda or Nissan would jump to serve it to increase their sales. apparently there isn't enough demand to justify the high expense of two body styles and the sedan is more popular.

I suppose I could blame the shortage of MTs on greedy dealers and manufacturers who want to force me to buy a more expensive AT. However, I am more inclined to blame Starbucks.

But kinetic energy is being lost. Doesn't matter if the car has to be brought to a stop anyway, but if the driving situation requires little or no compression braking and the computer orders a lot, the car will slow unnecessarily and fuel will be consumed bringing it back up to speed.

I use compression braking a lot. Sometimes I use a little, sometimes I use a lot. Sometimes I coast with the clutch disengaged for maximum distance. A transmission that always applies medium compression braking is not as going to achieve the same thing.

But you are satisfied with it making the decision mindlessly. I agree I do not want the car making that decision. I want to make it myself and shift the transmission accordingly. My brain is the powerful computer and my eyes are the optical sensors.

Not true. I use compression braking all the time with my foot off the brake. Descending grades would be a prime example except that I live in Chicago and there are no hills. Instead, I am the only guy in crawling rush hour traffic not flashing his brake lights every 20 feet. It is way more fun than an AT if you play the game.

I am only pointing out that this feature has a down side. It might not bother 95% of drivers, but it would bother me.

Yes, but how many is that? If they had sold that many, it would still be on the market. Don't get me wrong, I don't think they are any less reliable than a conventional Civic AT, but I don't think there are enough out there to really know for sure. They certainly aren't terrible like the GM CVT.

that's unsafe btw.

no i'm not. i am satisfied from extensive experience over many years in all conditions and many many miles, clearly based on extensive research on driver usage and testing of driveability, that programmed engine braking algorithms used in electronically controlled automatics are highly proficient and effective. as you would know if you'd driven one.

i don't want it making an independent decision. they're programed to make *dependent* decisions based on how the driver is asking the vehicle to behave. which is what you want.

you can - use the shift lever.

but the two are apparently unable to work together to get the ass on into a dealer to test out the new fangled machine the mouth is criticizing.

no, it's true. autos engine brake, just like a stick. and you can engine brake without the foot brake too. we've discussed that repeatedly.

see above.

but your assessment of the whole situation is flawed - it's assumption and speculation /not/ based on either experience or sufficient knowledge. modern autos shift to engine brake. they do it flawlessly and just as a normal driver would shift a stick. and the degree of braking depends on how hard the driver's braking - /that/ is /more/ sophisticated than the average stick driver.

Why?

It can have all the algorithms it wants but it doesn't know whether the hill is long or short. It doesn't know whether, at the bottom of the hill, there will be the beginning of a steep ascent or a freight train crossing. It just goes for some predetermined drive ratio, oblivious to what is outside the window. I chose the level of compression braking based on information the computer just doesn't have.

No, I want to tell it what to do. I don't want it to infer what I might want the transmission to do based on what I am doing with the throttle and brake. At least a conventional AT gives you some direct control over this. I would like to believe that the CVT does also, but I don't really know. It should assume that I want minimal compression braking unless I specifically signal otherwise. If it wants to assume I want more compression braking if I am at least moderately on the service brake, that is OK. But if I take my foot off the brake, it should resume minimum compression.

Not clear on how you control the CVT, but the article implied that it automatically went to some medium level of compression braking. It wasn't clear that you could over-ride this and force less or more compression braking.

That's just it. I don't test drive cars unless I am considering buying one. If I get the opportunity, I would gladly take it. But I don't think I would ever consider buying one if an MT were an option.

But you said it happens *only* when you are foot braking. I thought you were referring to the CVT. I know how a conventional AT works.

It may or may not be more sophisticated than the *average* stick driver but it is no way more sophisticated than what *I* do. I can't decide what gear to be in without assessing the situation outside the windshield. No matter how powerful the computer and elegant the software, it is a blind driver. I use engine braking instead of the service brake. To have it mindlessly aping what I do with the service brake is hardly sophisticated.

It doesn't allow engine braking for one thing and it's illegal for another.

On 1/14/2007 4:23 AM Brian Smith spake these words of knowledge:

FWIW, neither of those responses have anything to do with safety, although I suppose you could make a tenuous case for the first, given some not-obvious assumptions about manufacturer's intent to provide and buyer's intent to use engine braking.

RFT!!! Dave Kelsen

For one thing, the fact that coasting doesn't allow for engine braking is a safety issue. For the second, the fact that it is illegal makes it a valid point.

you just want to express your opinion, not discuss merits of either system. that's fine if the opinion is informed, but since you have no experience and apparently don't wish to pay any attention to fact, then your opinion isn't opinion, it's mere prejudice.

i'm sorry i wasted my time.

Saying that coasting is unsafe because it doesn't allow engine braking is like saying that not pushing on the brake pedal is unsafe because it doesn't allow frictional braking. Engine braking is available if I want it, just like frictional braking. When and where each is used is a decision to be made by the driver based on the situation. For example, coasting on glare ice is a lot safer than compression braking, especially on a rear wheel drive car. Even more so with the advent of ABS.

As for laws against coasting; they may vary from state to state. Here is one form Maine:

An operator, when traveling on a downgrade, may not coast with the gears of the vehicle in neutral. [1993, c. 683, Pt. A, §2 (new); Pt. B, §5 (aff).]

Now notice that it specifies coasting *on a downgrade* *in neutral*. Coasting on level ground or coasting down a hill with the clutch disengaged but the transmission in gear is legal. I would agree that this is generally a bad idea and certainly bad if you are riding the foot brake. However, if you are coasting down the hill with the clutch in, not using the foot brake, and you are satisfied that your speed is not excessive, what is the problem?

The only purpose of such a law is to give them another charge to throw as some idiot who causes a wreck because he doesn't know how to control his vehicle. Can you imagine being pulled over because you were coasting down a hill?

And just to be sure we are perfectly clear on this point, I use more compression braking and less frictional braking than about 95% of drivers in the same situation. But I don't use any braking where it is not needed.

I stand by my original position; if it doesn't know what is happening on the road ahead, it cannot possibly make the best decision regarding operation of the transmission.