VW quantifies MPG gain of 4- to 2-cylinder deactivation at 31 mph = +16 mpg
 

For those of you (and there have been a few) who have dreamed of and/or actually tried your hand at DIY cylinder deactivation (mechanically), VW has provided data on the fuel economy improvement achieved by switching from 4 to 2 cylinders at various speeds in their 1.4 L gasoline engine which will be in this year's "Polo Blue GT" model:

Driving at a constant speed (load):

(Note baseline mpg for putting these figures in perspective: the car is rated either 50 or 52 mpg on the Euro cycle depending on transmission choice.)

• @ 50 km/h (31 mph) in 3rd or 4th gear = up to 16 mpg improvement

• @ 70 km/h (43 mph) in 5th gear = 11 mpg improvement

• On the European combined driving cycle (ie. variable load, and switching between 4 & 2 cylinders): about 8 mpg.

Source: Volkswagen Polo Blue GT Is First Car With Four-Cylinder Deactivation Tech


The article doesn't say how VW accomplishes deactivation, but presumably it's by shutting both the intake & exhaust valves on the cylinders.

EDIT: found a video demonstrating VW's approach...

www.youtube.com/watch?v=yxjGTBF-dVY

EDIT 2: here's VW's press release (edited to remove some unrelated topics & shameless marketing-speak):

IMHO it would be more effective to disable the appropriate injector and leave the valves alone, it seems to me that by shutting both intake and exhaust valves the piston would essentially be creating a vacuum causing even more issues.

Leaving both valves open would pump oxygen into the exhaust stream and mess with O2 sensor.

I think just leaving intake valves open would work better, if they deativated the same 2 cylinders all the time and the pistons are 180 off each other I wonder about connecting the intake runners between them to push the air back and forth between the 2 cylinders.

Leaving the valves closed basically uses the air as a spring. The piston uses energy to compress it, but the air pushes the piston back down on the other side of the stroke. Leaving the valves open would waste the compression energy.

wd, rooster: there are articles here and online explaining cylinder deactivation. Would the manufacturers close off both valves if closing one or none worked better? Really? You are missing a major part of the whole theory and that is reduction in pumping losses. Think about why air compressors require so much power.

I've played with this at home and was hoping someday they'd get it to work on a 4 cylinder. Mine sounded like it wanted to start but it just couldn't quite do it. Of course one of the key things might be to start on 4 then switch off at higher rpm as there just isn't quite enough inertia to get 'er going on 2, but I had to try starting on 2 because my deac scheme was under the hood, not on the go.

Thanks for the figures metro; some rough extrapolating says maybe +15% could be expected on average... that's about what DoD for larger engines gets too... so for the Tempo I'd be looking at +5 or 6 mpg. Nice, but perhaps I won't be pulling those two pistons after all. :/

Frank: maybe you needed another battery inline on the starter circuit, so you could have spun the engine up to a higher RPM with the key. :)

Ha- it was spinning over pretty good. Gee I wish I had a 48 volt system! errrrrr.... Next time I'll get Cletus to gimme a tow up to 55 mph or so and then see if it starts/stays running.

We all know that 2 cylinder engines start, run, and work. I think if mine had more flywheel it would have gone. Or, if it had a 2 piston delete. Or an active deac system for high rpm deac.

rooster: I haven't seen it mentioned (or, I forgot :rolleyes: ) about if the timing of the valve closing event is critical. I would think that since piston rings don't make a PERFECT seal, that cylinder pressure- or vacuum as the case may be- would reach some sort of equilibrium in short order. After all that piston will have gone up and down thousands of times in the space of one minute...

I inadvertently found that a Volvo B18 engine will start and run on two cylinders. Fortunately, it was the middle two cylinders.

Rush fixit job just before the welding supply store closed. Made it to the store, then hooked everything up to get home on four cylinders.

Mine'll run on two if I just pull the spark plug or injector wires, but I went the distance and disabled the valves.

for a 1500 cc engine, that is 15 percent or so.

which is EXACTLY the math we came up with in the bsfc thread.

pretty amazing stuff, because the math suggests it is 15 percent for ANY gasoline engine.

Found a YouTube video - (I embedded it in the first post of this thread) - of VW's mechanical approach for deactivation. As suspected, it's done via closing all the valves on the cylinder.

Or, direct YT link: Volkswagen cylinder deactivation technology on 4-cylinder engine introduced - YouTube

And... also found VW's press release with technical details. I added it to post one as well.

An experiment you can do: Run your vacuum cleaner. Put your hand over the end of the hose. The motor speeds up, meaning the the motor is working less. (You can verify this with a watt meter.)

Closing the valves lets the cylinder function as an air spring.

Honda has a system under development in which nanotech robots actually dissemble and reassemble the engine, placing the unused pistons and rods on a nicely padded shelf, temporarily. They have demonstrated process times down to 5 milliseconds -- fast enough for any road-going engine. The engine is automatically declutched from the driveline during the rebuild process: 1. Declutch 2. Engine stop. 3. Engine dissassembly/reassembly 4. Engine start, 5. Clutch re-engagement.

Although this works fine in the lab, they have been unable to find a place to put the parts shelf in a production car, what with the engine compartment being so congested.

It is believed that future implementations will include the option to re-ring the engine at high mileages.

actually with no air the fan cavitates out.

a better experiment. run your vacuum with an amp meter on the cord.

then plug the inlet until you see 5 mm of vacuum. then read the amp meter.

hancock
 

Interesting VW ACT technology, however, what about engine vibration frequencies with only two cylinders activated?

Also the engine big end bearings are designed to carry bearing loads for four cylinders so will what about wear on big end shell life on the cylinders 1 and 4 during deactivation?

The wear life of the deactivation mechanism? How many times in the average journey will it operate?

VW will have to re-assure buyers of the Polo Blue GT 1.4 L engine live or give a special warrantee?

with the appropriate 2 cylinders deactivated at the same time the vibration would not be an issue.

The bearings will last longer with less hp wear as it would be traveling gently a great deal of the time.

vw name is so good they have it made and will not have too assure anybody of anything, unlike some other car companies.
I like it !

hancock
 

Thanks Ken for your reassurence!

However with any new technology there has to be 'due diligence' by that I mean not only engine test bed testing to build up sufficent technical confidence on the application and on the actual wear parameters since there will be 'sliding' wear.

The valve operating mechanism in any 4 stroke engine is where special attention has to be made to protest 'sliding' surfaces. For instance the introduction of roller followers on cams.

So no hydrodynamic protection and wear protection will be based on the lubricant antiwear properties (ZDDP) and metalurgical considerations such as choice of material and surface treatment to the splines that move the 'collar mechanism' also how this movement interacts with the cams surfaces?

Additionally, I'm sure VW would have conducted field trails to assess wear in the real world. All this testing takes time to complete before the engine can obtain VW sign off and approval.

So I belive in this case where VW is introducing new technology, there needs to be warranty protection.

Worry about nothing much?

I expect that VW will have done massive testing mileages and thousands of hours on the dyno before going to production. I imagine this has been under development for years and personally I wouldn't have many qualms about buying one.

The unused cylinders will have reduced wear. I wonder if it would be worth the added expense of making all 4 cylinders deactivatable?

Didn't think my spell check would like that word. :D

Doing so would extend the life of the engine, assuming it spent a good deal of time in 2 cylinder mode. The ECM could keep track of hours run on inner/outer cylinders and balance it out.

Don't think it would make a huge difference, and it's not usually wear of the valvetrain that kills an engine.

I don't see any big scary red flag on this, engine building is a pretty well established science and I trust VW's done their homework. They do a 10/100 powertrain warranty on the touareg, which would probably be appropriate for this too.

I'm curious though, wouldn't vacuum pull valves open a bit even if the cam wasn't pushing them? Especially with the help of manifold pressure on the other side of the exhaust valves..

If it pulled in air it would/should only be 1 time because it's way easier for air to get in than to get out..

All the pistons are still going up and down- that's a good deal of it right there. Today's engines have better cylinder materials than the play-doh that's been used in the past. I've seen cylinders in a Ford 3.0 with 150,000 miles on 'em that looked like the day they left the factory- no ridge, vivid crosshatching, etc. Contrast that with the cylinders of a small block Chevy from anytime pre-'80s or '90s- 100,000 miles on 'em and they probably need to be bored .030 to clean 'em up. :(

Ever tried to compress a valve spring?

I wonder if anyone has tried a hydraulic coupling that splits the crankshaft and lets the idled pistons sit still?

I've pondered such a thing and if you could get it to index to one position when they are coupled that would be great. If not, I'm thinking it would have bad, bad, BAD, bad vibrations (it's giving me bad vibrations).

I think a big issue is that you deactivate cylinders on the same throw (for a 4cylinder, either the middle pair or the end pair) you can't split half way as you would get uneven firing. If you rearranged so you had it splittable in the middle you'd have terrible dynamic balance when running 4 cylinders, and even worse once split.

Hi guys, new ecomodder here :)

Just as a thought, would shutting off the intake and exhaust ports on all cylinders under overrun (gliding at high rpm) conditions decrease pumping losses and allow you to travel further without dropping into neutral? I suppose the gains would be marginal though.

I think this may be the key to the arrangement. Since they chose cylinders 2 & 3, then not only is there an air "spring" there are two air springs which are 360 degrees of chank rotation apart. The spring effect is doubled, symetrical and complimentary.

Don't know that VW has done so, but it isn't difficult from a design viewpoint to provide pressure lubrication to the splines. One only needs an oil galley through the center of the the camshaft. I suspect that VW would have seen spline lubrication as an issue early on.

Does it have to be inline? What about splitting into two opposed twins (a la BMW Boxers), or two 90 degree twins (like Ducati or Moto Guzzi)? Each half would have perfect primary balance, and each could have their own counterbalance shafts to reduce secondary vibrations?

Possible, of course, but unlikely for a major manufacturer to adopt such a layout when *most* platforms are engineered for in-line configurations (speaking from a European perspective here!)

could this work on a turbo deisel? with an electric fuel cut off solonoid on , say , the two inner injector pipes. Which would be the best 2, inner or outer cut off. How would the turbo react to only working on 2 cylinders.Would a fuel cut off solonoid work on the high pressure side of a Bosch mech injector pump. The engine is 2500cc, so I wonder if a landrover would drive (on highway) on 1250cc without overheating?

AFAIK there is no such thing as an ICE with perfect balance; it changes with rpms too. It may be minimal, it may be acceptable, but I think if the two are not in synch you will know it!

In that case, we need a four cylinder "boxer" engine that has two different displacements, for example a 1L pair and a 1.5L pair. - then use a slide valve in the intake/exhaust manifolds that turns the larger pair into a supercharger for the smaller pair during deactivation... ;)

Thanks guys, yes we can all make our own decisions! Interesting that Ford are about to launch the 3 cylinder 1.0L gasoline engine for the Ford Focus. A different technical approach to VW but coming up with the same sort of answers on fuel economy and CO2 emissions but with clever turbocharging and reduced engine friction.

Hancock

Go common rail, and then turning off the fuel is easy--just turn off the injector. As far as the turbo goes, I think I'd be okay. I'm not sure about the overheating concern--you'd be putting less heat into the coolant with half the cylinders.

Once again, I think that I have a genius idea, look into it, and it has been tried and disproven, and then figured out by VW (or at least someone).

In this and other threads you guys talk about vibrations from having an odd number of cylinders firing. Are one-cylinder engines inherently rough? What about those three-cylinder metros that you know and love?

I just kept reading about how ICE engines are most efficient at 90% capacity, but they are never [responsibly] used like that. I do not have any problem with the idea of accelerating hard to highway speeds, but in theory, couldn't one lonely cylinder maintain speed?

Honestly, I do not think this is necessarily one of those things where if it worked automakers would try it. Sure, maybe we cannot figure it out, but do you think that in a few years Volkswagon or someone else might?

Yes, one-cylinder motors are harsh. They usually have very large flywheels to help damp that out to some extent.

The three-cylinder Metro engines are not that bad. In large part because they are designed specifically to be three-cylinder engines, with combustion events equally spaced around the 720-degree four-stroke cycle.

If you take a four-cylinder engine (combustion events every 180 crank degrees) and just turn off one cylinder, you have three "hits" and one "miss", instead of three evenly-spaced "hits". This makes the motor run rougher and makes the vibrations harsher.

If you disable two opposite cylinders, you get alternating "hit" and "miss" events, which should be less vibration than the 3/4 engine.

The best candidates for cylinder deactivation are engines with a bunch of evenly-spaced cylinders, such as V6, V8, and V12 engines. You could shut down one bank of the motor and effectively make an I3, I4, or I6 engine. If the firing order is set up correctly in the first place, all combustion events will be evenly-spaced and not too far apart, and the engine will be reasonably smooth.

Honda has done the V6 single-bank deactivation. Several manufacturers have deactivated various parts of V8 motors, with varying levels of success.

-soD