New concept variable displacement??
 

I know there must be at least 5 automotive engineers monitoring our threads here so here is a thought for you.
Honda I believe sucessfully breached the electronic valve issue delving into variable valve timing.
Has there been an atempt to get this next step of variable engine displacement?
An example would be a 90 degree V6 where 2 cylinders are 4" and 4 are 2.5".
utilizing cylinder cut out and injector deactivation with all cylinders @ same stroke length The torque could be doubled for acceleration then dropped for cruising speeds and economy.
couple that to a hybrid with regen and electric boosters like formula one uses.
It could teach todays sports cars a nasty little lesson while providing 50+mpg???????
Not magic power just better utilization of it, with much less waste!

...balancing "nightmare" trying to sling different sized/weighted pistons up & down, not to mention the vastly different MEP values from the two different piston/cylinder volumes.

Does this question need an engineer?
 

The only degrees I have are my temperature, and a buddy tried to fly this idea past me. I raised the same questions you did: how in the world would you balance it? It doesn't take advanced edumecationing to know this would be a dog's breakfast.

I put it to my friend: why not just make all the cylinders work as well as you can possibly make them? Dropping out a couple of jugs when you don't need them is well and good, but mixing cylinder sizes in one engine feels like creating a problem for the intellectual challenge of then fixing it.

SAAB had a design a long time ago for a variable compression engine; one side of the top end was hinged from the bottom end, and a hydraulic servo would move the top up and down to bigger- and smallerize the combustion chamber as necessary.

You'll notice that SAAB didn't follow up on it, instead spending their time developing their brilliant turbo engines, a similarly effective and much more achievable method of making an engine behave as if it were two different mills. Those were the days.

Unfortunately this thought moves in the opposite direction of the KISS mentality. While simple in theory, this is far too complex a design in reality to be implemented into the late stages of a fully matured technology.

I really dont see a balance issue Ive seen old compressors where 1 cyl was ICE probbably 2 stroke other was compressed filter air.
It would make sense to perhaps do like a citroen v5 engine and drop 2 during low resistance.
I still like the big and small bucketts thought though.
1.9ci to 3.5ci just sounds rite.

Variable compression ratios aren't new, there are scads of different designs for it, mostly by small companies that seem to have difficulty proving that they are worthwhile designs. Your idea seems similar to the Split Cycle engine, but I believe balance issues would be a problem with this particular implementation.

A year or two ago in my Automotive Engineering International magazine, I remember seeing that Honda had a functional prototype with variable compression that looked promising. I haven't heard much about it since, I imagine it's still in development.

...the SAAB engine mechanically varied all four cylinders at once via a "tipping" head assembly as I vaguely recall.

...but, turbocharging accomplishes the same result much cheaper and controllably.

That one was wild.

But really... why go for difficult to implement variable displacement when variable trim turbos give you variable virtual displacement already for a fraction of the cost and complexity?

Boy, maybe I am missing something on the variable displacement, but why not have a couple (or few) of v-twins connected by a clutch? It can't imagine balancing issues could be any worse than the current crop of cylinder deactivation autos. Valve shutoff is moot. The clutch should last quite a while. You could even make one of the engines a lot simpler than the other because it wouldn't have a very big affect on emissions because it would only run during acceleration.

Old Tele man nailed it. Turbocharging is a simple way to do what you are asking.

As far as balance goes, it is amazing what can be made balanced. I recall that the Honda GP motorcycle engine for the RCV211 bike (2002-2006) was a V5 with three cylinders on the front bank and two on the rear. While Honda never released Bore X Stroke numbers, a few people found out through hearsay that the front cylinders each had a different displacement than the rear cylinders.
Honda RC211V - Wikipedia, the free encyclopedia

Yes it would be far from the simple ideal that is truth but It seems like something ICE will end up with in 10 -20 years, since rotary and turbine are still scoffed at too.

Well, I don't know, spam from morons is right up there too.

Better to have very high cpmpression ratio with reductions made by valve timing and or lift-overlap. I think that is what Mazda is doing with the skyactiv engines. 14 to 1 on pump gas. Probably only 14 to 1 rarely, but 7 to 1 at MAPs of 50% of atmospheric pressure. Add a turbo to that compression ratio! One of the keys to their design is the 5 separate injection pulses in every combustion stroke, occuring after TDC with no posssibility of pre-ignition since the fuel is not there to ignite prematurely.

regards
Mech

Caller ID! In our case on the TV screen and the phone.

regards
Mech

Another one I always wondered about was using valvetrain trickery, fuel and spark cut to turn the engine into a 6 or 8 stroke engine. I think Cadillac's Northstars did something like this to manage overheating, although the valves stayed open so there were still pumping losses.

Cadillac did it back in 1981. It was a disaster and almost put the company out of business. The engine was a 368 cid. At the time they called it modulated displacement or the V8-6-4. They actually run smooth......when the valve activation and deactivation works correctly....that's not often but it was the computer. It ran a basic TBI fuel delivery system and solenoids that would lock valves under certain conditions. They didn't have any balancing problems but then again the family tree that those engines came from were beast.

A lot of the problem with those engines was when the valves would open. Since it was fogging the manifold with fuel when cylinders reactivated it would deliver fuel very unevenly causing bogging...bucking..

With modern technology you could iron out the problems. They put out a prototype a couple of years ago called the Cadillac Sixteen that used it. Don't think they are putting out a sixteen cylinder engine anytime soon.

I've wondered this as well. A pair of 90 degree V twins connected by clutch.

Such a system would undoubtedly be more efficient, but, I do see a few potential issues.

1. It's more expensive. It needs two separate ignition systems, and ECUs. Probably two exhaust systems as well.

2. Possible NVH issues. Even though 90 degree twins are inherently balanced, two of them running right next to one another might sound like crap when out of phase with each other. Perhaps this could be dealt with with some engine control wizardry.

3. Emissions. I suspect these would need separate exhaust systems which means one of them kicking in and running cold/dirty frequently.

Um... there are lots of engines now with cylinder deactivation. The current methods for cutting cylinders works much better than the old Cadillac system. GM learned that lesson, and Active Fuel Management (AFM) has been readily accepted since about 2005 IIRC. Dodge has it on their new "Hemi", Honda has it on some V6s.

The new thing Justme was suggesting is to have different bores on certain cylinders.

This is a case where 1 + 1 != 2.

Coupling engines together is done somewhat regularly in land speed racing. Inevitably, they get less power to the ground than one engine with the same power as the two engines combined, even with a solid shaft coupling the cranks together. So, you might have two 100hp V-twins that only gets 160hp to the transmission.

It would save fuel when running as a 2 cylinder, but it would be less efficient than a conventional 4 cylinder when the power is needed.

I don't know that a land speed record vehicle is really applicable to a fuel economy technology. They are building those for brutal conditionsand 10+ times the horsepower. A passenger car simply needs low NVH levels and less than one tenth the torque handling. Cold startup is irrelevant because the coolant would run through both blocks, thus heating both engines. I don't see why one ECU/ignition system couldn't handle both engines. It is still making four cylinders fire. It would just have to be a little more complicated to program. Of course it would be more complicated/expensive. So are hybrids and diesels, yet they are building plenty of those. Hybrids are also why I think this will work. They have made planetary transmissions and others (Accord hybrid, for instance, used a four speed auto) receive torque from two sources without massive losses.

Of course, when I think of what seems like a good idea, I usually refer back to this question, "If x is such a good idea, why didn't somebody smarter than me do it already?" The pessimist in me thinks that maybe it was dismissed because its image wasn't green enough. You and I both know hybrids have taken of so fast in large part due to their green image, when much of the technology in hybrids (like active radiator blocks, aero, etc.) could have been put on cars long before hybrids hit the market. But a green image wasn't as valuable to car makers then. Still, it is likely that there are other problems.