10-15% reduced consumption after "engine split" (physically deactivating 2 of 4 cyl.)
 

Hi guys! This is my first post at this forum. My primary language is swedish so my english may look a bit odd here and there.

I had a small 1981 Fiat 127 with a 4-cyl 1050cc OHC and four-speed gearbox. -A tiny and old-fashioned budget car with fuel consumption like a modern medium sized car. In Europe we don´t use "mpg" to describe fuel efficiency. Instead we use liters per 100km. My Fiat used to consume at least 6,5.

Some years ago I got this brutal idea to reduce fuel consumption by ripping out two pistons from a 4-cyl engine. I thought the modification would reduce the piston- and bearing frictions and increase the combustion pressure or actual compression ratio. Two years ago my already quite worn-out Fiat engine seized in one of it´s big-end bearings and it was not worth the money for a proper repair or engine replacement.

I thougth "what the hell..." and ripped out 3:rd and 4:th pistons with the engine still in place in the car. I also modified the camshaft and lifters to "kill" the valves not needed. The lubrication orefices on the crankshaft for piston 3 and 4 were welded shut. I finally re-used the cyl-head gasket with some "permatex" gasket seal.

To my big surprise and hapiness the engine did not only start and run, it kept on running for over 40.000 the following year and was still running well when I sold the crappy car as spareparts. I consider my theories proven but the reduction in fuel efficiency turned out less than I expected. A reduction of "only" 10-15% at unchanged avrerage speed must be called a success regardless of my expectations.

I have been driving old cars with verry weak engines for many years but with an increased interest for fuel efficiency and eco-driving I learned a lot that year. I now drive a 1975 Saab 96 V4 that is much bigger, about 100kg (220lbs?) heavier but with much better aerodynamics. Maybe one day I will try to repeat my "engine split" with this much more suitable car but since my old Saab is the family car I have to be prepared for some yelling from my wife and children... :o I guess however that the Saab V4 engine (a 1500cc/65hp V4 from Ford) have better margins than the tiny Fiat engine (1050cc/45hp)

Anybody who havre hearde about this kind of crazy and brutal modification? I suppose a fifth gear will achive at least the same improvement in fuel efficiency but as a relatively easy "quick-n-dirty fix" on an old car with four-speed gearbox it can definitely be a way to reach good results. Combine this with improved aerodynamics and reduced weight and it can be a true success!

Here´s my projetc blog (mainly in swedish but with a lot of photos):

Projektblogg - Fiat 127 engine downsizing

Two movies of the car:

http://youtube.com/watch?v=GcKNt6QxEYE

http://youtube.com/watch?v=vjQ1DVT8Dq4

(since I´m new at this forum I don´t seem to be allowed to post URL:s befor my fifth post so you have to copy and paste)

It has been discussed and I think somebody did it or is going to do it. I was considering it pretty seriously for a long while but the bottom line is, as long as the car runs well, I'm not all that motivated to mess with it, especially since some of us have predicted the potential econo improvement to be less than 20% which for my car would be less than 7 mpg.

I'm surprised you took out #3 and #4- did it shake and/or sound bad? I'd take out pairs 1-4 or 2-3.

Seems to be a bit tricky to post an URL as a new member. I got a message that URL:s can´t be posted before my fifth post. Gess I will overcome that problem soon... :)

I'm currently working on a similar project. Inline 4 cyl gasoline engine. Cylinders 2+3 run in the conventional manner. 1+4 are steam driven using original exhaust valves as steam inlet and original intake as steam exhaust. Traditional radiator is substituted with a heat exchanger that sponges waste heat from engine coolant. Hot water is then pumped into a water jacket that surrounds most of the exhaust system raising it above boiling temperature. The steam engine purist cringe over this but there isn't enough real estate under the hood to accomodate a perpose built steam engine. I did a similar project a few years back using a radial inflow steam turbine mechanically linked to the crank via a cogged belt. 47MPG highway on a 26MPG epa. 150 rear wheel HP vs 118 advertised HP at the crank. Biggest problem was generating quality dry steam to eliminate water slugs that erode turbine blades. Reciprocating steam engines can tolerate some slugs without catastrophic failure. Have to fabricate custom cam core for this to happen but its in the works.

Frank -

Is it possible that JohnAh's Fiat engine was "pre-disposed" towards the pistons he picked for removal or is it a case of "all inline fours are the same"?

EDIT: Ok, the YouTube says the engine is "special" in terms of balancing?

CarloSW2

I don't see videos; maybe his engine has balance shaft(s). Still...

Frank -

Here are "text blurbs" from the video :

CarloSW2

I probably would have only removed the one piston, but good idea for getting some extra miles out of a broken car.

You would have either put it back in or taken another one out within 10 miles! It's that awful.

This I'd considered once on a Geo Storm, taking out the center cylinders (had a broken rod) and running on just the outers (1 and 4). Ended up selling the car to a guy who thought he'd "just buy a new engine" and "drop it in", then ended up junking the car when he priced NEW engines for it instead of listening to me and figuring out that engines from about 100 other cars bolt right in.

That's awesome! I had wondered if anyone ever tried this...

Thanks for posting - great project.

I made your links active.

The seized big-end bearing was on cylinder 4 so I could chose between keeping 2 and 3 och 1 and 2. If I had kept piston 2&3 the engine would be a more common twin with "symmetric" firing order of 360 deg. On the other hand it would also get the same problems with mechanical balance as any single cylinder engine. In that case I would have to add some counterweigths at the crankshaft but would still have had vibrations at any speed.

I also have a smaller Fiat 126 with a 600cc paralell twin in the rear. The only reason this engine SEEMS to run smooth is because it rests on a big soft coil-spring. The engine itself is constantly rattling itself to pieces. It´s air-cooled with at lot of sheet-metal that cracks everywhere if you push it to the maximum. (which you mostly do because it only have23 hp...)

A single cylinder och 360 degree P-twin simply can´t get good balance unless a balance-shaft or similar solution is added. This is why I chosed to go for the 180 degree configuration on my Fiat 127. The two counter-acting pistons doesn´t give perfect balance either but the need for counter-weight is less critical I think. Since the 1050cc straight four is quite long i block and it´s crankshaft have five bearings the weight of the block will reduce the rocking vibrations. I guess the front-drive gearbox at the end of the engine also add some extra momentum.

Anyhow, my somewhat unusual cylinder arrangement gave a well balanced engine at any speed, but due to the long separation of 540 degrees every second firing the engine´s power-strokes felt a lot at low rpm. I also guess that the car would have been slightly easier to drive if I had chosed the more common 360+360 config but I´m sure the vibrations at ANY speed would have driven me crazy and perhaps killed the engine.

It would be verry interesting to try the same modification but saving piston 2 & 3 to get a comparison.

I now want to try this down-sizing cruelty on my Saab 96 V4, but such an engine makes the selection of pistons to rip out a lot more difficult. The engine, a Ford V4 designed in the early 1960´s I guess, have an even 180 degree firing order with pistons in opposite corners travelling in paralell. -It doesn´t share the same crankshaft throws for two opposing cylinders as a traditional american V8 or a Harley V-twin. A V4 engine must have a counter-rotating balancing shaft to run smooth. This shaft is driven from the crankshaft by the same cogwheel as the camshaft and is made by a fiberous plastic. A few weeks ago my balance shaft stopped working (and so did the fan, alternator and waterpump since their driving pulley is attached to the balance shaft and not to the crankshaft -a stupid solution) Anyhow, this failure gave me the experience of feeling a V4 without balance shaft and this was a lot worse than the vibrations in my modified Fiat. It was quite nasty vibrations of higher frequency.

I realy have to think twice if I decide to repeat the "engine split" on my Saab V4...

Welcome to ecomodder, and that's some intro :thumbup:


I've been in Vallentuna last July - visiting Åke Jansson's aircraft collection along Väsbyvägen. :)

A few users have done so - or at least contemplated doing so.

And of course VW have recently introduced their 1.4L TSI with dynamic cylinder deactivation .

Wouldn't that make it possible to deactivate one side of the engine, and just use 1 bank of the "V" as a standing twin ?

Ooops, I made this reply to some posts at the first page of this thread without seing that I already had answered. But I´ll leave it here since people are always asking me why I did it the way I did and not like a more traditional two-in-line engine.

--------------

Frank, have you tried to run a four cylinder engine with only three pistons? Long ago I heared about a man who bought an old car with V8 engine that felt slightly "off". The buyer suspected something was wrong with the ignition system or perhaps a stucked valve. He finally put a thin welding rod down the sparkplug hole and ended up in the oilpan... -No piston at all! But that was a heavy slow V8. Removing a single piston from a four-in-line is probably something completely different that never appeared as an alternative to me.

When the idea of removing pistons first appeared to me perhaps ten years ago I thought of it as a way of saving fuel only. My experiences from single- and two-cylinder-in-line engines told me that if I hade chosen to keep the two pistons running in paralel without modifying the crankshaft counterweights I would have ended up with an enginen that SOUNDED smooth but with terrible vibrations.

Earlier in life I had a coupple of small Fiat 126 with a two cylinder in-line engine in the rear. These engines are designed for this like most 2-cyl motorcycle, but there´s no doubt that they still give nasty vibrations. The 126-engine is resting on a soft coil-spring in the rear. I once tried to replace the spring with a common rubber cushion (?) and got terrible vibrations.

A two-in-line two-stroke engine have a 180 degree crankshaft with the pistons going opposite directions. There still is a rocking imbalance but that is smaller than the total vibrations from a single cylinder engine of same size. I have a 600cc 12Hp Kubota diesel that also utilizes the 180 degree configuration. So choosing this for my Fiat engine modification wasn´t a difficult decission.

Now with some distance to this project I think of the "engine split" as one way of three possible ways to reduce fuel consumption almost the same way. The other two "similar" methods is to reduce gearing/engine rev and to perform "pulse & glide-driving". In all thre cases an oversized engine is operating at a higher combustion pressure than normal near maximum torque for most of the time its running. Having extra power when needed is definitely nice in some situations so the "engine split" as a way to save fuel only is the worst method of the three. But as a desperate repair it´s a nice comfort that it also save a lot of fuel.

Wow .
Welcome to Ecomodder

I haven't removed any pistons but I tried disabling 2 and then 1 on my 4 cyl Ford. The triple ran but sounded simply awful.

Running a four on three by removing the ignition cable reminds to well of a junkie car... I guess it can´t be any better if the piston is removed too...

Removing TWO cables can give a hint of how the engine will sound after having two pistons removed and there is a big difference if two paralel-going or two "oposing" pistons are disconnected. The engine sounds smoother with two paralel cylinders running but as long as all four pistons are still in place such a test will say nothing about the imbalance when the pistons are removed.

It would be interesting to modify two similar engines and compare them by driving the car. I guess the 360-engine will be easier to drive and produce less vibrations from the power-strokes at low rpm but the 180+520 engine will definitely be more pleasant in total.

I guess there are two reasons that an "engine split" will increase fuel efficiency:

Friction losses are reduced when the piston, conrod and valve mechanism is nearly halved. (crankshaft/camshaft bearings are still the same and so is the oil/water pumps) I noticed a huge difference in engine breaking after my "engine split", it almost felt as if I hade put the gears in neutral when I released the throttle.

Even if the reduction in friction losses most likely make a noticabce difference I think more of the improvement comes from the fact that combustion will be more efficient when the cylinder filling/compression is almost doubbled.

An interesting test would be to only disconnect the valves of two cylinders. On a push-rod engine (unlike the Fiat 127 i modified) this may be done very easy by removing the push-rods and/or the adjustment screws for the corresponding rocker arms. The sparkplugs should be left in place with their cables still connected. The cylinder pressure should even out and hopefully there will be no build-up of engine oil above the pistons...
Such a test should show how much of the 10-15% saved fuel is because of the reduced frictions or de improved combustion. This modification would be non destructive and reversible, but also keep the engina balance normal.

The possible problems I see by only disconnecting the valves of two cylinders is what happens to the engine oil on the cylinder walls and how to take care of the rattling cam lifters that may be difficult to remove on some engine designs without ripping appart the entire engine or at least the cylinder head.

That's what I did- pulled the rockers and the injector wires. It wouldn't start on 2 cyls but did on 3.

These are some very cool numbers. I'd like to know more on sizing of different exchangers and such of this cogged belt combination. The largest steam turbine I've found was 5 horsepower...and it was just the plans. What did you use? Thanks in advance
rubenova

an engine that will automatically disable half of its cylinders is the latest thing in big mills.. well maybe not such a new idea ..
look at this 1977 article about the Ford Inline 300CID Dual Displacement engine .. two in one engine .. 6 cyl when required and 3 cyl when on cruise

too bad there was not enough market acceptance in 77 for this to hit mainstream then .. i sure would like that dual displacement on my 91 300CID..

the link will put you on the power curve chart . you will have to scroll up a page or 2 for the whole article

http://books.google.com/books?id=FgE...0curve&f=false

Right now I´m thinking about a solution somewhat related to my displacement reduction project. -Instead of the standard solution with a 4-in-line two identical 2-cyl engines on separate gearboxex side by side! This would offcourse be heavier than with a common gearbox and separate cluthces but by having two separate systems I get a unique advantage: reliability! - If one engine or garbox breaks down I can still continue my trip. When extra power is needed both engines will run and also give better traction than with a differential between the wheels. While in eco-mode I can alternate between the two engines to wear them equal or just wear one out before the other.

Two gearboxes for half the load each should be about the same weight as a single transmission minus the weight for the differential. Perhaps more friction losses are added and definitiely more complexity to linkage and engine control. Offset traction while using one engine only may be a obvious problem. There must be an rpm limiter while using both engines in case one tire suddenly lose it´s traction while accelerating.
I don´t think this is a concept for mass production but for a DIY:er it may be interesting. It may also be suitable for "scrapheap challenge enthusiasts" like myself. :)

Now I just have to check if this crazy concept can fit into an old Saab 96! A unique feature with these cars are their freewheel. The Saab´s longitudinal engine/transaxle placement may be just perfect as long as there is enough room between the wheels. (these cars are quite narrow)

Well, the scientist in me got tickled after reading this article.

Yesterday, I disconnected the fuel injector harness and spark plug wires from the 2 center cylinders on my Saturn SL1. It starts, it idles (a little rougher than stock but not bad), it sounds like a wussy version of a v-twin. How does it drive? :turtle:

After driving it around a parking lot, i didn't think it would have enough power to maintain highway speeds. I was wrong. Besides the fact that it takes about 15-20 seconds to get up to 55mph, it actually holds 55mph on the flats in 5th gear at about 3-5mmHG on the vacuum gauge. Taking hills is quite scary, and i find that i have to downshift way sooner and use much higher rpms to limit the amount of speed lost.

I suspect the A/F ratio is really rich right now, given that air is passing through the center two cylinders without fuel, which is probably causing lean readings from the 02 sensor and dumping tons of fuel into the 2 operating cylinders. I'm sure my MPG is taking a hit overall, but it's fun to experiment!

The car stock with a brand new motor has 100hp and 114ftlb. Right now (182,000miles), the car probably has a 10% hit in performance due to wear. Given the experiences I'm having with this current half engine, and plugging numbers into the Aero+RR calculator on this site, I'd say the torque output is somewhere around 35ftlb or so at the wheels.

So this got me thinking about getting rid of the stock engine and transmission and installing a motorcycle engine that meets those requirements. Some sources say the Saturn Sl1 SOHC engine weighs about 200lbs, and the MP2 transmission is probably about the same. I'm thinking a 800cc +/- in single or v-twin configuration would suffice. I'd also save about 200 lbs.

@theycallmeebryan: Nice to see that my crazy project can inspire! I did´t think about the possible problem with clean air being pumped and blended with the exhaust. You are probably right about the sensor reading lean mixture making the injectors dumping a lot more fuel than needed in the remaining two cylinders. Dammit! I was hoping to get a some answer about how much of the fuel savings is comming from the harder work in the remaining cylinders and how much is saved by reduced piston friction.

Is there perhaps an easy way to block the intake manifold or to disengage the valves? Don´t remember if I´ve written this before but I want to try removing all push-rods for two cylinders in my old Saab. Removing the push-rods is easy but the question is what will happen when the valve tappets get lose? Removing them is a LOT more work but perhaps they can be lifted up and locked in place by magnets or a thin strip om metal.
Doing the same trick with an OHC-engine demands most likely a destructive operation.

The test is however very interesting. If fuel consumption can be reduced by 5-10% with all piston still in place there may be a way to modify the engine so it can go from standard mode to "split-mode" while driving.

If your motor is a pushrod engine, you could remove the pushrods, but your concern about the tappets getting loose is valid. If it's an overhead cam design, you'll have to remove the rocker arms.

Otherwise, i suppose you could remove the intake manifold and cap off the ports to the cylinders you are turning off, using a solid gasket or something.

I suspect that you could achieve a cheap conversion by correcting the A/F ratio and putting the plugs and injectors on a switch.

Good project. Keep the posts up.
Welcome.

Could this be done on a honda civic 1.5/1.6L motor? Wouldn't leaving the pistons in create engine knock?

@Lethedethius: What do you mean by "engine knock"? I do see a small risk for slow oil-buildup ontop of the pistons even if the average pressure should be atmospheric when all valves are shut. The best reason for this fear is if one valve have a small leakage. I guess this can easily be checked by removing the sparkplugs now and then.

I know nothing about the Honda engines (and verry little about any modern engines) in detail. I would love to see my "split" experiment on a modern engine with injection and also one with turbo. What may make a reversible modification difficult on an OHC engine is if removing the tappets may not give enough clearence between the cams and the valve shafts. This was the case with my Fiat but since the engine was wrecked anyway all I had to do was to cut the clearence needed with my angle-grinder! :-)

On an OHC engine with rocker arms I suppose it may be quite easy to make a reversibe mod by removing the rockers only. (provided there´s no hole where oil starts to squirt like crazy when parts are removed...)

If an engine is to be modified "all the way", still with a chance to turn back, I suggest that the crankshaft oil orefices is shut by another method than welding. There´s always a risk for small drops of metal entering the crankshaft when welding and even if it´s a junk-engine I suggest to first insert some sort of metal plugs before welding starts.

I once read in a military handbook that a seized piston or bearing can be temporary "fixed" by removing the piston and then plugging up the oil-holes in the crankshaft by some layers of tape compressed by ha hose clamp! This is a realy dirty fix thay may both hold just around the corner or forever... I wouldn´t do this if I didn´t have a reliable oil pressure gauge! -A warning lamp may not give the early warning needed to save the engine if the hose clamp slowly gets loose.

Think I wrote it before, downsizing the engine is one of three ways to increase efficiency by increasing compression/combustion pressure and reducing ammount of engine friction per driven distance. The same thing can be done by changing gear ratio (modifying/replacing the gearbox or change to much bigger wheels) or switch to "P&G driving".

One thing that may lessen the improvement of fuel economy from the engine downsizing is fuel enrichment on a carburetted engine. Some (most?) carburettors give a richer mixture when operated near full throttle and since the engine will be weak the driver may be tempted to use a heavy foot. This will NOT increase power other than at very high revs. On my Fiat I noticed that if driving at full throttle I got no higher average speed but most of my modification´s 10-15% improvement were lost. The easiest solution to this problem is to put a wedge between the pedal and the floor.

I don´t know how an injection engine will behave but I suppose the airflow sensor will average out the pulsations from each intake stroke so the system will think you are driving with four cylinders on half the load, avoiding the fuel enrichment regardless of how much the pedal is pressed down.

I guess a carburettor will also average out the the pulsations at higher revs to a more continous spray but here the fuel enrichment may be mechanically activated by the throttle opening and not by actual airflow.

I also see a risk for imbalance in fuel mixture between the two cylinders if using a 180 degree config as I did on my Fiat. -If my theori is right the "second" cylinder will get a slightly richer mixture since it gets some of the fuel spray started by the "first" cylinder. Since it will take 540 crankcase degrees until next induction stroke the carburettor spray may have time ease down more than on the following 180 degrees. When I compared my sparkplugs I saw no sign of the cylinders getting different mixture but there may still be something to this theory...

Interesting.... The newer insight had once advertised that it had cylinder deactivation. It was never clear if it infact would run as a 2 cylinder varable cylinder such as only 1 or up to 4 or if it was all on or all off like the tradational fuel cut, but in this case turning off the rocker arms by vtec as it has dual vtec.

The car sounds like a singer sewing machine at ide with no load at operating temperature and since I havfe a straight pipe exhaust it sounds like my 2 cylidner geo that had a bad cylinder. Then when you touch the gas or turn on the ac the engine comes to life and sounds like a tradational 4 banger.

Guess I could unplug an injector for a start since I got the IMA backing it up and a 4 kilowatt reservce of power for a few hours of driving/?

A few days I got an idea on how to improve engine balance when converting från a straight-four to paralell twin: Use a 360 degree configuration and leave the unemployed pistons in place! To reduce friction losses the pistons should be modified by cutting holes in the piston crown and removing the rings. The pumping losses are at a maximum about "halfway" between full throttle with normal valve lift and no valve lift at all. When introducing holes in the piston crown this will also introduce internal pumping losses! The holes must therefore be as large as possible. Perhaps the piston skirt should also be modified to reduce friction even more.

I have seen single cylinder motorcycle engines with a balancing device working as at counterweight to the piston, reducing the vibrations in the same way as a "perfectly symmetric" 2-cyl boxer would do. My passive modified pistons should act in a similar way. The big challenge may be how to cut the pistons for minimum pumping losses and friction while keeping the maximum weight.

Good job.

Now you have half the engine (cubic).
Less airflow resistance in engine, less speed, friction...
After all the math, you are getting correct numbers.

Soon will come the day that we will be running on one super giant cilynder with hi compression special aloy. Or 2 opposed linear... Mark my words :-)

What you have done with this engine is indeed a cruel thing. Last time when I was replacing a ignition coil in my Renault, I came up with an idea: what if I add a little electronic circuit which will switch off fuel injectors in some particular moments when power demand is low, for example when you've already stopped and you are waiting for a green light but you for some reasons don't want to stop the engine. I know that I will not reduce friction loses, but think about that another way: if there are only two cylinders working and they have to produce the same amount of energy (to power up engine accessories) they will be heavier loaded - so they should work more efficiently, am I right?
If I build a circuit which steals signal from gas pedal position sensor I could switch two fuel injectors quite often, for example in situations when gas pedal is pressed less than 15%.
Another think is to change which cylinders are switch off, if I do that quite often no cylinder would get colder so whole engine would be always ready for higher power demand. What do you think? I can build such a circuit:)

Thanks for posting this, I have a few comments.

only 5-10% reduction in fuel consumption just isn't enough motivation to justify the work, in my mind. By disabling a couple cylinders, but not removing the 'dead cylinder's' friction doesn't seem worth the trouble. If you leave the pistons in, you need to maintain some valve action. If you don't, the piston is compressing air needlessly. Worse, it has no place to go. Ring friction is also a big deal.

Most of the disabling a cylinder or two FE experiments I've read about, have started with a worn out and/or broken engine. I'd like to see someone test a good running engine on a dyno, then remove two pistons and run it.

Pushrod engine: there is a couple ways of dealing with loose lifters after removing pushrods, pistons, permanently disabling a couple cylinders. If you completely remove the lifter, on many engines, you must close off the lifter oiling port or replace the lifter with a plug of somesort. If you remove the camshaft, and precision grind the lobes off round, you can just leave the lifter in place. Another way to remove pistons without having to m9odify the crankshaft, is to modify the connecting rods. Saw the beam and upper end completely off, grind it smooth, and re-install what is left with bearings. That will control the oil leak from that journal. If you don't want it to turn, just add a little .003 shim stock under the bearing shells of the modified rods.

Rather than cutting holes in pistons, and still having that unwanted friction from the rings, you could just throw away the piston and rod assembly, and add weight to the crank throw. A simple bolt together steel 'split ring' that weighs the same as a piston and con rod assembly would do it.

However, it will still run rough. It's the unbalanced 'power pulses' that will be responsible for that. Lack of a power stroke from the dead cylinders. But if you deactivate 2 cylinders from a 4 cylinder engine, such that a cylinder fires every 360 degrees, it should run reasonably smooth at higher rpms. Idling, it will still vibrate like a tractor.

American car companies had been trying for years to smooth out 4 cylinder engines. Compared to a V8, most customers had historically disliked the 'feel' you experience from fewer pulses per revolution of a 4 cylinder car. Internal engine-smoothing 'balance shafts' were popular for a while, but they are not good for FE or power. Most modern 4 cylinder engines balance shafts have been replaced with these engineered HUGE vibration damper systems, formerly referred to simply as 'motor mounts'. My guess is the engine tuned mass dampers and vibration isolation systems on modern cars are mostly responsible for the smoother feel you get from a modern 4 cylinder car.

Remove 2 power pulses per revolution, and that 'system' might need some rethinking to keep the car from shaking like a Harley Davidson motorcycle :thumbup: