Cylinder Deactivation

[onegammyleg]

Originally Posted by theclencher

You better share your findings with GM, DaimlerChrysler, and Honda- they're doing it all wrong!

I stand by my last post.

From what I have read , auto manufacturers deactivate the cylinder after the firing stroke and use the exhaust gas in the chamber as an easily compressable spring.
This works fine , for a while.
In time the exhaust gas leaks out (valves and ring leakage) and it starts to compress air , then the advantage is gone., and now you start loading up the other cylinders.

My answer was in reference to Bruce's question ?What if you were to merely block off the ports on the intake manifold by sandwiching a steel plate between the intake manifold and head rather than disabling the valves??

As Bruce would be vac'ing + compressing air immediately it wouldn't work.
My suggestion to remove the plug to allow air to vent in and out easily would provide an improvement by not trying to compress the air.

But as far as existing cylinder deactivation is concerned , with only a 5 to 10 % increase in FE it is hardly enough gain to warrant such a complicated system.

[omgwtfbyobbq]

Air movement does incur a pumping loss, but the biggest contributor, what people talk about wrt gassers, is the pressure difference between the crank case and the closed cylinder as it moves down. At low load there is less air in each cylinder, which means the delta pressure between the crank case and cylinder at the bottom of it's stroke is significant, with the crank slowing significantly before the air/fuel is ignited because of this.

Gammy is right, although I don't think the exhaust and intake valves need to be blocked off. Just remove the plugs from the cylinders you don't want to run and make sure there is no fuel getting to them. Actually, leaving the exhaust/intake valves functional would be an advantage, since this lets in more air initially, so as the piston moves down there's already more air in, so less has to come through the spark plug hole which may reduces any losses from the smaller (but not static) pressure difference between cylinder/crank case while the piston is expanding, not to mention that more air will get in because the other two pistons need more air/fuel to do the same amount of work as four did during the same time span. All in all, I think you could easily see a 20% increase in FE if this worked. Do eeet!

Something else I just though of, get a normal gasoline PFI engine from the junkyard, and set it up for TBI/SAFI (megasquirt or whatever). This way, you can put a solenoid on each of the fuel injection holes that will open whenever you want. At a cruise you can fire one cylinder per rotation sequentially while opening the other cylinder's solenoids/cutting fuel to them. I'm not sure if this would result in some funky vibrations, but since you can fire a different cylinder/combo of cylinders each engine cycle, maybe you could work something out for best fuel efficiency/least vibrations.

Anyway, as for your first experiment, cutting fuel to two cylinders, you should see some increase in mileage. Removing the injectors from those holes should net a larger increase in efficiency. If you had a nice long stretch of road next to a gas station where you could go at some low speed in top gear this would be a great experiment, checking the increase in efficiency after fuel deactivation and then after pulling the spark plugs.

[omgwtfbyobbq]

Originally Posted by theclencher

Intuitively that seems all wrong.

Think about it this way. Diesel engines are already BTU corrected by the EPA for mpg figures. And in the same car with the same transmission diesels get ~30-40% better efficiency, partially because compression ignition is a little more efficient, but mostly because the throttle is always wide open. In fact, on most there isn't a throttle. Diesels always suck in enough air/exhaust gasses to fill up the cylinder totally, so when the cylinder is at the bottom of it's stroke, the pressure in it is about the same as the crank case pressure. But in gassers, there needs to be a correct air/fuel ratio, so there can only be so much air drawn in, which is metered by the throttle, cam duration, and valve size. Since most of the time, the car needs much less than full power, there isn't a full amount of air drawn into the cylinder, and when the cylinder is at the bottom of the stroke, the pressure in it is much less than the pressure in the crankcase. The entire time it was traveling down, if the pressure is greater in the crankcase, then this is effectively resisting the cranks forward momentum, and reducing efficiency.

Going to the variable displacement idea, if the engine only needs 20hp at 65mph, by only using say, four out of eight, the throttle can be opened wider since those active cylinders need twice as much air to burn twice as much fuel (or remain closed and act as springs if vibrations are a problem), and since the other four aren't burning any fuel, they'll suck in twice as much air too, w/o any probs, and the difference in pressure between the cylinder/crankcase will be much smaller with more air in all the cylinders. I think the biggest hurdle was making the transition and operation with fewer cylinders smooth/reliable.

One more example is the Atkinson cycle. Toyota keeps the intake valves open longer so that a larger portion of air can be pulled in, then closes them once the cylinder has moved up and pushed enough air out for the air/fuel ratio to be correct. By letting more air in the pressure difference between the cylinder/crank case is smaller, so less negative work is done.

Originally Posted by Wikipedia

The Atkinson cycle may also refer to a four stroke piston engine in which the intake valve is held open longer than normal to allow a reverse flow of intake air into the intake manifold.

Originally Posted by theclencher

Your engines must run on different principles than mine! The intake valve would NOT be open longer as it is controlled by the cam- on non-variable cam timing engines there would be no change. I don't think this pumping loss thing is registering yet... or the "air spring" effect of a closed-off cylinder either. Why don't you pull your plugs, un-plug your injectors, test your FE and get back to us?

That's a typo, I meant the intake valve would let in more air because the throttle is open wider. To allow the two operating cylinders to make double the power you need double the air so you could go the same speed compared to all four firing at that speed. Bleh, that's a mouthful. I'd love to screw around with this, but I only have a carb'd pickup and a diesel that are mine to play with.

Originally Posted by theclencher

I'm thinking if I removed two of my injectors the vac leak would screw up the "functioning" cylinders- it probably wouldn't even start!

Huh? I meant that as an experiment with an engine retrofitted with throttle body injection. You are using MPFI, one injector per cylinder, correct? The point of my brainstorm was to have solenoids in the injector holes and fuel/air coming from the TB to allow for selective cylinder deactivation in a certain pattern. All you should need to do is disable two of the injectors so they don't fire, which should be the largest increase in efficiency since you'll be cutting pumping losses in half. And by removing the spark plugs from the two cylinders where fuel is deactivated, this allows the pressure in the cylinder to always ~equal the pressure in the crank case, even if you're at a speed where the other two cylinders aren't drawing as much air in as they could. The first step should increase efficiency the most, and removing the spark plugs should help a bit more.

[onegammyleg]

Originally Posted by theclencher

Exhaust gas or air in the cylinder = no difference.

The claimed FE increases range from 8 - 25% depending on conditions.

Wrong.


http://www.fueleconomy.gov/feg/tech_engine_more.shtml
Potential Efficiency Improvement: 7.5%

http://ezinearticles.com/?Cylinder-D...ing?&id=226193
Although typical gains range in the neighborhood of just 5-7%,

http://www.worldcarfans.com/news.cfm...r-deactivation
Preliminary testing of the 2007 Chevy Impala equipped with the 3.9L V-6 with AFM indicates an estimated 20 mpg in the city and 29 mpg on the highway ? improvements of approximately 5.5 percent and 7.5 percent, respectively.

http://www.edmunds.com/advice/specia...1/article.html
On the regular EPA mileage test, DoD delivers fuel economy improvements of 6-8 percent

http://en.wikipedia.org/wiki/Active_Fuel_Management
EPA tests show a 6% to 8% improvement in fuel economy

http://www.auto-report.net/j30avcm.html
While such cylinder deactivation strategies are expected to improve fuel consumption by around 5%

http://autospeed.drive.com.au/cms/A_2618/article.html
By keeping the intake and exhaust valves closed, it creates an ?air spring? in the combustion chamber ? the trapped exhaust gasses (kept from the previous charge burn) are compressed during the piston?s upstroke and push down on the piston during its downstroke. The compression and decompression of the trapped exhaust gasses have an equalising effect ? overall, there is virtually no extra load on the engine.

http://en.wikipedia.org/wiki/Active_Fuel_Management
In order to deactivate a cylinder, the exhaust valve is prevented from opening after the power stroke and the exhaust gas charge is retained in the cylinder and compressed during the exhaust stroke. Following the exhaust stroke, the intake valve is prevented from opening. The exhaust gas in the cylinder is expanded and compressed over and over again and acts like a gas spring.

http://www.carterdodgechrysler.com/n...dge-magnum.htm
Deactivation occurs during the compression stroke of each cylinder, after air and fuel enter the cylinder. Ignition then occurs, but the combustion products remain trapped in the cylinder under high pressure, because the valves no longer open. No air enters or leaves. During subsequent piston strokes, this high-pressure gas is repeatedly compressed and expanded like an air spring,

[omgwtfbyobbq]

I think theclencher's take on mileage increase is accurate. The EPA tests have the car bouncing around in terms of speed, so who knows how often DoD is actually activated. I'm guessing that a four cylinder with ghetto DoD could see as much as a 20% increase cruising at 45-55mph because it'll always be on unlike commercialized systems by GM, etc... that are designed for the overall driving experience.

From one of the articles Gammy posted

EPA tests show a 6% to 8% improvement in fuel economy, but real-world highway use promises even larger gains.

[onegammyleg]

Originally Posted by omgwtfbyobbq

The EPA tests have the car bouncing around in terms of speed, so who knows how often DoD is actually activated.

The highest figure that I saw was 20% improvement and that was on an suv with a huge engine at steady highway speads under ideal conditions.
It dragged it up to 22mpg , which is still S&/T FE.

I never saw any 20% improvments when talking about small 8's 6's ,,and 4s.

For a budget deactivated gheto car why not take out the piston and rod , block the ports and add ballance weights to the crank pins.
Then you have real cylinder deactivation with far higher gains.

[omgwtfbyobbq]

Because it's easier to just kill two injectors, have two cylinders firing that require twice as much air from the throttle, with the two dead cylinders getting twice as much air as well... Compared to pulling apart the engine to have what would a bit more efficient (no friction from movement and no pumping losses from pulling the air in/pushing it out, even though these pumping losses are small compared to the usual pressure difference imo). I think killing two injectors and removing those spark plugs will see a far bigger increase over stock compared to the increase from putting weights on the crank pins/blocking it off over killing two injectors and pulling the plugs.

Granted, how much pumping losses are reduced depends on gearing, cylinder size, etc... But if large SUV can see a 20% increase at a steady cruise, I don't see why two cylinder operation for a large four cylinder car wouldn't result in similar economy gains at a steady speed. And, unlike the usual DoD systems, this one can be activated/deactivated at will by putting switches on select injectors. I think more than a 20% increase is possible because the FE minded driver can always have the system on.

We need someone with a scanguage or supermid to figger this'un out!

[onegammyleg]

Originally Posted by omgwtfbyobbq

I think killing two injectors and removing those spark plugs will see a far bigger increase over stock compared to the increase from putting weights on the crank pins/blocking it off over killing two injectors and pulling the plugs.

Your dreaming. !
Removing the mechanical parts and rebalancing is the ulitimate solution if you wish to take that route.
You have no frictional or pumping losses whats so ever.
How can simply removing the plug off an injector equal that . SHEESH.

[thisisntjared]

ogl: you are not alone in the issue, but you have covered all points i would try to make haha.

[omgwtfbyobbq]

Originally Posted by onegammyleg

Your dreaming. !
Removing the mechanical parts and rebalancing is the ulitimate solution if you wish to take that route.
You have no frictional or pumping losses whats so ever.
How can simply removing the plug off an injector equal that . SHEESH.

I never said it will equal that. But it will be the greatest increase w/o pulling apart the engine. By cutting fuel to two cylinders we cut pumping losses in half, which should have the biggest impact. We can then pull those two cylinders/balance, and remove friction losses, which are small at the usual rpm a fuel economy minded driver travels at, and pumping losses, which may still be larger than friction even after they were originally cut in half. But, by pulling the spark plug, we would remove almost the same amount of pumping losses versus blocking off the cylinder, since pushing air in and out isn't what causes inefficient low load operation in gassers, lack of manifold/cylinder pressure is. So, for the record, I'd go with cutting fuel to two cylinders and that's it. But I think pulling the plugs would be a nice compromise since they would cut pumping losses almost in half again, w/o pulling the engine apart. I mean, how much do you think we'd gain from halving the friction at 2k rpm? I'm guessing not nearly as much as almost quartering pumping losses.

This engine for example, is probably around the 500g/kwh area at 55mph in top gear.

By running off of half the cylinders, we double the torque made by the other half, and double the air flow to all, probably going from the 500g/kwh to the 300g/kwh area. Now, we can also pull out the cylinders that aren't firing to cut the remaining friction and pumping losses in half, but if the minimal BSFC is ~240g/kwh (actually less since we'd only have two cylinders), how much more of an increase in efficiency are we gonna see going from ~300g/kwh?