late model ford head - so what heck is this?????

[DaveMcLain]

It's a crutch to make up for the fact that they are now committed to a 4 valve head , yet the average power and fuel mileage they are shooting for could have been had with a much cheaper to produce 2 valve head!

But what car shopper these days would be interested in only a 2 valve per cylinder motor if it's not a V8 , right?

Uh, what? This doesn’t make any sense. Please tell us how a 2 valve head would have made more average power and fuel mileage?

The comment insinuating engineers are fools for DI is also laughable.

I don't think that they are foolish but I do wonder how a lot of those problems with severe buildup in the intake tract on DI engines ever got through their test programs without being noticed.. How did that happen?

[peejay]

It doesn't happen to all engines. I've pulled apart 90k mile DI engines that were no different than what you'd expect from port injection at that mileage.

That said, what happens in testing doesn't always correlate to the real world. Like the Hemis that eat valvetrains if you let them idle all the time, like most people do. I keep hearing about multi hundred hour WOT tests, including tests where the engines are cooled down to freezing and then run out to max power as soon as oil pressure stabiizes... but how many test idling for a few hundred hours?

[pamotorman]

the new 755 HP corvette uses both port and direct injection to get enough fuel.

[ptuomov]

I found this paper interesting on the topic of tumble:

SAE 2016-01-0692

This paper describes the key requirements for engine design to
generate high tumble flow with high flow coefficient by using both
simulation and experiment. It is important that the air flow of
tumble is straight and runs along the wall of the combustion
chamber near the exhaust valves for high tumble flow efficiency .
As a result, better performance can be achieved by changing the
layout of the valve angle and the intake port. The generated
tumble flow improved by 10%.

What this paper says that the angle of the flow from the intake port needs to be about normal to the exhaust valve face. In effect, the exhaust valve face and the combustion chamber roof around the exhaust valve is an important flow guide / diffuser. A spherical dish also promotes efficient tumble flow. That makes sense to me, whether its true or not.

For most port angles, efficient tumble generation (according to the paper anyway) requires either shallow port angle or large included valve angle. If the compression ratio requirement isn't very high, this sort of requirement can be met efficiently. If, however, the compression ratio requirement is high, then one will need to either have shallow port angles or have a dome on the piston. At that point, additional tumble generation devices may help. An example (b) of figure 5 has a "choked" port design similar to the Ford head photos earlier in the thread -- by my understanding.

As to the carbon buildup, it is my understanding that almost all of it is caused by the crankcase breather pushing oil into the intake manifold. In a port-injected engine, the sprayed fuel washes most of the oil away before it burns onto the port and the valve. In a direct injected engine, you're SOL if any oil makes it to the intake manifold. I'd look into various catch cans, baffling, and air-oil separators to mitigate those problems.

[ptuomov]

It doesn't happen to all engines. I've pulled apart 90k mile DI engines that were no different than what you'd expect from port injection at that mileage.

I think the main determinant is whether the engine injects any oil into the intake from the crankcase breather system. If oil makes it to the intake, you're SOL in a direct-injected engine.

[ptuomov]

Design a nice high flow cylinder head and then block a portion of the bowl to introduce turbulence to improve low end torque. Why don't 4v head designers add some swirl to a design that favors tumble? DV adds swirl to a 4v on his Polyquad patented design with good results. I always thought the ports should be cast so swirl and tumble was produced. Low end would improve and top end would not be changed much if at all.

It is my understanding that swirl is inefficient compared to tumble in a pentroof 4V head. They can get whatever mixture motion very efficiently just by accelerating and/or directing the flow out of the intake port.

[ptuomov]

It's a crutch to make up for the fact that they are now committed to a 4 valve head , yet the average power and fuel mileage they are shooting for could have been had with a much cheaper to produce 2 valve head! But what car shopper these days would be interested in only a 2 valve per cylinder motor if it's not a V8 , right?

A two-valve small displacement car engine wouldn't survive the fuel economy regulations, emissions regulations, and functional requirements by the consumer. Turbocharged three-valve engine is a pretty strong competitor for the four valve head, however.

[turbo2256b]

Last I knew of ford was totally against swirl since the first few years of the 4.6 that were high swirl port designs.
If my memory serves correctly High swirl helps more in top end performance, tumble better for economy, mid range and cruse.
Each engine design for best performance does like a balanced relationship between tumble and swirl the amounts of either can also depend on usage

[4vpc]

It doesn't happen to all engines. I've pulled apart 90k mile DI engines that were no different than what you'd expect from port injection at that mileage.

I think the main determinant is whether the engine injects any oil into the intake from the crankcase breather system. If oil makes it to the intake, you're SOL in a direct-injected engine.

SOL?

The coking up of ports is from the EGR.

[ptuomov]

It doesn't happen to all engines. I've pulled apart 90k mile DI engines that were no different than what you'd expect from port injection at that mileage.

I think the main determinant is whether the engine injects any oil into the intake from the crankcase breather system. If oil makes it to the intake, you're SOL in a direct-injected engine.

SOL?

The coking up of ports is from the EGR.

$hit out of luck.

A friend of mine has been looking at those direct infected used engines and is convinced that the coking of the valves and intake ports is caused by some engines puking oil thru the breather into the intake. Now might be correct or might not be.

[hoffman900]

There are lots on swirl and tumble in the Honda R&D library. Free to sign up to read.

[MadBill]

Regardless of the cause, the accepted cure appears to be adding port injection to the mix.

[hoffman900]

Regardless of the cause, the accepted cure appears to be adding port injection to the mix.

That's for higher rpm running, though my DI Hyundai daily does just find getting to 6500rpm. Granted, it's not like each cylinder displaces a lot either.

[peejay]

I think the main determinant is whether the engine injects any oil into the intake from the crankcase breather system. If oil makes it to the intake, you're SOL in a direct-injected engine.

SOL?

The coking up of ports is from the EGR.

$hit out of luck.

A friend of mine has been looking at those direct infected used engines and is convinced that the coking of the valves and intake ports is caused by some engines puking oil thru the breather into the intake. Now might be correct or might not be.

Well, there's a hundred-odd years of direct injected Diesel engines, many of which had functional PCV systems, that didn't have problems.

EGR I can believe. One emissions problem that DI gasoline engines have is particulates. If that backfeeds up the intake, that's definitely going to cause problems. (EGR is often/usually done "passively" with variable cam timing rather than an actual valve.) This is the main reason why we're seeing engines that are hybrid direct and port injected. The port injection helps cover the operating regions where DI doesn't work as well emissions-wise.

[ptuomov]

Most diesel engines don’t turn the types of rpms that make oil shoot out from the crankcase breathers. Many gasoline engines do.