Just wondered if anyone had considered trying direct injection on an engine using the oil injection ports to house fuel injectors. Not as good as the curtis wright setup but far more doable.

Bill

 

I was thinking about that exact thing for some time. I cant wait to actually try it out.

 

Seems like that would be a good position for the primary charge, but the 2ndary charge should be injected in the compression "stroke" if you are looking to take advantage of direct injection benifits.

-edit- I think...

 

Not quite true, as there is considerable opportunity to inject after the inlet has closed. You're right that its not an ideal location but the potential gains certainly make it worth trying.

Bill

 

The oil injectors are in a horrible place for use in direct injection. The idea behind direct injection is to get a "stratified" charge; you want a pocket of rich mixture surrounding the spark plug, and then all other areas extending out from there can be super lean, so that the overall mixture if it were evenly distributed would be very lean, hence better fuel economy. So the best place for the injector would be as close to the spark plug as possible.

Additionally, direct injection is designed more for economy than performance. The power gains would not justify the expense of the added hardware and not to mention that you'd have to figure out some kind of ECU to control it all.

 

Audi has been using direct port Injection on the R8 for 3-4 years and they have been DOMINATING. In the next few years more and more car will have them, its just that the injectors cost about ten times more. Direct injection rotary would be incredible, next to no detonation, simply watch the EGT's.

 

Why not use the injectors from a Diesel? you can often pick up a knackered one which still has a good engine.

You could use all the bits from the diesel fuelling, but it would have to be from a V6 to get the correct timing. You could drill a hole to the side of the spark plugs to fit the injectors, and then do some modifications to the injector wiring so that they get the correct signals (each injector will get the signals for 3 injectors for the sake of timing).

A (relativily) cheap way to get direct injection!!

 

This is a different form of direct injection. Due to the design of a rotary, you can inject the gas directly into the chamber during the intake phase, and not worry about compression acting on the injectors, because that prt of the engine never sees compression. The premise behind this would be different that for high pressure direct injection. First, you could mirror polish the ports, as well as the whole intake tract, and not worry about fuel clinging to the walls and throwing off the a/f ratio. Polished ports flow more, but it causes problems when fuel is flowing with the air. So sacrifices in flow have to be made to keep the fuel from dropping out of the air, and clinging to the walls. Hence the rough texture of manifolds. Precise metering would be easier here, because no fuel would be lost due to dropping out of the air.

High pressure direct injection has its advantages, but cost is NOT one of them. And do you really want 1500 PSI worth of fuel running through out engine bay, with the history of engine fires these cars really have? I would not trust a pieced together system for anything. Theres too much to lose, and too little to gain. Also, it takes precise control. Youre not going to get that unless you have the budget Audi has...

 

It would seem there is some confusion over the various forms of direct injection.

Stratified charge has been done on rotaries, mainly coming from the ingersol-Rand/Curtis Wright developments, where there is an injector pointing at the spark plugs. Can't recall the URL but if you search on omnivorous rotary you should fine some pictures of a SCORE engine.

I am more interested in the fact that, by injecting into the chamber, you have the possibility to increase the VE of the engine. Back in the 1940s it was realised that you lose up to 10% air charge due to the presence of fuel. So there is more power to be had. Whether the gains outweigh the work required remains to be seen.

Now practicalities will limit this, but there is a gain to be had. Its also a handy place to inject water.

 

RPI lost their website, and it looks like they are about to lose their shirts next.
http://finance.yahoo.com/q?s=RPIN.PK

 

bill: I tried to take a picture of a housing that I did this to but my 2nd generation digital camera sucks hard. I never built an engine like this but the thought did cross my mind for the same reasons that you brought up. My idea was to use a staged fuel injection system where these ran at low rpms and then a seperate set kicked in later at higher revs. The oil metering injector holes have enough metal around them to easily drill out for a fuel injector to fit. It would be an extremely easy thing to do. Rotary Performance sells weld on fuel injector bungs that completely enclose and mount the injector. Just weld this to the housing and it would be done easily. On my little drill experiment the injector tip sits way back in there about a half an inch or so. With some work I'm sure I could bring it in much closer. Anyways I'd love to hear if someone has actually done it and made it work. It would suck to go to all the trouble and have it fail. Oh well, I think the idea is sound. I'd use a fairly small injector to keep fuel droplets smaller and rely on bigger ones farther up the manifold.

 

i disagree i think donkey kong was the best game ever.

 

 

errr....

YES!!

 

Not quite true, as there is considerable opportunity to inject after the inlet has closed. You're right that its not an ideal location but the potential gains certainly make it worth trying.

Perhaps you misunderstood what I said since you agree and disagree w/ me in the same sentence above...

It is my understanding that direct injection configurations work best in their "power" mode by injecting a small amount at first as the intake charge is ingested into the engine to cool the hot turbo temps and then later late in the compression stage to get best dispersion w/ out whetting on engine parts AND get a stratified charge for easy ignition.

On a rotary this requires two seperate injector locations per rotor housing.

 

one in the inlet, and one near the spark plugs

 

Not quite. The injector near the plugs is to allow stratified operation, but won't effect max power. For power its more to do with getting the fuel into the chamber without taking up space in the inlet. In theory this alone can get about 10% more air into the chamber.

In a turbo it is more complex as you want some air to cool the charge as, up to a point the increased density from the charge cooling outweighs any loss of air charge due to the presence of fuel. Current theory suggests that this cooling fuel is best injected higher up the inlet, although I would like to test the theory sometimes. Suspect I'll never get around to it.

Bill

 

The longer the fuel is in the engine the more fuel is going to condense on the engine internals- this is going to be more pronounced in a rotary where the rotor is moving the air/fuel physically and causing more turbulence and more engine internal surface area for the air/fuel to contact.

Therefore, you have to inject more fuel to get the desired amount of fuel at the moment of ignition. I think you can easily see how this pertains to this statement by you.

Back in the 1940s it was realised that you lose up to 10% air charge due to the presence of fuel.

I believe your cooling "pre-charge" should be small enough that the atomised fuel is fully vaporized by the heat of the intake. This cools the intake temps and reduces the volume of the air (allowing more in) w/ out adding any volume of fuel.

Likewise, by injecting the main charge once the mixture is compressed there is instant fuel vaporization from the heat of compression so there will be no condensation (not to mention no loss of air intake volume as in earlier injection schemes).

So, injecting a primary cooling charge small enough that the fuel is fully vaporized (injection % based on intake temps) and then a 2ndary charge after the intake has been compressed means there will be no loss of air charge volume due to displacement by fuel.

Also, the stratified charge is easier to ignite (more power). You know how ignition systems add power to our cars so well due to our terrible combustion chamber dynamics.

Stratified charge will also have a more quick and complete burn (more power). If you want a longer burn, then inject more fuel now you have a quick complete burn of even more fuel (more power).

Please point out any discrepencies between our thoughts, as you still seem to agree to all my points while disagreeing that staged direct injection pattern will make more power than single early direct injection pattern.

I am confused

 

Stratified charge engines are for economy NOT power. That is my point. You also need to read the definition of stratified, as it is very hard to ignite, which is why you squirt a cloud of fuel around the plug to provide the initial kernal.

You also want the fuel to vapourise DURING compression as you then decrease the work required during the compression phase.

I am only interested in the effects that allow an increase in volumetric efficiency rather than improving BSFC. Most DI research focusses on improving the lean limit.

Bill

 

And the last reason staged direct injection will make more power is it allows for a substancial increase in compression ratio at the same knock resistance.

By injecting the second fuel charge very late in the compression stroke it cools the air/fuel charge significantly just as running lower compression would.

12.0 as opposed to 10.5 in the production Mitsubishi GDI 4G93 vs the MPI 4G93.

Edit- BTW you should check out Mistubishi's GDI engines design as they have both an economy mode and power mode that switches over like V-tech for injectors...

 

I am well aware of how the GDI works and the different modes. It would appear that you are getting some of the basics of the thermodynamics of the fuel/air cycle confused. And in fact the GDI has 4 operating modes not 2. Perhaps you need to take another look at the design?

This late injection provides no more cooling than early injection AND you have to have put all of the energy in to compress the less dense charge, so increasing pumping losses.

There are other reasons why GDI uses a late injection which are of little or no interest to this application ,not least because I am not interested in injecting at 2000PSI.

Bill

 

What are the 4 modes Mitsubishi GDI uses?

I have read that it uses 2 modes. "Ultra-lean combustion mode" in light throttle application and "superior output mode" when under load.

This late injection provides no more cooling than early injection AND you have to have put all of the energy in to compress the less dense charge, so increasing pumping losses.

The 2 stage Direct Injection (DI) should provide same cooling in the intake stroke since there is only so much heat energy in the intake air to vaporize the atomised fuel (or you need a more efficient intercooler).

As I stated earlier, idealy in 2 stage DI you would 1st inject just enough fuel to fully vaporize in the intake stroke thus cooling the intake charge. Any more fuel injected and it will remain atomised during the intake stroke and well into the compression stroke- so displacing potential AIR intake (as you yourself pointed out earlier in the thread- up to 10% displacement in a conventional piston engine)

Then, the 2 stage DI injects the second pulse towards the end of the compression stroke. Since the intake charge is greatly heated when compressed the fuel vaporizes quickly cooling the charge. This enables a higher compression ratio (or more boost pressure at current CR) to be used since the detonation threshold is greatly retarded.




So, what I am understanding you are saying is that you believe maximum power is made when you use DI soley on the intake stroke.

This is simply not the case because to get the same cooling affect of the 2 stage DI you must inject

1) enough fuel to fully vaporize and cool the intake charge (just as w/ 1st stage of 2 stage DI)

2) then enough fuel that by the time the intake charge is carried over to the other side of the engine for compression there is still enough fuel atomised to provide the same compression charge cooling as the 2nd stage of 2 stage DI.

This much fuel can certainly be injected in the intake stroke, BUT since we know

1) rotaries run more rich than piston engine (especially boosted) at the same detonation resistance- so more fuel will need to be injected.

2) the atomised fuel will be physically transported accross a much greater surface area of hot rotor housing and side housing and rotor face as compared to piston engine and so much more atomised fuel will be vaporized cooling these hot engine parts instead of staying atomised to later cool the compression charge- so more fuel will need to be injected.

Therefore it can be seen that w/ a rotory well over the afore mentioned piston engine 10% of the intake charge could be displaced by fuel if it is injected in the intake stroke. Since liquid (atomised fuel) cannot be compressed whereas the intake air charge can be- this xx% represents an even greater loss of intake air charge potential in a (properly intercooled) forced induction rotary.

You stated-
I am more interested in the fact that, by injecting into the chamber, you have the possibility to increase the VE of the engine. Back in the 1940s it was realised that you lose up to 10% air charge due to the presence of fuel. So there is more power to be had. Whether the gains outweigh the work required remains to be seen.

Then you contradict yourself by stating the most power is made when you inject all the fuel in the intake stroke... I don't understand.

There are other reasons why GDI uses a late injection which are of little or no interest to this application ,not least because I am not interested in injecting at 2000PSI.

2000psi may be typical of diesel direct injection, but the Mitsubishi gasoline direct injection is *only* ~750psi I believe. Still, I can understand why that high pressure would impede a do it yourself DI set-up.

 

There are 4 modes of operation. 2 are considered transient 2 are considered steady state.

You are confusing WHEN you inject fuel vs WHERE you inject fuel. you can do DI during the intake stroke. You can also do it after inlet has closed and compression has started. DI means you inject into the housing not that you inject during the compression stroke.

You need to go away and think a little more about the underlying thermodynamics of this. Then I will answer your questions

 

Not to be a *****, but it's VTEC. V-tech is a freaking phone maker...

 

Bill: I found this picture of the Racing Beat 3rd gen Bonneville engine. Look closely, direct injection!!!

http://cp_www.tripod.com/rotary/images/pg22_07b.jpg