DavidBoren

@ gtfoxy, thank you for your contribution to this discussion. You have given me a lot to ponder and research.

I do know about the piggyback injector controllers that used to be popular. But all the systems seemed to run cruising injectors and hooning injectors. I'm talking about a system firing all 16 injectors, all the time.

Think of the fast 102 intake manifold. It is generally regarded as a good intake, and LSx motors seem to do well with the length of runners on the fast.

Now imagine you were using this fast 102 on your engine, with something like 50# injectors.

Okay, now cut all of your 50# injectors in half, so you have sixteen 25# injectors. And now mount a 25# injector in the original injector port, and install another 25# injector at the half way point of the intake runner.

You still have the same amount of fuel being sprayed for each combustion event, it's just broken up between two injectors to try promote vaporization/homogenization.

To my knowledge, very few, if any, of those older systems were designed to do what I'm talking about. I'm sure they could be programmed to do so, but I cannot recall anyone running the system I am referring to.

But I can't be the first person to think of this, and there's probably a reason it's not in use.

gtfoxy

Motec makes what you are looking for to run 16 injectors. Myron introduced me to those back in the mid-90's.

kingtal0n

iTrader: (1)

What I am getting at is this.

At room temperature, I have 2uL of water in a tube. It being finely constrained to a tube, does not evaporate. Rather, the interaction with the walls of its container and interactions to itself as water is wont to do holds it down, keeps it from turning to a gas. It will eventually but very slowly. Water molecules also leave ice cubes in freezes- ever notice they shrink? Stuff gets frostbitten because water molecules escape the surfaces even of solid ice or frozen "puddles" of water. This is a great example because it shows properties of liquids like water that have great attraction to itself is still able to easily escape and diffuse as a gas.

Now consider If I were to have 2uL of water sitting on the table in front of me, that it would evaporate very quickly. Add some temperature and whip the air up around it- and it will evaporate even faster. And still this is water, with very polar self-adhering properties.

Now instead, try it with gasoline. It will be gone before you blink, turned gaseous and finding the corners of the walls in your house, several feet away from the table, diffusing through the air rapidly.

Injectors open, a tiny quantity of fuel comes out such as 2uL or 5uL, it is going to evaporate and rush away from it's point of entry on a millisecond time scale or better. Furthermore, the advantage of keeping it in a collected puddle on the intake valve is that less molecules are allowed to escape, and this would be especially important in a short runner intake where a gasoline molecule could travel a "long distance" (1 foot) in a matter of milliseconds (the time frame between firing an injector at idle may give it a chance)

So I dont care how we get 5uL of gasoline to the tip of an injector. let me drip it there with an eye dropper. Given the blink of an eye, it will be gone, evaporated, gaseous. It might not have started that way (out of the injector no doubt, is a liquid state) but it does NOT take very long to send it up completely, when the quantity is so tiny.

Larger volumes 80uL+ will not evaporate as quickly. I am sure the manufacturer would prefer that it sit and wait at the valve, as close to the cylinder as possible, where it can get inside completely as quickly as possible (the whole "puddle") and begin diffusing inside the cylinder to the ends of that container (just like it seeks the corners at the ends of your walls in your house when it evaporates there)

DavidBoren

Manufacturers DO want the whole puddle right there waiting... So much so that now they have a puddle pocket on the piston and inject the fuel directly into the combustion chamber, for better or for worse, it eliminates the whole argument of where the fuel goes after the injectors. It also eliminates the question of which physical state of matter the fuel is in when it enters the chamber.

I'm not saying it's the best way, but the industry is heading that direction... And it does, indeed, eliminate some of the variables pertaining to fuel vaporization/homogenization.

kingtal0n

iTrader: (1)

SO then you answered your own question about multiple injectors, spray quality, vaporization, etc... i.e. it clearly does NOT matter if an injector sprays a sloppy messy puddle all over the walls and intake valve as this is what they seem to go for. A fine mist leaving an injector would be carried away, and take very little temperature from the valve, etc...

In the past (2003) I've raised fuel pressure in the hopes of finding some measure of fuel economy due to the same idea "better atomization" I tried adjustments from 45psi to 85psi of fuel pressure, during cruise situations, with OEM 370cc/min and 440cc injectors and found no difference at the pump (measured accurately, filled up,mileages , re-fill up, divide) If anything, engine vaccum went down, as driving the pump harder drew more current, the fuel is getting hot faster, and cheap fuel pumps like a walbro do not enjoy long duration sustained high pressures.

Darth_V8r

iTrader: (4)

So if I'm understanding you correctly, the real benefit to fuel injection is not the atomization, but the very precise metering of fuel?

Then, the only real benefit to direct injection would be that you can time the injection to always occur after EVC, meaning all the overlap in the world wouldn't matter?

joecar

That's my understanding, EFI allows very precise metering of fuel under all conitions, along with a whole host of other supporting functions.

DI means you can time fuel anywhere, regardless and exclusive of overlap (which you will need anyway in order to be able to pack in air at the higher rpm's, the point being that overlap still matters, and possibly even more so now that it can be extended without pushing fuel right thru); the very high pressure (relative to compression pressure) that DI uses allows the fuel charge to be completely delivered in a much shorter time compared to port injection, i.e. in the time between EVC (and continuing on thru IVC) and spark ignition or shorter; and of course DI allows any additional fueling for other purposes (comustion chamber protection cooling mode) regardless of valve and spark events.

joecar

My point was that with DI you can now have as much overlap as you want/need and not lose any fuel.

Darth_V8r

iTrader: (4)

That's what I tried to ask, thank you. So, also, theoretically, one could run a very large injector for a fuel "burst" into the cylinder after EVC to get max fueling even at low duty cycles?

I know it's a bit off topic, but its fascinating.

Darth_V8r

iTrader: (4)

I have another question, going back to some prior posts regarding timing:

In a perfectly ideal engine, one would fire the spark at 13 ATDC, and it would burn completely instantaneously, generating max pressure in an infinitesimally small time. In reality, the burn takes some measure of time, so the spark is fired earlier to compensate. So, here is the crux of the question I'm trying to formulate:

In a higher compression engine, the pumping losses are higher for the actual compressing the charge portion of the pump, but the ignition timing is actually retarded in response to higher compression charges. Earlier ignition timing leads to earlier fuel burn, which must be fought by the piston on its last bit of travel BTDC.

Do you think these losses cancel each other out, so that a higher compression engine with a later spark and a lower compression engine with an earlier spark experience the same pumping losses, or do you think that the higher compression engine still overall has the higher losses?

Darth_V8r

iTrader: (4)

The attached file is Overlap time vs RPM. I should have converted to ms, but it is in seconds for now.

The equation derived to a perfect:

Overlap Time (s) = Overlap degrees / 6 / RPM, with an R-squared of 1.000000.

Seeing it like this, I think it is easy to see why the BSFC is so horribly high at idle and off idle RPM and then comes into normalcy around 2,000 RPM. BSFC hits it's best mark at 3200, which is not peak TQ, but it is what it is. This is the point at which OL drops to 4-ms.

Doubling the overlap, takes the 4-ms point out to 6400 RPM. Perhaps this correlates with the general finding we all see that increased overlap increases top end power and helps carry power post the peak longer. I know that has to do with scavenging and filling of the cylinder, but there comes a point in the RPM band at which the OL time is so short that there is not enough time to short circuit fuel, but long enough to fully evacuate the cylinder. Delaying this sweet spot would certainly seem like the entire power curve would shift to the right as a result.

gtfoxy

I don't know how realistic 13*ATDC is, it would have to be moving north of 25,000rpm to do that, bu in a really efficient chamber, with the right stuff going on, 0-3* ATDC is possible. Figure your overlaps at that rpm... See where it takes you. If your doing that you have to remember residual temps are going to drop like a rock because you're using the energy. Doesn't need all the overlap. It goes hand in hand & RPM is the key.

While you're at it run some torque figures at that kind of RPM. See how easy it is to make power.

Overlap is definitely part of the equation, but it is dependent on a given lobe timing at a given duration split. What is really required here depends on the residual energy after the power stroke & how much you need to keep, on a given exhaust flow capability

4-stroke, 2-stroke, it doesn't matter. Valve curtain area is valve curtain area.

VE is what you're after if other stuff is lacking. It's a trade-off game.

Fullblast

iTrader: (17)

I did a bit of research, and a chevy performance website said the main parts that go out on a stock rebuild are the wrist pins of factory pistons. Is this a common occurrence or something I should worry about after over 400.

They also said something about replacing the plastic oil connector tube.

My mechanic said the stock rods are good for 600:-P +.

I'm just wondering if I need to upgrade to aftermarket pistons. I really don't want to if it's not necessary.

joecar

!...?

Do you want your own thread for that instead of this one...?!

Darth_V8r

iTrader: (4)

But, is 25000 even realistic for s reciprocating engine? Especially a V8 with a non-planar crankshaft? I mean, that is jet engine territory.

joekneale

Super good discussion here - this should be stickied.

joecar

Possibly yes, but only if it's size is in the range of a matchbox...

so no it's not practical.

gtfoxy

Not quite that small. It has been done, to 21,000rpm, as I said at 21:1 compression on 91RON, in an 80cc 2-stroke. It made 50HP, BTW.

Now imagine 8 of those comprising a 640cc engine cranking out 400HP. Or a 16 ciylinder. At 1280cc's making 800HP, NA, on 93 pump... My head swoons at the thought. One day I will do something along those lines, but in the aviation field. Keep in mind, planted in reality, I'm just a nobody. Just like the Wright-brothers were nobody's.

What is import is the dynamics used & learning from them. Then working within the limitations of what you are playing with. Being able to gain points in CR will pay big dividends if you can get the limitations of doing so reduced.

The compression limits shown here aren't the highest possible, but at the same time not totally attainable in an LS. But still being able to increase CR by 40-50% could be. Regardless of whether or not it would spin even 15K isn't the point. It's how much more torque, on less fuel, it would make. Even if limited to say 8500RPM. That is my minimal goal.

joecar

But at 21K rpm how much torque is it making...

Darth_V8r

iTrader: (4)

Ok, but two strokin it, the valves are stationary. I could feasibly put the valves very near TDC so static compression is 21, and dynamic compression is still only 8. And it's a very light engine at 80cc, so pumping and acceleration and friction losses are much less.

Cut the stroke in half and acceleration is cut in quarter. That's one reason If I was to build a high revving engine I'd go wide bore, short stroke. Think a 6.2 with a 4.8 crank. Thing would take boost like a ****.

As far as aviation goes, the air would be very thin, so not sure you'd get the air in and out without a turbo at those speeds.