Turbos to replace mufflers

[Pskhaat]

If I remember right, its the duration and lobe separation that kills your MPG,

Correct, it's the valve overlap really, that does it, which is driven by duration and seperation.

Help me out here, for the lay and on an injected/ECU: if you're overlapping too much are you saying your dumping too fuel out and causing a lean situation in the cylinder on natural aspiration? BUT you're breathing better (at least on natural aspiration), so there has to be tradeoff, yes? Just thinking in my head, too much overlap + too much advance = less fuel per given volume of air?

 

[R_Lefebvre]

Scott, I want to answer your question but don't quite know what you're asking. Let me know if this answers your question.

Overlap is when both valves are open simultaneously. Back in the day, this usually meant much of your intake charge could go right out into your exhaust without being burnt. That's why it was bad on fuel and emissions. However, with the advent of massive computing power at the design centers, they've made huge improvements in the way the gasses move around in the combustion chamber. This means they can have more overlap than they used to, without wasting as much gas. Still, you can only do so much. Well.... then we can get into direct injection, where the fuel can be injected directly into the combustion chamber after the exhaust port is closed...

By advance, do you mean spark advance, or cam timing advance?

 

[Pskhaat]

... of your intake charge could go right out into your exhaust without being burnt

Exactly. Which even with (non-direct) injected engines meant that fuel:air ratio was low and would run a touch lean, or at least as I understand the effects of overlap camming, and thus you'd enrichen the mixture (or increase duration of injection) which would lead to even further fuel use?

 

[ntsqd]

Keep in mind that these effects are RPM dependent. 2nd Order harmonics, Column Inertia, all of that stuff that I only partially grasp. So a lot of overlap in cam timing will have an ideal RPM range where it runs exceptionally well. It's just that anywhere out of that range will suffer.

I do not think that the mixture would run lean. Assuming near total homogenation there would be a power and mileage loss, but the A/F mixture remaining in the combustion chamber would be roughly the same ratio as that which went out the exhaust valve.

If the homogenation process was less than successful then it could go slightly lean due to the fuel's greater inertia, but the various bends and turns in the flow path should result more in wetted surfaces than outright lost fuel. Depending on the location of those wetted surfaces the fuel may or may not be lost.
In an old school engine this wouldn't matter for a couple of reasons. The first being that the carb would have been calibrated for roughly a 12:1 A/F ratio for max power, and second this would be happening at an RPM where the engine wasn't happy and therefore hard to fully load w/o drivability issues.
In a newer, electronically controlled engine I can see this potentially being a real headache to program around.

 

[R_Lefebvre]

We're starting to get into the areas where I hesitate to speak with any authority, as it's at the edges of what I know I know.

I do not think that the mixture would run lean. Assuming near total homogenation there would be a power and mileage loss, but the A/F mixture remaining in the combustion chamber would be roughly the same ratio as that which went out the exhaust valve.

That's my understanding at all. Assuming the charge is homogeneous (which may or may not be true), then charge going out the exhaust won't cause the engine to run lean.

If the homogenation process was less than successful then it could go slightly lean due to the fuel's greater inertia, but the various bends and turns in the flow path should result more in wetted surfaces than outright lost fuel.

You're assuming that the fuel injector is well upstream of the intake valve, and that the fuel remains in droplet form. Modern engine design strives to put the injector as close to the intake valve as possible. And the goal of fuel injector design is to make sure the fuel is vaporized, not remaining in droplet form.

However, with older engine designs, cold engines, or bad fuel, it can happen where the fuel sticks in liquid form to the intake port walls. However, it can't really build up to much extent as it would be just drawn out when the air rushes by. If this happened, it would come off the wall in large drop form, rather than mist, and thus more likely enter the combustion chamber in liquid form. That's bad, because there's a good chance it won't get burned completely, and then will go out the exhaust, still unburnt... so yeah, I guess you could "lose" that fuel.

In a newer, electronically controlled engine I can see this potentially being a real headache to program around.

I know my Pectel T2, no longer state of the art but is was about 10 years ago, I have an adjustment called "Injector End Angle". It's a 1D table just broken down into RPM ranges. The idea is, I believe, to have the injector close just before the intake valve. At low loads (and thus low injector on time) the injector on squirts fuel in the last bit of air column entering the combustion chamber. Therefore, any air that went out the exhaust port during valve overlap, would have no fuel in it. However, you can only cut it so close to the closing of the intake valve, so the fuel actually has time to travel with the column into the CC. If the valve closed before the fuel went in, the fuel would sit near the valve, and would likely go in at the start of the next intake event.

At high loads, the injector on-time can approach 80%, and injection end angle is pointless, since the injector is spraying fuel long before the intake valve opens.

You can imagine it would take a lot of dyno time to sort this out. I just left it on the default setting... for better or for worse. This is mostly an emissions thing. I just make up for any lost fuel with more injection time based on my reading from the O2.

12:1 isn't necessarily peak power. Assuming detonation isn't an issue, peak power is typically found at around 13.5:1. You just need a bit more fuel than air to make sure that every molecule of O2 that went into the CC finds some H or C to mate with. You don't want any more than that. However, in the real world, detonation is a problem. You need to balance that perfect chemical composition off against the increasing tendency to detonate. You could run 13.5:1, but would have to retard timing a lot to avoid detonation. Or, you could run rich, 11:1, with lots more timing. The actual peak power will be found somewhere between the two extremes. Again, lots of dyno time.

I just set mine for 13.5:1 at moderate loads (no detonation risk), 12:1 at heavier loads, and 11:1 under full boost. Timing is set to the detonation limit by ear, then backed off a bit. The resultant horsepower... well it is what it is. This is a fairly conservative way to tune.

And don't forget, detonation is most likely to happen at the peak torque point of the motor. You can lean it out a bit, or run more timing below and above. Lots of dyno time.

This is dyno time that a lot of people don't pay for. And if your tuner is aggressive to give you a big power number, without the investment to make sure it's safe... well that's why so many "tuner cars" go Kablouie!

 

[robert]

That all depends on the turbo you select and how everything is set up. I have though of doing a similar set up for my Tundra, with the right turbo you wouldn't be producing much "boost" around town and cruising, therefore not dumping fuel...but when you open it up you will be putting down a considerable amount of power.

Besides turbo lag = poor mans traction control

I'm aware of that, however the site that was posted seems to be showing a turbo for high performance. Maybe great for desert running, not so great for slow speed travel.

I'm also aware that cast parts can be extremely well made, but there's a reason most performance engines use forged parts.

 

[ntsqd]

No point in using a more expensive forging where a casting is appropriate for the loads on the part. Hyper-eutectic cast pistons have replaced forged pistons in a lot of none to low boost applications, they're more than up to the job and they seal better, particularly when cold, than do forged pistons due to their tighter piston to cylinder wall clearances.
It is also much harder to forge complex shapes like a turbine or compressor scroll than it is to cast them.

Castings get a bad name from poorly done sand castings. That is just one of many methods for producing a casting, and done well a casting's grain structure will easily be a superior part.