Turbo Setups? (POLL: Which is best?)

[Bryan8412]

Okay I'm confused on the concept of different turbo setups, so please if someone could have patience and help clear my ignorance I'd be grateful:

Okay, I obviously know a single turbo is a unit of forced air induction which compresses air which allows for more gas to be mixed with the air in the chamber (even more with an intercooler). But, different setups come into play due to turbo lag, because the turbo gets its power from the wasted exhaust energy and It takes a little while for the fan to begin spinning correct? But, when someone says "Twin-Turbo" is it an implied Sequential Twin Turbo? Or is there both Twin-Turbos, and Sequential Twin-Turbos? I ask because I thought, or assumed, the concept of a sequential twin-turbo would be better, but as I researched how they worked, I constantly came across pages with instructions on how to REMOVE this system from vehicles which instill this technology stock! I understand the concept: A much smaller turbo, probably with a ceramic fan and other attributes to give it low-end and quick turbo is used for the low RPMS, and when that fan obtains dangerous RPMS, instead of going to a wastegate it dumps the exhaust smoothly to another, much larger turbo, which handles the high RPM boost giving you a constant turbo speed with little lag. Am I incorrect in anything thus far? Also, is a wastegate used in a seq. TT? Now, IF a just "twin-turbo" exists, how does it work, and why does it or a single turbo benefit over a seq. turbo which I THOUGHT removed the flaws of the standard turbo. If I'm way off please explain the various setups and which one is better. Thanks for listening to my ignorance ;-)

bryan

[texan]

These are all good questions, though we've been over some of them in one way or another before I'll try to hit the topics again here for your specific question.

As far as normal production car turbo engines go (at least gasoline turbocharger setups), there are three basic designs. The single turbo we see so commonly, the twin parallel setup, and the twin sequential alternative. Which works best is a long standing argument depending largely on engine size and layout, efficiency of exhaust routing, and intended use. A drag car for instance does not need a very flexible engine power curve, one can primarily focus on a small RPM range and setup the car accordingly to operate in said range. A street car contrastingly needs a very flexible engine with not to much emphasis placed on any one RPM point's power, but I suspect these realizations are something you already understand. So that leaves us with engine configuration and exhaust layout to cover, and exactly why one turbo setup may work better vs. another in a given application.

When you look at engine layout and exhaust routing, the problem of single turbos is that they don't work very well on V engine configurations. Here you have two banks of cylinders which almost certainly exhaust not into the center of the V, but towards the outer edges, meaning you have two exhaust routes separated by quite a bit of space and solid metal. Routing an exhaust system to tie the two together is not very thermally efficient, in that quite a bit of heat (energy) will be radiated into the engine compartment and away from the turbo's turbine blades. Neither is a good thing, and as such a single turbo system is from the manufacturer's point of view likely of out the question. So now you have two options, either a parallel or sequential twin turbo setup. The sequential actuall represents an even worse option here, since not only will both banks' exhaust output need to be merged, but they'll need to be merged into two turbos and not one. Hence the only real option is a parallel setup, whereby each turbo is fed by one bank of cylinders and in turn feeds it's compressed air back into the system. And if you look at what production systems are out there, this is exactly what you'll find. Its efficient in terms of exhaust routing and thermal losses, its fairly simple in overall complexity for this type of engine, and its very effective at increasing power output.

Alternatively, let's look at an inline 6 setup. You still have all three options, but again at least one can be looked down upon as a good setup for a production car with street use as it's primary function. The single turbo setup has a serious problem here... it ties each cylinder into a high pressure exhaust path. With inline 6's, the best possible exhaust manifold design breaks the engine down into two three cylinder setups. This is due to the overlap of ehxaust strokes which occur in an inline 6 configuration, where coupling all cylinders into a single high pressure exhaust tract means every cylinder's exhaust stroke is subjected to the highest possible pressures, and therefore doesn't function as efficiently. A better system would be one that functions as two three cylinders, where street oriented cam timing would nearly completely separate the pressure surge from one cylinder's exhaust stroke from any other. So it's not hard to see that some sort of twin turbo system is required here, but which is the best?

Well, that largely depends upon what the engine's power requirements are and what technology is available. Sequential twin setups were conceived in the early 90's, when the trickle down efects of Formula 1 turbo setups were still coming of age. In such a state, there were clear benefits to having one small turbo which spooled quickly and another to keep providing boost on the high end, but that benefit is quickly being reduced by further advances in technology. When the Toyota Supra TT and the 3rd generation Mazda Twin Turbo were being developed, turbo technology clearly showed an inability to produce a compressor which worked efficiently at both low and high RPM flow ranges. Therefore complex exhaust manifolds were designed where the overall engine flow would progressivley spool up one turbo and then another to provide sustainable boost amounts at virtually any engine RPM. Nowadays, however, parallel turbo setups can mimmick these low RPM boost characteristics while providing superior top end power, all in an overall simpler package with more efficient exhaust routing. Simplicity is the mother of reliability, so from a manufacturer's perspective the current choice is clear cut.

Lastly, take a look at an inline 4 cylinder. It is small enough not to get much advantage from 2 turbos, has all exhaust ports on on side and is often packaged into tight engine bays. What this calls for is a single turbo setup, anything else takes up too much space and the airflow levels required are easily within the single turbo range.

However, the single most important thing to remember here is that basic design is not nearly so important as tuning and execution. Any setup properly designed should work well, and all of them will have a positive impact on power output. Engine management is also a key issue, most Honda people haven't figured out yet that the stock ECU is completely incapable of handling a turbocharged motor. In fact, it's slowly becoming evident to many of us that turbocharged Hondas just don't make sense as a street car, they are too expensive and difficult to properly setup.

[454Casull]

Wow, great post texan.

Can twin-scroll turbos be substituted for a pair of turbos? (I believe they have two intakes)

Here you have two banks of cylinders which almost certainly exhaust not into the center of the V, but towards the outer edges

Is it possible to build an engine with the exhaust and intake valves switched around?

[texan]

Actually, twin scroll turbochargers simply have two turbine housing inlets (more correctly a divided single inlet), which is all about conserving exhaust gas energy and minimizing exhaust backpressure. Go here to learn more.

Yes, its definitely possible to build a V engine with the exhaust ports dumping into the V. It just doesn't make any sense in passenger car applications, but it has been done in a few racing applications.

[Bryan8412]

thanks for the info ;-) But now i have a couple more Q's, just quickies tho:

1. So the supra is an inline 6? i thought it was in V formation but maybe i thought wrong...
2. How are turbos on rotories, ive heard issues with them which is why mazda's making the rx8 naturally aspirated.
3. How are they on aluminum engines and how do they differ from aluminum cast blocks? I've heard issues regarding Hennessy's turbo charging vipers..
4. Why are they not efficient on an inline 4? I was asking in regards to the Lancer Evo. coming to the U.S. which comes on board with a single turbo.
5. How expensive would it be to get a custom turbo setup installed on an engine? For ex: If someone bought something like an evo and wanted to put a seq. on it to reduce lag how much would it cost to have it done or do it yourself even?

and last Q (sorry im too curious :-( )

6. What is the "perfect" turbo engine? A big inline-6 with new-TT tech, a small 4 with a big single, or what? Thanks for your help you really know your stuff!

bry

[454Casull]

It just doesn't make any sense in passenger car applications

Why not?

What is the "perfect" turbo engine?

That's going to be a toughie. What do you want to use it for?

[texan]

Bryan8412

1) The Supra uses a 3.0L inline 6, not a V6. The Nissan 300ZX used a V6, but both the Supra and Skyline use inlines.

2) Rotaries are actually very well suited to turbocharging, given their three exhaust pulses per exhaust port (per 3 revolutions). Additionally, all newer rotaries but the upcoming Renesis have used direct exhaust ports, which makes the pulse both very loud and very high in energy content (the new one will use a side port, which is preferred for noise levels, emmisions compliance and tunability). The addition of a turbo to a direct port rotary had two advantages... one it could make great use of that very high energy exhaust pulse, and two it quiets the thing down a good bit. The problem with using turbos on rotaries basically comes down to cost (which is what the rotary is supposed to be all about; fewer moving parts means lower overall cost) and heat management (rotaries are very prone to heat related failure, and the additional heat of a very potent intake charge just gets that much harder to deal with). There are a few other points to be made here, but I think this answers the question well enough.

3) Aluminum engines are absolutely fantastic in terms of weight vs. rigidity, but it is true they aren't as long lived under stressful conditions. They are also more expensive to manufacture, but generally the pros outweigh the cons in today's market. Unless of course you are trying to build a high specific output forced induction engine that will last forever, in which case iron is the best overall block material to use. The Skyline GTR and Supra TT both use cast iron blocks, and are revered for their strength and longevity under the harshest of conditions. The 4G63 Eclipse & Evo motors also use cast iron blocks.

4) Turbochargers are very efficient in virtually any type of application (assuming the setup is designed properly), I don't recall saying anything bad about turbos on 4 cylinders. I will again state that anything other than a single turbo is 99% a waste of time though, unless you are trying to break the maximum power output level per liter or something. There are singular turbos that can supply enough air for over 1000hp running gasoline, if you are looking to make more than that you need to start with something larger than a 4 cylinder anyways IMO. Using two turbos in any configuration just ups the complexity, cost and fitment issues, it won't accomplish much else. If you are looking for serious power just off idle, you need a bigger engine and likely positive displacement supercharging. That's just not what turbos are designed for, but you should still be able to see the boost threshold point of a very well thought out design be around 2500 RPM on a 2L 4 cylinder. And really, on small motors like these 2000 RPM is the minimum usable RPM point in terms of acceleration, so you're only 500 RPM away from boost at worst. If you want big time power below 2500-3000 RPM, you'll need a much bigger engine.

5) It's not the expense I'd worry about as much as the plausability. Especially on something like the Evo, where the turbo motor is thoroughly worked out and very near it's optimal state of tune. On most other motors, a well thought out kit complete with modified engine management (when necessary) is still something that should be installed by a professional IMO. Designing one of your own would mean you have the background, technical know-how and r&d facilities to make something a reality... otherwise you are much more likely to make things worse than better when playing with forced induction.

6) There is no "perfect" engine, though the Evo 7 and WRX STi are probably as near to that as anything. It depends upon intended application, performance goals and many other variables as to what makes a great engine, and very rarely will you find any standout which is able to perform well under a variety of circumstances. This is especially true in forced induction motors, where their power delivery flexibity is largely dependent upon compressor choice. Large NA motors are naturally more flexible and potent, but of course have the common drawbacks of size, weight and fuel efficiency (the the latter is less and less a drawback in today's technology).



454Casull- Why not is primarily because of space considerations. The exhaust system is most properly setup to run under chassis for the length of the car, where it exits out the back. Having an exhaust system that goes up and over the block makes this a harder proposition. Additionally, it's a lot easier to package the larger and more complex intake manifold where there's room; directly on top of the engine and largely in the space of the V. Throwing it out to the sides means you have to have two manifolds with either two inlets or one exceptionally convoluted one (on a street car), and in addition the engine package is going to get pretty damn wide. This presents a problem as the incursion of the front or rear suspension system into the engine bay will either limit needed space to design an efficient intake setup, or you'll sacrifice suspension performance and geometry to fit the wide engine package. With a flat engine you at least have one that's not very tall, with a reverse flowing V you'd have one that's very tall, very wide, and likely very complex in it's intake and exhaust manifold routing. While this doesn't go into every last reason it makes little sense to build such an engine, it should shed some light on the subject.


As always, hope this helps explain things, peace.

[Bryan8412]

thank you very much for the great response. the 3 exhaust pulse being from a 3-rotor rotory, or are they all 3 rotor-engines and i got that wrong? but thanks for all the responses. of coarse i could avoid all this exhaust and heat problems by looking at supercharging :-D But turbos are more efficient when done right so it looks better. thanks again

[ivymike1031]

Each rotor has three faces, and all faces are used simultaneously, which means that each rotor behaves similarly to 3 cylinders in a piston engine. Each of the faces is at a different part of the cycle at any given time.

[texan]

As usual, ivymike1031 is dead on. Also realize that the rotor spins at 1/3 eccentric shaft speed, which is why I state 3 exhaust pulses per 3 revolutions. One complete turn of the rotor will have one rotor face (basically our virtual cylinder) in exhaust pulse, one face in compression or combustion cycle and one in intake.

[Bryan8412]

ok thanks guys, wow i was definently brain farting if i look at it, it makes sense it wouldn't follow one side of the rotor it'd go all 3 at once :-P

Thanks and sorry for curing my slip of intelligence hah

bryan

[Polygon]

Originally posted by texan
Bryan8412
4)Turbochargers are very efficient in virtually any type of application (assuming the setup is designed properly), I don't recall saying anything bad about turbos on 4 cylinders. I will again state that anything other than a single turbo is 99% a waste of time though, unless you are trying to break the maximum power output level per liter or something. There are singular turbos that can supply enough air for over 1000hp running gasoline, if you are looking to make more than that you need to start with something larger than a 4 cylinder anyways IMO. Using two turbos in any configuration just ups the complexity, cost and fitment issues, it won't accomplish much else. If you are looking for serious power just off idle, you need a bigger engine and likely positive displacement supercharging. That's just not what turbos are designed for, but you should still be able to see the boost threshold point of a very well thought out design be around 2500 RPM on a 2L 4 cylinder. And really, on small motors like these 2000 RPM is the minimum usable RPM point in terms of acceleration, so you're only 500 RPM away from boost at worst. If you want big time power below 2500-3000 RPM, you'll need a much bigger engine.

I am glad someone finally too the time to explain this. Two is not always better. Also, if you are concerned about turbo lag you can see if you can get a VNT turbo, it has two impellers, one for low speed and one for high speed. I believe that Garrett made these and they aren't widely used. So you might be hard pressed to find one for your car or application. Personally as far as "the best" goes, I think that Garrett makes some of the best turbos hands down. I have a Garrett T3 on mine with almost 150,000 miles and the turbo still puts over 14 PSI. And this 2.2L is great as well, one of the best 4 cylinder engines IMO. It can take 30PSI stock. I think all that can be explained has so I'll shut up now.

[Holyterror]

To repeat the old adage with a little modification, "There's no replacement for displacement, except technology." Just as engines in general benefit from continued research, turbochargers too have really come of age. Variable nozzles are rather obscure for some reason. Variable turbine geometry seems to be limited to diesels, although I don't see why. There are turbos with electric motors that act as motors to keep the turbo spooled at low RPM and act as generators to keep the turbo from overboosting - sending power back to the battery and eliminating the need for a wastegate at the same time. But the most ground-breaking technologies are being used by the masses. I would have to say that ball-bearing turbos are the best thing since sliced bread. They're less prone to oil starvation problems (or so I'm told) and the major benefit: reduced turbo lag! A well-matched modern ball-bearing turbo can often outperform yesterday's sequential turbo setups. And it's less expensive, and less complicated (and thus more reliable) to boot.

There may still be some merit to other setups. I believe that twincharging (sequential supercharger and turbocharger) still has some value. There's really not many other ways to build an engine with an 8000 RPM-wide powerband. Unfortunately, very few tuners have gone this route (MKI Supercharged MR2 owners are the only ones I know of), and there are even fewer factory examples. Nissan did this to good effect with the March Super Turbo, but no serious, high-power engines of this configuration exist to my knowledge.

Rotaries and turbos are a good combination, provided that the car has good cooling (unlike the FD3S RX-7), because rotaries don't stand up very well to detonation.

Polygon, it's nice to see somebody else can appreciate American 4 cylinders. Granted, a lot of them are very lack-luster motors, but I can think of several Dodge motors than respond well to boost. My loyalties lie with Nissan, but I give credit where it's due.

Just like we no longer need 7 liters of displacement to make decent power, we no longer need overly-complicated twin turbo systems. That said, few things get my heart started like the sound of a couple of huge turbines on the boil. :smoker2:

[Polygon]

Yeah, Dodge has some of the best 4cyls, especially compared to Ford or GM, IMO anyhow since Chrysler has turbocharged the most cars out of any manufacture I believe. Nissan is my favorite Japanese manufacture; they have the best engineering of the big three over there.

I also thought of something for reducing turbo lag. I am not sure if it was already mentioned, if it was I'm sorry. You could use a 50 shot of N2O just to eliminate turbo lag. Just before you floor it, push that button and the Nitrous will cause the impeller to spin. That way your turbo is spooled up a lot quicker. I just remembered that.

[Holyterror]

Computerized nitrous systems such as the Venom kit are designed to do this. You just tell it how long to stay on.

I don't think Chrysler holds the record on turbos. The most turbos in the States, probably. But remember that in Japan, practically everything Nissan makes comes in a turbocharged flavor (quite a few cars are turbo only). In fact, the first production turbocharger had "NISSAN IKI JAPAN" stamped on it. But Chrysler is far ahead of other American manufacturers in implementing turbos into their engines. It's refreshing to see a maker opting for forced induction in an N/A crazy country.