Why Bi-Turbo V6es ALWAYS suck.

[dwightlooi]

Have you ever wondered by all the Bi-turbo V6 engines are always less impressive than you expect? Specific Output always seem to be worse compared to 4-cylinder counterparts. Maximum Torque seems to always arrive later and the engine always seem a tad lethargic. In short, they always seem to SUCK relative to the 4-cylinder turbo engines the same manufacturer puts out. They really SUCK compared to Inline-6 legends like the Nissan RB26 or the contemporary BMW N55.

For example:-

GM made a 2.0T (LTG) with 272 hp @ 5,500 rpm with 260 lb-ft at 1,700 rpm. That is 136 hp/liter and 130 lb–ft per liter. This same engine is later tuned to permit more mid-range boost with torque rising to 295 lb-ft @ 3,000 rpm (147.5 lb-ft/liter). This engine operates on 18 to 22 psi of boost and is able to get to full torque at 1,700 rpm or 3,000 rpm.

At the same time, GM made the 3.6 Turbo (LF3) with 420 hp @ 5,750 rpm with 430 lb-ft @ 3,500 rpm. That is 116.7 hp/liter and 119.4 lb-ft per liter. This engine operates on a paltry 12 psi of boost but can’t seem to get to full torque until 3,500 rpm. Throwing really fancy and expensive stuff like Titanium Connecting rods and turbine wheels at the engine helps a bit. But, even the uprated 3.6TT (LF4) still took 3,500 rpm to reach 18 psi. A boost level the ubiquitous Malibu engine has no problem hitting at 1,700 rpm. Why?

REASON

The reason is very simple. And, it is not something you can conquer with expensive turbos or fancy materials. Just look at the diagrams below...

In a 4-stroke engine, the exhaust valve opens once every two rotations of the crank for about half a rotation of the crank.

In a 3-cylinder engine, or any bank of 3-cylinders, exhaust flow to the turbo is interrupted for significant periods with the exhaust valves closed on ALL cylinders. A V6 Bi-turbo engine is essentially two Inline-3s with each turbo is feeding off one bank of 3-cylinders.

In an Inline-4, one exhaust period is always beginning while another is ending. Hence, while exhaust flow diminishes, it never really falls flat completely.

In an Inline-6 the periods overlap generously ensuring a continuous exhaust flow to the turbine that does not diminish appreciably throughout the entire cycle of the engine.

SOLUTION

The solution is to use ONE Turbo for both banks of cylinders in an V6. But that is problematic. You either have to route the exhaust from one side to another -- which is both a packaging nightmare, a source of leaks and a measure which causes it to lose a lot of energy (heat) getting to the turbo. Or, you can use a Hot Vee design with the exhaust exiting in the valley of the Vee. A Hot Vee is problematic in the V6 because most of them are 60 degree engines which has little to no room in the Vee for a turbocharger. In addition, a single large turbo will stick through the hood in a Hot Vee. Hence, a Hot Vee is seldom used. When it is used, it is employed in a 90 degree V6 with two turbos -- which defeats still results in interrupted flow to the turbines. Hence, V6es continues to suck in turbocharged applications -- then, now and (probably) into the future!

Edited January 25, 2019 by dwightlooi

[riviera74]

Or automakers could do what BMW has done for decades: I6 and then add a turbo.

[A Horse With No Name]

Excellent technical answer.

[dwightlooi]

1 hour ago, riviera74 said:

Or automakers could do what BMW has done for decades: I6 and then add a turbo.

The I6 is very long and undesirable for that reason. A compromise is the VR6. Remember that? The staggered bore 12 or 15 degree V6? That keeps all the exhaust on one side too. Intake and exhaust port geometry, as well as the slant top pistons are sub-optimal, but not significantly so.

 

[Robert Hall]

36 minutes ago, dwightlooi said:

The I6 is very long and undesirable for that reason. A compromise is the VR6. Remember that? The staggered bore 12 or 15 degree V6? That keeps all the exhaust on one side too. Intake and exhaust port geometry, as well as the slant top pistons are sub-optimal, but not significantly so.

 

The length of an I6 is not an issue for RWD vehicles.   Not undesirable at all. 

Edited January 25, 2019 by Robert Hall

[Drew Dowdell]

There is still no replacement for displacement.   A naturally aspirated V8 with variable displacement can "switch on" its off cylinders substantially faster than any normal turbo can spool up. 

[dwightlooi]

1 minute ago, Robert Hall said:

The length of an I6 is not an issue for RWD vehicles.   Not undesirable at all. 

It is if you want to keep the entire engine behind the front axle for balance purposes... You'll either have to have a very long hood and wasted wheelbase length (which does not go towards a roomier cabin), or you'll have to stick the engine past the front axle and have sub-optimal balance.

It is significant enough for Nissan to abandon the 2.6 RB26DETT for the 3.8 VR39DETT.

[Robert Hall]

Hasn't been an issue for BMW, and Mercedes is going back to straight 6s.   The main driving issue for V6s seems to have been having 6 cylinder engines that can fit in FWD transverse engine applications. 

Edited January 25, 2019 by Robert Hall

[dwightlooi]

1 minute ago, Drew Dowdell said:

There is still no replacement for displacement.   A naturally aspirated V8 with variable displacement can "switch on" its off cylinders substantially faster than any normal turbo can spool up. 

Yes, and a supercharger has (practically) zero lag too. But that is not the point of this thread. The point is that a V6 bi-turbo is architecturally inferior to other turbocharged arrangements. Displacement or not, you'll have the same problems turbocharging a 4.7L V6 as you'll have a 1.8L V6.

[Drew Dowdell]

1 minute ago, dwightlooi said:

Yes, and a supercharger has (practically) zero lag too. But that is not the point of this thread. The point is that a V6 bi-turbo is architecturally inferior to other turbocharged arrangements. Displacement or not, you'll have the same problems turbocharging a 4.7L V6 as you'll have a 1.8L V6.

I'm talking more about the attempts to replace V8s with Turbo V6es of a lower displacement.   The Cadillac 3.0T and Lincoln 3.0T are both nice enough engines, but they're no V8.  I'd much rather have a 6.2 V8

[dwightlooi]

1 minute ago, Robert Hall said:

Hasn't been an issue for BMW, and Mercedes is going back to straight 6s.   The main driving issue for V6s seems to have been having 6 cylinder engines that can fit in FWD transverse engine applications. 

It is an issue they pay a price to deal with. A price in smaller cabins. A price in a heavier engine. A price in a heavier car with unnecessarily long wheelbases.

As far as the M256, that is an S-Class only engine for now and not even on US bound S-Classes. Looking at the C-Class engine bay it may take a platform change to fit it in the C-Class.

2 minutes ago, Drew Dowdell said:

I'm talking more about the attempts to replace V8s with Turbo V6es of a lower displacement.   The Cadillac 3.0T and Lincoln 3.0T are both nice enough engines, but they're no V8.  I'd much rather have a 6.2 V8

Well, the C8 has a big displacement, Naturally Aspirated, Pushrod V8 for you with 500 hp (give or take)! And, it is probably the cheapest engine on offer. Expect at least 17/30 mpg (already achieved on the LT1 with AFM on) from it which is no worse than anything else making 500 hp either.

[ccap41]

On 1/25/2019 at 4:20 PM, Drew Dowdell said:

There is still no replacement for displacement.   A naturally aspirated V8 with variable displacement can "switch on" its off cylinders substantially faster than any normal turbo can spool up. 

"normal" is key here because twin scrolls have gotten great reviews for no noticeable lag and electric turbos the same. 

[Drew Dowdell]

Just now, ccap41 said:

"normal" is key here because twin scrolls have gotten great reviews for no noticeable lag and electric turbos the same. 

I was specifically excluding electric turbos, but twin scrolls are still slower than cylinder activation, even if it is hard to notice.

[ccap41]

We're talking fractions of a second so it's essentially negligible differences. 

[dwightlooi]

4 hours ago, ccap41 said:

"normal" is key here because twin scrolls have gotten great reviews for no noticeable lag and electric turbos the same. 

You do know that this is physically impossible, right? Turbos will have lag, period.

Why? Because they build boost by pumping more air than the engine ingests and building up positive pressure. It takes time to go from vacuum to 15 or 20 or 25 psi just like it takes time to fill up a balloon. It is impossible to do that immediately just like it is impossible to make boost by revving your engine in neutral (where the unloaded engine can rev up faster than the turbos can move air into the manifold).

Twin Scroll turbine housings do not eliminate lag they are there to solve an intake problem. You see, in a 4-cylinder engine when the exhaust valves on a cylinder opens towards the end of the power stroke, another cylinder still has its exhaust valve open towards the end of its exhaust stroke. To make matters worse, that cylinder also has its intake valves opening as the end of the exhaust period overlaps with the intake period. This means that the high pressure exhaust gases exiting the cylinder at the end of the power stroke will push into the cylinder ending its exhaust stroke and beginning its intake stroke. This hammers intake aspiration as well as rob the turbo of exhaust pressure that otherwise would be applied towards spinning the turbine!

There are two solutions to this problem. The first is to have essentially no intake/exhaust overlap -- most traditional turbocharged engines do this. But, this means the engine won't breathe right above about 5000 rpm and it negates whatever advantage you might gain from a DOHC 4-valve head. The second is to segregate the exhaust flows from cylinders 1 & 4 and that of cylinders 2 & 3 -- that's a twin scroll, twin volute or simply divided turbine housing. You never see twin scroll turbos in bi-turbo V6 engines or inline 3 engines because it is unnecessary and USELESS -- exhaust periods on a 3-cylinder or a V6 bank do not overlap!

 

Edited January 29, 2019 by dwightlooi

[ccap41]

3 minutes ago, dwightlooi said:

You do know that this is physically impossible, right? Turbos will have lag, period.

Did you miss the part where I said "no noticeable lag". There's a difference between it being non-existent and it being very minimal. 

[Drew Dowdell]

5 minutes ago, ccap41 said:

Did you miss the part where I said "no noticeable lag". There's a difference between it being non-existent and it being very minimal. 

His point is that you can't do twin scroll on a V6.  So the "no noticeable lag" on the V6 doesn't apply. 

[dwightlooi]

12 minutes ago, ccap41 said:

Did you miss the part where I said "no noticeable lag". There's a difference between it being non-existent and it being very minimal. 

You must not be very good at noticing then! Even on the VW-Audi 1.8T (circa 2000) where the turbo is making a paltry 7~8 psi (very low by today's standard) you'll notice it. You'll notice it particularly during part throttle application. You put in 70% throttle and it feels like the throttle takes 2 seconds to roll from 30% to 70%. Even in a laggy engine, full throttle lag is never an issue because you know its there and you know the power coming. The problem with handling a turbocharged car through the corner is that when you apply part throttle mid corner it is as if you are working the throttle via a rubberband. It is non-linear and it is difficult to control. Instead of applying the throttle you need and rolling on more as you straighten out the wheels, you must apply the throttle eventually want a second or two ahead and time the torque rise match your corner exit. In a way, it's like doing differential equations while driving.

In any case, if you have a V6 you are not getting a twin scroll.

Edited January 29, 2019 by dwightlooi

[Drew Dowdell]

1 minute ago, dwightlooi said:

You must not be very good at noticing then! Even on the VW-Audi 1.8T (circa 2000) where the turbo is making a paltry 7~8 psi (very low by today's standard) you'll notice it. You'll notice it particularly during part throttle application. You put in 70% throttle and it feels like the throttle takes 2 seconds to roll from 30% to 70%. Even in a laggy engine, full throttle lag is never an issue because you know its there and you know the power coming. The problem with handling a turbocharged car through the corner is that when you apply part throttle mid corner it is as if you are working the throttle via a rubberband. It is non-linear and it is difficult to control. Instead of applying the throttle you need and rolling on more as you straighten out the wheels, you must apply the throttle eventually want a second or two ahead and time the torque rise match your corner exit. In a way, it's like doing differential equations while driving.

Partial throttle is where turbos usually fall down verse larger displacement/greater cylinder count engines of similar peak power. 

[dwightlooi]

16 minutes ago, Drew Dowdell said:

Partial throttle is where turbos usually fall down verse larger displacement/greater cylinder count engines of similar peak power. 

And, the reason is simple...

When you open the throttle, the naturally aspirated motor has to fill the volume between the throttle body and the intake valves -- aka the plenum -- from vacuum  to atmospheric pressure. This is normally very fast. It can actually be faster than the throttle plates can move if you have the throttle plate(s) mounted at the intake port. Remember the the S65 4.0L M3 engine with the individual throttles in the velocity stacks for each cylinder? Or, heck, remember the individual side-draft carburetors on the intake ports of motorcycle engines? 

The Turbocharged engine has to first do the same thing and fill up the volume to atmospheric pressure, make exhaust energy, spin up the turbines to spin the compressor, start increasing that volume to the working boost pressure. That takes significantly longer. To make matters worse, the volume between compressor and the intake ports is also a lot more voluminous on a turbocharged engine. It usually goes from one side of the engine to the other and includes additional volumes like the intercooler.

Edited January 29, 2019 by dwightlooi

[Drew Dowdell]

@dwightlooi Would something like Fiat Multiair for BMW Vanos remove that since there is no more throttle plate restriction? It does seem to increase fuel efficiency.

[dwightlooi]

24 minutes ago, Drew Dowdell said:

@dwightlooi Would something like Fiat Multiair for BMW Vanos remove that since there is no more throttle plate restriction? It does seem to increase fuel efficiency.

Oh, there is plenty of restriction. Whether you are choking the engine with a throttle plate or by reducing the intake valve openings, you are choking the engine. If that is not the case, you'll have unintended acceleration ,which will be very bad!

The only time there is no restriction is in a Diesel Engine. In a Diesel Engine, there is no throttle and the valves always open fully. The engine simply injects more or less fuel to make more or less power. This is half the reason why diesels are so efficient -- you are never choking the diesel engine. But, it runs very lean when you drive gently and running very lean makes a lot of Oxides of Nitrogen which -- unlike Carbon Dioxide -- is actually bad for the air. This is why you have nitrogen storing catalysts on modern diesels. Because these cats get poisoned by sulfur, that is also why low sulfur gasoline is a must for some of these diesels. Or, you can inject piss into the engine -- Mercedes Benz's Bluetec is simply a nicer name for Urea Injection.

What Multiair and Valvetronic (not VANOS) do are that they eliminate the vacuumed space between the throttle and the intake ports. This makes throttle response technically superior as there is no plenum to bring up from vacuum to atmospheric pressure. I am not sure it makes the engine more efficient although the engine holistically might be for a host of other reasons.

Edited January 29, 2019 by dwightlooi

[Drew Dowdell]

7 minutes ago, dwightlooi said:

What Multiair and Valvetronic (not VANOS) do are that they eliminate the vacuumed space between the throttle and the intake ports. This makes throttle response technically superior as there is no plenum to bring up from vacuum to atmospheric pressure. I am not sure it makes the engine more efficient although the engine holistically might be for a host of other reasons.

 

Fiat make the following claim about the efficiency:

"MultiAir technology can increase power (up to 10%) and torque (up to 15%), as well as reduce consumption levels (up to 10%) and emissions of CO₂ (up to 10%), particulates (up to 40%) and NOx (up to 60%) when compared to a traditional petrol engine."

[A Horse With No Name]

Twin Turbo Big Blocks on the other hand...

[Drew Dowdell]

@dwightlooi  Would a hot-V twin-scroll setup for the turbo help a V6? 

[dwightlooi]

2 hours ago, Drew Dowdell said:

Fiat make the following claim about the efficiency:

"MultiAir technology can increase power (up to 10%) and torque (up to 15%), as well as reduce consumption levels (up to 10%) and emissions of CO₂ (up to 10%), particulates (up to 40%) and NOx (up to 60%) whe compared to a traditional petrol engine."

That is what they claim. Ask yourself two things. Firstly, how much of that is due to Multiair itself and how much of that is due to everything else Fiat put in the engine -- say Direct Injection or using tiny displacement engines (like their 1.4L). Secondly, how does cutting off the action of the intake valves partway through the cycle accomplish all those things better than VVL systems that... well... offer long and shorter duration valve actuation with advance and retard of the timing?

[dwightlooi]

1 hour ago, Drew Dowdell said:

@dwightlooi  Would a hot-V twin-scroll setup for the turbo help a V6? 

As explained in the original post, a Hot Vee SINGLE TURBO design will provide uninterrupted exhaust flow to the turbine. A twin-scroll turbine housing will be preferred for intake breathing efficiency. Hot Vee Singles are rare because the exhaust collector bridging both sides and a single larger turbo means the turbo sits really high and interferes with the desired hood line. It has been done though... the Duramax 6.6 is an example of a Hot Vee Single Turbo. It makes 445 hp @ 2,800 rpm and 910 lb-ft @ 1,600 rpm (it's a diesel).  Note: Notice how tall the entire turbo setup is and why it may be a problem in a sports car or a sedan.

However, a Hot Vee Twin Turbo design like Mercedes' M276 DELA30 (3.0L Twin Turbo; C400/C450/C43 AMG) provides no advantage apart from packaging (which is itself debatable). I find Mercedes' coining of the term "Pulse turbocharging" ironically laughable as it describes the interrupted exhaust problem of V6 bi-turbos so succintly. IMHO, conventional flanking turbos can be located just as close to the exhaust ports as a turbos in a Hot Vee, and flanking turbo designs like GM's LGW (3.0L Twin-Turbo; CT6) performs just as well.

Edited January 29, 2019 by dwightlooi

[Robert Hall]

Note that the Duramax is a diesel, though.  Is anyone doing single turbos like this for a gas V6?

[A Horse With No Name]

11 minutes ago, Robert Hall said:

Note that the Duramax is a diesel, though.  Is anyone doing single turbos like this for a gas V6?

I don't think so. And the Duramax does not have the hood heights issue. Pickups have lots of room under the hood.

[dwightlooi]

16 hours ago, Robert Hall said:

Note that the Duramax is a diesel, though.  Is anyone doing single turbos like this for a gas V6?

Nope, but they should for a mid-engine sports car. I think ford wasted a good opportunity with the Ecoboost V6 Ford GT. These have a lot of height in the engine bay if you don't mind seeing the engine in the rear window. The Audi R18 Le Mans race car is a hot vee single turbo 3.7L V6. But that is (oddly enough) a DIESEL Le Mans race car, and that doesn't constitute a "production" engine.

The C8 is unlikely to use a hot vee single setup since it is a V8 and that is as good as an I4 which is good enough.

Edited February 1, 2019 by dwightlooi

[Drew Dowdell]

Where do the turbos sit on the Subarus and is the flat 4 or 6 any advantage?

[A Horse With No Name]

I can't see flat as being any advantage.  

[dwightlooi]

2 hours ago, Drew Dowdell said:

Where do the turbos sit on the Subarus and is the flat 4 or 6 any advantage?

LOL... It sits in a terrible location that is unreachable, unserviceable and very far from the exhaust ports! Basically, it's behind the engine, below intercooler and above the transmission. VERY long pipes connect the turbo to the exhaust ports below the engine. If you straighten out the pipes the turbo is about as far from the engine as the front seats in a car. For emissions purposes, Subaru even stuck a small catalytic converter between the exhaust ports ahead of turbo for the further pair of cylinders. Another main catalyst is, of course, located after the turbo. It all amounts to a mess that is both inefficient and a nightmare to work on. Also, with the intercooler above the turbo and fed by a hood scoop, when the car is stationary, all the heat rises up to heat soak the Intercooler. Congratulations! BTW, if that pre-cat breaks all that debris goes turbine...LOL! (See photos)

That is why all the STi WRXes always have inferior turbo response compared to the Inline-4 Lancer Evolution. It takes 4,000 rpm for the STi to make full boost and it is rather placid 14.7 psi.

Edited February 1, 2019 by dwightlooi