The Perfect Turbocharger? Porsche Did It First

The Porsche 911 Turbo was the first gas car to use variable geometry turbos!

How Porsche perfected turbochargers: the Porsche 911 Turbo did it first! Variable geometry turbos were first used on diesel engines, starting in 1997. But it took nearly a decade to get them fitted to a gasoline engine. In 2006, the 997 Porsche Turbo was the first gasoline production car fitted with variable turbine geometry turbos!

In this video we'll talk about the problem with turbochargers, the various solutions out there, how variable geometry turbos fit into this, why VTG turbos are rare, and details of Porsche's new 911 Turbo S 3.7L boxer engine. It's producing 640 HP and 590 lb-ft, thanks to two variable geometry turbos. Enjoy the video for all the details!l

Engineering Explained is a participant in the Amazon Influencer Program.

Don't forget to check out my other pages below!

[EngineeringExplained]

The whiteboard's back! I think arguably electric turbos are up there as the best (like what's used in F1), but it's of course a much heavier solution, adding a large electric motor and battery to the mix. As a more mechanical solution, this seems pretty ideal; what do you think?

[dhess]

First person I’ve ever seen who launched a 911 Turbo S and didn’t have a huge smile on their face. This guy’s a pro’s pro; an engineer through and through.

[dracobutwhatever]

Everytime I see the whiteboard, I get a dopamine kick

[dr.richardmarrotte]

The best explanation of turbo charging I think that I've ever heard. This turbo redesign is truly special.

[jonathanmatthews]

The white board returns!

How I missed you old friend!

[wompa]

This dad appreciates the joke.
I know faster spooling and smooth power makes for a “better” driving experience. But, man, I’ve yet to drive anything as fun as my Dad’s 87 Shelby Charger. Stomp the gas and hold tight to the steering wheel. Cause when that turbo hits, the torque steer is unreal!

[DC-rnfc]

Oh my ... that rings a bell ... when I wrote my Diploma-Thesis, I did a lot of simulation to improve response of a miller-cycle gasoline engine. Having a variable valvetrain (I used the lift arrays of BMW's Valvetronic) is a key element here, but only so long as the steady-state operation at full load is not at the limit in regard to CA50... Having this at like 40 deg ATDC in steady state ... means you won't. Having it there in the transient operation helps - you can use a longer and higher lift, basically just swinging over a bit, and still avoid knock. Miller cycle works best, of course, with max. boost pressure and shortest intake duration. Adjusting the cams to more overlap helps also, but don't be fooled - miller-engines have poor transient performance still, compared to a similar one.

A VGT, of course, helps here, too - and in some cases, the lower backpressure at max power makes up for the rich mixture used in order to avoid overfrying the turbine. Man, I loved that ... setting up cam timing and valvetronic overswing, ignition angles, and the operation of VGT versus time, and handing it back to a PID-control at the right time ... makes the response a whole lot better - but still, we have less exhaust mass flow, less exhaust temperature and need more boost pressure, equalling a higher speed of the turbocharger's rotary parts, means that you can't help but, unless your boost system is a bit more complex, a miller engine will have poor response.

Porsche enabled something here that bears a value many seem not to see ... introducing this to gasoline engines a decade before it entered real mass production (think EA 211 of Volkswagen), ... anyway, a hybrid drivetrain comes to mind, where the lag may be covered electrically, and also, the turbocharger(s) may use electric power to spin up faster, ...

... great memories... And now, I do industrial equipment for something totally else. Also interesting, but not of any mattering here....

Great video, as kinda always!

[thereissomecoolstuff]

As a teacher I want to let you know how much I appreciate you using pens with consistent discharge flow. That shows the sign of a professional educator. Eliminating frustrations in learning is what the good ones do...now what were we taking about?

[gustavmeyrink_.]

7:13 It is called a 3.8L because that is how motor sport works, there is no rounding down ie up to 3700cc it is a 3.7L and from 3701cc it is a 3.8L.

[curtaustin]

I had a dim recollection of "PM2000", then googled it to find it was similar to something I worked on the early 1980's (MA956). Yes, very special and unusual, made only by the company I worked for. Somewhat surprised this sort of alloy is still used, after advancements in more conventional superalloys that are easier to prepare. Yes, they are known as superalloys, a well-deserved term, though most are based on nickel. The only way to get a fine enough dispersion of the oxide in an oxide-dispersion strengthened alloy is to literally beat it into powdered metal in a ball mill.

[torino]

Chrysler had their VNT, Variable Nozzle Turbos from Garrett in their Dodge Shelby CSX Shadows back in 1989.

[coltonhill]

“The rotating-vane VGT was first developed under Garrett and patented in 1953. One of the first production cars to use these turbochargers was the 1988 Honda Legend; it used a water-cooled VGT installed on its 2.0-litre V6 engine.”

[minivanin]

Some models of turbo fwd dodges in the late 80s/90 had a variable turbo

[crztank]

The honda legend 2.0 v6 turbo (jdm) used vgt turbos back in the late 80's. It was rudimentary, using 5 vanes. They called it the wing turbo

Edit: just googled it to be sure, there is an image available for anyone who needs proof

[Sir.VicsMasher]

0:36 I thought the Dodge Shadow was the first production car with a variable geometry turbo. 1989 Dodge Shelby CSX VNT

[rdspam]

Variable geometry turbos: 1988 Honda Legend (Japan), 1989 Shelby CSX-VNT, 1992 Peugeot 405 T16, Porsche’s own 2007 911 Turbo, 2015 Koenigsegg One:1.

[OxBlitzkriegxO]

For the US market, the '89 Shelby CSX-VNT was the first variable nozzle turbo, that's literally part of it's name.

The Turbo-IV made full boost (15 psi) at 2, 100rpm and produced 205tq and 175hp at the crank. The engines dyno around 140whp and 175wtq. Upgrading the engine with this turbo was problematic because it was advanced for the time (aka failure prone) but also because the car used the VNT as the wastegate. Installing a larger exhaust meant the turbo was easy to over boost.

Regarding the failures, the earliest turbos tended to bind up and freeze the vanes in place due to rust and carbon build up. An update was released by the engineers to move the vanes back and forth upon start up to prevent build-up. That seemed to cure the failures.

Sources: Allpar and various Chrysler Engineers from articles on the aforementioned site.

[ManuelBTC]

Jason, congratulations on everything you've accomplished. You're an inspiration.

[Intrepida]

I've heard of variable geometry turbos but didn't know exactly how they worked. It amazes me these units can reliably stand the heat and rpms they will routinely live in.

[phildem]

I had one in a E46 BMW 320D and it was a game changer! Best Diesel of it time, period. Unfortunately, VTG turbos are subject to carbon build up and need to be disassembled and cleaned at high mileage.