Variable Geometry Turbochargers And Gasoline Engines

[rx-line]

Fellas!

I have been contemplating how it is I will gain my Low-Mid range power back on the RB20. Of course even with boost sub 5k rpm she's not doing much, but without boost you might as well equip your car with chess board for you and your mate.

After thinking about quick spool valves, nitrous and high compression pistons(built bottom end) I started looked into VGT turbos (Variable Geometry Turbos) as most of you already know about.

It turns out we are on the cusp sort of speak with these turbos and there relevance in production vehicles.

Diesel engine makers and large North American companies have been equipping their production vehicles with these VGT turbos for last 5-10 years. Unfortunately "coking" (carbon build up) in diesel engines can cause failure to units and their veins.

Some initial fears with these VGT units on gasoline engines was the increase in EGT's and impact it may have on the vein system. However Volkswagon and Porsche among others (Honda - 1980's legend) have been using these units on gasoline engines in recent times. With the ability to build boost in neutral we are dealing with a new era in turbo charging. For me this alone eliminates a need for launch control as power can be had off the line!

What I have come across so far as a viable replacement for my Hy35 is the Holset HE351VE with it's simplistic and rudimentary design. It seems very unlikely to fail from heat. The only concern i have short of converting it from electronic vein control to pneumatically actuated would be that the unit is still far too large for the RB20, as the HY35 is laggy enough. My concern would be that the only gain would be more flow on the high end in which case I would not use. The theory behind these units really make sense for the RB20 but I am unsold on the idea that the HE351VE is the unit for this engine.

I am interested to know of other users on here using these turbos with the RB's!. And for those of you that are unsure of what I am speaking of please see the photo below.The holset design is a bit different with far less moving parts.

Any comments, concerns or discussion is welcomed.

[zebra]

All very well...or you could just fit a 2.5 or 3.0....or 3.4

[rx-line]

All very well...or you could just fit a 2.5 or 3.0....or 3.4

I would have to put it in the boot seeing as there is an engine in my bay.

[rx-line]

Also,

There is a little info here on a sister site.

http://www.gtr.co.uk/forum/154333-first-big-variable-geometry-turbo-rb-engine-tuner-help-needed.html

[AngryRB]

nissan navara's have variable vane setup i believe...

[rx-line]

nissan navara's have variable vane setup i believe...

The thing about Diesel turbos is that the factory models are very very small for the displacement in comparison to a gasoline engine. The hy35 (HE341) comes from a 5.9L. So a 2.5L diesel turbo seems quite small (Garrett GT2052V)

A twin setup would be interesting i suppose

[Dose Pipe Sutututu]

Hypergear have one.. but not in production, just a prototype for them to ticker with...

[GTSBoy]

The petrol ones in use have rather exotic materials associated with the vanes and some clever engineering to prevent the unavoidably high EGTs from causing problems. Diesel variable turbos are just not in the same league.

The control systems have to be fairly thoroughly thought out and it's nowhere near as simple as just sensing a pressure or two and using a pneumatic actuator powered by boost to move shit.

Consequently, short of a well thought out achievable control system and access to turbos that are actually suited to petrol EGTs, we're still not rally there yet when it comes to doing aftermarket/retro/DIY style systems.

I think in this day and age of arduino boards and so on that it should be possible to come up with a suitable control scheme that you would actually have a chance of writing the code for and being able to interact with the real world with. Use some sort of electro-mech servos to achieve linear or rotary actuation for the vane system. Use fairly readily available pressure sensors for control inputs, probably also with tacho input from the ECU. But it all sounds like a lot of work if you're starting from scratch. OK if you like a project, not so much if you just want to achieve a goal. I haven't looked at what people have actually been trying in this field in the last couple of years though, so perhaps you'd want to do some research.

I would say that the smartest thing you could do with an RB20 if you want some response and low end torque would be to supercharge it. Just plain supercharge if your power goals are not excessive. Otherwise a decent twin-charge setup using a well matched blower that would give you 10 psi off idle and a f**koff big turbo (well, big for an RB20....say something based on a 3582 or something similar) to really leverage the 1st stage compression from the blower. There is info on here and on PF regarding Stocky's VL Commodore with that setup on an RB30. It was a bit of an animal.

[Dose Pipe Sutututu]

^ or just drop in a RB25DET NEO for the fraction of the price and make more torque down low, more torque everywhere and more power everywhere too.

[jiffo]

GTSboy has covered the salient points well.

Yes they've been in diesels for many years and almost broke Cummins. Owners ditched the VGT in favour of old fashioned waste gate style and real engine life returned. Precise control of the thing, even within the tight confines of a truck's usable rpm is extremely difficult and of course carbon build up etc does not make for long service intervals.

Still OEM on small capacity diesel cars where an owner who gets over 200,000k's thinks it's amazing. Us old time diesel owners aren't interested in anything until it's done 1,000,000 trouble free.

The complicated sequential turbo setup on the 2JZ supplies seamless power all the way to the top, but it must weigh 100kg !!!

[DRF72]

have you looked at what Aerocharger have been doing? http://www.aerocharger.com

[zebra]

Yep....

$16000 USD, to make a commodore (Pontiac G8) go 11.08 at 130.....

errrr......seems like really good valve considering with just a cam change it would probably run 11's too.........or for 16k you could have a built motor and a promod88 and run 8s......

[hypergear]

Hypergear have one.. but not in production, just a prototype for them to ticker with...

yes got few working prototypes that actually makes very noticeable differences. They are best working with VCT motors. I've tried them with RB26 engine, they wanted surge, as in there are too much air for the engine to swallow with low RPM. Result wasn't as good as when they were on my RB25. I think those systems needs to be individually developed on given engines.

[rx-line]

The petrol ones in use have rather exotic materials associated with the vanes and some clever engineering to prevent the unavoidably high EGTs from causing problems. Diesel variable turbos are just not in the same league.

The control systems have to be fairly thoroughly thought out and it's nowhere near as simple as just sensing a pressure or two and using a pneumatic actuator powered by boost to move shit.

Consequently, short of a well thought out achievable control system and access to turbos that are actually suited to petrol EGTs, we're still not rally there yet when it comes to doing aftermarket/retro/DIY style systems.

I think in this day and age of arduino boards and so on that it should be possible to come up with a suitable control scheme that you would actually have a chance of writing the code for and being able to interact with the real world with. Use some sort of electro-mech servos to achieve linear or rotary actuation for the vane system. Use fairly readily available pressure sensors for control inputs, probably also with tacho input from the ECU. But it all sounds like a lot of work if you're starting from scratch. OK if you like a project, not so much if you just want to achieve a goal. I haven't looked at what people have actually been trying in this field in the last couple of years though, so perhaps you'd want to do some research.

I would say that the smartest thing you could do with an RB20 if you want some response and low end torque would be to supercharge it. Just plain supercharge if your power goals are not excessive. Otherwise a decent twin-charge setup using a well matched blower that would give you 10 psi off idle and a f**koff big turbo (well, big for an RB20....say something based on a 3582 or something similar) to really leverage the 1st stage compression from the blower. There is info on here and on PF regarding Stocky's VL Commodore with that setup on an RB30. It was a bit of an animal.

Hey I hear ya, A project with a goal is where I am always at. I do not find this daunting at all, it will work (And work good at that) if the proper unit is selected. There is three ways to control these either with a boost sensing actuator (wastegate actuator), aftermarket standalone controllers or like you have said custom arduino boards.

The Holset HE351VE Design uses veins that are connected via a telescopic ring. (See picture below), I have even seen a rotary with some of the highest egt's of any engine run this turbo for some time. Other diesel designs however do seem liable to excess egt's. Most of the gasoline turbos are using titanium bits form what I have researched.

*Edit - including picture. Also the turbos usually come with the electronics/servos and just require some sort of computer to tell them what to do.

The turbo actually has guides for the fins on the exhaust housing.

Edited December 9, 2015 by rx-line

[rx-line]

have you looked at what Aerocharger have been doing? http://www.aerocharger.com

I have, we have a motorcycle/machine shop and the aerochargers have been round for a while.. As of now a Turbo I4 Kawasaki in the 8 seconds and a supercharged harley soon to be in the same class. None of which use aerochargers atm.

[Dr. Surge]

Honnestly, imho, i wouldn't advice an vgt turbo on a petrol engine. As you say porsche ect uses them. But they don't last as long as a normal turbo, even porsche doesn't use them anymore in the new porsche turbo, if i remember correct, they're using a twin setup now ...

[GH05T]

much easier ways to get your torque low down, like a 25.....but thats boring.

ima go grab some snacks for the watching.... continue your progress.......

[rx-line]

much easier ways to get your torque low down, like a 25.....but thats boring.

ima go grab some snacks for the watching.... continue your progress.......

Although it would the first RB25DET in an FB rx7 i woud think, I like the RB20det and already has proven itself to be faster than vettes and camaros. Plus I have a spare RB20 anyway. The 25's go for around $3k with trannys here. A variable vein turbo will not cost me much as I will just sell the near brand new HY35.

The ported head I'm building should compliment more boost down low as well.

Here's a rotary with the HE351VE, I think he blew the engine recently.