VNT or Twin Scroll
Turbo Mini Forums

I have a GT1752 for my next 998 Turbo. Now this may be a bit late in spooling, probably coming in at 4500rpm at best, which is a bit late.

There seem to be a couple of options that could help. Firstly the GT1749V which seems to be the VNT equivalent of the GT1752 and which can be controlled to boost much earlier.



The other option is the MHI TD04L-12T, twin scroll turbo with very similar capabilities.


The twin scroll is supposed to spool up quicker. There is also the option of gating the turbine inlet to use a single volute at low revs.

Any thoughts.

Edited by Paul S on 18th Oct, 2011.

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

id go for the gated twin scroll. i feel the total flow path on the vnt is possibly a restriction at the top end , they allways seem to have a farly small outlet , but a bit of working out the areas and comparing needs to be done to be definate , it may just be an optical illusion .

also , looking at that twin scroll, it looks a small a/r anyway ,so may work well as it is .also consider the 1749 non vnt .also remember no one has sourced a smaller a/r for the 1752 and tried it , it may be that with a .33 or whatever is available it would be great on a 998.

Medusa + injection = too much torque for the dyno ..https://youtu.be/qg5o0_tJxYM

I like the idea of a gated twin scroll, but I'm a bit concerned about how well the gating would flow,

I'm erring more towards the twin scroll.

Two issues:

1. Making the gate.... if it is needed at all. Difficult fabrication.

2. I can't simulate it, so I don't know if it will affect the optimum exhaust runner length. There will be less pulses, but I don't know which ones will go. Scratch head.

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

Without the gate I cant see there being a massive benifit, I think you might see boost a couple of hundred rpm sooner amybe 3-400 max, gate it an i'd expect to see boost 1000rpm sooner maybe more, thats just a gut feeling though.


The VNT yould make for a simpler manifold though,

Edited by Joe C on 18th Oct, 2011.

I was wrong about the simulation. I am now running the model with twin half size turbos just to see what happens pulse wise.

One of the main advantages of the twin scroll is that you should be able to run more overlap. We shall soon see, at least in theory.

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

Would you need to make the exhaust gas control valve? There are some oem parts out there..
Side mounted twin scroll sounds beautiful :)

Yes i moved to the darkside

Instagram @jdm_brett

I'd say VNT all the way.

Big cc diesels have no problem making power with them.
and worst case scenario if the turbine was too restrictive with VNT, boost would climb. Fit an external wastegate to bypass it then.

I dont see any benefit in running "more overlap"
Many of the best turbo setups run no overlap ( ie most 911 Turbos )

9.85 @ 145mph
202mph standing mile
speed didn't kill me, but taxation probably will

Or fit a larger one and no wastegate.

In the 13's at last!.. Just

Medusa + injection = too much torque for the dyno ..https://youtu.be/qg5o0_tJxYM

Well, I've simulated the twin scroll with LCBish length exhaust runners. I've had to use twin half size turbos and wastegates, but it seems to be quite representative

The simulations have shown about the same peak power, about 0.5kW lower, with a slight drop in torque at lower revs.

What I can see is that there is a nasty pressure spike at the exhaust valve when it is open, particularly at lower revs. This is what is hurting VE/torque.

However there is a very useful pressure dip available about 150 degrees before, so I just need to adjust the runner length to make use of it.

All purely theoretical of course.

Onwards and upwards.

Edited by Paul S on 18th Oct, 2011.

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

Fastest 998 mini in the world? 13.05 1/4 mile 106mph

www.twin-turbo.co.uk

A big diesel may only rev to 5k absolute max, but don't forget they very commonly use really high boost levels, 2 bar not out of the question, so the volumes of air and hence exhaust involved are considerable.

It is said in the automotive business (where I work) that VNT for gasoline engine are not a valid option as of today.

On a gasoline engine, the vanes and added mechanic has to cope with much higher temperatures than with a diesel engine.

I have in mind the followings :
Gasoline : 1050°C
Diesel : 700°C

Ever wondered why VW built its hi power TSI with a supercharger AND a turbo if a VNT was available and suitable ?
Twice the parts is twice the price and twice the hassle.

I would stay clear of VNT's ... except maybe if I got given the turbo for free !

Rusty by nature

the vane thing is a consideration for production stuff, but if you dont mind looking at the naes periodically theres no issue,

Ive not had any vane issues and have a few miles under my belt.

A bit more palying with the simulation. I've managed to reduce the pressure spike with longer runners, but this has shifted the peak VE down the rev range as you would expect.

Also double checked but there is no sign of the pressure spike on the simulation of the same engine and the GT1752.

One thing is for sure.... if you want to make the most from the Twin Scroll turbo, then you need tp do more than just adapt it to a "normal" setup.

I think it would like larger diameter runners to reduce the pulse amplitude. Might test that out tomorrow.

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

Rusty by nature

its had its share of abuse, pretty much used as a daily drive too.

Joe,

I do tend to agree with Advantage, that it's a bit of an unknown using VNTs on petrol, especially long term.

Having had two 1749V (deisels) apart recently I can't actually see why they sould be adverse to petrol, but I don't know what material the vanes are made of.

You've also said that the control of the vanes needs some better understanding ???

So I'd say a bit more research is required.

EDIT - spelling

Edited by Rod S on 19th Oct, 2011.

Schrödinger's cat - so which one am I ???

Well, the simulated engine does not like separate runners to the turbo. Tried everything, longer, shorter, fatter, thinner and still have the pressure spike during the exhaust stroke. I say pressure spike, it is actually a shock wave followed by a pressure spike that causes reverse flow as the piston moves up the bore.

It's only theoretical, but does underline why collector design is so important. Remove the collector and things take a nose dive.

Edited by Paul S on 19th Oct, 2011.

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

your fancy softwware makes me drooling ...

Rusty by nature

Just a thought. Say that the GT17 was available in a twin-scroll, what would you expect the boost threshold to come down to?

Also, because of the size of the compressor, I always worry that with a smaller turbine, or one that will spin up quicker, there is a risk of running in the surge zone; especially on a small bore.

Fastest 998 mini in the world? 13.05 1/4 mile 106mph

www.twin-turbo.co.uk

Schrödinger's cat - so which one am I ???

Either option could push the turbo into surge at low revs, but we are still guessing at the shape of the compressor maps. That is always going to be a problem with a turbo that is selected to give a good headline hp.

The theory of how twin scrolls reduce the boost threshold is interesting. Apparently it is down to pulses. Instead of the normal four successive pulses hitting the turbinn every 180 degrees, you get 2 pulses in each volute hitting the turbine every 360 degrees. You get half the average flow, but you still get the same peak flow in each volute. So the same peak flow in a volute half the size is going to generate twice the peak velocity. They say that this will spool the turbo quicker.

Unfortunately, I cant model this as the software uses a quasi unsteady-state model for the turbo. Although the software evaluates flows, temperatures and presures throughout the intake and exhaust every 2 degrees of rotation, it assumes a fixed turbo speed throughout the cycle. It does a power balance on the turbo shaft to determine the turbo speed. It does adjust the amount of wastegate opening every 2 degrees though.

This has been an interesting exercise, but I'm not convinced myself that it is worth spending an extra £400-500 yet.

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

Dazed and Confused....