Fitting higher flowing Garrett turbos - what's involved

[scientist]

I'm in the MHI camp too: whilst they may not be as good turbos as Garrett GT series, they are damn good turbos, and represent excellent value for money. More importantly, they are incredibly easy to hybridise...again for very little expenditure. A properly specced GT is going to have equivalent spool and more eventual power than an MHI unit, and cost more. Whilst that is brilliant for those of us who want uber-huge power and normal driveability, it would appear we're not there yet: TD04s and TD05s seem to be able to deliver enough power to damage drivetrain and possibly engine components, whilst spooling early enough to remain civilised. A TD05 turbine is probably about on the limit, I would suggest, but there may be margins in the map for more.
A pair of TD05-16Gs should just see you into 600bhp territory.

I would tend to disagree on the power potential part. Problem is, most people compare OEM MHIs against aftermarket Garretts.

Take for example, the TD05H-16G7 on the evo 3. The record on that turbo is 490whp @ 31psi boost. true it was pushing it well beyond its efficiency, but unlike some of the garretts it HELD up with such heavy abuse for a few racing seasons.

I definitely wouldn't go garrett route on most of these cars, since there are bolt on MHIs...which can be rebuilt. Just look at the prefered set of turbos on 300ZXs and 3000GTs in the US....there are quite a few running 9s on a pair of Evo 3 turbos. Even JUN opted for a TD05-18G for the 900whp 350Z

[Turbo_Steve]

er....presumably that's a PAIR of TD05-18Gs?

And whilst I agree with your comments pertaining to the T series turbos, the GTs are a ball bearing cartridge and really do spool earlier for the same power.

[Davezj]

Yes, the turbos are definately TD-03's, i have mine i a box, and the numbers confirm the info on the list attached. They come with the 7g compressor wheel, and it is steel not ceramic as some people on OZVR4 thought for a while.

i don't mean to put a spanner in the works, but hasn't everybody just said the impretza has TD-04 as standard and these are goodone to go for.

That is all well and good.

However the table posted up shows TD05H-16G-7 for the WRX/STI/Forester.

where did this info come from and is it reliable?

Is the standard Impretza turbo TD-04 and WRX/STI TD-05 or are we all wrong about the TD-04.

just an observation, i am not having a go at anyone, i appiciate all the effort that goes into finding all this info and at some point hope to make use of it.

[bradc]

early ones came with td05's, most of them however came with td04's

[Turbo_Steve]

er....and some of the later ones....Subaru turbo choices were complicated dependant on model, target market and vehicle shape / gearbox.

The later ones that aren't a WRX generally have an IHI turbo. Technically, this should fit an work well (pair of VF34s anyone?) but they generally cost a lot more money!

[Davezj]

I will get my coat this to LA LA LA LA LA!!!!

i will just read!

[Madhav]

Also don't overlook the kkr 280 turbo, good for an easy 300kw atw and spool up time is as good as the standard turbos (or so i hope). My car is currently in the workshop getting a pair fitted, they need an adapter plate and a new thermo fan, but otherwise it seems to be ok. They will be in within 3 weeks max, I'll keep everyone posted.

[Eurospec]

The scooby td04 13t turbo is what people are normally refering to right? It came on quite a few versions and you want the one without the rotated entry.

I agree that these tubs will go on to stock manifolds, since they have very large hotside foot plates , meaning you can get the space to drill them and get them to bolt up. I dont think of it as a straight forward mod though!

Wierdly the front one will be harder than the rear, but i would think it will go with a thinner fan, or maybe a push fan etc.

Now Gowf has genuine td04 13g's on his and that makes lets say ball park 400 bhp. Since they are mitsubishi turbos it required the custom construction of manifolds, but i guess you could get a td04hl15g and drop those on to the same manifolds. the physical size is very similar. I've pushed over 550bhp out of a set of those on a 6g72.

The scooby turbo way will give rod bending ability in a 6a13 i would think. We use hybred 13T's ont gtos all the time and pretty much i can say that at 1 bar boost, on stock intercoolers, warlbro pump, 550s, pressure reg, fuel lines, and map2 i can make 400bhp. small increases in boost increase the power quite a bit. 1.5 bar will make nearly 500, but you run into charge cooling issues and resultant knock before getting there. 450 really needs a front mount etc on those.

So what i'm still not too sure about is just how far stock rods will go before they are getting pushed to their limits. Look at a 6g72 bottom end, its huge compared to a 6a13, and for daily driven cars above 450bhp, we reccomend forged in those.

Cheers,

Ben.

[Mante]

I can understand why you may want to stick with twin setup but why?? its harder to tune a twin, cost more and requires alot more space..

the trick is finding a setup (compressor size/hotside-cold side) that takes everything you want to achieve and go from there. Mine, literally instant boost, great mid range and holds way past redline.

Twin?? why bother. Put it this way all the "big boys" i.e supra's...skylines.. etc use single, YES I know they have inlines vs the v6 but its all done for a reason.

[djb160]

I think it's more of a keeping the car as it "is" sort of thing. I mean lets face it if actually cared about power efficiency that much we wouldn't be driving VR4s

[Turbo_Steve]

Twins spool better. It's that simple.

If you were talking about a sequential setup, Mante, I'd agree with you.
But Parallel twin turbos will always spool earlier than a single large one, even with twin scroll housings.

That said, there is a grey area, where a single turbo that is slightly larger than a pair of twins will spool roughly the same but tend to make more power.

However, for a daily driver, a single TD05 turbo delivers roughly the same performance as a pair of TD04s on these engines, but goes on to make less eventual power. A TD06 will spool significantly later.

Big single conversions are all about ultimate power: I've driven many (many many) big single conversions, and whilst they are fantastic drag cars, and work well on the highway, they are totally unsuited to driving on European roads (hence why they are so unusual over here) as you are forever changing gear trying to keep it in the power band. Of course, a smaller turbo could be fitted.....but then you lose out on top end power.

Manufacturers go to the expense of fitting paralell turbos because it makes a better drivers car.
Most VR-4 owners are in love with the sheer responsiveness of the engine (which is excellent) and don't want to sacrifice that for more power.

Valmes has a pair of TD05s on his, and after watching his videos posted, I wouldn't want to go that large: they are definitely slower to spool than the factory turbos, and it's right on the threshold of being "noticeable" whilst driving normally.

That said, the factory turbos do seem to be too small: Mitsubishi were trying for a small engine that felt like a much bigger one - instant throttle response throughout the rev range and effortless power with no thump in the back.
Which they achieved. Most owners want to retain this, and the TD04s or GT2554Rs seem the best way...marginally slower spoolu up but with much higher eventual flow.


As for being harder to tune....how?
Quick comparison to get to 400bhp.

Both setups need:
Intercooler
Fuelling
Inlet
coolant and oil lines
Boost control.
ECU Remap / piggyback / similar
Exhaust.

However, whilst the twin turbo setup will just require the turbos 'bolting into place' the single turbo will require:

Extensive manifold fabrication (transverse V6, so lots of piping)
Longer runnels leading to increased under bonnet temps
Longer runnels leading to increased spool times
Longer runnels increasing likelihood of cracked piping (seen this lots)
Completely reworked inlet piping
Most people find they have to change the throttle body to the other side of the inlet manifold to make room.

Assuming you go for MORE than 400bhp (and risk the rods which are an unknown): So lets say 600bhp.

Pair of TD05-18Gs will fit with some persuasion (radiator / fan changes).
Still using factory manifolds.
Will spool roughly the same as TD05-16Gs....which are about the limit of spool for a daily driver IMO. So gearing & rev limt need be unaffected.


Single turbo will be...hm...we're looking at a GT40R here, or an MHI TD06/07.
That's a big turbo to find room for.
Significant increase in oil flow & cooling required for one of these.
Much later spool as this is a big turbine to turn: on a 2.5L we're looking at 4000rpms at least.
Later spool means being caught "off boost" on gear changes. Rev limit needs to be raised (effectively doubling the risk of damage to the internals with every 500rpm it is raised) or the gear ratios need to be changed (expensive, boring).
You also now have a single turbo spooling up hard as soon as sufficient load is available, which generally increases the shock loading on all components.



There are also the efficiency arguments:
A pair of larger (but not huge) turbos will remain in their efficiency band, and at lower temperatures, than a single turbo that spools as early and flows the same. So two turbos need less intercooling, and potentially less fuel to soak up heat in the bores.


So I am going to have to strongly disagree that a single conversion is the way to go unless you're building a car purely for the out-and-out power.
In which case, I wouldn't start with a VR-4 LOL!


I hate to say this, but the i6 vs v6 thing is a big deal as well.....i6 can do a single turbo with minimal (even less!) pipework. V6 (especially transverse in a cramped engine bay) means a LOT of fabrication work.
That said, once you have converted to single, upgrading the turbo is a much simpler task, but that assumes you will keep changing it?

[Turbo_Steve]

In fact:
http://en.wikipedia.org/wiki/Twin-turbo


We're all trying to stay with:

use a light pressure set up where the turbos are designed to output less boost but spool earlier, however, this set up sacrifices top end power

Trying to stay towards the lower / middle efficiency band of smaller turbos will give a significant increase in power with marginal sacrifice in spool charecteristics, especially if the transition is mapped smoothly.


A big single turbo goes from NA to +ve pressure and is difficult to hide the spool up without sacrificing power until quite late in the rev range.

[Mante]

So I am going to have to strongly disagree that a single conversion is the way to go unless you're building a car purely for the out-and-out power.
In which case, I wouldn't start with a VR-4 LOL!

Sequential setup is even more of a hassle for tuning purposes but like you said it is GREAT once it is all completed.

It is ironic you mentioned 400 bhp, how about 368 awhp and 363 lb That was accomplished on crappy 93 octane or 98 MON as I've been told with a single Gt35R (not off the shelf made specifically for the 6a13) that spools around 3500 making it extremely driver friendly. Limited to those numbers because the internals are stock 100%) I've seen and read about Valmes setup, the shop that did my setup wasnt impressed (nothing against his car) with the power or overal layout. FTOLTD?? ( if I misspell it sorry) did a single conversion years ago, same overal dislike for the outcome.

Full explination of my setup dyno sheet, 1/4 runs etc etc..
My Vr4 site

I want to touch on the myths you mentioned..

Extensive manifold fabrication (transverse V6, so lots of piping)
Longer runnels leading to increased under bonnet temps
Longer runnels leading to increased spool times
Longer runnels increasing likelihood of cracked piping (seen this lots)
Completely reworked inlet piping
Most people find they have to change the throttle body to the other side of the inlet manifold to make room.

-manifold fabrication: true but these cars are 10 years old there are hairline cracks all throughout the manifolds SO why not replace them all while everything is apart ? I intended in using my factory ones but they were cracked pretty badly.

-longer runners= increased heat underhood: Ok any tuner should have a vented hood at this point. With a single setup it can (and should) be positioned under that vent to allow proper cooling

-increased spool: Not if research is done and you dont just toss out turbos. Alot of people walk in saying "I want a GT958359453245R to fit in my honda civic" lol Ive seen I've driven some cars with setups just like this flat out scary.

-cracking runners: same as stated above, there are methods to combat this. heat coating treatments and the BEST WAY. Finding a fabricator that can actually weld!! Alot of people think they can weld but there beads are an eyesore which leads to premature cracking.


-the last two (reworked piping and changind inlet direction) All I have to say I will let this photo speak. I honestly havent seen a setup as clean and makes as much power then mine (not braggnig going by numbers)

Single conversion photo

As for expense, well with a single turbo you dont have to buy two but that money you save will be used on piping so it works out to be EVEN!!

Now to play devils advocate, you were right that a twin setup should spool sooner BUT again that is if you grab something off the shelf.. "Lets take two evo turbos and use them.. wait lets rebuild them first" or scooby turbo's or 3000gt turbos. You will accomplish getting power but the setup was originally made for another car in mind.

Nothing is more appealing then seeing two turbo's when you pop the hood, you will always get the "ooooooH" factor for onlookers lol

[Mante]

I think it's more of a keeping the car as it "is" sort of thing. I mean lets face it if actually cared about power efficiency that much we wouldn't be driving VR4s

If someone wanted to keep a car as it is then no modifications would be done, routine maintance and drive it..

[djb160]

after stock turbo boost characteristics 3500 feels real slow and doesn't feel like a daily driver at all. You'd probably get used to it in time, WRX owners don't seem to mind that sort of lag.

[Mante]

after stock turbo boost characteristics 3500 feels real slow and doesn't feel like a daily driver at all. You'd probably get used to it in time, WRX owners don't seem to mind that sort of lag.

Do you even know when the stock 9c's are full boost?? 2200 ish so 2200 is acceptable but 3500 is slow?

[mpau009]

Do you even know when the stock 9c's are full boost?? 2200 ish so 2200 is acceptable but 3500 is slow?

I think DJB is referring to another single conversion from NZ, which has a GT3540 mounted. I would doubt that his setup would be on 'full' boost at 3500, but it is making some really good power for the early stages of its tuning.

I decided to go for the TD-05-16g from an Evo6 for my conversion, but with longer headers than your setup from what i can see. Not quite the extent or intent of your build, and i havent had it properly wound up yet, but i am happy with the compromise between spool and boost at 10 psi.

I reckon the main difference in feel is the way a single comes on boost. It is easy to take the td-03/7s for granted for that kick in the back off the line..

Still i would love a ride in yours or Tylers single setups.

Mante is there a decent build thread of your car? i was aware of it, but i would love to see some photos of the ideas you used. Heres a pic of mine, cobbled together from ideas kicking around here and personalised a bit. Im hoping to get some numbers back before christmas.

[djb160]

Yeah thanks mike that was what I was meaning, sorry if I confused you there, I should have explained a bit better.

[mpau009]

i don't mean to put a spanner in the works, but hasn't everybody just said the impretza has TD-04 as standard and these are goodone to go for.

That is all well and good.

However the table posted up shows TD05H-16G-7 for the WRX/STI/Forester.

where did this info come from and is it reliable?

Is the standard Impretza turbo TD-04 and WRX/STI TD-05 or are we all wrong about the TD-04.

just an observation, i am not having a go at anyone, i appiciate all the effort that goes into finding all this info and at some point hope to make use of it.

Oh yeah? weird, i dont know much about the subaru stuff..

I just had my turbos out, and i put the numbers off them into google to see what i could find out, and that came up. It may be BS, but i am pretty sure the VR4 info is correct, the numbers match up anyway..

Were there some single turbo STI subarus? or are they an aftermarket kit maybe it could be something to do with that. Otherwise with TD-05s there would be some formidable forresters out there

[Goku]

You just get the TD04 turbo's off the 98- WRX's and they preyy much bolt right up, with a little adaption.