Fuel type actually isn't relevant. There is X amount of CFM required for Y horsepower. That is where the power rating comes from. Like I said, we'll see how it goes. The compressor is 9mm smaller than even their GTX wheel and they're claiming similar power. And the compressor map only goes to 165k rpm. Which isn't exactly "off the charts" crazy for turbine speed, especially since the wheels are small.


Don't get me wrong, I'm excited to see what these can do. Because if they can light at sub 4k rpm's and make 600whp; I'll be buying a turbo kit. I'm just skeptical. They do have a large inducer though, the 660 has a 54mmx67mm compressor. My red is 54.8x76.

That's like saying moving from standard steel valves to stainless steel exhaust valves will result in higher power because you're not melting your exhaust valves anymore. I have development experience with the family of engines you're referring to and I can assure you that the material change isn't the sole determining factor for a hotter tune on the RS vs the mustang.

Furthermore, heat is not indicative of power. If you crank the timing way back you'll increase exhaust temps, does that mean you gain power? If you're getting temps hot enough to approach the melting temp of a turbine wheel then I can assure you that you're waaaayyy off to the right of the compressor map. Or you're not throwing enough fuel at the engine.

Also, inconel is not one material, there are many different grades of it and it's progressively gotten better over the past 10 years. Nowadays there are inco grades that have melting points 1300+°C so the use of mar-m doesn't necessarily mean it's a superior material, rather an improvement over what Garrett was using before, which means this can't really be related to turbos made by other companies. (Inco and mar-m are both high nickel steels so their properties are pretty similar).

I can't verify, but I recall the wheels in the GT/GTX were Inco713 and MarM247 was common in Garrett turbine wheels. I want to say MHI used MarM246. My memory is a bit hazy.

Notice I didn't say result, but allow for higher power and durability. The same way stronger connector rods don't make more power but allow an engine to make more power for a longer time without blowing up. Since you've done development work on the engines, you've probably seen turbine wheels that were over-temped and had the inducers of the blades fall apart.

And you are correct, EGT is not the sole indicator of power. And as you noted, timing and A/F ratio affect EGT. I remember my 2005 Evo ran richer than 10:1 stock at WOT. Lots of fuel thrown at it right? And of course, going leaner will increase the EGTs up until the A/F ratio is about stoich.

So let me rephrase, the MarM alloy Garrett is using has a stated temperature limit of 1050C whereas the previous Inconel alloy was ~980C. And this doesn't mean that if you go to 1055C, the wheel will suddenly fall apart, but the more the temp limit is exceeded the lower the fatigue life. Take the Inconel Garrett was using and run it at 1050C and it will have a much shorter fatigue life than the MarM wheel. Sure, the inconel wheel will work at 1050C for a period of time, it'll just be much shorter than compared to marm. Same deal with the GT/GTX ni-resist turbine housings vs. the stainless steel housings used in the G-series.

Situations in motorsports where higher EGTs are encountered include when they're trying to maximize fuel economy at WOT and also anti-lag. Time attack guys also like to spin the hell out of turbos, way off the right side of the compressor map.

Garrett typically tests the turbos up to the speed limit of the compressor wheel (though not always). And often times, the speed limit of the compressor wheel is significantly lower than the turbine wheel. There's a reason all the Garrett compressor and turbine housings are certified to contain a wheel burst:
http://www.motoiq.com/MagazineArticl...ntainment.aspx

But it's safe to take the max speed line on the compressor map as the speed limit for long-term fatigue life. Exceed the speed limit and expect shorter fatigue life.

So I've been in the no "f$#ks to give" department when it comes to cars. Last time I was looking, the EFR turbos came apart at the first overspeed they encountered. Not even like a massive overspeed condition where the car blew a charge pipe at peak power or anything like that... just a "I upped the boost a little and the turbine exploded." Did something change and some how the EFRs became reliable?

That was crazy to me, as I spent years in the DSM world where I literally disconnected my wastegate line to tap out boost every weekend to go out an slay every contender I could find. I never missed an opportunity to rip through the gears like an *******. Yet...I never had a Garrett or MHI turbo fail on me and I guarantee I was overspinning the hell out of them at 4500' elevation...hell I did canyon runs up to 9000' in elevation.

With regards to these turbos... If these are $2500 turbos, then nope, not interested. $1500 for a GTX was hard to swallow. I'll take a journal bearing turbo with the dial set to 11 for $1000 and deal with 500rpm more lag. Don't get me wrong, I love the technology advancement, but when a single turbo cost more then the rest of the engine build, it's getting a bit absurd.

But 11hp/lbm of air is pretty reasonable on a modern motor (at the crank). An 4G63 with some decent head/intake work and a good manifold will be similar. Crap manifolds and stock head won't do it though.

EFR turbo's still do it they are not designed for that ****. BlackE I believe had a 9180 go out due to a surge problem I believe its documented here .

There is the GTW line which are still really solid turbos. Just lower grade materials, older turbine aero, and no water cooling (can't remember if none of them have water cooler or some actually do). So just have to be cognoscente of those limitations. They are perfect for more budget builds that are primarily fun street cars not doing 20-30min road course sessions.

I actually give BW a ton of credit for developing the TiAl turbine wheel. Don't forget MHI had them in the TME Evo6 and those were prone to shooting the TiAl turbine wheel out the exhaust. MHI also used TiAl turbine wheels on the Porsche Panamera. Those turbos also had failures and Porshce had to issue a stop build order on the car; turbine wheels were falling off allowing oil to get into the hot exhaust and catch fire. The first round of EFRs, they did have really really high failure rates. Like nearly every one. But now, they seem to be quite reliable. The only OEM production turbo I know of using TiAl is the IHI on the Ferrari 488. Just a little bit of history, Garrett had ceramic turbine wheels way back in the day in the 1990s and they were super light. Nissan used them on the Skyline and/or Silvia. But they also shattered like glass when hit with FOD. These G-series turbos took a different approach, reduce inertia by reducing diameter while maintaining the same flow.

I'm really looking forward to results from these. Also, when the next size turbine wheel comes out, things are going to get rowdy!

I agree Garrett is promising a lot from these G2560 G2567 turbos and their quoted figures do look optimistic based on wheel sizes and trims .


I'm also not a fan of Vband fittings to mount on an exhaust manifold .
Full Race has outline diagrams of these turbos a shows a T4 TS mounting .
I hope Garrett follows Borgs ideas and has a few T3 mount housings with integral gate as well .


Also I'm interested to see what the next size up will look like turbine wise .
I'm guessing that there will be a G30 version and that the GTX3584RS will possibly become the G35 .


It is good to see that Garrett have finally got off their butts and developed more modern turbines for the performance industry . The GT28/30/35 have been around for a LONG time , think around 27 years .
I'm going to have a stab and say that Garrett may have changed the material to MarM because it may have qualities other than temperature advantages over Inconel . I suspect that the hub area of these turbines is less bulky than the GT ones , so with possibly more blade area they appear to flow more for the same approximate overall dimensions as a GT28 NS111 turbine .
It seems they use slightly different machining processes for the compressor wheels as well , again probably less hub area than the GTX wheels for their major diameter .


Time will tell .

First of all sorry for my English. I've just ordering the new g25-660 as a part of a turbo kit build on my specifications. I know this is a risk because nobody (that I know) has tried this turbo on an evo, but the flow chart looking very promising for 550hp (crankshaft) applications. Now is time to waiting for the shipping :D

Keep us posted! I was going to go down the MHI route but am holding off my turbo purchase because i'm waiting for real world figures on these:D.

Ironic I made 547whp on the first gen gt3076r no X. Since then with more octane and a few tweaks added another 35whp at my local dyno.
Just as I stepped it up to a TS g2gtx3071 (CB long block 2.0L) this comes out... Well hope to have maxed out 3071 results coming in the next month to see how things go.

That's got to rip your undies, though still will be interesting to see how the G2 GTX3071R goes as that still looks like a decent thing in it's own right!

Very very very interested to see how the G-series go, they really could be serious performers.

i was buying my Garrett G series 660, and received. I am waiting the sheapey built turbo kit to put on my Evo 8 RS.