What turbine wheel and A/R housing is just as important as the compressor The GTX 76mm compressor wheel really needs the larger GT35 turbine wheel. Are you trying to do 550whp on pump gas or E85? I just noticed your T4 twin-scroll manifold setup. If you're on pump gas, The EFR7163 is too small as is the GTX3071R (Gen2 flows a bit more than the 7163). The GTX3576R should do it. It looks like the only turbine housing option you have is the 1.06 A/R which should be fine. If you're talking E85, the GTX3071R can do 550whp.

 

It's going to be 550whp on e85! Wow I had no idea, I called ATP to be sure and their reps told me that the turbinewheel/compressor missmatch is a 3067r thing (not to be confused with 76r). I will be going with the 1.06AR as that's the only divided housing they offer.

 

No, the 3076 is a mismatch. Talk to the owner George at ATP. He tested the 3576 vs. 3076 on the EvoX like back in 2008 and picked up 50whp with no loss in spool. Actually, the 3067 was created because the 2867 is a mismatch. The 3067 made 25whp more with the same spool. If you're only targeting 550whp on E85, go with the GTX3071R. It will be at the edge though, maxed out.

 

I am thinking of going a completely different direction. I want to explore the G30-660 with a .82 AR OPEN T4 (https://www.atpturbo.com/mm5/merchan...ry_Code=G30660

How do you guys think this will fair? Will I be shooting myself in the foot with the divided T4 full race manifold to the open T4 turbine housing, will the .82 AR choke up top?

 

I like it! People have pushed 0.82 A/R housings over 700whp, it won't be an issue. It's hard to say how laggy it will be though because I think that's an ATP cast housing and not an official Garrett housing, so no turbine flow map. Eye-balling it though from the picture, it looks like about the same volute size as the tried and true Garrett T3 flange 0.82. I think you've found a pretty damn good solution though allowing you to keep your current twin-scroll manifold. Full-Race has been selling what's know as the 1.5 setup on STIs for a long time, twin-scroll manifold with open volute housing. Honestly, it's basically like a Tri-Y header with the 1-4 and 2-3 merging and then into the main collector being the turbine housing. Also note the G30-660 is the 60mm turbine wheel with the 67mm compressor wheel, so it's the new version of the GTX3067. It should be very responsive (better spool and transient response than the GTX3071R) and meet your power target.

 

Why not the G30-770? It's got a great compressor match. Even though it's the new 3071, Its compressor map is more like the GTX3076R.

Apparently, we think along similar lines. I have an ATP undivided T4 housing in 0.82 a/r for a GT3076R sitting in my garage. The plan was to dyno it and my S300SX FMW back to back. Need to replace my clutch first though.

Tim

 

Good question, for my build i will be limited to 550whp. This is because of my clutch and stock ap2 trans. In fact people have dropped second gear with less torque/hp BUT, I will be banking on boost by gear to keep numbers lower in first and second to preserve the longevity of the trans. So you might be wondering why keep the ap2 trans at all? Well frankly I enjoy the shifter feel, its a great transmission. So to compensate I would like to have the most quick response and fastest spool. Although the 770 is rated for more power, and the 660 and 770 have nearly identical compressor maps, the only difference being the 660 has nearly 15,000 additional compressor speed (which should translate to perhaps at 12-13% faster response) which should aid in my goals.

But I might be talking way out my a$$ here lol

 

Spoke with ATP again, they advised against it claiming the gtx3576r with a divided 1.06AR T4 is the right direction.

I am still thinking about going against their recommendation and grabbing the g30 660 open .82AR T4 as an expensive test, haha, and perhaps in the future, if they or someone else does make a divided housing all I would need to change is the housing instead of needing to replace the entire assembly.

Hopefully, the .82 doesn't choke! That's my biggest concern.

 

Correct me if i am wrong but you and ATP are battling out the notion of divided exhaust pulses vs merged exhaust pulses into the exhaust housing of the turbocharger right? I think your intuition about the merged G30 will be dead on with how boost will come on.

The new g series have optimized compressor technology and different alloys that should make them superior. If you want to stop at 650 ish i think the g30-660 should be an excellent option. If you want more headroom i would opt. for the g30-770 like others have suggested. Either way the G30 SHOULD be the more efficient choice vs the older GTX stuff from the technical documents i have read from Garrett...

 

Well initially the concern was if my manifold is divided and the exhaust housing is merged than either you won't see the benefits a true fully divided system would provide (and I believe this is true), BUT even if we were to go full divided with the 1.06A/R gtx3576r, would the open .82AR G30 660 still outperform? I suspect so. I think the divided manifold -> merged exhaust housing g30 660 will still provide better transient response and faster spool than the traditional setup, I suspect the technology is just that good. And in the event that say Garret, ATP or another 3rd party comes out with a compatible divided t4 exhaust housing all i'd need to replace is the housing and not whole assembly.

Apologies if this is all confusing!