Jump to content
SAU Community

Recommended Posts

Hi guys,

I've been looking into T3 turbine housings for the GT30 series turbos (specifically the GTX3067) which will fit to the standard manifold. Response has been high on my priorities, so the 0.63 IWG housing was originally considered, although was always a bit concerned about boost control.

Always thought it would be nice to find a turbine housing which sat between the 0.63 AR and 0.82 AR housings commonly used. I was pleasantly surprised to have stumbled upon a couple of options which might do the trick.

First one is the Cosworth 3-bolt outlet IWG housing (first pic below):

http://www.atpturbo.com/mm5/merchant.mvc?Screen=PROD&Store_Code=tp&Product_Code=ATP-HSG-075&Category_Code=GTH

Seems to have a really decent sized wastegate port too:

http://www.turbobricks.org/forums/showthread.php?t=103756

This housing is stated as having a 0.72 A/R, so sits nicely between the two "normal" housings. Seems to me like it could be a good balance between response, top end and also boost control. Downside is I would have to go with a less common outlet flange.

Second option, is this housing which has an external wastegate flange (second pic below):

http://www.atpturbo.com/mm5/merchant.mvc?Screen=PROD&Store_Code=tp&Product_Code=ATP-HSG-043&Category_Code=GTH

Quoted from ATP website:

"Special turbine housing, T25-EWG-44, T25 inlet flanged, GT28 style 5-bolt outlet. External gated (44mm flanged), flows like a full T3. Available for GT28RS/71R/76R, full size 60mm GT3071R/GT3076R/GT3082R turbine wheel and Stage III (76 trim) turbine wheel. Allows full T3 flow but on T25 flanged manifold. Only available in .72 A/R. .72 A/R is built to GT30 geometry so this A/R is comparable to one that is right in between a .63 and .82 A/R GT30R housing."

Note that although it's listed as T25 it can be optioned with T3 inlet. Also interesting here is that this can be gotten in form which fits the cropped GT3071-WG turbo, so may be a decent option to get HKS 2835 Pro S type performance at half the cost?

Overall second option is less appealing to me because it's another step away from stock looking, plus more expensive due to cost of wastegate and extra plumbing of dump.

Wondering if anyone has had any experience with either of these (especially Cosworth one)? Or even just thoughts on how these might perform would be very welcome?

Hoping the collective intelligence of SAU might shine some light on this.

Cheers,

Mark.

post-83859-0-80345900-1383819783_thumb.jpg

post-83859-0-40757000-1383819792_thumb.jpg

Link to comment
https://www.sau.com.au/forums/topic/434671-gt30-turbine-housing-alternatives/
Share on other sites

  • Replies 40
  • Created
  • Last Reply

Top Posters In This Topic

Thanks for the input lads.

Wolverine, guess I've been a bit worried about the response of the 0.82 housing. But equally concerned about boost control with the 0.63 as you've highlighted. Was kind of hoping the 0.72 Cosworth would be a good balance. Admittedly have maybe been a bit hung up on trying something different, e.g. like with the GTX3067.

Scotty, didn't realise ATP housings were crap too. Cheers for the heads up. Read about the AVO / Sonic housings, but not ATP.

So it really does seem the Garrett 5-bolt outlet housings are the only viable IWG option?

I have an ATP twin scroll housing on my GT3076 at present. I had an open Garrett 0.82 on the GT3082.

There isn't much difference between the two (ex gate) on the road. The casting of the ATP housing was ok unlike their T4 based housings (perhaps what Scotty is referring to?) but internally and externally it is a lump of metal that houses the good bits just fine.

In IWG config I wouldn't stuff around with a 0.63 just in case you get boost control issues. Results for GTX3067's are like rocking horse poo on RB's so if you feel like being a test pilot the 0.63 would be the way to go. It might be ok, it might not.

ps. looking forward to seeing your results

I just haven't seen a good result from many ATP housing variations, and they look like they were cast in a kindergarten compared to my Chinese knockoff Tial. They have a good range of housings though.

My mates Subaru specific ATP GTX3071 housing was missing a fair chunk of power compared to my old setup. Every time I hear about ATP it is generally someone complaining about low outputs and poor quality, which is why I now make my own.

As much as I'd like to go external gate, it's most likely overkill for what I'm after. Looking at mods in your sig Scotty, I think we're playing in different leagues, haha.

Part of the reason for the GTX3067 was to try and get GT3071 like peak power (260kW+), but with full boost by no later than 3,500rpm. And to try something new. But yeah paranoid about boost control with the 0.63. Not sure if you guys saw, but pic below is of the wastegate port on the Cosworth housing. That's a US Quarter Dollar (24mm dia) he's holding. Port looks decent sized to my eyes.

Based on the comments, I think I need to do more research. Last thing I want to do is have this (Cosworth) housing be more restrictive than the little Garrett, or worse, crack and fail on me. If the design is good, but ATP castings are crap, maybe I need to look at getting a second hand OEM Cosworth housing from somewhere...

Will definitely post up results when done, but to be honest probably isn't going to be until early-mid next year.

Appreciate the input lads.

post-83859-0-71139800-1383863914_thumb.jpg

Stao does large port wastegates like that, but his are machined. I can't see how the puck would seal properly in that pic...

Can you get the flange? You wouldn't want to have to make that one...

You will need a heavy duty actuator which adds to the cost.

External gates offer much more flexibility with boost settings, and controller setups. They aren't just for high boost applications, although you would need to go external to control boost well at 25+psi imo. If you ever go e85 you will be kicking yourself if you didn't think about it at least.

do it, external gates are mad...

In all honesty, it's better that you direct the heat & exhaust pressure from the rear housing via the external gate. I've tuned a few of my mates car and I find that cars running external gates like more timing (due to less exhaust back pressure) and also better if you're hammering it on the track as the housing doesn't get as hot as an internally gated housing (just my theory).

and the bonus, you can run a screamer pipe on the gate

I'm only theorising but my gut feeling is that the GTX3067R would probably need the Garrett GT30 0.63 turbine housing to spin it fast enough to get boost at reasonable engine revs . From what I researched , the fella at GCG agreed ..

More later been chased away by missus , cheers A .

I'm only theorising but my gut feeling is that the GTX3067R would probably need the Garrett GT30 0.63 turbine housing to spin it fast enough to get boost at reasonable engine revs . From what I researched , the fella at GCG agreed ..

More later been chased away by missus , cheers A .

The plot thickens... look forward to hearing more on this.

do it, external gates are mad...

In all honesty, it's better that you direct the heat & exhaust pressure from the rear housing via the external gate. I've tuned a few of my mates car and I find that cars running external gates like more timing (due to less exhaust back pressure) and also better if you're hammering it on the track as the housing doesn't get as hot as an internally gated housing (just my theory).

and the bonus, you can run a screamer pipe on the gate

Hehe. Even if I did go external would plumb it back. It's going to be predominately a road car so avoiding unwanted attention is a priority. What are the chances of getting an engineering certificate for an external gate set up?

Stao does large port wastegates like that, but his are machined. I can't see how the puck would seal properly in that pic...

Can you get the flange? You wouldn't want to have to make that one...

You will need a heavy duty actuator which adds to the cost.

External gates offer much more flexibility with boost settings, and controller setups. They aren't just for high boost applications, although you would need to go external to control boost well at 25+psi imo. If you ever go e85 you will be kicking yourself if you didn't think about it at least.

Yeah the flange is a concern. Haven't seen one off the shelf, but was planning to talk to my mechanic who's pretty nifty at fabricating manifolds and the like. Plan is to get the "big can" actuator, which is 18 psi I believe.

Good food for thought re: external gate. Maybe I really should be looking at a housing with a more standard outlet. That way if (when) I decide to upgrade again later, can get a larger housing to bolt up rather than having to make a new dump pipe again.

ive tried both the open 0.82 external on a gtx3076 and a twin scroll 0.82external on a gt3076 and if doing things again id opt for an open garrett housing, and make it a 0.63 internal to save money and drama's, Mafia's internal 0.63 worked well from reading his old thread but needed a larger flapper..

I think the 0.82 is not really suited to the street, a 0.63 with ethanol/meth is best match for street I reckon

anyone know if the xr6 ford style rear could be used with an xr6 turbo dump?

Edited by AngryRB

I just haven't seen a good result from many ATP housing variations, and they look like they were cast in a kindergarten compared to my Chinese knockoff Tial. They have a good range of housings though.

My mates Subaru specific ATP GTX3071 housing was missing a fair chunk of power compared to my old setup. Every time I hear about ATP it is generally someone complaining about low outputs and poor quality, which is why I now make my own.

Different engine = different result so it is hard to compare IMO. Was it the same dyno?

I don't think I got a low output from mine but always hard to tell unless we do a back to back swap. Certainly I would have used a Garrett made housing IF one was available in TS at the time. Then again I would have just re-used the open housing I had and saved myself a fair bit of stuffing around.

I'm only theorising but my gut feeling is that the GTX3067R would probably need the Garrett GT30 0.63 turbine housing to spin it fast enough to get boost at reasonable engine revs . From what I researched , the fella at GCG agreed ..

More later been chased away by missus , cheers A .

I think you might be right but if there are boost control issues then the cost of changing is a pain in the arse. External gate it would be a lock.

Different engine = different result so it is hard to compare IMO. Was it the same dyno?

Same dyno, same response from the tuner. He said the compressor is maxed, I said 'how can it be maxed if I am running 80kw more.' Unfortunately it is hard to pinpoint exactly what is restricting flow as everything is restricting flow to some extent.

With a large enough turbo you can push through some of these restrictions. (as the turbo is the main restriction in all turbo setups.) Strangling the engine with a small rear is not a good way to get power or response. If you want a more responsive turbo, go for smaller lighter wheels like a gtx28 or SS1pu. It will probably still get close to 300kw if all the restrictions are sorted.

Something funny happening there with an 80kw drop.

I couldn't agree more with matching the turbine and compressors. In this case a 0.63 with good boost control is close enough but there is no doubt a 0.82 IW will be fine.

Not helping the OP much now but 0.63 with the right wastegate flapper is probably worth being a test pilot.

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now



×
×
  • Create New...