Turbine housing question
 

Okay my new manifold is T3 based so my next turbo will have a T3 turbine housing, but in selecting turbine A/R, I had a question, hoping that someone out there has some input based on experience rather than just theoretical numbers as it is next to impossible to find a turbine map for a GT wheel inside a T3 housing. Info to follow:

I will be using a GT2871 turbo for my next setup, BUT I will be getting the T3 turbine housing option. So considering that the T3 turbine inlet is 136% greater than the T2 inlet (38mm x 49.3mm vs. 44.7mm x 57.2mm from Garrett catalog) and considering that the turbine wheel is the same regardless of selected housing, I was wondering if anyone had any comparisons between the different A/Rs T2 vs T3 housings.

For example, I know that on Brad's ~1.7L his T25 0.64A/R turbine housing spools to ~15psi at 4400rpm based on his MS datalogs (with a boost leak no less) so I was wondering will the 0.48 A/R T3 housing be comparable or too small/restrictive or will the 0.63/0.64 T3 housing be about the same.

FWIW, I'm shooting for ~325-350rwhp and would like the broadest powerband possible (aka early spool, relatively speaking) so if the 0.48A/R T3 housing would still make it to 350 and provide earlier spool, I'd opt for that, but if it's going to choke off the top end then I'd go bigger.

Sure, I could just go with the 0.64A/R, but if the 0.48A/R can get me to my goal AND give me an extra 500+/- rpm of spool, then even better. Granted this may be mentally masturbating over a very small detail, but it's all about selecting the "right" turbo for your goals and the devil is in the details.

how much are the turbine housings? maybe you can buy one and then later get the other housing.

Yeah, I thought about that as well and new housings are ~$150-175 + shipping and then ceramic coating it is another ~$50 or so. Granted that's always an option but it's also an extra expense and effort that I'd prefer not to go thru.

Also forgot to mention, that the 0.48A/R option only comes on the 4-bolt T3 housing whereas the 0.64A/R option comes on both the 4/5 bolt housings. Not a big deal, but the 5-bolt leaves the T3/T4 replacement/upgrade/options open in terms of changing turbos later. :vash:

From my reading, id go with the .64a/r, espically with a 2871, thats a somewhat small turbo.

Just FYI, my T3 has a .64 a/r houseing, and I get full boost (10 psi) at 3750 rpm in 3rd gear.

I got your PM, and I don't have any hard data for you. My gut says to go with the larger housing, unless you are using an external wastegate, in which case a 40mm wastegate and the smaller housing could make it viable to make plenty of power while not constricting flow terribly. This is where a larger external wastegate and bring some balance to the equation.

Mark

PS - the 38 mm external wastegate may flow enough as well.

from what i can determine, the .86 a/r t25 housing is comparable to the .63 a/r t3 housing. both have max efficiency at 21 lb/min. that being assumed, the .63 a/r t25 housing and .48 a/r t/3 housings should be comparable as well at 16 lb/min. doing the math with a 1630cc motor and 8500k, that puts you at 14-15 lb/min of exhaust flow.....

All 3 of my .48 housings on the shelf are 5-bolt, so where did you get your info from?

is for t25 turbine (54mm vs 59mm). i assume ATP gets them machined specifically for such.

I have an extra .48 housing with a big ol f'n crack in it, if you want to get the .63 and then want to try the .48 it is in good enough shape to run a while to see if you like it or not. It's standard 5 bolt.

Are you going to track this car? If you are, go with the larger turbine because it should help reduce underhood temps a bit.

I intended to buy a t3s60 with the .63 turbine but the company I bought from was out of the .63 turbines so I had to go with the .48. I plan to "upgrade" to the .63 some day.

If Tom can run his turbine at max efficiency without choking out, more power to him (figuratively speaking).


I run a .48 myself. My engine produced 13 lb/min of airflow volume out my exhaust at 7000RPM. no reason to rock turbine designed for max 21 lb/min when the .48 is maxed at 16 lb/min.....


it's one of those compromises, while the .63 will probably produce more power per psi, it will spool slower coupled with his tubular manifold and produce a torque curve he doesn't want. If the .48 can satisfy his power goals and reduce the spool 500rpm, then again more power to him.

Thanks for the offers and info. As for the 0.48 A/R only being offered in the 4 bolt T3 housing, that's the only way ATP offers that A/R for the 2871.

If I had a 1.8L, then without hesitation, I'd rock the 0.63/0.64 but considering the 200cc less displacement + longer runner tubular manifold + turbine flow estimates based on other T3 turbos, I think that the 0.48 will be better matched for my goals. From further research, I'm pretty sure a 0.48 A/R T3 turbine housing should flow very similar to the 0.64 A/R T28 turbine housing. So if Brad got 15psi at 4400rpm with the 0.64A/R T28 housing then I should be in a similar ballpark. Sure max power can be fun but so is spool/max boost before 5k rpm, regardless of redline. Granted if I was shooting for 375-400, then I would go larger, but my power goals are more modest. ;)

If you are getting a mani made from scratch, to me it seems easier to get a standard T2 turbine, that way you know what the flow rates are or at least it would be less of a guess. Why go to a T3? If a .48 T3 flows like a .64 T2, they why not use the T2 and simplify things?

Agreed that T2 footprint would have been easier, but either the manifold was already partially made/assembled or he already had all the materials, but it was being sold as a T3 manifold with extra cost to adapt to T2, and it didn't make sense to do that to me since ATP had T3 housing options available. Not 100% sure, but what's done is done.

t3's are bomb....now its easier for him to upgrade to a 3076 or a t3/t4.

Yeah that's true. Seems to me to be a lot of serious coinage to be dropping on a kinda unknown combo (2871 + .48T3). If tom picks it right, it'll be sick though!

says mr two gt15s.

why not just use a 60 trim or even a super 60?

he's looking to make OVER 300rwhp...s60 would be reaching limits at 270-300rwhp.

a 60-1 or sc61 would get him over 300rwhp wouldn't it? whatever term you want to call it anyway.

which is a very bad idea

Those are a bit big for my application. Based on the compressor maps of the 56 trim 2871, 18-20psi will put me right in the center of the middle efficency island while also providing reasonable spool. Any number of turbos can get you to a peak power number, but I didn't want to be running out of steam with the super 60 or disco potato and I didn't want a oversize turbo (relatively speaking) like the 60-1, SC61, T04E 57 trim for a 1.6L that would spool really late giving me a very small powerband. Having had several turbo kits and the opportunity to drive many turbo cars, I'd say more important than just the final peak numbers is the shape of the torque curve and power delivery. My 2 cents.

whatever term? you mean which turbo?!

T3 Super 60 = T3 Super 60
T3 Super 60 :ne: T3/T4 60-1

T3 Super 60 = 60mm compressor wheel
Garrett 60-1 = 76.2mm compressor wheel
60mm :ne: 76.2mm

T3 S60 = 60mm
T3/t4 60-1 = 60 trim
60mm :ne: 60 trim

T3 Super 60 = 275-300HP
T3/T4 60-1 = 600-640HP
300HP :ne: 640HP

:noes: :noob:

ha ha, be nice, I have never needed a turbo that big so got them confused!!

Been thinking about this a bit more and making a big assumption, here's my take:

The assumption that I'm making is that since the turbine wheel stays the same, that the overall radius of the T28 and T3 housings for the same turbo stay pretty much the same.

Making the above assumption, the turbine A/R is simply a ratio of cross-sectional flow area to the distance from the center of the housing. So to make the math easy, let's say that the 0.64 A/R T28 housing has an inlet cross sectional area of 100 units squared. So, at a radius of 100 units from center, the cross sectional area would be 64 units squared.

Since a T3 housing has a 133% larger inlet area, the beginning inlet cross sectional area would be 133 units squared. Then, at the same radius of 100 units from the center of the housing, the cross sectional area would be 63.84 units squared with a 0.48 A/R (0.48 * 100 * 133/100), and 85.12 units squared with a 0.64 A/R (0.64 * 100 * 133/100)

Thus, assuming that cross sectional area can be taken as a surrogate for flow, the 0.48 A/R T3 housing mathematically should flow very similar to a 0.64 A/R T28 housing, and a 0.64 A/R T3 housing mathematically should flow similar to a 0.86 A/R T28 housing.

Yes, trying to convince myself that I'm making the right choice. Yes, it's called rationalization.

Assumptions suck.

I don't know if I would assume the A/R of the T3 housing (or the T2, or any turbine ever) is the same starting at the inlet to the turbine. Just because the hole at the inlet flange is the T3 size, doesn't mean 1" into the turbine it's still the same amount larger than the T2 turbine. My GT1548's have a T2 flange but neck down very quickly after the flange. If I were ATP and making custom cast turbines, I would copy the geometry of the original, except with the modified flange attached. So .64A/R T2 would be the same as a .64A/R T3. But that's my logic. If they are taking standard T3 turbines and machining them to mate to the GT28 center section I dunno then.

I would call and talk to some Guru at ATP.

I may have misunderstood your point, but the shorty tubey should spool better than a log mani.

Thanks for the suggestion. I did e-mail the tech guys a bit back and finally got an answer today pasted below. No turbine flow map, but better than nothing! Woohoo!

"It's just as you suspected, T3 of the same A/R is bigger than T25.

T3 .48 has similar turbine flow numbers as T25 .64 A/R.

This is due to the two being different "framesizes" and so the scaling of
the dimensions are different.

Thank you."

Tom's is not a shorty tubular. I'd expect it to spool slower than a log / shorty tubular, but make more top end.