Its ok, I've found something useful...

Basic T3

Basic T4

 

There are two different T4 sizes on the inlet...

 

and they are....?

 

I will attempt to explain this...

forget the flange as a tell-tale. Mr Mitsu is right in saying there are two T4 flanges, but I've only ever seen a "T4" flange in one size - Garrett makes some T4 housings with a T3 flange, so I suppose you can say the T4 has two flanges available, T4-sized and T3-sized! Seems confusing, and it should, so don't worry abou that for a few seconds, and lets look at the important stuff... GAS FLOW!!!

(all the quoted numeric data is from the 2002 Turbonetics general catalog)

T3 stage 1 = 1.918" exducer, 2.319" major (a true "T3" turbine)
T3 stage 2 = 2.122" exducer, 2.559" major
T3 stage 3 = 2.229" exducer, 2.559" major (T300 turbine?)
T3 stage 4 = nonexistent
T3 stage 5 = 2.439" exducer, 2.798" major (T350 turbine?)

T4 N trim = 2.071" exducer, 2.922" major
T4 O trim = 2.296" exducer, 2.922" major
T4 P trim = 2.544" exducer, 2.922" major
T4 Q trim = 2.693" exducer, 3.111" major

phew!

ok, so keep in mind that a T3 exhaust housing is sized for a T3 turbine, and a T4 exhaust housing is sized for a T4 turbine...

also, keep in mind that A/R is a relationship between area and radius; saying "0.63A/R" does not imply a capacity to flow any particular amount of air, because it only defines the relationship between area at radius for that housing.

lets get back to the numbers...

If we take the T3 turbine major diameter to be 2.32", then the maximum cross section of a T3 0.63A/R housing can be worked out quite easily.

(2.32") / 2 * 0.63 = 0.731 square inches max.

If we take the T4 turbine major diameter to be 2.92", then the maximum cross section of a T4 0.63A/R housing can be worked out quite easily.

(2.92") / 2 * 0.63 = 0.920 square inches max.

So, for a given A/R RATIO, the actual flow area depends on the design of the turbine/housing family as well...

(incidentally, a 0.63 housing is not available for a T4, but a 0.58 is)

The other thing to consider is the turbine circumference. A T3 @ 2.32" has a circ. of ~7.29" inches, while a T4 @ 2.92" has a circ. of ~9.17" - this means the gas will flow through a longer path with the T4 turbine for the same turbine RPM when compared to a T3; the flow path for a T4 turbine is ~25% longer than the flow path for a T3 turbine. This is a bit of a generalization, because the air is directed away from the circumference by the turbine blades, but at low engine speed (when there is insufficient energy to really spin the turbine with some enthusiasm), all of these things are factors to consider because they all affect spool characteristics. You could get a turbine housing with the same cross-sectional area for a T3 and a T4, but the T4 would spool a little later due to the longer scroll length caused by the larger circumference. Now add to mix the increased leverage on the shaft provided by the larger diameter turbine wheel, and it should spool quicker because there's less load? well, maybe, but that's going to be very hard to prove or disprove. What about the fact that the larger diameter turbine will be more capable at trying to muscle around a large diameter compressor wheel (like say a T61?) when you start trying to crank up the boost?

...all of that numerical nonsense gets thrown out the window when you get to some critical RPM though, and the T4 makes WAY more torque than a puny little T3 could ever make because the flow paths involved with the T4 are so much larger than those in a T3... hit 5000rpm in a T4 and tell me why you'd want a T3? (I suppose "driveability" is a valid reason, and you will undoubtedly get better driveability and response from the smaller T3 setup)

blah blah blah

I'm losing my train of thought here, so I am bowing out - I hope you can gain some insight from my babble

 

Thanks for the information.

I think I get what you're saying.

I also found from having a look around that there is a divided and not divided flange for the T4.

 

510rob that was very informative and i don't think i've ever had it broken down like that before. it's another piece of a huge puzzle - choosing the proper specs for a turbo.

thank you.

 

so on a T3/T4 which side uses which and what trim/stage is typically used?

 

T3/T4 hybrid SHOULD be a T4 compressor powered by a T3 turbine.

If you don't go too wild on the mismatch, these setups can give you good spool characteristics, and make lots of power - take the 50 trim T04E compressor that pairs up undeniably well with the T3 "stage 3" turbine setup and a 0.63A/R turbine housing. It's hard to argue that it won't work when so many people have had such great successes with that combo (lots of guys dynoing over 400HP at the wheels!). On the other hand, if you paired a 50 trim with a standard T3 wheel ("stage 1"), I'd say the smaller turbine is choking the crap out of the compressor so it won't work very well.

Some people prefer the T04E 46 trim compressor over the popular T04E 50 trim unit and say it has faster spooling at the expense of slightly lower top end potential.

I happen to think the T04E 50 trim compressor (and I'm sure lots of people would agree) is the magic compressor - VERY high efficiency, with a VERY wide operating range, with VERY high flow capacity = sounds almost too good to be true! ...but it isn't. But I would call it a "max effort" street unit that is capable of BIG HP numbers at the sacrifice of some delay in spooling response. Pair it up with a stage 3 wheel and it will spool reasonably well. I had talked to Turbonetics about pairing one up with a Stage 5 turbine, but from the practical reports on this site, it seems unnecessary (adds cost for little gains, as well as increasing "lag" a bit)

Some people have reported surging problems with the 57 trim T04E compressor when they get up over 15psi on a stock motor - this matches what can be seen on the compressor flow map so it shouldn't really come as any big surprise.

Some people go for the 60 trim T04E for their own reasons, but if you lay all the flow maps against each other, the one map that stands out is the 50 trim unit.

(Well, now I've committed it to internet print, so let the flames begin! All the lurkers can now tell me why I'm wrong, which is fine, but please have some firsthand proof or data to back up your ideas/opinions - I like learning, not needless arguing)

 

Ahhh, yes I just asked my friend who has a 240SX S14 with a T3/T4 on a built KA24DE engine. He has a T3/T3 with a 0.63AR and 50 trim. Im assuming this is the most commin set-up?

 

510Rob, your math is wrong, but your overall explanation is still a pretty good one.

http://www.turbobygarrett.com/turbobygarrett/tech_center/turbo_tech102.html

The radius is not the radius of the wheel, but the radius from the center of the wheel to the center of the cross-sectional area.

There is actually a bigger (not smaller) T4 family then the T4 you are use to hearing about.

 

Hmm, good info. I am looking fot a smaller turbo to bolt on to my greddy t78 manifold and dowmpipe.

 

99 GS-T, the numbers are not necessarily incorrect if I say the error was in my quoting the numbers as "max" instead of "mean" (that's my way of BSing out of the fact that you are correct, but the original numbers are still valid for a different position in the housing! hehehehe... Either way, cross-checking each other's work is always a good thing for a healthy community debate/discussion so I'm glad you caught that)

With that in mind, I ran some "clean" numbers in Excel (well, a little bit more clean anyway)

The statement "0.63A/R ratio" implies an infinite relationship, but the particular turbine's major diameter must govern the maximum useful size for the given A/R relationship. For a T3 turbine with a major diameter of 2.32", the centroid radius of ~1.75" gives a radius that equals (centroid radius - centroid area's own radius) equal to 2.32", and at that point, the nozzle area should be 1.103 square inches, and 1.185" centroid diameter

blah blah balhb adlknfmgas blah blah blah...

So, here are the numbers (numb-ers? hmm, interesting how some words have their root words disguised like that!) ok, so it's a bit late, don't blame me

T3 housing, 0.48 A/R = ~0.8 sq in. maximum flow area
T3 housing, 0.63 A/R = ~1.1 sq in. maximum flow area
T3 housing, 0.82 A/R = ~1.5 sq in. maximum flow area

T4 housing, 0.58 A/R = ~1.2 sq in. maximum flow area
T4 housing, 0.70 A/R = ~1.5 sq in. maximum flow area
T4 housing, 0.84 A/R = ~1.9 sq in. maximum flow area

well, someone better check these ones again - maybe I'm full of crap... ...again!

 

I've always been curious about the curvature of the center of the cross sectional flow area in the turbine housing. Your post is honestly the first post I have ever seen come up with a relationship between wheel diameter and the path radius of the housing. If I'm under standing your thought pattern correctly, you are basically assuming that the cross sectional area is a circle and the circumference of the cross section touches the edge of the turbine wheel?

One thing about all of this, the T3 housings don't all use the same wheels, but the housing is the same. In other words, if you get a "GT 4-bolt" turbine housing in the .82 A/R, it's the same size, regardless of the wheel that it is cut for. Now, if you count the nozzle area as part of the cross section, when you go up to a larger wheel, you effectively reduce the cross sectional area. So, does Garrett count the nozzle area as part of the cross sectional area in finding the A/R ratio? OR do they just use the scroll area and the nozzle is not considered as part of this cross sectional area?

Hum, you know, I just had a T3 .82 A/R 4-bolt housing show up today to replace a .63 A/R. Maybe I'll go do some measuring just to see what kind of numbers I come up with on both of the housings.

One thing is for sure, these .84 A/R T3s are SO much larger then the .63 housings.