T3 Divided Turbine housing

[thilaksharma]

Since I was kept at the waiting list in my local turbo shop , I came upon a tuner who was shopping for turbos. I was going for a 57Trim , and was told that it wasnt in stock so I have to wait a week for garrett to ship in the turbos. The tuner instead was getting himself a

t3/t4 setup
t04s compressor housing
61mm inducer , exducer i couldnt measure for the compressor side
compressor antisurge holes like turbonetics style

but what concerned me most was Turbo Lag , that the reason i dint want to purchase anything above 57trim trim. I did ask for the tuners opinion , he said , the turbo he is purchasing runs on a divided 0.63 t3 turbine housing. And he does recommend to fabricate a turbular manifold that pulses cylinder 1-4 into one turbine inlet and 2-3 into the other turbine inlet of the divided housing. He also did say , the lag was bearable in a 2.0L car as he gets 23psi by 4000rpm.

Any opnion guys , i also did google it up , and found this article by turbobygarrett


The concept is to DIVIDE or separate the cylinders whose cycles interfere with one another to best utilize the engine's exhaust pulse energy.

For example, on a four-cylinder engine with firing order 1-3-4-2, cylinder #1 is ending its expansion stroke and opening its exhaust valve while cylinder #2 still has its exhaust valve open (cylinder #2 is in its overlap period). In an undivided exhaust manifold, this pressure pulse from cylinder #1's exhaust blowdown event is much more likely to contaminate cylinder #2 with high pressure exhaust gas. Not only does this hurt cylinder #2's ability to breathe properly, but this pulse energy would have been better utilized in the turbine.

The proper grouping for this engine is to keep complementary cylinders grouped together-- #1 and #4 are complementary; as are cylinders #2 and #3.
divided turbine housing

Figure 7. Illustration of divided turbine housing

[GVR4592]

The TO4E 50, 57, and 60 trim all spool about the same and make similar power, but the 50 trim does it more efficiently. The other two are just a waste in my opinion. The turbo your tuner has, sounds like a GT35R or a similar variant.

Divided housings are nothing new, although I haven't seen very many divided .63 T3 housings. ATP turbo makes a divided .78 T3 housing though.

[Defiant]

The idea of dividing the housing and keeping the "pulses" segregated sounds like a good idea. Except, the "pulsing" isn't what drives the turbine. The turbine is driven by the heat energy.
There may be some payoff in this additional complication. As little as it's being implemented, it must be a small one.

[Steve93Talon]

Originally Posted by Defiant

There may be some payoff in this additional complication. As little as it's being implemented, it must be a small one.

EVOs are divided. New turbo Minis are divided. JDM WRXs are divided. Most newer Diesel truck turbos are divided. All modern, larger T4s are divided. All the high end exhaust manifolds like Shearer's top-mount are divided. I think it's been proven to be the LEAST complicated way to improve spool up without sacrificing top end. Just look at an EVO. Full boost at 3k and enough power for most people to go 11s with bolt-ons on pump (and a handful in the 10s even!).

[Defiant]

Yeh. How many are seeking divided turbos to put on their DSMs?

[definitiveno]

When you find a complete kit for our car let me know. I would have to respectfully disagree with defiant and say there's not a lack of want from the dsm community for divided/twin scroll setups however as it's been explained to me by several of the big 4 or 5 running the dsm parts and fab world that there's more returns for investing R/D into the evo market. Hence why alot of the fabricators that build for us offer such kits for the Evo's and not DSM's

[90laserRSfwd]

to the OP, when I get into DSM's along time ago, I didn't want a gt35r or anything large because I always thought, man I have to wait till 4,000+ RPM for the boost to come on. Now, being the owner of a gt35r, the lag isn't a big deal. When your making 250-350hp you def. don't want to wait till 4K+ to see boost. But when your making 500+ @ the wheels, lag becomes less and less of a worrry. I don't know if you've ever gotten the chance to drive a well tuned 500+ awhp car, but the lag doesn't even come to mind. 1st and 2nd gear bang off the rev limiter in a hurry. 0 to 60 is around 4 seconds, and 0 to 100 is not much over 6 or 7. (I'm talking about MPH not KPH by the way)

[Steve93Talon]

Originally Posted by 90laserRSfwd

0 to 60 is around 4 seconds

More like half that

http://www.sjdsm.com/tech/0_60_calc.htm

[thilaksharma]

Originally Posted by GVR4592

The TO4E 50, 57, and 60 trim all spool about the same and make similar power, but the 50 trim does it more efficiently. The other two are just a waste in my opinion. The turbo your tuner has, sounds like a GT35R or a similar variant.

Divided housings are nothing new, although I haven't seen very many divided .63 T3 housings. ATP turbo makes a divided .78 T3 housing though.

yes , you are right, its a mistake , i went back to the turbo shop, it was a 0.82 t3 divided and not a 0.63

[Steve93Talon]

Originally Posted by thilaksharma

yes , you are right, its a mistake , i went back to the turbo shop, it was a 0.82 t3 divided and not a 0.63

Still doesn't make sense. There is no .82 divided housing for a garrett T3 that I'm aware of, only the ATP .78 as indicated above.

[bpestilence]

Originally Posted by Defiant

Except, the "pulsing" isn't what drives the turbine. The turbine is driven by the heat energy.

No it doesn't. A difference in pressure is what makes a turbine, any turbine, spin. If you were to pipe high pressure, cold air into a turbo it would still work. Divided housing turbines have been around in large turbos for a long time, and they do work well. By dividing the runners you maintain a smaller, higher pressure area on the turbine wheel to do more efficient work. This is also why higher compression engines spool turbos faster.

[thilaksharma]

Originally Posted by Steve93Talon

Still doesn't make sense. There is no .82 divided housing for a garrett T3 that I'm aware of, only the ATP .78 as indicated above.

TurboByGarrett.com - Catalog

TurboByGarrett.com - Catalog

TurboByGarrett.com - Catalog

[definitiveno]

You said you were talking about 57 trims/50 trims though (T series turbo's), I assume thats why the poster said "garrett didn't make them"


Those are also GT turbo's where he said there are no T3 housings with those specs, I assume he meant T Series turbo's. Those are GT

[thilaksharma]

definitiveno : sorry mate i will try to get a pic , i do shop at an autorized garrett dealer, will take the pics for u guys

[thilaksharma]

i tried looking for t3 divided on the net, seems like all came from atp and fp

[123bobby123]

Originally Posted by bpestilence

No it doesn't. A difference in pressure is what makes a turbine, any turbine, spin. If you were to pipe high pressure, cold air into a turbo it would still work. Divided housing turbines have been around in large turbos for a long time, and they do work well. By dividing the runners you maintain a smaller, higher pressure area on the turbine wheel to do more efficient work. This is also why higher compression engines spool turbos faster.

The exhaust heat is resonsible for most of the kinetic energy that drives the turbine/rotating assembly in a turbocharger. This is becasue there is a large temperature drop across the turbo and this equates to a large pressure drop across the turbocharger. You can think of it as the opposite of what is happening in the compressor, the air is compressed and it heats up. Now the exhasut veolocity does of course contribute to the kinetic energy, but as I said before the mojority of this energy comes from the exhasut temperature drop. This is one of the reasons why you cermaic coat or wrap your exhasut manifold, to keep the exhasut as hot as possible when it enters the turbocharger.

Now as far as getting the turbo spinning with high pressure shop air, I'm not sure if you could get it upto proper operating speed, but you would not want too, becaue the temperature drop would cause the moisture in the shop air to freeze out and wreck the turbine wheel or freeze the turbine wheel to the housing, which is not good.

At school, in my senior design project, we test turbocharger bearings and we also are designing a test stand to test the turbocharger with compressed shop air. The hardest part of this was to see what tempearture we needed to get the air upto before it entered the turbine.

Bill

[thilaksharma]

Originally Posted by definitiveno

You said you were talking about 57 trims/50 trims though (T series turbo's), I assume thats why the poster said "garrett didn't make them"


Those are also GT turbo's where he said there are no T3 housings with those specs, I assume he meant T Series turbo's. Those are GT

I thought I was wrong i went up to the turbo shop again and took the pics
here u go







this 60-1 retails at USD 720 , its made in local asia garrett plant (not china knock off)

[tkelly27]

Originally Posted by 123bobby123

The exhaust heat is resonsible for most of the kinetic energy that drives the turbine/rotating assembly in a turbocharger. This is becasue there is a large temperature drop across the turbo and this equates to a large pressure drop across the turbocharger. You can think of it as the opposite of what is happening in the compressor, the air is compressed and it heats up. Now the exhasut veolocity does of course contribute to the kinetic energy, but as I said before the mojority of this energy comes from the exhasut temperature drop. This is one of the reasons why you cermaic coat or wrap your exhasut manifold, to keep the exhasut as hot as possible when it enters the turbocharger.

Now as far as getting the turbo spinning with high pressure shop air, I'm not sure if you could get it upto proper operating speed, but you would not want too, becaue the temperature drop would cause the moisture in the shop air to freeze out and wreck the turbine wheel or freeze the turbine wheel to the housing, which is not good.

At school, in my senior design project, we test turbocharger bearings and we also are designing a test stand to test the turbocharger with compressed shop air. The hardest part of this was to see what tempearture we needed to get the air upto before it entered the turbine.

Bill

What school do you go to? I'm not going to say you're wrong, you obviously work with the thing. However, I've got my thermodynamics book in front of me, turbines are modeled as isentropic (geez, it's been a while, isn't that the right word?).

"Heat transfer from turbines is usually considered negligible..." and if you think about the amount of insulating air around the thing, it makes sense.

Also, it says that the change in pressure at the same temperature causes a change in enthalpy, and the kinetic energy (opposite of the compressor which accelerates the particles, the turbine decelerates them).

Furthermore, EGT gauge manufacturers advise that thet the sensor can be placed up to 2" from the turbine with a drop of only 200 degrees F
http://www.autometer.com/productPDF/2650-1088.pdf
"If the exhaust manifold can not be removed, install the probe 1-2 inches after the turbo exhaust outlet (Exhaust gas temps could drop over 200˚ when installing after the turbo)."

If I'm reading that right, that's the difference from 1-2" from the port vs. 1-2" from the turbine outlet... So I'm not really sure where this huge temp drop comes from.

It's been a while, go easy on me

[123bobby123]

Originally Posted by tkelly27

What school do you go to? I'm not going to say you're wrong, you obviously work with the thing. However, I've got my thermodynamics book in front of me, turbines are modeled as isentropic (geez, it's been a while, isn't that the right word?).

"Heat transfer from turbines is usually considered negligible..." and if you think about the amount of insulating air around the thing, it makes sense.

Also, it says that the change in pressure at the same temperature causes a change in enthalpy, and the kinetic energy (opposite of the compressor which accelerates the particles, the turbine decelerates them).

Furthermore, EGT gauge manufacturers advise that thet the sensor can be placed up to 2" from the turbine with a drop of only 200 degrees F
http://www.autometer.com/productPDF/2650-1088.pdf
"If the exhaust manifold can not be removed, install the probe 1-2 inches after the turbo exhaust outlet (Exhaust gas temps could drop over 200˚ when installing after the turbo)."

If I'm reading that right, that's the difference from 1-2" from the port vs. 1-2" from the turbine outlet... So I'm not really sure where this huge temp drop comes from.

It's been a while, go easy on me

Hi,

I go to Virginia Tech, they got a engine lab/dyno room there which is where I and the project team I'm on do our work.

Isentropic means that that the flow is reversible and adiabatic which means there is no heat transfer coming from the turbine (adiabtatic) and no losses in the system due to friction (reversible). This obviously is not real and is an idealization. When working on turbines in power plants and stuff like that you can usually assume that there is a little amount of heat transfer becasue of the immense amount of insulation they have.

Now for turbochargers, I believe that you can not assume them adiabatic because heat transfer is driven by temperature difference, heat always flows from high temperature to low temperature. So like the exhaust is 1300-1500 degrees F and the outside air is like maybe 150 degrees in the engine bay, and the only thing between the two temperatures is the turbine housing, which is a obviously metal. Also when drving down the road the air is moving very fast and is most likely very turbulent in the engine bay so this would make the heat transfer out of the turbine even higher.

For our project we are going to bring the turbocharger upto operating speed (140,000 rpm) using compressed air heated to a certain temperature. We are going to try to get the air upto about 400 degrees F and that should be fine for running it considering that we want the outlet temp to be no less then 32 degrees Fahrenheit. We came to this temperature by assuming that heat transfer out of the turbine was zero, we did this becasue of the smaller temperature difference from air inside the turbine (maximum of 400 degrees F) to air outside the turbine, and we also might ceramic coat the turbine housing to ensure minimal heat transfer.

You are an engineer right, I figured since why the hell else would you have a thermo book, haha. If you look at the enthalpy values for say 400 degrees F (I pick this becasue this is what we are trying to use) and then compare that to the enthalpy value for air moving at 450 ft/sec. You will see that the enthalpy value for the air veolocity is very small when compared to the enthalpy value for the temperatures.

In your thermobook there are many ways to deal with turbines, you are looking at just one. If you look at other ways to do it, you will see that it will say that velocities can usuallly be ignored, which I just described to you. I would try to tell you where to look in your thermo book but I dont have mine since i'm back home for the holidays.

If this explanation isnt good enough I can try to post this part of our final report, but it is like 10 pages long and I didnt think you would want to read all that.

Oh and for the EGT, it says that it could drop over 200 degrees Fahrenheit. So 200 degrees F or greater, which represents the temperature drop through the turbo. Now like you said the temperature drop through the turbo isnt massive but it can be quite large.

Hope this helps. Merry Christmas. Sorry for spleling errors, haha.

Bill