T4 vs T3 and Divided vs Non-Divided Turbines

[jgkurz]

Hi Folks,

I have started a project to build a divided turbo header for my L28. My goal is to build boost sooner but still achieve more peak HP. I know this sounds like an impossible project, but I believe that modern turbo improvements coupled with flow improvements in my engine will net positive results. Today I'm at 535bhp/455rwhp with a VERY modified T3/4 hybrid at 26psi boost. I would like to achieve 500rwhp but at a much lower boost, say 18psi. This should also help lower intake air temps.

 

I know that turbo spool is affected by many things, but the turbine A/R size and the turbine wheel are the largest contributers. For the sake of conversation, let's say I have a GT35R with a .84 A/R T3 non-divided turbine that hits 20psi at 3800rpm.

 

How will spool and peak HP compare if I use a .84 A/R T4 non-divided? How about a .84 A/R T4 divided turbine?

 

Does the A/R translate across T3 and T4 turbines in that 20psi would still be achieved by 3800rpm with the same A/R turbines?

 

Does a divided turbine with the same A/R and flange as a non-divided turbine allow earlier spool but less peak HP capability? It seams to me that divided turbines might be more restrictive.

[thehelix112]

A divided manifold will yield improvements regardless of the turbine, as what it achieves is keeping the pulses from the engine from interferring with each other as long as possible. Of course this is even better with a divided turbine housing as well. I run a divided manifold with the GT3582R with 0.82 A/R T3 turbine housing.

 

A T4 housing has a larger inlet, so the airflow will be slower for a given volume, so the boost threshold will be higher.

 

Dividing the pulses in both the manifold and turbine housing I would expect to bring the threshold down a small amount.

 

I have no idea on the quantities of these increases/decreases, and hence now idea how a divided T4 would compare with a non-divided T3 housing.

 

Dave

[ktm]

A T4 housing has a larger inlet, so the airflow will be slower for a given volume, so the boost threshold will be higher.

 

It is my understanding from reading Maximum Boost and How to Tune and Modify EFI that the primary method of spooling a turbo is from the heat differential across the turbine and not so much the exhaust gas velocity.

[jgkurz]

It is my understanding from reading Maximum Boost and How to Tune and Modify EFI that the primary method of spooling a turbo is from the heat differential across the turbine and not so much the exhaust gas velocity.

 

Don't you mean pressure differential? The hotter the exhaust the better the velocity.

[jgkurz]

A divided manifold will yield improvements regardless of the turbine, as what it achieves is keeping the pulses from the engine from interferring with each other as long as possible. Of course this is even better with a divided turbine housing as well. I run a divided manifold with the GT3582R with 0.82 A/R T3 turbine housing.

 

A T4 housing has a larger inlet, so the airflow will be slower for a given volume, so the boost threshold will be higher.

 

Dividing the pulses in both the manifold and turbine housing I would expect to bring the threshold down a small amount.

 

I have no idea on the quantities of these increases/decreases, and hence now idea how a divided T4 would compare with a non-divided T3 housing.

 

Dave

 

Thanks for the response Dave. Just curious, have you considered running a divided T3 turbine to improve boost response? Also, do you have any pictures of your manifold? I've never heard of anyone using a divided manifold with an L28.

[thehelix112]

It is my understanding from reading Maximum Boost and How to Tune and Modify EFI that the primary method of spooling a turbo is from the heat differential across the turbine and not so much the exhaust gas velocity.

 

ktm,

 

I thiiiink the two are related.

 

The larger the volume the slower the air flowing through it according to the venturi effect.

 

The larger the volume the lower the pressure and heat according to the ideal gas law.

 

Thanks for the response Dave. Just curious, have you considered running a divided T3 turbine to improve boost response? Also, do you have any pictures of your manifold? I've never heard of anyone using a divided manifold with an L28.

 

When I got my GT358R there was no divided turbine housing available? Is there now? I have considered extending the dividing material at the flange out and into the turbine housing though. Obviously this wouldn't be a perfect seal, but better than as is.

 

I would also be wary of putting non GT-generation turbine housings on a GT35R. The GT housings are far superior than older housings due to the increased nickel content, especially so the GT35R. If the option had been available I definitely would have gone for it!

 

I'll try to dig up a pic of it. I didn't make it, I got it from an experienced TiG welder who made it for his 240z race car.

 

Dave

[ktm]

Edit: A quick refresher shows that it is indeed more temperature driven than exhaust gas velocity.Gas flow plays a role as well, but the idea is to maintain high exhaust gas temperatures. As the high temperature exhaust gas expands, it drives the turbine. The greater the temperature differential across the turbine, the faster it will spin. Velocity is a result of high temperature (as well as cross-sectional area for a given flow rate), not the other way around.

[jc052685]

you should look at a second gen rx7 turbo. they have a valve that closes one of the ports forcing all of the flow through just on side at low rpms thus increasing the velocity.

[Tony D]

JGK, you should send JeffP an E-Mail.

He's running a GT53R and is boosting at 3400. Can't remember if it's a .82 or .63A/R turbine he has. And the 18psi numbers you are mentioning sound familiar.

[jgkurz]

JGK, you should send JeffP an E-Mail.

He's running a GT53R and is boosting at 3400. Can't remember if it's a .82 or .63A/R turbine he has. And the 18psi numbers you are mentioning sound familiar.

 

Thanks Tony. I've paid close attention to Jeff's build. He is using a GT35R with a .63 undivided T3 turbine housing. Somehow he's gotten 658bhp out of that turbo. It almost defies physics.

[Careless]

Edit: A quick refresher shows that it is indeed more temperature driven than exhaust gas velocity.Gas flow plays a role as well, but the idea is to maintain high exhaust gas temperatures. As the high temperature exhaust gas expands, it drives the turbine. The greater the temperature differential across the turbine, the faster it will spin. Velocity is a result of high temperature (as well as cross-sectional area for a given flow rate), not the other way around.

 

Somewhere in that book, it explains how a reversion damn on the exhaust side of a millimeter or two could help retain heat next to the turbine wheel from within the manifold rather than flowing back into the ports. It helps as the manifold pressure/gas velocity builds up it will push out one way (turbine to exhaust), rather than both ways (back into port, and through turbine/exhaust).

 

I was reading it on the can yesterday and made note of it because i didn't port match my exhaust ports, which made me feel good again (and even better after I was done).

 

just thought i'd mention.

[jgkurz]

Somewhere in that book, it explains how a reversion damn on the exhaust side of a millimeter or two could help retain heat next to the turbine wheel from within the manifold rather than flowing back into the ports. It helps as the manifold pressure/gas velocity builds up it will push out one way (turbine to exhaust), rather than both ways (back into port, and through turbine/exhaust).

 

I was reading it on the can yesterday and made note of it because i didn't port match my exhaust ports, which made me feel good again (and even better after I was done).

 

just thought i'd mention.

 

I'm going to have to re-read that book. I do some of my best work on the can...

[Careless]

I'm going to have to re-read that book. I do some of my best work on the can...

 

yeah i hate when I get compressor surge though.

[Bernardd]

John, have you measured the backpressure before the turbine inlet with the current exhaust manifold? When I max out my current setup, assuming the motor withstands it, I'll be looking at doing the same. My goal is a bit different in that I want similar spool but more power.

[Clifton]

A T4 housing has a larger inlet, so the airflow will be slower for a given volume, so the boost threshold will be higher.

 

I don't see how. The slightly larger flange volume and only flange volume will be filled faster than measurable and then it's just. If the rest of the mani, or collector match the T4 flange the volume isn't an increased just at the turbo flange. The air isn't going to be slower. You also need a volume of air to spool a turbo. The more you move across the turbine the better. A larger flange isn't the problem, well maybe for .001" second.

[Clifton]

It is my understanding from reading Maximum Boost and How to Tune and Modify EFI that the primary method of spooling a turbo is from the heat differential across the turbine and not so much the exhaust gas velocity.

 

 

You could spool a turbo with 10* air if the volume is there.

[ktm]

You could spool a turbo with 10* air if the volume is there.

 

Yes, you can spool a turbo with 10* air if the pressure is there. However, you would not be spooling as quickly as with 100* or 1000* air.

[Tony D]

On industrial turbos on stationary powerplants, you will watch TIT and TOT for a differential to see turbocharger efficiency...

[Clifton]

Yes, you can spool a turbo with 10* air if the pressure is there. However, you would not be spooling as quickly as with 100* or 1000* air.

 

You sure about that? It's not a gas turbine. Just a turbine that needs a pressure drop accessed it.

 

Either way 500rwhp isn't going to happen with 18 psi unless it has a fairly large hotside. JeffP is running his turbo off the map to make what he is making and it isn't making 500 rwhp at 18 psi. Look what TimZ is using, GT42. If you want fast spool use a smaller turbo if you want big HP use a larger one. If you want both, toss the L28 and get something larger and have everything.

[Tony D]

Pressure drop is a function of pressure and flow present, guys. Temperature drops as work is accomplished, wether the air is 10* or 1000*...it's the premise on which most commercial air separation plants function.

 

"JeffP is running his turbo off the map to make what he is making and it isn't making 500 rwhp at 18 psi."

 

?