mike - what about the 997? The most advanced gasoline turbocharger i am aware of is on this car, and mfg'd by borgwarner.

i dont doubt you blew up BW turbos, but in our experience we have not had a single customer turbo failure come back to us - not one. FWIW a 360 degree thrust has disadvantages over the 270 degree - the only advantage is more surface area. If you run clean, high viscosity synthetic oil i see no reason why you would wipe these 270 degree thrust washers out.

edit: philthy - good finds, im sorry i can not share some of the documents i have

 

I have to disagree with Garrett not being on the leading edge. I simply think that other companies are just on the edge of the marketing front FOR THIS INDUSTRY.

Billet construction doesn't mean **** for what we do. From a performance standpoint, billet construction improves Low-Cycle Failure fatigue life and NOTHING else. We really don't load a turbo in that range, as it would be similar to stop and go traffic hitting full boost almost instantly. Think large diesels that are constantly starting and stopping, like garbage trucks and buses.

Billet construction is more cost effective for producing small batches though, as there isn't the initial investment in tooling required for casting. Because of this, it does allow a manufacturer to go from concept to production quicker. That doesn't mean that concept is any good though.

As for market follower, just an FYI, that billet GTX4294R was designed and produced back in like 2002 from the references I've been able to dig up. Which I beleive is further back then BW or anybody else was offering billet wheels?

BINGO which is why I keep asking for references. I'm sure the data is out there, the question is how do you get it?

 

sti's generally run higher oil pressure.

 

Argh, I wrote a long reply but it timed-out, so this is abbreviated. The reason there is all the 'billet' wheels out there is because the small companies don't have the resources to tool up to cast a wheel. And 5-axis machining is getting cheap now. Spend $100k to make tooling for one wheel before even making anything? Or spend a $100 bucks machining one. OEMs have the volume where it makes sense financially to cast wheels, small turbo companies do not. And all those small turbo companies are probably just copying OEM wheels and resizing them.

Also, just 'billet' wheels are not any stronger than cast UNLESS the billet they start with is forged. I know the OEMs (Garrett, Borg Warner, MHI, IHI) used forged blanks to start with. The OEMs have been making forged fully machined wheels for quite a while.

As for the GTX wheels, those were designed in 2007 from what I researched.

For data related to twin-scroll, I feel there's enough data out there to logically reason the advantages of it. Reference, BMW N63 and S63 engines. Same basic foundation, 4.4L V8s, direct injection. Differences being open voluted vs. twin-scroll setups. The S63 with twin scroll makes more power, more torque, and more torque at a lower rpm, from bigger turbos.

The fundamental idea of separating cylinders that are next to each other in the firing order for twin-scroll is the same basic idea of NA performance header tuning.

Anyways, I had more written before but gotta run. You can go to the SAE website and try searching for papers with keywords: twin entry, twin scroll, divided.

For the hardcore engine data you want, only the engine manufacturers will have it, and they don't like sharing much!

 

http://www.youtube.com/watch?v=zDWA_4wpobc

This shows 2 advances. I am not tempted by the power of the Twin Scroll side.

BUT

I am tempted to put the TB pre turbo and let the turbo free rev in vacuum between shifts however. I mean it only made 1000whp per liter or so, and 80psi needs something to help it spool. There are SAE papers on this as well. The vacuum on the compressor is nothing new, but I know of ZERO people other than JR and I that have even talked about it. Anyone else?

1980 > most engines in 2010

 

spdracerut, I could be wrong on the GTX stuff. I have a reference from 2002 though where they tested a conventional 6 blade + splitter against an 11 blade compressor wheel in a GT42 based turbo. The turbo was for a garbage truck and the focus of the design was to reduce the induction noise that you get with a splitter blade and ported shroud. The 6 blade wheel was cast, the 11 blade was billet. The billet wheel offered about 10% increase in airflow with similar spool as a secondary benefit.

Maybe it is a different design, but it sure sounded like they were talking about what would later become the GTX42 wheels.

JB, I've been very interested in it. Got a couple references on that topic as well. Actually, the idea that I liked the most was actually a 2 valve system to maximize response. You have a throttle plate on the inlet of the compressor, but then you also have a one way valve that bypasses the TB and turbo/intercooler. The pre-turbo TB stays closed when you first get on the throttle and the air short circuits around the turbo straight into the upper intercooler pipe. The TB before the turbo then gradually feeds in. The net effect is near N/A throttle response and it also reduces the pressure drop of the system when the turbo is not producing boost. As the turbo comes on, the 1-way valve begins closing from the pressure difference and the turbo starts taking over. The controller to pull it off would need to be electric though and you would need to base it off TPS, engine RPM, and turbo shaft speed.

The benefit of the 2 valve system is the single valve helps keep the turbo spinning fast once the speed is already high, on shifts and on-off-on throttle modulation. The 2-valve system helps spool the turbo even when it's been spinning slowly.

 

Pardon me for not being more clear, 'FOR THIS INDUSTRY' was my point. LOL


You do what I did to get access to Ford EEC-IV programming in the early '90s. You get to know someone on the inside who has access.

 

don't know if it will help, but there are some pretty pictures, mostly a battle of memories
http://www.efi101.com/forum/viewtopic.php?t=5586

 

11-blade, yes. Same blade design? No At least that's what I was told at PRI.

As for putting a valve in front of the compressor.... I'm not really liking that idea. Couple things: first, you need a speed sensor on the turbo to make sure you don't overspin it. Next, by putting a valve in front of the compressor, you're reducing the pressure at the compressor inlet. This, in effect, means the compressor is operating at a higher pressure ratio and can put the compressor into surge at low flow rates.

A modern solution to achieve the same goal is anti-lag or VNT. Anti-lag, not so turbo friendly. VNT... only one turbo on the market designed for gasoline. Of course, compressor surge is horrible for turbos too; it'll destroy the wheels and bearings.

 

OT
i'm surprised the anti lag method of turning the turbo into a mini gas turbine via fuel injection pre turbo hotside didn't take off as it was turbo friendly i.e no pops and bangs

 

Ah, the IHI 'Rocket'. I could never find detailed info on exactly how it functioned. The angle that it's welded to the manifold makes it seem like flow goes out, not in. But the only way that it makes sense in how it works is if it injects fuel and air.

 

Porsche and Ferrari played around with it at the same time BMW was making 1500whp. It was originally labelled as Hyper Bar and is talked about in either Turbocharging 101 or Max boost. Bleeds air from the compressor directly, injects fuel/ignites, and uses the turbine as the combustion chamber.

aaron

 

03 GSR you said this 3 years ago.... Amazing how time changes things.

Scorke

 

Yes, but I think he's looking for hard data. All those are true in theory (and probably application), but no one has done a strict A/B test with limited variables on this platform. That's all he's asking for (I think).

 

I am going to go through my notes and start a separate thread, but from my understanding of the papers it doesnt operate at any pressure ratio since its free revving in vacuum (this can be a shaft speed problem very fast obviously). Its the sudden transition from vacuum back to being a compressor that can be the problem.