30

Project [Rotary] FD RX-7: Part 4 - Turbo System

by M-P Spierer

 
Designing and installing the turbo setup is one of the most enjoyable aspects of any build. There are many decisions to make when planning out a turbo setup and each decision significantly affects subsequent decisions. It is a balancing act of desired goals and undeniable trade-offs. Many questions need to be answered like, am I looking to maximize mid-range response or top-end power? And, how much response will I be giving up if I choose one turbine housing A/R over another? These are questions that have different answers for every build. For this RX-7 we are trying to maximize response around a specific power limit without breaking the bank.
 
The beginning of the turbo selection process for this project started over a year ago when power goals were first being determined. A key factor in this process was defining a power limit that would still allow for good reliability track session after track session. With proper cooling (charge air, coolant, and oil), good tuning, and an auxiliary injection system running 100% water (to be featured and installed in a future article), a power limit of 450 wheel horsepower was chosen. Now, I say power limit because this is the maximum horsepower that we will be tuning for to keep the heat generation and engine stress levels in check, but not necessarily the power level that the car will operate at. In actuality, the true target wheel horsepower for normal operation is 425 (an even 500bhp at the engine if we assume 15% drivetrain losses). Either way, we are asking the 13B-REW to produce roughly two times as much power as it did when it came out of the factory in Hiroshima, Japan over 20 years ago.
 
Once the power goal was established, turbo selection was next. If you remember back to the first article of this project, the car originally used a ball-bearing Garrett T04R (often referred to as a T04Z). This was a great turbo for its time and honestly would have suited the power goal just fine. However, turbo technology has come a long way since the T04 series of turbos were popular and there were good reasons change to the new turbo...the Borg Warner S362 FMW.
 
 
The Borg Warner S300SX FMW (S362) is technically in the AirWerks family of Borg Warner turbochargers, but unlike most models in the AirWerks lineup it uses a billet compressor wheel similar to what you might find in an EFR turbo. There are not very many billet compressor wheel, journal bearing turbos out there, but Borg Warner has filled that void with several FMW Airwerks turbos ranging from 57mm to 82mm. I like to think that these turbos are for the people who want an affordable and extremely robust turbo with some of the cutting edge turbo technology found in an EFR series turbo.

Forged aluminum is inherently stronger than cast aluminum, so the benefit of a billet (forged) compressor wheel is that it can be made lighter than its cast counterpart without sacrificing strength. This is the same basic principle that applies to wheels as well. Now if we compare numbers between the old T04R and the new S362 FMW it is plain to see one of the benefits of changing turbos. 

The Borg Warner billet compressor wheel has a 61.4mm/83mm (inducer/exducer) and is capable of flowing ~74 lb/min of air. The T04R's larger 66.7mm/84mm cast compressor wheel could flow ~75 lb/min of air. Basically, with the Borg Warner you have a turbo with a smaller and lighter compressor wheel that has similar airflow capability. The smaller and lighter compressor wheel has a lower moment of inertia and therefore should accelerate (spool/respond) faster than the larger and heavier compressor wheel.

 
Shot of the turbine side of the S300SX FMW. The turbine housing features a fully divided T4 inlet and 0.91 A/R. It has also been modified by Full-Race to have a 3" V-band discharge instead of the Borg Warner full marmon discharge and is ceramic coated. All the quality check paint marks shown in this photo leads one to believe that quality control is very important to Borg Warner.
The crown jewel of this turbo has to be the billet compressor wheel or "Forged Milled Wheel" (FMW) as Borg Warner calls it. This compressor wheel features Borg Warner's "Extended Tip Technology" and borrows much of its design directly from the new EFR series of turbos.
For those who are compressor map literate, here you go. I expect the turbo to operate with a maximum pressure ratio between 2.25 and 2.5. This should bring me right across the widest part of the 76% efficiency island and the entire map in general. The peak flow at this pressure ratio is ~66 lb/min which should be good for a realistic 450-475whp in a rotary powered RX-7.

For the exhaust manifold we didn't need to look past Full-Race. Their manifolds are used on some of the fastest cars around in every discipline imaginable. Their reputation coupled with the fact that they offered an off the shelf manifold with all the features required for this build sealed the deal. It was important to have a true twin-scroll manifold for the RX-7 because the exhaust pulses of a two rotor engine allow for full utilization of twin scroll technology and the resulting benefits.

Full-Race also installed 1/8th NPT bungs in each runner so that we can monitor exhaust gas temperatures or pre-turbo exhaust backpressure while tuning. Monitoring the exhaust backpressure and reading dyno plots will allow us to determine if the 0.91 A/R turbine housing was the correct choice or if the 1.00 A/R option would have been better.

 
Full-Race has an incredible reputation for quality and performance. The manifold used in this build is twin scroll with a T4 turbo flange, dual 44mm wastegate flanges, an EGT sensor bung on each runner, and has been ceramic coated.
Page 1 of 5 Next Page
Bookmark and Share
Comments
Dusty Duster
Dusty Dusterlink
Monday, March 31, 2014 9:37 AM
I love the attention to detail on this build.
Chuck Johnson
Chuck Johnsonlink
Monday, March 31, 2014 12:09 PM
Borg Warner, Turbosmart, and Full Race Motorsports what an awesome combination! It's good to see that M-P has good taste in mountain bikes, cars, and turbo systems! :::thumbs up:::
Dusty Duster
Dusty Dusterlink
Monday, March 31, 2014 12:51 PM
Why didn't you go with the EFR turbo?
M-P
M-Plink
Monday, March 31, 2014 2:17 PM
Thanks Chuck.

Dusty: Three reasons I didn't go with EFR. First, I purchased the turbo about this time last year while the EFR's were all but impossible to get your hands on. Second, the EFR turbo was a bit cost prohibitive even if it was readily available. Third, the Gamma-Ti turbine wheel used in the EFR turbos has an exhaust gas temperature limit that is very close to what rotary engines typically produce and I wasn't comfortable making my car the guinea pig to see if it would live or not.
Hoss
Hosslink
Monday, March 31, 2014 2:26 PM
What kind of oil pressures are you expecting? I would hate to see that nice turbo prematurely blow its oil control seals do to excess pressure and flow.

I'm also excited to see what kind of engine management you are going with. I know that same MAC valve is what Adaptronic recommends for use with their ECUS. I hope to be testing one next month.
Dusty Duster
Dusty Dusterlink
Monday, March 31, 2014 3:45 PM
I didn't realize that rotaries had EGTs that high!
Dennis
Dennislink
Monday, March 31, 2014 3:51 PM
Looks really good! I like the use of flex sections on everything including the wastegate outlets.

Are you planning on bracing the turbo at all? That's a lot of turbo with a decent moment arm off the exhaust studs. Seems like it might lead to cracking in the future. I know you're limited by the firewall, but that conical section right at the turbine outlet is really short/steep. Seems like it would be best to keep it smaller and then have a nice taper out to 4" where you have more space.

Uncatted exhaust smells the best IMO. Especially when you're burning corn!
M-P
M-Plink
Monday, March 31, 2014 4:45 PM
Hoss: I expect maximum oil pressure to be around 90psi. The high oil pressure is the reason for the -3AN feed hose rather than the more common -4AN used on journal bearing turbos.

Dusty Duster: Turbo rotaries operate with EGT's around 1650-1700F (~900C). Several people in the rotary community are experimenting with them, but since I am going for reliability out the gate it was too risky of a venture.

Dennis: As far as I have ever seen/heard, there is no bracing needed for the turbo using Full-Race manifolds. As for the 3"-4" transition, I agree that it is much shorter than optimal, but like you said there isn't much option with the firewall. Add to that the fact that I wanted the wastegates plumbed back in after the transition, but not under the car and my decision to keep the transition I had was made. May try and change that up one day, but for now it should work fine.
Dan DeRosia
Dan DeRosialink
Monday, March 31, 2014 10:40 PM
Hm; had wondered about the S362 FMW turbo on a rotary given the relatively modest boost levels, but now that you point it out yeah, the map is pretty wide around 2.5 P/R. Neat stuff; looking forwards to results.
Adrian Avgerinos
Adrian Avgerinoslink
Tuesday, April 01, 2014 4:16 PM
In regards to the use of firesleeve on that oil drain line: Because everything is currently clean and unused, I propose wrapping the exposed ends in self-fusing silicone tape (A-A-59163/MIL-I-46852C) to keep moisture, and more importantly, oil/grease out of the fiberglass innards.

Though I admit I haven't taken a blowtorch and tested the theory, it seems to me that oil soaked fiberglass mat would make an excellent media to create a sustained fire.

(To be honest, I'm actually rather surprised that I don't see silicone tape used with more regularity on automotive applications like this one.)
M-P
M-Plink
Wednesday, April 02, 2014 4:25 PM
Adrian: The firesleeve actually came with some self-fusing silicone tape that I just neglected to install. Thanks for the heads up...I'll make sure it is installed before the engine goes back in the car.
Chris333
Chris333link
Wednesday, May 14, 2014 5:58 PM
Hi, you mention the use of inconel studs here for the rotor housing to manifold flange. Do you happen to have the part number for these studs? I am currently doing a turbo build on my mx5, and the 9YA9-21-008 studs which I believe are the correct ones, are a whole lot cheaper than TSE and other inconel offerings out there by aftermarket companies. Thanks.
Post Comment Login or register to post a comment.

MotoIQ Proudly Presents Our Sponsors:

© 2014 MotoIQ.com