M-P's Custom Turbo NSX Build

Well...I guess it is time to start my build thread. If you had asked me two and a half years ago when I bought my NSX, if I ever thought I would modifying my it, I would have said "No, that's what the RX-7 is for"....I guess I was wrong.

To start, here are some details about the car.

Year/Color: 1992 Formula Red
Mileage when car was purchased (5/7/09): 64,500
Mileage at start of project (2/2/12): 85,000
Current Modifications: GT-One N1 Exhaust, Topspeed Pro1 Headers, UNI Filter, Downforce Intake Scoop, TR-Motorsports 17/18 Wheels



I have been planning this build for the past month and am about finalized on nearly all the details. Along with the custom turbo setup, I am going to be installing a new dash, head-unit, speakers, gauges, and other small items.

The Plan:
- Build a custom twin-scroll exhaust manifold and utilize two 38mm Tial MVS wastegates.
- Use a water to air intercooler
- Eliminate the VVIS
- Run AEM EMS and their built in boost controller

Parts that I have already purchased/already own and will be installing before anything else:
- RPS Single Disc 6-puck clutch with modifications (Used/Rebuilt from Costa)
- Dali Billet Coolant Tank (Black)
- Dali FRP Dash
- Dali Short Shifter
- Gruppe-N Bead Blasted Shift Knob
- Kenwood KVT815-DVD In-Dash Monitor
- Kenwood KV-4100 Navigation

I'm planning on ordering most of the turbo components and the material to build the exhaust manifold on March 1st...then the real fun will begin.

Sounds like fun!

Are you doing a twin scroll being separated bank-wise (like Autowave), or a twin scroll being separated cylinder-wise (never been done on an NSX to my knowledge)?

Dave

Mac Attack said:

Sounds like fun!

Are you doing a twin scroll being separated bank-wise (like Autowave), or a twin scroll being separated cylinder-wise (never been done on an NSX to my knowledge)?

Dave

Click to expand...

Separated by bank.

I don't see any benefit to doing it separated by cylinder. The firing order is 1-4-2-5-3-6, so the exhaust pulses will alternate per bank which will be perfect for the twin scroll turbo.

M-P

You're correct, but each bank will have to have exactly the same primary/secondary lengths in that case. I don't think any aftermarket headers truly have this, but I could be wrong. Oops, just remembered you're making your own.:smile:

Besides discussing A/R's and a limited discussion on intercooler pressure drops, I don't think anyone on here has publicly displayed knowledge of our VE curve and factored that in to their turbo decision making. It's nice to hear what you're doing and I hope you share some of your turbo sizing information please to help others out.

As an engineer, you'd like the old Turbocharging the Internal Combustion Engine by N. Watson and M.S. Janota. These are out of print and very expensive to buy, but try to find one from a college or vocational library near you.

I like the technical and theoretical aspects of things, but in the end, I just decided to do a semi-infinite A/R design that also doesn't require a wastegate by design either - twin Aerochargers.

Dave

Mac Attack, it sounds like we could have a lot to talk about. I have been recommended the Watson and Janota book on several occasions, but at nearly $600 on ebay...I just can't justify it. I also have the entire NC State University library close to my house (apparently the second largest library on the east cost behind the Library of Congress) and they don't even have it. If I could find one for under $200, I would buy it in a heartbeat.

I do have a VE curve for our engines (C30A) that was done buy Honda. I believe it to be quite optimized and not very realistic, but figure the curve is at least similar in shape. For my turbo matching I have been using a VE that is 5% lower than the values plotted on the map that I have attached.

This is the map that I am referring to....



I have also been in close contact with Raffi/Geoff at Full-Race for my turbo decision. The biggest problem is with turbine maps (or lack there of) and having to trust the vendors on turbo selection because of them. They have been a huge help and I believe I have my turbo selected....I will go into this later.

Just for extra info, you guys can check out my thread at MotoIQ that talks about the effect of twin-scroll on a 6-cylinder.

http://www.motoiq.com/forum/aff/12/aft/1278/afv/topic.aspx

Hi,

Mac Attack said:

I don't think any aftermarket headers truly have this, but I could be wrong.

Click to expand...

Well... i didn't measure them myself, but one of the advantages of the GT-One F1 headers were that all their primarys have the same exact lenght.

Nuno

M-P said:

Mac Attack, it sounds like we could have a lot to talk about. I have been recommended the Watson and Janota book on several occasions, but at nearly $600 on ebay...I just can't justify it. I also have the entire NC State University library close to my house (apparently the second largest library on the east cost behind the Library of Congress) and they don't even have it. If I could find one for under $200, I would buy it in a heartbeat.

Click to expand...

Nash Community College about 45miles from you has a copy:
http://www.worldcat.org/title/turbocharging-the-internal-combustion-engine/oclc/9071850?loc=11111&tab=holdings

UnhuZ said:

Hi,
Well... i didn't measure them myself, but one of the advantages of the GT-One F1 headers were that all their primarys have the same exact lenght.

Click to expand...

Yeah, there was a nice header thread a few years ago that had some good info in it. I think the GT and maybe the Fujitsubo headers are truly equal length, but they are $$$. Autowave has a great design on their custom headers with equal lengths going to their twin scroll turbo. I'm going to have a similar design for my rear header bank with my Aerocharger.

Dave

I read somewhere on prime that the Topspeed Pro1 (DC Sports replica I believe) headers are actually equal length primaries per bank. Meaning that cylinders 1-3 have equal primaries and cylinders 4-6 have equal primaries, but primary length for cylinders 1-3 don't necessarily equal the length of the 4-6 primaries.

It might not be 100% optimized this way, but it will suit me just fine. I do not plan on building new headers, but rather a y-pipe similar to what is already used in turbo kits on the market today. The differences are that I am going to:

- Run a true twin scroll setup with twin wastegates and a fully divided exhaust manifold
- Make sure that the overall length of bank 1 equals the overall length of bank 2 when they enter the turbine housing
- Have a proper wastegate flow path (i.e. not tee the wastegates straight off the exhaust manifold...this is important because I have calculated that at 8psi, nearly 50% of my engine flow will be directed through the wastegates)
- Save a ton of money