Thanks for being my ChatGPT and finding answers to my question. The above posts helped, very much appreciate it!

 

I ran the 7670 for roughly 1000 miles on a very small ported engine and around 4-500 miles on a stock reman. Unfortunately I was only able to drive it about 50 miles max on the pettit ecu as it kept randomly going into limp mode due to a reason I couldn't figure out. It ran great at 12-13lbs tailing off to 11ish lbs in upper rpms. The majority of my time on the 7670 was on the stock port reman with PFC. I modified fuel amount a little through the commander but otherwise a mostly base map. Car ran incredible with little to no knock running 13-15lbs that would tail off to 11-12psi in higher rpms. The butt dyno says very low 300hp level at the wheels. The stock twins feel faster up to 12psi but I think the edge starts to go to the 7670 around that 13-14psi mark. I'm sure it is a riot at 18-20lbs as it was just starting to wake up at 13-15lbs.

Knowing what I know now, I would have opted for an external wastegate as I am old school and prefer the simplicity. I still have the 7670 setup and will put it on another FD in the future.

 

@djseven
Excuse my ignorance but... why would an external wastegate by simpler? Seems like one more thing to setup and plumb.

 

I'm be curious if anyone has done any testing or has an inkling of boost threshold on an 8374 with one of the Indy 1.45 turbine housings. My last foray with such a large turbine housing was on an undersized 9180 and 3 rotor so I may be a bit optimistic.

 

pretty sure that rx72c has commented in other threads about results with the 1.45 A/R housings, I don’t recall it being so positive but my recollection could be off.

edit: here’s one

https://www.rx7club.com/single-turbo.../#post12552782
.

 

The internal gate actuator has about 1/16" clearance and is not designed to fit on my most FD manifolds without making a special bracket. If it isnt lined up just perfectly, it doesn't want to fully open or close all the time. Then there is the spike/creep/tail off issues that would be easier to remedy with a larger external gate. Once again, the keep it simple stupid motto works wonders on a FD that is being thrashed on. Its all things that can be worked around and dealt with, but can get expensive and time consuming in a hurry.

 

Thanks for the post. Still wish there was some real info on the specifics of boost threshold, should probably PM him

 

I just want to expand on this a little bit because I think this explanation conflates system planning with component simplicity. Compared in isolation, an internal gate turbo is simpler than an external setup. It packs more capability in a single and overall smaller package. If there was a variant of the 8374 on the market that offered a larger internal gate that could flow enough bypass to hold low boost pressures on our cars, I think most people would opt for an internal gate model vs external; because it would then truly be simpler. But with that not being the case, your explanation outlining how an iwg turbo complicates the overall system by needing to deal with boost in other ways becomes the reality.

In no way trying to say you were wrong about anything, just trying to add some more color.

 

There is plenty of room to port the EFR 8374 IWG- just got done doing it.

Easier to port than stock S5 turbo WG ports, about as hard as a 2ndary and aux intake ports of a 6 port NA port job. 1/8" shafted carbide on a flex-shaft and the smallest sanding drums are the tools of choice.

Mainly though the EFR 8374 boost creep looks to be an issue with the cast shorty manifiold (as other manifold vendors have remarked) and where the turbo WG passages are.

Specifically, the S- bend on the rear rotor's exhaust runner puts the rear WG runner (which otherwise has best priority) in the turbulent "dead" low velocity flow area of S-bend.

You can look up CFD of S-bend flow and easily see what I am talking about.

I chose the Greddy FC cast IWG manifiold since it has the S- bend on the front rotor's exhaust runner which puts the front WG port in the high velocity flow off the S-bend and a straight shot from the rear rotor's exhaust runner into the higher priority rear WG passage.

 

Somehow I had avoided commentary pointing to the TBlown Cast Shorty manifold as a contributor on this issue. But after thinking about this and looking up CFD of S-Bend like you mentioned, this is an interesting thing to think about. To your knowledge, has TBlown chimed in on this? Also, for the FD specifically, what would be the guidance on an alternative manifold for the 8374iwg? Are you aware of an off the shelf option or would a custom manifold be the only path for someone looking for a manifold that better suits the higher priority rear WG passage?

 

Full Race used to make an off the shelf IWG manifold option, but it is no longer available. I just got my 8374 iwg and am looking to have a custom manifold fabricated because I'd rather not run the turblown shorty and can't find any other options. From all the reading I've done, the boost creep issues seemed to be using the turblown setup.

 

I’m assuming you mean this one:



.

The bend is actually worse than it appears above. It’s more appropriate to call it a kink rather than a bend. And totally unnecessary with a casting. It’s like they just duplicated the fabricated version without considering how that can so easily be avoided and fixed with a simple mold change.
.

 

I should have been more clear, I am using the Greddy FC3S cast IWG manifold on my FD3S.

Off the shelf the Greddy FC cast IWG manifold puts the compressor housing really close to the FD "frame rail".

To get the compressor all the way against the intake manifold instead so I can run stock motor mounts (without turbo smashing into unibody) I had the manifold turbo flange and turbo flange milled 3/32" on the front (angled) and have spherical cup washers to load the studs straight.





The restriction of the manifild S-bend on the front rotor exhaust while not ideal will help balance the heat load of the usually hotter running rear rotor.

The turbo set-back puts the heavy turbo several inches back (chassis dynamics) and also allows room for a 6" velocity stack right off the turbo with a 20" long cone filter straight off that. Very nice.

The Tomei Arms WG actuator fits with the stock EFR WG bracket as shown.

Is the Greddy cast FC IWG the best EFR manifold?
No, its the compromises I chose for what I wanted. There really wont be a best manifold. Each has issues and advantages.

I wanted
durability (cast)
response (short runners)
limited boost creep w/ IWG (runner to WG priority)

 

The Greddy cast FC IWG manifold has the "s-bend" more as a "z-bend". This provides less degrees of bend off the exhaust port and the into the turbo flange while providing the longest straight section between them which minimizes the flow issues associated with "s-bends"

 

Here you can see what I mean by having the "s-bend" on the front runner aligns the WG passage in line with high velocity exhaust flow.

2nd and 3rd pic shows how you can stick a zip tie from the WG ports straight out the exhaust ports.

 

Greddy "s-bend" into front WG port versus (flipped Greddy) to show the rear "s-bend" manifolds put the rear WG port in the blue low velocity dead zone of flow.

 

Or you can use the turblown long runner cast manifold and cap off the ewg ports.

 

still pretty much the same on all those long-short runner manifolds, more than one compromise going on with them all.

nice porting job!
.

 

I had thought about running the turblown long manifold with IWG and EX WGs (so the EX WG open to relieve boost creep at high RPM, but you get more gas pedal without the BRAAA!).

I love the visual of front/ top mount turbo placement, but this build Im trying for throttle response and low boost to mask the turbo powerband for more of a easy to drive/ intuitive experience.

I really want NA 3 rotor, but my budget isnt there..

 

Has​​​​ there been any back to back testing that shows long runners increase lag? For piston engines, a properly tuned exhaust header will have better turbo response than a log manifold.

 

a 13B-2R has perfectly phased exhaust pulses in two distinct flow streams; equal length and fully divided is ideal. Since they’re not actually collected as on a multi-rotor/cylinder engine, you don’t have to get into any of that collection phasing, and shorter is better as long as you aren’t making flow compromises imo. The dual 90°elbow bend posted earlier as a compromising example relative to the IWG placement as was discussed.

As you can see with that S-bend diagram, two 45° elbows is going to flow a lot better than two hard 90°s, or appropriate variable radius where space permits with more organic blending such casting or 3D printing (also talented metal working, though rare).

Or even with multi-axis CNC now, it’s not necessary to butt an elbow out on the face of the 1/2” or thicker flange, the opening can be cut to allow the elbow to slip in and fit right up to the gasket face, gaining the flange thickness for making a more appropriate bend. Or custom CNC turbo flanges with radius machined in rather than straight openings. Which can make all the difference with a tight fitment.



.

let me just throw all urban legend out the door by stating that an EWG properly placed on a turbine housing is just as effective as an IWG … more so even given the size difference and that it can be merged back into the main exhaust stream at a less disruptive location than right into the turbine outlet. It’s been done on here before, but not seen as often as it could be.

The turbine housing EWG can even be a single divided for a skilled fabricator and it’s not so risky now given SS housings, 321 & Inconel/Nickel etc. material options being more readily available in this age. There are plenty of them being successful in cast iron housings even.

Having been a welder from out of a high school votec program, to acquiring a engineering degree later than most, and working the last 27 years specifically with stainless fabrication, my view is not so tainted for thinking a cast solution is required. It just has to be done properly with the correct procedures and materials.
.

 

not many people actually do tests like that. i do remember way back in the mists of time Rice Racing insisted that the longer ones were better, and then he switched to the HKS T04Z manifold because it was better...
the current HKS cast was redesigned to be more like the T04Z on the inside.

ironically its what Blue had on his previous ride

 

I just heard from someone who has done this testing and generally shorter is better in a rotary, however the difference in transient turbo response and energy losses from a 18" runner vs a short runner is negligable in most practical road race and street applications.

 

what’s optimal can vary by the criteria:


lighter, less forward CG, less heated surface area exposure

vs

ease of routing/installation, ease to access/maintain, and so on


like fabricating a manifold; choose your flavor:

18 Ga 625 Inconel - king of champions where cost is irrelevant

Sch 5 T321 - light and strong, not all that expensive now (or 16 Ga OD tubing, ~same thickness in the sizes we use)

Sch 10 T304 - sufficient, a little cheaper and easier to find

Sch 10 T321 - reinforced concrete poop-house

Sch 40 ANY - concrete galoshes to keep your feet warm during the trip to the bottom of the sea.
.

 

Regarding short runner VS long runner turbo manifold and the often ignored low volume VS higher volume turbo manifold.

TL/DR-
Yeah, it depends

i chose my words very carefully when I said I chose short runner turbo manifold for better THROTTLE response with EFR 8374.
_____________

Turbo response and throttle response will of course go hand in glove since the turbo compressor wheel inertia is an impediment to pre throttlebody intake flow.

A BOV that is open under vacuum helps tip in throttle, but on the responsive EFR turbos we are talking about boost (and so closed BOV) around 2,000rpm depending on load.

There are clever ways around helping decouple throttle response and turbo response (ITBs, cold side boost control, drive by wire, etc), but a short runner turbo manifold is a simple solution that has worked for OEMs.

I care how the engine/turbo behaves off idle and even free reving without load because it is something I have to deal with every time I stop or downshift.

Throttle response is something that wont show up on a dyno nor even in any performance test (given the driver can learn to drive around driveability issues- left foot braking, partial clutching, etc).

Some people will really enjoy getting the most performance out of a difficult to drive car. I used to, but now even a 400-500rwhp RX-7 is pretty slow accelerating compared to modern modified cars- so, I feel like "why bother".
___________

A short low volume exhaust manifold will improve throttle response/turbo response ONLY at very low exhaust flows.

Low volume/short runner turbo manifold is something one might use if they are happy with 350-400rwhp out of an EFR 8374 and enjoy easy to drive cars.

At higher exhaust flows a long runner turbo manifold can improve turbo response as it is easier to package in equal lengths, less restrictions and provides more volume for pre turbo exhaust expansion (we have interal and external combustion engines).
So, if you want lower boost rpm threshold on a BIG turbo where you are without any appreciable boost till 3,000rpm- you might find longer runners help, especially with a bridgeport or semi-p port dumping more air into the exhaust.

This long runner bigger turbo car would be perfect for a race car where the driver has the skill, gearing or DCT to keep the rpms over 3-4,000rpm everywhere he wants performance.

At higher exhaust flows low volume/short runner manifiolds are simply a restriction over an equally well designed larger volume/longer turbo manifold.

So, its a balance of feel and turbo sizing. If you are building an 800rwhp 2rotor you are already commited to sacrificing throttle response/turbo response at very low exhaust flow levels like from a stop at idle or even 2,000rpm.

I used 800rwhp example because if you can live with only 650-700rwhp an EFR 9180 or 9280 is basically a small responsive turbo for a 2rotor compared to anything besides a smaller EFR.