zyounker

From what i read, you want to collect the three tubes at about 13 inches. and put a flange on it. Effort should be made to make it equal length. so the 2 that are closer together should go towards the 3rd that is seperated.


Or you could go tri-turbo and the exhaust is more simple, but the intake will need to be collected



-Zach

Red-Rx7

Ah yes.. the Manifold.

Tripple turbo eh? lol That would be pretty nifty, but also expensive... What about practical too?

Talk about an intake plumbing nightmare, and exhaust too.

I was thinking more along the lines of a manifold my alamo, like this one:



Thats the one Tito made for Derricks car.

Hmm... tri turbo.

Mike

zyounker

That is pretty much was what i was trying to describe

But i would never heat wrap a header unless i didn't want it to last.


If you want to reduce heat do ceramic coating..


Tri-turbo would be a bitch in an FD.. But it is do-able, i have seen pics.


-Zach

$150FC

tri-turbo's been done in a supra, but what's the point? any advantage you get in spooling by a smaller turbo is lost by only having a third of your exhaust gas going to one turbo

Jedon

Just to answer the question of where the 240ci came from, it is the 118ci multiplied times two to get piston engine equivalent figures since the rotors spin once for two revolutions of the e-shaft. A 13b rotary is usually considered to be in the same class with 2.8L 6cyl engines for comparable power output. Please correct me if I am wrong here.
-Jedon
ps. so you end up selling your 60-1 kit?

Red-Rx7

Hey Jedon!

No I never did. I still have all the stuff here.

I do appologize for never getting it all worked out with you, for I have been ill.

I will PM you.

Thanks,

Mike

Evil Aviator

Well, first of all, the factory service manual rates each 13B rotor at 40.0 cu in, so 2 x 40.0 = 80.0 cu in.

The 20B uses 3 13B-sized rotors, so 3 x 40.0 = 120.0 cu in

The standard equation for determining the volume flow of a 4-stroke piston engine is:

volume of air in cu ft / min = (engine rpm x engine cubic inch displacement x volumetric efficiency) / (1728 x 2)

Notice that ominous "2"? Where does it come from, and what does it mean?

Engine displacement ratings are based on the total volume of all pistons (or rotors). In a 4-stroke piston engine, each piston must make 4-strokes (duh) to fire once. The crankshaft rotates 2 times for each 4 strokes of the pistons. Therefore 2 crankshaft revolutions / 4 strokes per power stroke = 1/2 ratio. From this we see that a 4-stroke piston engine only fires 1/2 of its displacement per crankshaft revolution. This is where the 2 comes from in that formula, indicating that the 4-stroke piston engine's displacement must be divided by 2 to yield its displacement per revolution. For example, a 4-stroke piston engine rated at 400 cu in displacement fires only 200 cu in (half of its pistons) for each revolution of the crankshaft.

The rotary engine fires all of its displacement per eccentric shaft revolution. OK, follow along - a 20B has 3 rotors, each of which is rated at 40.0 cu in. So, 3 rotors x 40.0 cu in per rotor = 120.0 cu in total rating for the engine. Each rotor fires 3 times per rotor revolution, but the eccentric shaft rotates 3 times per rotor revolution, so 3 / 3 = 1 / 1. Therefore, what we end up with is an engine which fires its entire rated displacement for each revolution of the eccentric shaft.

Obviously, this presents a problem when trying to use the 4-stroke piston engine forumla with rotary engines. Therefore, you have 2 choices:
1) Simply remove the 2 from the formula, and then enter the rotary engine's displacement
2) Double the rotary engine's displacement, and use the formula as-is. This is where the 20B 120 cu in x 2 = 240 cu in.

BTW, note that the "2", which is actually a "1/2" if you look at the formula, may also be represented by a "0.5" in the numerator, or even pre-multiplied by the 1728 conversion factor so that you see a "3456" instead. It's all the same, though, and depends on who wrote the equation.

Oh, and I'm still looking over those numbers - that's a lot of Shiite to go through, and I don't use the same method as that web site, so it's going to take a while.

Evil Aviator

So do you want to contribute any of your flow rate knowledge to help out Red-Rx7, or am I the only person here who has a clue?

Kurgan

So spill it, what is the CFM of a non ported 20b? i guess I'll have to find out how much CFM my motor makes since it has a pretty aggressive port....

Greg

Please do tell us the stock airflow in cfm for the 20B and Cosmos 13B RE.

10sec rx7

a 1.32 will choke a 20b, i run a 1.32 on my 13b and it is at full boost by 3700rpm, on a 20 it will have full boost by 2400 and be choked by 6000,
i would got up to something like a 1.52 rear with a TA45 compressor,
get a turbo made to suit your needs, dont go buying one of the expansive t66 t76 etc turbos, just get a garrett TA45 cheap and very powerfull turbo's.

soul assassin

my head hurts

Red-Rx7

Lol.

Why? Can't you offer some valued math eh?

Mike

Greg

I agree, I think a 1.71 would be ideal for high rpm power but maybe less realistic for space and fitment.

mmmmmmm GTB88 with 1.5 turbine housing ona 20b = :jerkoff:

Greg

Still waiting for that airflow. I was speaking with j9fd3s and he is going to determine the stock airflow of a 20b after testing on his running 20b. After speaking with mike there is 2 ways to calculate the airflow of the 20 b.

1) Finding out the airflow of the 13B Cosmos Engine (more restrictive exhaust ports) and dividing it in 2 and multiplying it by 3 or you can come up with the airflow based on Air Fuel. We Will see. Nick Id imagine your porting to be around 30% more than the stock ports so you can factor that in. Id give don a call and ask him how much of an air flow improvement his massive street porting would improve it in %. Id also consider dowl pinning the motor a little more as the 20b is more prone to twisting than the standard 20b and with the horsepower numbers you are probably going to be making its gonna be twisting Lata yall, BTW Im drunk as ****.

Evil Aviator

Air-Air intercooler efficiency depends on its design, the turbo airflow rate, and the vehicle speed. A good bar & plate air-air intercooler will have about 60-90% efficiency. Also note that the intercooler and associated plumbing will cause a pressure drop, so although the air will be cooler, it may also lose up to 2 psi of boost on its way to the manifold.

Yes, the efficiency islands are different, but also note that the compressor rpm is different. Also, if you look where the surge line meets your desired boost level, the smaller T66 allows the potential for max boost to be made at a lower lbs/min airflow rate, which means that full boost will be reached at a lower engine rpm than with the T72. If you reverse-engineer your numbers for the flow rate you can calculate the engine rpm.

No, that is a compressor map. To determine the lag, you would need a turbine map which took into account the engine exhaust, engine load, compressor inertia, etc. Some things are best left for the trial & error method.

BTW, I looked at your numbers, and they appear correct as per the web site. However, I have some issues with that web site. For one thing, I believe that the ^0.263 exponent is supposed to be ^0.283, but I've been out of college for a long time now, so maybe I'm wrong. There are some mathematical errors, and I'm not quite sure if their temperature compensation is correct when fed back into the engine cfm. I really wish I had time to reverse engineer their numbers, derive the equations, and put it back togeter more coherently, but that would probably take me a month. Corky Bell's books are pretty good, but they have some errors, too. At least now I don't feel so bad when I make mistakes.

Evil Aviator

Man, a lot of help you guys are, lol.

OK, I guess we fall back to the Evil SWAG with the slide rule...

Given 20B-REW = 276bhp @ 6500rpm as per Mazda
Given 13B-RE = 227bhp @ 6500rpm as per Mazda
Guessing ISA conditions for Mazda's dyno numbers
Given most engines best power AFR = 12:1 on pump gas
Given Air = 0.0765 lbs/cu ft @ ISA
Guessing .64 BSFC @ 6500rpm

Using the Evil homemade redneck equation that would probably make SAE cringe:
Airflow lbs/min = BHP * (BSFC/60) * AFR

20B-REW Airflow = 276bhp * (.64 / 60) * 12:1 = 35.328 lbs/min
20B-REW Airflow = 35.328 lbs/min / .0765 = 461cfm

13B-RE Airflow = 227bhp * (.64 / 60) * 12:1 = 29.056 lbs/min
13B-RE Airflow = 29.056 lbs/min / .0765 = 380cfm

j9fd3s

iTrader: (3)

actually we know that in the car the afrs are more like 10:1, i dont think this changes your numbers much.

mike

Evil Aviator

Geez, how the engine function at 10:1 AFR? That's RICH! Maybe that's why you aren't getting a whole lot of hp out of your engine?

Anyway, the reason I posted the formulas was for people to use whatever numbers they like. Plug N Chug

Anybody have better BSFC numbers?