I've been around rotaries for some time now and haven't seen too many aftermarket intake manifolds. My experience with every other engine I've worked on has shown that the difference they can make is quite significant. One thing the plenum needs is volume, and the stock 13B plenum has none.
In our spare time we'e built a manifold that takes care of a few of the shortcomings... has volume, larger ports, less restrictive flow path, and utlilizes perhaps the most popular throttle body upgrade (Q45). This also keeps the TB in the same location, allowing for the use of the existing IC piping.
There are a few members on here that have shown great results using custom intake manifolds, and as no one makes once, so why not...

 

no offense but lots of people think they are making things better. Just cause it looks better dosnt mean it is.

Fastest rotary in the world uses factory inlet manifolds. I personally make 700rwhp on factory cosmo inlet manifolds.

 

Nice numbers no doubt, but have you tested your cosmo setup against any other intake manifold (other than stock 13B)? I agree that there are big numbers to be had with stock manifolds (we're making 620 rwhp on pump), but that doesn't mean there isn't room for improvement. Didn't Crispeed pick up 40 rwhp with just a manifold swap? What did ErnieT gain with his? Chibara claimed a 100+ gain at the 1,100 mark (was any of that even verified or just rumors??).
This manifold hasn't been tested yet, so I can't yet claim as to what sort of gains we will see. However the countless number of manifolds we've done have all seen significant gains (dyno tested/flow tested...) over there stock counterpart.

 

I will say this: that's the best looking aftermarket intake manifold I've seen for a rotary.

 

Nice job

 

It's shiny. So what. What's the volume of the plenum? What's the ID of the runners. How long are the runners. I've seen how real manifolds are made. They're rapid prototyped and tested, then remade and retested, and so on until the best compromise is found (simulations and equations will not give you the best design). Then the real one is made. Since this is in the tech section I'd like to hear more about your design and what testing you have done.

 

The last post states that this manifold has not yet been tested... Also, what constitutes a "real" manifold?
The basic principles used to fabricate a number of our other manifolds (1UZ/2UZ V8's - 32% increase in flow over stock, less variance across all ports... 2JZ manifolds @ 335 cfm where the Veilside/Virtual Works flow 280's/315's, again with less variance, and so on...) have been carried over to this one. Larger volume, shorter runners, less restrictive, free flowing....
The plenum volume is 1.5 litres. Runners are 8.25" shorter than the stock. Runners draw from a common plenum, stock manifold really has no plenum. This manifold eliminates the immediate directional changes inside the TB. Uses a less resctrictive/simlpified TB.
Am I going to rapid prototype a number of manifolds to figure out what I already know will improve upon the stock design given fairly simple goals? Not any time soon.

 

Howard Coleman's a techy, he's in WI maybe he'll test it for you.

Maybe before his Texas mile.

Ken

 

I assume this is for an FD, so if you increase the size of the runners of the upper intake manifold, and do nothing on the LIM, how will it increase flow?

 

nice job.

it is always a win for the FD to see new stuff, new thinking.

i raced/built/engineered a no holds barred 2 liter overhead cam Ford four banger that ran 4 discrete runners to the 4 cylinders and Weber DCOE 45 carbs w adjustable velocity stacks. NA of course.

all the tuning was done on a Stuska engine stand dyno.

while my setup had NO commonality whatsoever between cylinders the runner length was absolutely critical.

Brake Specific (and with it torque/hp) varied enormously and depended on having the runner length and velocity stacks tuned to the quarter inch.

it is no different under pressure.

the rotary generates more abrupt pressure waves than a piston engine due to the manner in which the ports close (as well as larger area) so it should be even more sensitive.

none of this is meant to be critical of the manifold, rather to point out that it is probably just the starting point in the journey to optimization.

the initial manifold, ideally, should be adj for runner length and plenum volume. and if you ever get it right it will fly.


hc

 

I must say from a visual perspective, this manifold is very appealing. As usual listening to Howard will always yield rewards and like he says, it's great to see more things being added to the rotary parts option lists.

One thing I would say though... if you end up going forward with this manifold, I would also make some that uses a stock TB configuration. This manifold and the rotary-works TB would be a killer combo(me thinks). Better yet, why dont you do a test with that combination just to see what results are achieved.

Here is a test sheet I think alot of us would like to see:

Stock lower : Stock Upper: Stock throttle body
Stock lower : Stock Upper: Rotary -works TB
Stock lower : Your intake upper: stock throttle body
xcessive lower : stock upper: stock throttle body
Stock lower : Stock upper: q45/ Rotary -works TB
Stock lower : Your intake upper : q45/rotary-works TB
xccesive lower : your intake manifold : q45/rotary-works TB

This might sound excessive, but everyone claims that their part make so much power, and at this point I think test are now the only thing that can be used to help sell these parts anymore.

just my .02

 

Flow numbers alone don't mean anything. A 3" ID tube flows more than a 1" ID tube. That doesn't mean it's better.

An increase in flow at what RPM? All rpms? 7000 rpms? Are you measuring solid state flow?

 

Trots 88, flow can be increased without increasing the size of the lower intake runners. Idealy they would be changed as well, but if the air flow going down the runnder is smoother and at a higher velocity you will increase the efficiency of the engine. Also the runner lenth and plenum volume affect the flow regardless of pure pipe diameter. This has a lot to do with resonance of the intake piping that howard and the others refer to. The idea of having more volume in the case of the rotary is to insulate or muffle some a resonance affect so there is not such a dramatic affect on the flow to one runners as a shockwave is comming back up the adjacent runner. As for testing a back to back dyno test would be idea(spoke to howard at length about this today). Finally a benefit of using this manifold is being able to keep the same diameter pipe and run it into the manifold in a smooth transistion with the q45 tb. The Greddy elbow makes a bend that is way to sharp and not needed on most cars as we no longer have the stock intercooler location. This TB and manifold setup would allow for a smooth mandrell bent pipe to be matted to it and not have unecessary sharp bends. Good thoughts guys a like the conversation about the manifold.

 

here is a link to a similar thread in the 3rd gen general, I posted a couple picture so of other manifolds that we've done over the last 7 years and there are also some other comments.

https://www.rx7club.com/3rd-generation-specific-1993-2002-16/new-intake-manifold-912818/

 

An increase in flow at what RPM? All rpms? 7000 rpms? Are you measuring solid state flow?

So you didn't design the manifold with the intention of the pulse wave from one rotor pushing air into the other rotor?

 

hopefully with the way the manifold is designed it should see a hp increase top end. It needs to be tested, you can keep speculating all you want. I talked to howard about having it tested back to back agaist the stock one within the next month or so. And to answer your question your resonance has to be tuned to a certain rpm when the pulses work the most effieciently. Once you are out of this range the violent pulses from the other runners are not helping feed the engine but create turbulence that hurts flow. There is no perfect manifold designed for max flow at all ranges. This will need testing as stated several time in this thread

 

As usual, Howard hits the head on the nail. The rotary pulses are nothing like those ina piston engine and the intake/velocity rules of thumb simply do not apply. In aircraft applications where there are no abrupt throttle changes there has been a lot of experimentation with rotary intake manifolds including length, diameter, taper, plenum size, etc. and they are STILL working on it. In an automotive application where throttle and engine speed changes are very rapid these pressure waves are even harder to deal with, which is one of the reasons why Mazda spent so much time redeveloping them. You cannot simply enlarge the plenum and runners and a better intake make. At the minimum I would say that there would need to be some CFD work done taking into account port timing/pressure waves/runner length and dia/ plenum shape ( I can actually see an equal pressure type plenum working well) and a metric assload of time and engineering. Still it would be cool to see something out there for the 13B engine that performed better than the OEM.

Below are pictures of various equal pressure intake manifolds used by Audis over the years in motorsports, and I think the design would work very well in a rotary application, just again would take a fair amount of development to get perfect, but maybe not too much development to get really damned close. These would especially work well for super-high RPM race engine with big ports.

 

How do the intake rules of thumb not apply? Compressed air travels down a runner, enters an intake port (head or iron, still part of the runner), and either enters a combustion chamber, or is reflected back (closed port) by a valve or rotor (face or side depending on port location).
How the air is affected at this specific point (valve traveling to closed position, or rotor rotating to closed position) is different due to the geometry of the port opening, the rate at which the port is closed, the geometry of the valve vs. a rotor face, and the path of movement the valve and rotor face take in relation to the intake port/ compressed air path. This will give you a resonance that varies in sensitivity, as Howard has mentioned, however the same general rules of thumb do apply.
IF we wanted to get specific to an engine, which we will, we would take into consideration the specific rpm range desired, the intake air temperature at the specific rpm range (which affects the speed at which the resonance travels), specific port size/opening time... all which give you your resonance speed at a given rpm... and from there determine the theoretically "ideal" runner length. From there we tailor the rest of the manifold (plenum shape - to aid in resonance reflection, runner taper, thottle body entry angle, injector location, etc...
Again, as stated earlier, this manifold is designed to improve top end flow (while sacrificing some torque, standard) while using the stock upper manifold as a base point. This is simply designed to increase the the plenum volume, shorten the runner length, and smooth out flow by eliminating bends found in the stock manifold, and allow for a constant diameter/smoother transitioning intercooler pipe by eliminating the Greddy elbow with the use of a common TB size. It follows the general rules of thumb, and in doing so should net results similar across a large range of motors. We will see shortly how this plays out in the real world!

 

No sir, thats not true, at least not the fastest 13Bs. Some of the fastest 13Bs have proven that a modified RE intake manifold performs, and everybody else uses a carb style intake.

 

It's always amusing to read about how some people think the rotary is so unique that the "basic rules or thumb" don't apply. It sounds like you're on the right path. Can't wait to see some results. I hope you have time to compare this design to stock. I'd like to see how much low end is lost and how much top end is gained. What are you using as a test platform? Specs? Relevant mods?

 

Part looks well designed and built. Since you are in Oshkosh, maybe you can pay a visit to Howard for some testing of the manifold. I would really like to see how it performs on a 500whp setup when compared with the stock setup.

 

PAC performance's MX-6 20B uses a stock intake manifold. Whether it was modified or not, I don't know, but it is the fastest rotary in the world currently. Now, seeing how the 20B and the 13B-RE manifolds are very similarily designed (with plenum volume), I believe that it would beneficial for high RPM/power over the FD's longer runners.

Another thing to mention, many of the people running the carb style intakes do so largely because most of the fastest 13B's are semi-pp.

 

I've talked with Howard yesterday and our plan is to do a back to back with his intake manifold, with the last turbo in his test (whichever that happens to be I don't know... maybe Howard can jump in here). So... it is in the works to test this, back to back, on a real dyno, and provide some real numbers. All we need now is to source a Q45 TB and get it up to GB before go time!

Thanks, Scorpion. We're shooting for just that... and looking at the turbo selection in Howard's lineup, we just may see some 500+ rwhp dyno's with this (again... thats the plan for now).

 

I am talking strictly 13Bs, sicne the manifold posted is for a 13B.


And thats also incorrect, carb style manifolds have been used since the 70's by most of the fastets drag cars.

 

Rx72c was talking about the fastest rotary, held by a 20B; its manifold is very similar to the 13B-RE manifold, which was also mentioned. And we are discussing Rotary engines in general, not just 13b's, so whatever man.