Hey guys, I have some big plans for when I get my degree in mechanical engineering. I plan to build an NSX with either a 20B BP/PP or my own custom 4 rotor.

Basically, I'm wondering about e-shafts for 3 rotor and 4 rotor applications, I hear mazda sells the 3 rotor e-shafts (but for some ridiculous price like a couple grand) and I'm more than positive they would never sell a 4 rotor e-shaft. I also think Racing Beat still makes 13J's or 20B PPs or something like that (I can't find it anymore).

How did people/groups like mad mike, scoot and others make a 4 rotor e-shaft?

Also I know it's a really bad idea to modify the rotor housings, but how difficult would it be to install a third spark plug (per housing) if I did all the math and timing junk and had a stand alone that could support it.

Also can you have PP's and the side ports at the same time on all rotors?

Yes I'm going all out with this and I have LOTS of time to gain information and plan it out, because I don't plan to even start this until I'm in my 40's

 

Compared to aftermarket E-shafts, that is phenomenally cheap. I was looking into it, very briefly at that, when I saw a quote for $8k for the E-shaft and necessary adapter bearing if you aren't going to use the 20B's wide center housing (which is a roughly $3k part nowadays!). This doesn't include the necessary machining to a stock center housing in order to house the bearing.

Why not? They sold 13Js, they had to have sold repair parts for it.

The 13G was the MFR 3-rotor, it wasn't anything Racing beat did.

 

we've been researching the 3 and 4 rotors for years, being in your 40's before its done might actually be true!

i assume the people making e shafts use a lathe and a mill... no really google "rotary engine" by kenichi yamamoto. the formulas for designing the multi rotor cranks are in there.

installing the 3rd plug would require adding some material for it to actually bolt too, ive seen it done, some of the JDM tuners used to sell em. if you can do the crank, the 3rd plug should be really simple. as far as getting an ECU to run it, you're on your own.

you can have PP and side ports at the same time, but why? the PP is only at a disadvantage at part throttle and low rpm, plus on a 4 rotor, you'd have to cast your own housing, which is possible but takes time and isn't cheap.

 

When doing engine swaps, you need to be more focused on making sure the engine will actually fit in the car. Shortened 3 rotor may work if it's not longer than the v6 that's already in the NSX. I know the v6 is mounted sideways and is not vertically mounted so the engine can only be so long before you hit the frame. If you can make it work, I would love to see it. If you like Hondas an easier vehicle would be an S2000. There's a ton of room in that engine bay. I was personally thinking of doing a 4 rotor swap in one of those.

 

I realize that there's a LOT of things to take into account and I'm not really concerned about the small things like "will it fit" and such right now because I'm trying to get a grasp of the core requirements.

I just chose the NSX as the platform because I know the chasis works as honda put so much time into engineering it, because it's mid engined long wheel base and the lines are phenomenal IMO (that and the fact there's a good chance they will be dirt cheap by the time I actually start this project). Also because I think the NSX is just so under rated, I think honda could have had amazing sales if they had a stronger engine in there (mind you I still love that V6 either way), I just think that a lot of people were comparing it to Ferrari's and other supercars and they's compare the numbers and see a measly 276 versus the 400+ with other supercars with a slightly higher price tag. Basically, the car was stuck in limbo

I'm not really set on 3 rotor or 4 rotor yet, I would preferably take the 4 but the difficulty factor/price is the main thing I'm concerned of (but I think I would the most satisfied with it). I think 400-600 rwhp is what I would be hoping for with a project like this (mind you it will be staying N/A just for the sound).

I believe if I do go with a 3 rotor I'll be sourcing a 20B. Does anybody know some of the main things holding back rotaries from reving high (Passed 10k)? I know the levels of friction/heat are immense at these levels, the level of aftermarket technology probably won't meet certain needs (in terms of processing speed/precision) and the weight of the rotors is probably another thing. What kind of material is used for the rotor bearings/e shaft rotor lobes? I'm thinking some sort of ceramic (low friction heat resistant material) would seriously help in high revving applications

I was also toying with the idea of using RX8 rotors/housings in a 3 rotor/4 rotor application, but it might be even more of a dick around...


Again, I'm just picking your guyses brains to see what the "average joe" in the community can tell me (average joe = somebody who doesn't have millions of dollars at their disposal for R&D). I don't think I'll have a problem with any of the math and calculations etc. as I'm doing mechanical engineering in hopes of actually doing stuff like this as my career.

__

Peejay:
Where do you think I might search around for some mazda racing parts etc.?

j9:
I meant not starting till i'm 40, haha. Can't find that book anywhere btw :S

t-von:
Not really set on honda, see above for choice of NSX. Was toying with the idea of an AW11 MR2 with a 13B PP in there today.


If you guys know of anywhere I can get some food info on 3/4 rotors feel free to share Also, videos and sounds are good too, always gives me something to do while in chemistry

 

the 20b is EASY. the cheapest way is a used JDM cosmo engine, or actually a non finished project.

the book? http://tinyurl.com/39brd5t foxed.ca is quite good..

we did the 3 rotor thing a million years ago, but we started down the 4 rotor PP trek in 2005, and we have a running 2 rotor P port, and some design drawings.

 

Yeah, I know, I hear the motor mounts are in the same location as FC's (This might be for the 13b re) or something of that sort, is this true?

Yeah, I tried googling "Rotary engine kenichi" and ever page I tried (including foxed) didn't work. Got it afterwords on http://www.rx7.net.nz/REbyKenichiYamamoto-1981.pdf

Gotta say, some nifty stuff in there. Good read.

How difficult was it to design the ports and what did you design them with? What kind of shape are they Os, Ovals, squares?

 

no the cosmo engines have different mounts, but people sell brackets

we copied the 787B port which if the SAE paper is to scale are HUGE x ginormous. the real trick is/was designing the sleeve, so that the machining on the rotor housing is as simple as possible.

paul used pro E to model the engine, but the program you use doesn't matter, either learn one, or use the one you know, they all do the same thing

 

yeah 20B's on ebay for like 3g's,.. with tranny and ecu and stuff. dirt cheap.

 

Anybody really know the benefits of certain peripheral port designs? Square/Rectangle vs circle? Would something like fluidynamics or thermodynamics give one a better understanding of something like that?

 

Some people probably do, heh. The kind of people who've done more in-depth experimentation on peripheral port engines tend to not talk about it too much though. It's sorta held that the more square shaped ones end up a bit narrower powerband; sometimes referred to as "qualifying" ports.

If you want to do an actual computational optimization of these things, I highly recommend as many fluids and thermo courses as you can get under your belt.

 

Sounds good, I figured I wouldn't get any real DIRECT input on it due to everything like that being so secretive and such. Thanks for the advice. Keep an eye out for the build in the far future. I will do my best to get it up here somehow

 

Take a look at Kiwi Re out of Australia. They make a complete kit that allows you to construct a 3 or 4 rotor and includes an Eshaft intermidate housings etc. Something to check out if your planning to put this together.