Anyone have any good turbo recommendations to be used with this?

 

*yawn* Is that the best you can do?

But seriously... Holy schnikeees. Always raising the bar that long haired Londoner.

I'd slap it in a N/A PP and let it howl!!

 

I recommend a Mohaghany coffin.


I have to ask if thats the real deal, because last time I inquired about a real Mazda MFR e-shaft, I was told close to $100,000 used and he wasn't joking.

 

Quit Teasing!

 

Aaron is going to compete with Scoot over who has the best quad rotor! lol

 

It's going to take him wayyyyy too long to get this going too, I bet 5 others beat him to it...

 

Nice, two 13B shafts joined together...

 

Hey Aaron....I hope that is a multi peice shaft.....cause it looks pretty awkward to have to keep sliding irons over all those lobes lol.

So why isn't it built yet? j/k!

 

That's basically the plan. Peripheral port NA with S5 rotors, dry sump, etc. I've had the intake design in my head for a long time and as straightforward and unoriginal as it is, it's going to be a very nice manifold regardless. The exhaust will be a bit more interesting.

It's not an original 3 piece Mazda shaft. The Mazda 4 rotor shaft is 3 piece with a main bearing to support each rotor. Most homemade and aftermarket shafts are two piece and use only three main bearings to keep the cost and length of the engine down.

This shaft is sort of a mockup or prototype for the final version. As such, this particular shaft will never see the inside of an engine. It was a very worthwhile exercise because it's really cleared up the methods of joining eccentrics. It's one thing to think about it for almost 7 years, but quite another to actually put your hands on some shafts and start tinkering. I'd say that any machine shop worth it's salt is capable of making a 4 rotor eccentric from two 13B eccentrics for less then $1000. The thick center bearing iron is actually the challenge, but not huge...

 

While I know this will probably be intended for Tiina (have to rename her!)... tell us, do you get an evil grin when you look at this and your newly acquired RX-5 cosmo together? That'd be one hell of an original pairing!

 

 

It's about time you brought this Into light. You've been dragging this Idea for awhile now. Take my 20B. Think for tomorrow - not last year.

 

Sounds cool Aaron, so how's this one held together? Is it a taper and key?

 

Definitely Tina's next engine (or perhaps 2nd next because I want to try a full bridge 6 port 13B). I have a functioning 13B from a GSL-SE that will go into the Cosmo after a rebuild and street port. I've been flip flopping as to whether or not to use the 20B since it's the modern Eunos Cosmo engine but then I have to cut and hack the car. Also how much power does it need really? The 4 port 13B it came with can't be more then 120HP and I don't think I've ever heard anyone say those cars are underpowered. What I don't want to create is a difficult to drive monster that has 500HP of unusable power. I want a turn-key car that almost anyone can drive, with decent mileage, quiet exhaust and a slight bit of go when necessary.

Mazda of course uses a taper and key which is probably the best way to do it. But it's hard for a "backyard engineer" to create a good taper that will be precise enough. In my opinion, the best way to create a backyard 4 rotor shaft is to use a set of splines. The Mazda piece already provides a thinner snout that will fit into the rear of the "front" shaft after the hole (which is conveniently perfectly centered...) for the oil passage is bored out. The end of the spline then seals with an o-ring. Most machine shops can create splines without any issue, as can better driveline shops. The key is keeping the lobe phasing at exactly 90 degrees but as luck would have it, the lightening holes in the lobes are very convenient as a reference for jigging.

 

What about a 13B-RE for the Cosmo? Would be mighty nice with a set of twins.

 

so is it a mullet or is that the lighting... at first i thought it was a massive camshaft, ive got alot to learn in this rotary game lol

 

That's another option but then I have to package and intercooler and all the piping. The frame rails in the Cosmo are very narrow, so much that the stock manifold only has an inch or so of clearance. And there's the mounting issue as well.

Looks like the GSL-SE 13B will bolt right in for the most part. I may need to make my own headers but that's no big deal. Also I'd like to run a cat so fuel injection is a must.

 

Aaron, seems you're just crazy enough to make it happen and make it work well. GL and keep us posted.

 

That's wicked! A four rotor shaft in Ontario! Haha.

I have a few questions about the "design". I'm assuming you're going to modify a single intermediate plate to hold the single middle bearing and you're probably going to run some sort of double sided stationary gear off that at the same location as well. So that means that single bearing is going to have to support the centrifugal force of the 4 off-balance rotors as well as the torque induced on the stat gears by two rotors. Now I haven't done any of the math to calculate the magnitude of these loads, but I would seem very cautious of leaving the single "short" bearing having to support it.

All of the four rotor shafts I've seen have four bearings. They lack the fifth bearing in the place where you have your centre bearing. This way, two rotors have a bearing on both sides, two rotors have a bearing on one side (like in an OEM rotary), and each bearing has a single stat gear.

How are the two shafts coupled?

I'm just curious as this is something I want to do as soon as I have a chance, and I want to know as much as possible before I tackle the project!

 

I've thought about that a bit as well. Scoots 4 rotor used 4 bearings, one bearing, a lobe, another bearing, two lobes, another bearing, the last lobe, then another bearing. That's a LOT of bearings and also would make the engine quite long. Though his was made up of 12A parts so that compensated for the length somewhat.

The original Mazda 2002 4 rotor design used three bearings with a majorly thick center section and a double wide bearing. They abandoned that for several reasons.

As we all know, the 787B had one bearing per rotor, but then again it was designed for endurance racing at constant high RPMs.

For a street driven 4 rotor firing at 90 degrees (instead of those "bastard" 4 rotors that fire like two 2-rotors) it should be fine. Any more bearings and the engine becomes very thick.

That said, the Kiwi-RE kit uses two "inner" main bearings, not unlike Scoot.

This shaft is a prototype mainly for exploring ways of keeping the phasing correct using backyard engineering. Interestingly enough that is very easy due to the balance holes in the lobes. They are on center and make it easy to jig. This shaft was coupled by cutting 3/4 off of the front snout of the rear shaft, cutting the flywheel taper off of the front shaft, and then boring the oil hole in the front shaft to accept the snout of the rear shaft. Then a threaded pin was used to secure.

A much improved version of that is to just bore the front shaft to accept the keyway of the rear shaft. That's probably the easiest approach.

My personal favourite would be to spline the rear shaft and front shaft to match, seal the oil passage with an o-ring at the front of the spline and then secure with pins or have a slight taper machined.

I've been flip flopping the past year whether or not to go ahead and machine everything I've had down on paper for years, or just buy the Kiwi RE kit. In the end, the cost would be the same but the Kiwi kit is proven and all the hard work is done. The only disadvantage of that is I no longer have bragging rights to say that I made everything.

 

There would be no difference in length since the bearings are all housed within the centre bore of the intermediate irons, or at least they are that way in all multi rotor applications I've seen. The length of the engine is determined by the types of rotor housings obviously (12A or 13B), but your 50mm intermediate plates are always going to be the same width whether there is a bearing in them or not. If you take a look at the centre plate bearing modifications that Jeff Bruce does, then you'll know what I mean. But I'm assuming you're going to be using a modified centre iron to house your centre bearing, a la Jeff Bruce (and Steve from MPS... if you need local help with the machine work from someone who knows these things inside and out), but I could be off on this assumption. Perhaps you have something else in mind as a bearing carrier that's different from what's be traditionally used.

That's a good idea, but how are you going to hold them together? The only option I can think of would be to thread the end of one shaft and have a long tension bolt running through the centre of the shaft. But then you may run into oil supply problems with that.

 

I was wondering the same thing...there shouldn't be any difference in the engine length if you used 5 bearings (whic his what I would want to do). Have a bearing supporting the e-shaft in between each rotor.

 

Are you referring to the center plates that would be used with two piece eccentric shafts?

It's a bit of a brain fart on my part, but by thicker I keep thinking of two main bearings side to side like the 2002.

Another one of my plans was to machine a thick center iron like the 20B and have the engine assembled in two halves (two sets of tension bolts both meeting on the center most plate. But for obvious reasons this is a stupid idea.

I've already got this one figured out. The original plan was to machine a new center plate from scratch, but its far easier to machine a double ended stationary gear with bearing to fit perfectly within the opening of a stock center plate. Then pin it from the sides much like the rotor stationary gears. It's then easy to feed oil to it via a line that follows the stock filler passage.

Just me being confused thinking of three different engines at once. Right now my mind is full of 4 rotor thoughts, going over what I've been thinking about years back and adding new ideas.

Long hollow bolt. If it is clearanced to fit within the larger bore, there will be no vibration issues.

The only reason I would want to stay away from tapers is that they are hard to machine for the backyard builder.

 

Nice to see you finally got that thing into one solid piece.

 

Eagerly looking forward to this one, but with this project in it's early stages now. What's going to happen the that short throw shifter you were working on?