ITT we fix the 13b engine

ITT we fix the 13b engine.

Other urls found in this thread:

liquidpiston.com/
en.wikipedia.org/wiki/Nikasil
twitter.com/NSFWRedditImage

I start:
>silicium carbite apex seals
>nikasil coated irons and hoseings
>higher compression ratio
>3 stage variable intake ports
>second oil system for chamber
>rotor deactivation

Offset combustion chambers, laser ignition, maaaaaaaaaaybe hybrid boost for low RPM.

liquidpiston.com/

>liquidpiston.com/
>70ccm
>3hp
>@10.000rpm

Most 4-strokes are better than that...

...

Simple. Add pushrods.

>liquidpiston.com/
I think its more about the physical size of the engine.

The 70cc model is 4lbs.

>pushrods
>in a valveless engine

New from WAT Racing Development, the same group that brought you the revolutionary Direct Exhaust Injection..

>Direct Exhaust Injection
A system like that actualy exists and is used to burn out diesel particle filters.
The fuel burns in the hot exaust and heats the DPF to regenerating temperature.

Oh yeah, I completely forgot that's actually a fucking thing for diesels.

I even took a test on that shit while studying EVAP systems, I really don't retain a single goddamn thing

This engine litteraly has apex seals in the housing.

add a 3rd smaller rotor with a smaller housing. When engine load is small, shut off the big rotors and just use the small one.
This would improve its mpg, i don't think you can really do anything about the apex seals or needing oil combusted.

You could use pushrods to activate the variable intake port system and the throttle bodies.

Attach the engine block to the flywheel so whole thing rotates. suddenly the power is doubled.

That actualy was felix wankels first idea...

or get a goddamn v8 faggot. that's the only way to "fix" the wanker rotary

workaround =I= fix

use a lada engine instead

VAZ-430, 270hp naturally aspirated two rotor

>VAZ-430
>prototype aircraft engine

Did it even last longer than a normal 13b-rew?

rotaries as aircraft engine usually see about 2000 hours between rebuilds which is roughly 240,000km of nothing but steady highway driving
they werent the main engines used in the helicopter though, were just take off engines. apaprently VAZ-430 was also used in some cars for russian authorities only, but were massively unreliable turds (this probably had something to do with lada not using tachometers and nature of the rotary needing to be revved out. and general russian build quality)

>Implying it needs fixing

ITT: People who have never owned nor driven an rx7.

It my be fun to drive, but the fuel consumption is pretty bad and the compression loss between 100.000 and 150.000km on RX-7 or 70.000-100.000km on RX-8 is a problem.

Silicon carbite apex-seals combined with a nikasil couated housing and a proper synthetic oil system would make it last longer.
A rotor deactivation system would make it more efficient.

>implying it's perfect

ITP: A guy who has never owned an RX7.

serious question:
could you combine free valve technology with a rotary engine?

>could you combine free valve technology with a rotary engine?

>A rotor deactivation system would make it more efficient
would only work in low load, constant rpm situations. which is where the rotary already has its best fuel consumption

My understanding is the housings are hard-chromed (not decorative chrome) which means they are 60+ on the Rockwell scale. Is that nickasil harder?

I just looked and it's 69 to 72 rockwell, so that's an improvement.

Owned a 80 4-speed and a 79 Savanna by the way.

Actualy it is the opposite, rotarys are shit under partial load, but insanely good under constant load.
If you have a economy switch to shut down a rotor, you can doubble the load on one rotor, wich would give you a decent fuel economy.
If that system would be on a RX-8, it would shut down one rotor in the city or cruising on the highway.
You could also implement that in the cruise controll system.
The last series of NSU RO-80 engines had silicon carbite seals and lasted 200.000-300.000km before compression loss started.
These seals where pure silicon carbite and not just coated like the normal ones.

Nikasil is actualy silicon carbite in a nickel matrix.

would be more efficient to add valves to do one cycle per revolution instead of 3 cycles per revolution at lower load, for better fuel economy.

The engine would run extremely rough and the rotoary piston might warp.
The fuel injection pattern would have to be insanely complex as well.
A 3-rotor engine with the economy rotor in the middle would be better since it runs even, it would distribute the heat evenly and it wouldn´t need a special fuel injection pattern.

What about a diesel/electric Rotary?

>diesel for muh tork
>electric for muh fewl

>boom

>diesel rotary
That was tried, it is shit.

Also rotarys have insane specific torque.

The problem with shutting rotors off is lubricating everything without making it choke when it has to start burning fuel again.

Honestly, a 'valve' system combining peripheral and side ports that can be opened and closed as needed would do a lot to improve the overall powerband. But until they find a way to lubricate the housings and seals without being a total loss system, it's dead in the water as far as emissions go.

Yeah, a three rotor with 2 end rotors deactivating is simpler and fine too.
I actually don't understand why that hasn't been done already. It's quite simple.

Without all the reciprocating mass in a rotary engine, and how poorly they engine-brake, a 3 rotor RX-8 that ran on one rotor under light load could have been an amazing car.
They even could have narrowed the rotors down a bit so it's 500cc per rotor or so for a total of 1.5L.

Just seems like the Renesis was lazy.
Added side port exhaust (some good, some bad. especially bad with the shared center port getting hot as fuck), upped compression, lowered rotor clearance with shorter apex seals, slightly different side seals to clear the side exhaust, and the multiport intake.
But besides that, it was very similar to the 13b. The technology was there at the time that they could have been more adventurous.


And yeah, you're right about it being rough with what I said. Though, since you'd have valves, you could let out much of the exhaust during the normal stroke, just letting some continue to expand in what would normally be the intake stroke, but then it's going to compress a third time on what is the normal compression stroke before it can continue to expand, unless you added a third set of ports on the opposite side of the intake ports to bleed again before that third compression. But either way you're only getting energy from a portion left over each time.
Regardless, that's the main cause of inefficiency in rotary: the very short exhaust stroke. The 16X had narrower and wider Rotors, for a longer stroke.

We have this thread every day
Only Mazda (if anyone) can improve upon the rotary. AND EVEN THEY CAN'T. They have the foremost technologist and a vested interest and they can't even meet simple emission standards.
Stop pushing this shitty outdated technology, it's never gonna happen. You might as well spend your time talking about good engines like the ls.

Doesn't the Renesis have oil injectors that lubricate both the apex seals and the sides of the rotor, regardless of combustion..?

>Honestly, a 'valve' system combining peripheral and side ports that can be opened and closed as needed would do a lot to improve the overall powerband.
Like a portal valve to increase intake velocity at low rpm and stratifying the charge, while holding back the exhaust from exiting too quickly at low rpm as well.

But periphrial ports are dead. Side ports would be simple to "valve" off, though. Rather than the Multi-side-ports on the renesis, you just need a flap that raises in the port to make it gradual like continuously variable intake runners (but right outside the port). Problem is making it aerodynamic enough and not blocking the water jacket, etc.

You could reduce the amount of injected oil if the engine isn´t running.
But resarting it shouldn´t be a problem anyway since it is already reving at a decent rpm due to the middle rotor.
It would also start faster due to higher starting rpm.
Peripheral ports have issues as well, you need even more oil injection with them to lubricate the apex seals.

The total loss system could actualy meet emiossion standarts if you use special synthetic 2-stroke oil.

Mazda can definitely meet emissions with their new patents (longer stroke, better aerodynamics, and a pre-cat). A Renesis with a pre-cat would meet emissions, as well.

Their problem is not being sure that the resulting car would be marketable.

The RX-8 was damaging to the brand, with all the recalls and people not changing the oil. Reliability and being easier to maintain is the big concern, not-not meeting emissions.

>fuel consumption is pretty bad
you know ill never understand this being an actual concern on a second/weekend car

sure your daily should focus on having good MPG but when it comes to the car you only take out on saturdays and sundays assuming you dont have other shit to do that day I could not possibly care any less what kind of mileage I get.

>being easier to maintain
Really not a concern

Only the techs care how easy a car is to work on, John Q. Public only cares about the dealership experience

>That was tried, it is shit
Nah it was decent. Could run on diesel, gas, jet fuel and methanol fine without engine adjustments, no matter how shitty the quality was. And could start in -32f weather normally without any assist.
Was a 5.8l per rotor behemoth though (output of two rotor config shown)

Renesis wasn't lazy, it's still a 13B, but with every trick Mazda learned over 40 years thrown at it. NA it put out more than the FC3S did stock for stock.

They do, but again you're looking at a total loss and that can build up if only relying on the pressure without combustion to really force it out.

I'm talking a full combo port, like taking a PPort iron and putting it between 6 port sides and valving it from there. Mazdatrix kinda did this on the exhaust side of their REW/SP hybrid but the exhaust was too big/not timed right to really make it work.

Whoever said that Mazda is the only ones that will really be able to fix the rotaries problems is probably right. The Renesis is a big indicator of that.

The 13b.MSP had some nice ideals, featuring the sideports instead of the peripheral ports to reduce injected oil.
The main Problem was to low oil injection and apex seals out of the wrong material.
Also the 13b-msp gave us variable ports, wich I think schould be on the hypothetical 3.rotor as well.

Basicly I would build it the outside rotors exactly like the 13b-msp with nikasil coated housings and silicon carbite seals.
The mid rotor would be a more narrow rotor with smaller intake ports .

>your daily should focus on having good MPG

wrong

Adding a second small single rotor engine and sticking it in front of the main engine would serve best
Kind of like a hybrid but all rotary power

Too complex to try and shut off a rotor
Why would you want to mimic blowing seals

Most people don´t have the money for a second car, so they focus on mpg as well.
The 3-rotor design explained in would be ideal for that.

The RX-8 wasn't a "weekend car".

It was supposed to be something you can daily drive and bring your kids or the children you abducted to school in on the weekday, but then also have fun with on the weekend.

>Only the techs care how easy a car is to work on
Okay well I don't really know of a car that's easier to work on than an RX-8. Easiest car I've seen to change the oil and change spark plugs on.

Even if that's only a 1 rotor engine, 650lb-ft of torque from a 5.8l rotor on diesel sounds awfully inefficient.

>Renesis wasn't lazy, it's still a 13B, but with every trick Mazda learned over 40 years thrown at it. NA it put out more than the FC3S did stock for stock.
Jee it's better than an FC, big deal.
If they had gone with 3 500cc rotors, maybe they could have gotten 275hp and closer to 25mpg with rotor deactivation. Then it might be a car whose sales didn't tank after just a couple of years.

You could keep the engine temperature pretty equal in the whole engine if you have only the middle rotor running in a 3-rotor.

...

>650lb-ft of torque from a 5.8l rotor on diesel sounds awfully inefficient
Doesnt matter

Wankel engines are shit for high compression ratios as on a diesel due to the form of the burning chamber.

How would you keep the engine temperature distributed equal on that engine?
How would you start the main engine?

>Basicly I would build it the outside rotors exactly like the 13b-msp with nikasil coated housings and silicon carbite seals.
>The mid rotor would be a more narrow rotor with smaller intake ports

Interesting idea. It'd be harder to balance under load, though.
If I'm not mistaken, the best way to balance a 3 rotor is to space them 120 degrees.
If you do that with less power generated by one of them, even if you balance it with a counter weight it's still going to be imbalanced under high load.

It would make the most sense to simply have all 3 rotors be narrower, for a total of 1-1.5L of displacement between all of them.
If you have 3 450cc rotors, you're still going to have a significant load on that single middle 450cc rotor when it's operating alone while the others are deactivated.

The downside is that you'd need quite large intermediate housings to feed the center rotor as well as the end ones under load, but with the same lightening techniques in the 16X, that shouldn't be much of an issue.
I think Mazda gets too autistic with the small packaging, and having too small of those intermediate housings makes the engine suffer in many ways.

I think this is retarded. Who even came up with that? Even if you went with mismatch sized rotors, you'd want the auxiliary rotor simply be *narrower* and just not uniformly struck down, in order to generate enough torque to do its job and have enough stroke for greater efficiency since efficiency is the whole point of that stupid thing.
Who even came up with that idea?

You're right about most, but the variable port deal is a pretty old design. Pic related.

Maybe they could have if they were given the capability to completely retool for that, as it is they had to take the 13B as far as they could go. And 3 rotors are usually reserved for the Cosmo.

nvm I see who came up with it.
Dude should stick to decades old suspensions and shit.

Actually, you can achieve higher compression with them. It means a more clearly "8" looking shape to the combustion chamber. They just won't work at high RPM, that way.

The power output over one rotation would be a little bit unbalanced, the engine itself wouldn´t be unbalanced.
A Wankel can be perfectly balanced wit counterweights.
NSU actualy made the wankel Spider single rotor.

The small like 300ccm rotor would make verry small side ports possible and be under a high load all the time, wich gives you a great fuel economy like 40 mpg+ in real life.
The other 2 big rotors would give you power on demand.
Since they always rotate with the small one, they would start pretty much instantly.
So you would have a 0,3L wankel for efficiency, and a 1,6L wankel for max power.

Since the ports for the middle rotor could be smaller, the Irons could be even smaller than on the actual 13b-msp.

that's a dorito

The combustion itself creates imbalance.
Combustion forces will be pushing harder against two of the rotors over the third. That will make it rough under high load, I'm certain.

Even sized rotors is so much simpler. You just need large enough intermediate housings so it can breath as well the end rotors.
Or, further, you change the design so you don't have problems with side ports in more-than-one-rotor configurations that are especially worse in more-than-two-rotor configurations.

It won´t be more imbalanced than a 2 rotor design and since rotaries are even in single rotor versions smooth as fuck, it won´t be a problem.

Peripheral ports are problematic due to uneven Apex-seal wear and oil loss.
The side port engines might be longer, but that isn´t much of an issue anyway if you have 4 rotors or less compared to 4 cylinder/V-8 engines.
The middle rotor shouldn´t breath as well as the outer rotors in my design anyway, since it has small ports and less chamber volume.

Dont get autistic about dimensions in some paint mock up
An aircooled single rotor engine the size of pic related makes 38hp and weighs 10kg. Torque is irrelevant. Thats all the power it needs to power the RX8 along the highway with greater efficiency.
Why make rotors narrower? Now that is retarded.

>Who even came up with that idea?
Not someone who thinks having a tiny little toothpick width rotor jammed in between two normal sized rotors is a good idea

Have you taken into consideration that the deactivation rotor will wear at a considerably higher rate than the others, leading to a maintenance nightmare?

>If I'm not mistaken, the best way to balance a 3 rotor is to space them 120 degrees.

And a 2 piece eccentric shaft and floating front housing.

It may wear out faster than the other rotors, but it will still last longer than previous 13b engines.

You can split all stationary gears instead of the excentric shaft as well.

>tfw want to find a european butterfly LIM and port the aux ports in my car so heavily they'll only be good above 6-7K RPM just to see how it drives but can't find the eurospec LIM and will have to modify my spare US LIM with motorcycle parts

about the same bar the revs

Holy shit how many strokes is that? Has one ever been made?

Use it as a range extender for the C-X75 with a dedicated oiling system.
>AWD electric torque
>Constant high engine RPM = great efficiency
>Lower service intervals
>The C-X75 gets built
Wins all around.

>serious question

wrong thread fucking kill me good luck with your rotaries

Some cams would be nice to

New from WAT Racing Development, the same group that brought you the revolutionary Turbocharged Naturally Aspirated V8™..

You faggots that bad at reading graphs?
>560 kW @ 3600 rpm
>1700 Nm @ 2200 rpm

>Veeky Forums
>smart

o i am laffin

>Three-rotor
>Direct Exhaust Injection
>Triple-turbo Twincharged (one turbo per chamber feeding a single supercharger)
>Generate enough chamber pressure to ignite and run on diesel fuel
>Nanodiamond-Ceramic apex seals
>GT-R-style plasma chamber coating
>oil additives to make the exhaust smell like Black Ice

Also a Quad-rotor Quad-turbo dual-twincharged R26B in a caged NA Miata

>silicium carbite apex seals
>i know fuck all the post
enjoy the increased engine block wear

Regards, someone who uses carbide tools to machine metal

It is a 4-stroke engine and yes, ther were prototypes.
>what is Nikasil
en.wikipedia.org/wiki/Nikasil

>>what is Nikasil
The reason the Porsche 944's engine was such a pile of fucking shit.

>Porsche 944
That was litteraly one of verry few Porsches without Nikasil coated cylinders.
It had Alusil cylinders instead of Nikasil cylinders.

..and why? the rotary is inferior to piston engines so who cares!!!

This. The only way to truly fix a rotary is to put an LS in it.

yes this. I'm glad 100% of Veeky Forums has finally come the a consensus on this issue.
Perhaps torqulets have learned?

>Perhaps torqulets have learned?
Yuros are still allowed on this board, you know.

moving seals should made from high carbon steel, ie. cast iron. Because they have oiling effect too. iron carbide have pretty good oiling effect so that they doesn't wear much. ( i'm mechanical engineer)

As long as the lubrication is good, you are correct, but as soon as it starves on oil, carbite seals are supriror.
NSU for example had cast iron seals, sinter metal/carbon seals and silicon carbite seals in their rotary engine.
The carbite seals lasted wuch longer than all other seals before.

Some people have done some funky shit with the RX8's engine.

Running the engine hyper-lean and the timing with a negative split causes it to run like an FSI engine.

The sweep of the rotor naturally creates a rich mixture, which can be ignited by the trailing plug. This ignition compresses the rest of the leaned mix to the point where it's able to ignite, then the lead plug fires. Big chamber area keeps combustion cool and nOX down.

Direct injection after intake ports close. At least at cruise/idle loads. Boost injectors at each port for more stomp as revs increase.

Add a centre bearing and multipiece eccentric for better oiling and reduced high-rpm flexibiblity.

Blown-down turbines coupled through to drive either the electric loads, or back into the engine to capture was exhaust heat

Water injection.

DIRECT WATER INJECTION

But actually water injection means you don't need to run as rich at a given boost pressure and helps with the massive heat problems.

In a Wankel engine you inject water to clean the exaust ports.

How do you figures this? Fuck off idiot

Come back when you won le mans gmcuck

The ports will be burned free, so it isn´t going to have exaust problems.
The silicon carbite apex seals and nikasil coating are more wear resistant than steel.

Therefore it should last longer.

Steam is great for keeping the engine clean as well. This is something that can be retrofitted unlike most of the suggestions here.

>The silicon carbite apex seals and nikasil coating are more wear resistant than steel.

Apex seals already have a Ni-SiC coating

See the second part of

Solved