Technical question on the Rotary Engine and high revs

[4rotor]

Hi all! my first post here - hope it is in the right spot!

We have been having a large discussion on www.ausrotary.com about the Rotary Engine vs Piston Engine.

Essentially what I would like to know is 'Why couldn't you rev a rotary engine to 20,000rpm' like Formula 1 engines? I would have thought there would have been a number of technical/design hurdles along the way in designing an F1 engine to make those revs - and reliably.

My understanding is that in some ways the rotary engine isn't as efficient as a piston engine (on its use of fuel/heat generated etc.)

One thing that had prompted me to ask this question was - that I know of a drag racing car in Australia that has revved at an average of 13,502rpm for a 1/4 mile run. It produces around 800hp at the flywheel, it is a 13B (Twin Rotor), Twin Turbo, MoTeC controlled etc. (has run 169mph on the 1/4 mile). The engine itself only has a few high-tech parts in it, but is essentially made up of standard Mazda bits.

Would there be a physical limitation on this 20K rpm being achieved?

[Chucky]

From my understanding, their is a problem with crankshaft (wobbleshaft, whatever you call them) bearing speed/longevity at engine speeds past 12,000rpm. I know of a guy who can get them reliably to 14Krpm and is working on more, but the cobustion inefficiencies catch up with you pretty quick. I mean, it gets tough to move that much fuel in a controlled manner.

With this in mind, the above mentioned rotary tuner has begun development of a quad rotor as it got too hard to get any more out of 2 or 3 rotors.

[4rotor]

Chucky! ooohh - who is that building the 4rotor?

But F1 engines have 100's of moving parts, surely they have lubrication issues as well? What if the crankshaft was made of another material? So you are saying the combustion efficiency of a rotary engine is much worse than a piston engine? which is another thing I don't understand as - the rotary does work on a similar (not exactly the same) 4 stroke principal. How much more inefficient is it - do you know?

[avsfan733]

They are also designed to turn at 20,000RPM

issues:
1: consider that an engine needs to be able to suck and puke its displacment * rpm so pretty quickly you end up with ports (effectively camshafts on rotaries) that are so big they cause dimensional instability and serious overlap issues between the four combustion cycles (which a rotary has the same as a 4cycle, they just look different

2 rotating weigth... even though it doesn't move around as much as the pistons in an F1 engine, a rotor is significantly heavier than a piston rod combination, its also much tougher to balance a rotary because of the complex nature of the rotor movements

3 they simply weren't designed with that in mind. There are some piston engines that redline at 100 RPM or so( consider big ships) there are others that will turn 35,000+RPM (RC cars/planes) its a matter of design, almost all of the rotary engines in the world were designed for Road cars...and a few for Le Mans (I believe they were actually a seperate totally different design but im not sure)

[4rotor]

Thanks avsfan733,

What if the ports could be altered? I know that the Le Mans R26B engine internally is almost the same as a standard 13B engine, they had the variable intake system to alter port timing.

What would a crankshaft in an F1 engine weigh? the eccentric (crank shaft) in a rotary engine is quite heavy I guess - but again different materials could be used to reduce weight.

In all of Mazda's 13B, 20B, 26B engines the amount the rotor moves around in the rotor housing chamber is the same. What if like on an F1 piston engine the movement of the rotors (lighter ones) was less?

[Chucky]

That's a very good point concerning the reciprocating weight. It would be massive compared to a state of the art recip. engine.

I'm talking about the bearing speed of the crank and not the rotational speed. Correct me if I'm wrong, but don't the rotors spin at double crank rpm?

[4rotor]

But the engine goes round rather than up and down, should count for something?

Pretty sure rotor's actually spin at 3 times the speed!

[Chucky]

Therefore they are turning at 60000rpm when you're trying to do 20000 crankshaft rpm!!

It doesn't matter in this case as the speeds just start taking the inertia into the ridiculous zone.

[kevinq]

The rotor turns at 1/3 eccentric shaft speed. The Wankel could be made to rev to F1 levels but would take a lot of development to reach the power/efficiency of the reciprocating engines.

The rotary's engine geometry would probably have to change (from that used for Mazda's 10A, 12A, 13B) to reliably achieve that kind of RPM.

Nothing's impossible with enough $$$

PS. I think Wankel's are banned from F1 anyway

[4rotor]

Thanks Kevin! damn, getting confused!!! got that wrong eh

Kevin - when you mention geometry, do you mean the amount the rotor moves around in the housing i.e. the bobble on the shaft?

[kevinq]

Originally posted by 4rotor
Thanks Kevin! damn, getting confused!!! got that wrong eh

Kevin - when you mention geometry, do you mean the amount the rotor moves around in the housing i.e. the bobble on the shaft?

There are a few variables that may need to be changed to suit different situations. The eccentricity, generating radius, bearing diameter and number could all be changed to keep apex seal speed, bearing loads, combustion paths etc within limits at high RPM. The cast iron rotors would have to be the first to go.

The combustion chamber surface area to volume ratio will not be optimal but may not be much worse than that of a high compression, large bore reciprocating race engine.

[Chucky]

Detonation control might be tough in a rotary at that rpm and the big combustion surface wouldn't help.

[kevinq]

I think detonation is more of a problem at low RPM (ie. more time to ignite the "end mixture") so I don't believe that would be a major issue.

Getting the combustion completed in time could require some work though.

[4rotor]

Do F1 engines have variable air intakes? I doesn't look like they do - I had a look at a video here: http://www.renaultf1.com/en/public/f...deo/17495.html it doesn't look like they do - as you can see the injectors right above the intakes (which don't appear to move?).

[kevinq]

The intakes are variable length in F1. I believe the injector position is fixed and the top of the trumpets move.

Get a copy of Peter Wright's latest book :

http://www.amazon.com/exec/obidos/tg...books&n=507846

Expensive but worth the $$$ IMO.

[4rotor]

Cheers Kevin!

[thebear]

The following is information on the latest version, installed in the RX-8 and may prove interesting:

The motor for the RX-8 is an extention of the 13B twin turbo (still in production in Japan). It is referred to as "Rensis". Much of the same tooling is used and the installed cost is said to be less than that of a 3L V-6. Displacement and internal dimensions are identical. The rotor weighs 14% less and the new sump is half the depth and weighs 3% less than the originals. A redesigned rotor apex seal is provided. The turbo motor was rated at 280hp/152ft# and the new motor is 250hp/159ft# but without the turbos and their complications. The greatest restriction on the old motor was the exhaust port on the periphery of the `block' casting. The port's area could not be increased significantly without spoiling the emissions profile. That has been solved by switching to TWO exhaust ports on the side of each rotor housing. This provides improved exhaust flow, better fuel economy and higher thermal efficiency. As a result, unburned hydrocarbons are retained and utilized similar to the operation of the Exhaust Gas Recirculating valve on a piston motor. The new motor meets European Stage IV emission standards and has better fuel economy. For the future, larger intake ports will allow 9,000 RPM without distress. Electronic throttles and a new catalyst complete the picture. Starting with 1967 type 10A motor, Mazda has produced approximately 1.9 million rotary engined cars. In 1994 and 1995 the ¥3.8 million price was equal to $47,500 (~£35,000) and put a BIG CRIMP in RX-7 sales.

[Carse]

Is this statement true?: F1 engines have pnumatic valves as standard valve springs dont opperate quick enough to get the fuel/air mixture in and out. These pnumatically opperated valves mean that the engine can rev so high as the valves open and close at such a high speed that the mixture can enter --> combust --> exit quicker than most engines so therefore revolutions can be high.

Also, each cylinder is individually controlled so any imbalances at certain rpm's can be dealt with?

A road rotary engine maybe couldn't rev to 20krpm because of harmonic imbalance etc and would blow?

[4rotor]

Originally posted by Carse
Is this statement true?: F1 engines have pnumatic valves as standard valve springs dont opperate quick enough to get the fuel/air mixture in and out. These pnumatically opperated valves mean that the engine can rev so high as the valves open and close at such a high speed that the mixture can enter --> combust --> exit quicker than most engines so therefore revolutions can be high.

Also, each cylinder is individually controlled so any imbalances at certain rpm's can be dealt with?

A road rotary engine maybe couldn't rev to 20krpm because of harmonic imbalance etc and would blow?

Not sure about the F1 statement, it sounds right.

I believe that harmonic imbalance is an issue, but if cranks and conrods can rotate at 20,000rpm like in an F1 car - I would have thought it would be possible to do this in a rotary engine?

[cannardd]

I think that in a pneumatic valve F1 engine there are still valve springs, but they are very soft.
The main closing function is done by the pneumatic system (the soft springs still hold everything in position when the engine is off though).
It reduces valve train load (no heavy spring to push against) so everything can be lighter, which in turn allows it to rev higher.

[avsfan733]

i am just about 100% sure that there are no valve springs at all in the f1 engines. the issue is not as much speed as it is force and harmonics of the materials

[Heeltoe]

Originally Posted by cannardd

It reduces valve train load (no heavy spring to push against) so everything can be lighter, which in turn allows it to rev higher.

it is no heavy spring to push againts that's right,
but it's an even heavier (higher) air pressure,
so you will have all the same valve train loads as in a spring type setup.

a spring is much more prone to resonance/spring surge, and when this happens, the spring ratio will go all over the place leading to valve float and even striking the piston.
metalfatigue will also occur fast to a spring subjected to the fierce movements of a wild formula 1 camshaft.
etc. thereby air/nitrogen, it is much easier to work with.

[Heeltoe]

Originally Posted by avsfan733

i am just about 100% sure that there are no valve springs at all in the f1 engines.

hmm, then there have to be a total prohibition on turning the crankshaft when the pressurized air is not connected !?
the valves will drop to full valve opening due to nothing pulling up again - daring

[boyracer]

As someone mentioned before, there would have to be a small set spring to stop the valves from dropping, but that's about it.
From my memory of old TV broadcasts they use nitrogen in a small 1-1.5 Litre container that feeds the entire pneumatic system. The biggest problem with this type of system was (is ?) sealing, so they count on some percentage of the nitrogen leaking out. I remember one race where a car had to stop and get it topped up, if I remember correctly it also had catastrophic engine failure about 4 laps later .

[thebear]

Perhaps a better analogy would be that of a two-stroke engine. The ports perform a similar function and control the "timing" of the charge and exhaust. As a previous poster noted, enlarging them would comprimise the "block's" strength and there is very little real estate availible to position them. A really `high RPM' configuration would also be very dificult to start up in the first place.