Rotary engine exhaust design considerations?

[jasonsaini]

thread71-235689

Some good information in the above thread, guys, thanks! Are there any experts that can expand on what some of the actual design considerations would be for a rotary exhaust system? I am most specifically interested in a Renesis side-port engine, but some ideas for all types of ports might be helpful as well.

Basically, we are about to build a custom header or two to dyno test and would like to know what sorts of things to try - primary length, collector angle/shape, what to do about the different side and center ports.

There's tons of papers and software written to aid in the design of 2 and 4-cycle piston engine exhaust systems, nothing for the various Rotaries.

Thoughts, ideas?

-Jason

 

[patprimmer]

Does the exhaust system know that the engine it is attached to is a rotary, or does it just see exhaust pulses at certain intervals tied to number of exhaust strokes per minute and firing order and spacing of the source of the exhaust gas.

Regards
Pat
See FAQ731-376 for tips on use of eng-tips by professional engineers &

http://eng-tips.com/market.cfm

for site rules

 

[jasonsaini]

There are some differences, and I don't think most software for exhaust design would work... what do you enter for Bore/Stroke? Surely there is some relatively complex math to determine the pulse size/velocity for a Rotary? Plus, the Renesis has absolutely no overlap, which makes it unlike just about any piston engine.

Surely there will be some tradeoff between primary length, peak output and area under the curve. I'm just looking from some general direction on where to start so we can mock up some test systems for the dyno.

Some people are saying longish primaries with a very long/gentle collector, others are saying no collector and just run three pipes out the back of the car? Can you get a longer collector than a 12deg anywhere, or will we have to make our own?

Or am I asking for proprietary knowledge and are just better off trying some stuff to make my own conclusions?

-Jason

 

[GregLocock]

I think you'd be better off rereading Pat's note. He is suggesting that in a very fundamental way there is less to this question than meets the eye. The zero overlap point you make is interesting, but not to the extent of invalidating the traditional approach for a four stroke.

Your exhaust doesn't care much about bore, or stroke. It does care about the volume of the charge that was intook.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[patprimmer]

Bore equates to rotor surface area and stroke relates to rate of reducing volume over the rotor. Neither have much to do with exhaust design. Displacement per rotor and port timing do have some influence. Port timing relates to cam timing. Displacement relates to exhaust gas mass and it's impact on back pressure and velocity.

The lack of any overlap reduces the value of pulse tuning to aid exhaust scavenging.

Piston engines often had no overlap 100 years ago. I think a little overlap became the norm about 90 years ago. Maybe the rotary is 100 years behind in development ;-)

Regards
Pat
See FAQ731-376 for tips on use of eng-tips by professional engineers &

http://eng-tips.com/market.cfm

for site rules

 

[ornerynorsk]

There was a company in the 90's that commercialized rotary valve heads for small block Fords. If I can dig up a link I'll post it.

As I remember, it made the 302 capable of ~14,000 rpm with the removal of the conventional valve train.

 

[ornerynorsk]

http://www.coatesengine.com/technology.html

Looks like they're still around.

 

[jasonsaini]

Okay, I definitely get the point that exhaust gasses don't really care how they were created, that wasn't my point. I appreciate the correlation of bore/stroke and the ideas of port timing and displacement, Pat - but I still feel there are answers out there. Seems to me that scavenging all the exhaust gasses would be super-critical to the next combustion event's efficiency?

There has to be an answer as to 'how' best to scavenge those gasses.

 

[patprimmer]

There is. Grind on the ports to get some overlap, then pulse tune to published formula using port timing in place of cam timing.

Regards
Pat
See FAQ731-376 for tips on use of eng-tips by professional engineers &

http://eng-tips.com/market.cfm

for site rules

 

[izzmus]

There's been some interesting work in the past few years regarding the use of expansion chambers after the collector, and crossover pipes in the header before the collector.

One of my acquaintences has a small rotary shop and a chassis dyno, and he has done a lot of experimentation with the collector shape. The angle of the collector is absolutely critical, especially as the port overlap gets larger.

Unfortunately I haven't seen any math for determining a first cut for any of this! At least, no math that matches up with real world experience. I wish I could offer more help than this vectorial, but it seems that the people involved are not keen on giving away the secrets to their $600 headers.