2 stroke thread (with occasional F1 relevance!)

[Agerasia]

I used to own a Kawasaki KR-1S 250. That engine was something else, I had to strip it and it was well engineered. It had such an intricate power valve system.
Some of the racing versions were well over 70bhp.
The standard version is still the fastest production 250cc ever made all these years later.

P.S. It had Nikasil barrels which although good were a pain. The KR-1S had a design fault with the compression ring retaining pin that meant it was exposed allowing the ring to turn. This meant that as the end of the ring met a port...it pinged out and wrecked those barrels.

[uniflow]

The rev limit on a twostroke engine is the time needed to get rid of the combustion pressure out the exhaust before the transfer ports open, documented as port time / area. In truth it's time / area / pressure, but with crank case pumping pressure on transfer is limited. A 125cc road race bike will rev to about 13500 before this exhaust blowdown time runs out. What then happens is transfer charge is stalled for a time before it can flow. This transfer time is already limited, the engine just run's out of time to do gas transfers.
So the only way forward, as the time portion is diminishing, is more area. Problem is we have run out of cylinder wall to put ports or more pressure ( super charge ).

I've got to go back to work, I'll finish this later.

[J.A.W.]

The rev limit on a twostroke engine is the time needed to get rid of the combustion pressure out the exhaust before the transfer ports open, documented as port time / area. In truth it's time / area / pressure, but with crank case pumping pressure on transfer is limited. A 125cc road race bike will rev to about 13500 before this exhaust blowdown time runs out. What then happens is transfer charge is stalled for a time before it can flow. This transfer time is already limited, the engine just run's out of time to do gas transfers.
So the only way forward, as the time portion is diminishing, is more area. Problem is we have run out of cylinder wall to put ports or more pressure ( super charge ).

Aye U, those final G.P. 125 mills were pumping a specific output of ~440hp/Litre..
.. something a naturally aspirated pump gas 4-stroke could not emulate..
..even at ~20,000rpm, where like - diminishing marginal returns - are in also in play..

[manolis]

Hello.

In Rolls-Royce Crecy design:



there is a big volume (actually a deep trap for burnt gas) above the exhaust ports: about one third of the cylinder capacity (the dead volume included). Reasonably, the Crecy runs with a lot of residual gas.

In order to avoid “short-circuit” between transfer and exhaust ports during the scavenging, the stroke to bore ratio must be increased, limiting the rev limit and the peak power.

The stroke of the sleeve valves seems long, like 1/3 of the piston stroke, and they cannot be lightweight.
By the way, in the 4-stroke sleeve valve Bristol Radials, stationary rings seal the upper end of the sleeve valves.



If you add such rings in the Crecy design, the length of the sleeve valve increases substantially and the cooling of the cylinder head gets difficult. But without rings you have poor sealing. On the other hand, if the sleeve valves can run without sealing rings, why not to be used ring-less pistons, too?

The crankcase “dead volume” (in case the design is modified to use the crankcase as scavenging pump) increases because of the space required for the sleeve valve.


Compare the previous to the following Tilting Valve Opposed Piston engine:



333 cc, bore 84mm, stroke 30+30=60mm
(same bore to stroke ratio with BMW's boxer R1200GS of 2013)
weight: 8.5Kp (19lb) without the exhaust pipe and the carburetor
height: 250mm



The scavenging is cross-uniflow: the inlet ports and the exhaust ports are disposed anti-diametrically. During scavenging the chamber is like two independent slim disks. The short stroke becomes advantage for many reasons. Think.

Thanks to the independence of the scavenging at the two halves of the cylinder, the combustion chamber gets compact, the squeeze increases, the spark plugs are more centrally located, etc:



For more: http://www.pattakon.com/pattakonTilting.htm

Do you know another design that fits better to extreme revs and extreme power density? I.e. to F1?

Thanks
Manolis Pattakos

[smoker250]

Thanks Uniflow, I should have figured that out for myself. I dug up a little on the Crecy, "Ricardo used an open ended or expansion sleeve which relied on the fit of the sleeve in the cylinder for sealing. The sleeve clearance was to a certain extent self regulating because of the effect of thermal expansion" .
It is a fascinationg experiment you are doing, no-one has done as far as I can tell, I can't wait to see how it goes.

At the conclusion of the article on the Crecy it mentioned that in Dec 1944 a mechanically supercharged unit achieved 1798 hp, extrapolated to possible 2500hp turbocharged. By that stage in the war one of the latest variants of the Merlin was already certified for flight at 2340 hp.

Smoker, I am a little slow on building the crank cases to suit this cylinder but I'm back on to the project again. I intend to try several different types of cylinder including a vairiant of Frits Overmars FOS cylinder.

UniflowI know what it's like, time just seems to evaporate sometimes. One good thing about your sleeve valve is that you could change the timing relative to the piston without too much difficulty rather than alter the ports themselves. I had to do a google search on FOS cylinder, interesting idea, you would think the transfer period would be too short, but since he has about twice the transfer area it must work all the same. He claims 29-31 hp from a 90cc Derbi, I'd like to see some dyno graphs. In the 1920's before schnurlle came along Villiers built some sort of similar engines with two diametrically opposed transfers and two diametrically opposed exhausts, the idea was to do away with the deflector and use a flat top piston.
From what I read they performed quite well but when loop scavenging came along everyone jumped on the bandwagon.

[J.A.W.]

Manolis, very interesting post, as usual..

..The Bristol sleeve valve you show was of course, running as a 4-stroke & undersquare ( long stroke)..

& yet, I have operated the Bristol Hercules sleeve valve - as set up on a single cylinder via hand crank..
( it is on public display at the RNZAF Museum in NZ), it is very smooth/easy in its function, with much less
resistance than turning over a cam/spring controlled poppet valves..

& Len Setright wrote that Bristol test ran their sleeves at up to 8,000rpm "...without anything going awry."

[smoker250]

Just a futher comment on the RR Crecy , it had one particular problem that the engineers could never solve, at high output the pistons ran at too high a temperature, with crowns breaking and ring lands collapsing. The reason was, it was impossible to conduct the heat away through the austentitic iron sleeve, which being the exhaust valve ran extremely hot itself.
After opening out the oil passage in the conrod as much as they could, the engineers concluded "it was impossible to pump enough oil throughto cool the piston sufficiently." That was about when they lost interest I think and went back to tinkering with souped up Merlins.

Manolos has kindly posted a picture of what looks like a Bristol sleeve engine, and some may ask why it didnt have the same problem.
Four strokes have a cooling induction stroke between each power stroke so the piston and sleeve get a chance to cool, whereas the brief transfer period in a Crecy type two-stroke provides little cooling (1). The heat flux on the piston, measured in KW / sec is much higher in a two-stroke, meaning more heat to disipate. Also in the four stroke stroke sleeve valve the critical part of the sleeve which uncovers the exhaust ports spends most of its time up inside the junk head protected from the heat and surrounded by lots of oil and metal to cool it. In the Crecy the sleeve was exposed to very hot gas for most of the cycle and wouldnt have got much relief from the 250 deg air coming out of the supercharger. The Top third of the sleeve probably ran HOTTER than the piston. What was Harry Ricardo thinking? Might have worked as a diesel, (higher pressures but lower temperatures and heat flux)which is what it started out as in the first place.
The Crecy was doomed anyway because if it took turbocharging to get decent power out of it they lost the thrust from the exhaust gas ejectors and the plane manufacturers were just realising that in a high speed pursuit plane( what the Crecy was intended for) exhaust thrust could be worth tens of miles an hour. Cant have your cake and eat it too. Four stroke sleeve engines like the Napier Sabre make higher specific output than Crecy ever could and at the end of the war was extremely reliable.I won' t say the Crecy was a turkey, but it was never going to fly.
(1)In small petrol two-strokes you do get some cooling of the crankcase and piston during the transfer phase due to the evaporation of the fuel.

[J.A.W.]

Ah yeah, Smoker.. ..you might want to do a bit more background.. ..you risk embarrassing yourself..

R-R were required to do Crecy by the Gov't - under wartime compulsion authority, & they were permitted to relax..
..only when Whittle's turbine work showed real results.. .. nothing to do with Merlin development..

& it was much to Ricardo's frustration - & he kept going on his research units, & making real power too..

Yes, they were really 'pushing the envelope' engineering-wise, but the Crecy concept was sound..
..current ECM/injector tech/metallurgy/tribology could cope.. ..no question..

See Jan 2011 'Aeroplane' for a Crecy overview.. ..per Ricardo..

"He met his goal. Astoundingly, he managed to coax 219hp/Ltr & a BMEP of 325lb/sq"...
Extrapolated to the Crecy, this would have resulted in a whopping 5,694hp."



All `40s hi-po piston aero-engines featured supercharging, & if high altitude performance was wanted, then
multiple stages were also necessary.. ..most of the USAAF aircraft flying over Germany were turbo'd..
These turbos provided altitude pressure compensation up to ~30,000ft..

DFI crankcase breathing 2-strokes forgo the latent heat of evaporation, & want more substantial cooling
circulation circuits as a result.. ..yet even the simple units fairly recently banned from G.P. bike racing
make un-boosted ( by exotic fuels &/or forced induction) 4-strokes look lazy, &/or frenetic..

[smoker250]

I have often wondered why nobody has built a small high performance sleeve valve 4stroke engine .A small bore engine should be able to rev to similar rpms to poppet valved engines.One thing sleeves share with Two stroke engines is that the smaller the bore, the higher the surface to volume and therefore usable port area, and the higher the potential performance, so a 125cc sleeve engine should be competitive with a valved 125. Maybe the rules of F1 mandate valves. There was a guy in Australia a few years ago who built a rotary valve racing engine, Bishop Rotary Valve,, it had impressive potential performance but I think rotary valves got banned before it ever saw the track. Same reason they banned 2 strokes in most classes of racing, the four strokes were outclassed.

[Richard]

Australian rotary valve sounds familiar, was that related to this one? http://www.f1technical.net/forum/viewto ... f=4&t=8671

[smoker250]

Ah yeah, Smoker.. ..you might want to do a bit more background.. ..you risk embarrassing yourself..

R-R were required to do Crecy by the Gov't - under wartime compulsion authority, & they were permitted to relax..
..only when Whittle's turbine work showed real results.. .. nothing to do with Merlin development..

& it was much to Ricardo's frustration - & he kept going on his research units, & making real power too..

Yes, they were really 'pushing the envelope' engineering-wise, but the Crecy concept was sound..
..current ECM/injector tech/metallurgy/tribology could cope.. ..no question..

All `40s hi-po piston aero-engines featured supercharging, & if high altitude performance was wanted, then
multiple stages were also necessary.. ..most of the USAAF aircraft flying over Germany were turbo'd..
These turbos provided altitude pressure compensation up to ~30,000ft..

DFI crankcase breathing 2-strokes forgo the latent heat of evaporation, & want more substantial cooling
circulation circuits as a result.. ..yet even the simple units fairly recently banned from G.P. bike racing
make un-boosted ( by exotic fuels &/or forced induction) 4-strokes look lazy, &/or frenetic..

No risk of embarrasing myself JAW I only do that when I have had too many beers and I'm presently saving myself for the Xmas season. The comment I made about tinkering with Merlins was actually a joke though.
The piston overheating issue with the Crecy was never solved and no reasonable solution was ever suggested. No matter how much coolant was circulated, the heat could not be conducted through the sleeve as it was at such a high temperature itself and also the gap between sleeve and bore was additional barrier to heat. I'm pretty sure the sleeve would have had to run in a bore sleeve of similar thermal expansion to work as a seal, cast iron or steel, so even more barrier to heat .
"Each piston required 35-40 gallons of oil an hour to cool it, an impossible amount" "At one point,following a series of failures, no engines were run for a period of eight months, an extroardinary gap in a wartime program" That was from July 43 till early 44.
I remember a lot of this from when I trained as a mechanic in the army in the 70s. Was very interested in 2 strokes even then, GMs, DDs, etc and read some detailed stuff on various aero engines.
One of our instructors was an aircraft fitter during ww2 and taught those of us who were interested a lot of stuff that wasnt in the course and some of my best mates were mechanics in the Air force.. I have a 40 page article by The British Aeronautical Comitte titled British Aero engine Development during WW2 I guess those guys should know. I have to read it thru again again but mainly the the gist was that the existing 4 stroke engines were improving incrementally, jet engines were improving very rapidly and the two-stroke project was not progressing very fast at all. Resources were redirected to more promising projects. One thing in its favour was that it had surprisingly good fuel economy, but I guess that is relative to a Merlin running flat out at an AF ratio of about 12 to 1 . I used to have most of Ricardo's books but they got lost over the years.
You are right about DFI two -strokes, Rotaxes 800E cranks out 169hp, pretty sure they have water cooled crankcase. The early version had a single ring piston and so many shat themselves they had to be recalled. Later ones are better but still regarded as a fragile engine. On a carb 30hp 125 or a 50hp 250 motorcross bike the piston gets pretty hot anyway so I dont think any manufacturer is willing to be first to go DFI in case they have problems. I have some technical stuff from Honda about a 250 engine that never got built because they went 4 stroke.
55 hp@ 11500 , surprisingly it has a reasonable power curve, (Igot hold of some dyno graphs) It uses pneumatic port injection. I used to be in contact with the guy in Japan who headed up the R&D for the two-strokes and he was not very happy when they pulled the plug and sent him off to work on the new generation honda car engines.

[J.A.W.]

& here's yet another variation on the theme..

http://www.cycleworld.com/2014/01/17/as ... orcycling/

[smoker250]

Australian rotary valve sounds familiar, was that related to this one? http://www.f1technical.net/forum/viewto ... f=4&t=8671

Not the same one I think Richard, If you do a Google search "The Bishop Rotary Valve, Development of an Australian formula one engine" Theres a really good 6 page pdf article about it They put them on a Mercedes V10 and cranked it up to 18,000 with no problems and reckoned it was good for 25,000 . Can you post pdf's on here ? I tried to post a photo yesterday but couldnt get it to work.

[uniflow]

Just a futher comment on the RR Crecy , it had one particular problem that the engineers could never solve, at high output the pistons ran at too high a temperature, with crowns breaking and ring lands collapsing. The reason was, it was impossible to conduct the heat away through the austentitic iron sleeve, which being the exhaust valve ran extremely hot itself.
After opening out the oil passage in the conrod as much as they could, the engineers concluded "it was impossible to pump enough oil throughto cool the piston sufficiently." That was about when they lost interest I think and went back to tinkering with souped up Merlins.

Manolos has kindly posted a picture of what looks like a Bristol sleeve engine, and some may ask why it didnt have the same problem.
Four strokes have a cooling induction stroke between each power stroke so the piston and sleeve get a chance to cool, whereas the brief transfer period in a Crecy type two-stroke provides little cooling (1). The heat flux on the piston, measured in KW / sec is much higher in a two-stroke, meaning more heat to disipate. Also in the four stroke stroke sleeve valve the critical part of the sleeve which uncovers the exhaust ports spends most of its time up inside the junk head protected from the heat and surrounded by lots of oil and metal to cool it. In the Crecy the sleeve was exposed to very hot gas for most of the cycle and wouldnt have got much relief from the 250 deg air coming out of the supercharger. The Top third of the sleeve probably ran HOTTER than the piston. What was Harry Ricardo thinking? Might have worked as a diesel, (higher pressures but lower temperatures and heat flux)which is what it started out as in the first place.
The Crecy was doomed anyway because if it took turbocharging to get decent power out of it they lost the thrust from the exhaust gas ejectors and the plane manufacturers were just realising that in a high speed pursuit plane( what the Crecy was intended for) exhaust thrust could be worth tens of miles an hour. Cant have your cake and eat it too. Four stroke sleeve engines like the Napier Sabre make higher specific output than Crecy ever could and at the end of the war was extremely reliable.I won' t say the Crecy was a turkey, but it was never going to fly.
(1)In small petrol two-strokes you do get some cooling of the crankcase and piston during the transfer phase due to the evaporation of the fuel.

Yes, what was Harry thinking ideed. One of the V twin test engines I belive did 1000 hours without burning pistons or sleeves. It was found later ( after the war ) that the Crecy had an oil feed problem to the center part of the engine ( I think you will find it was the center pistons that failed ). The test engines showed it was possible to produce this kind of power safely.

[J.A.W.]

According to Graham White's Jan 2011 'Aeroplane' Crecy article ..

"One of the V-twin test engines still exists at Ricardo Consulting Engineers".

& S-250 needs to realize that propellers were also then an issue for very powerful aero-engines.
Especially for fighter aircraft, & jet thrust turbines leap-frogged this problematic area.

White continues on useful Crecy features..

"Piston cooling technology befitted high-performance diesels.
Most modern diesels use the Ricardo derived 'cocktail shaker'
idea for this critical function... ...it can be said that the Crecy was ahead of its time."

&,

R-R engineer Stuart Tresilian proposed a..

"9-litre 16-cylinder X-configured Crecy-derived engine.
This little screamer was calculated to produce 2,500hp at 7,000rpm."

R-R as a company, ( by contrast to Napier) then presented as none too receptive/successful with
cutting edge designs, doing better with the steady development of their conservative V12 mills..