I don't understand how some planes ended up being so fast (1 Viewer)

[Koopernic]

Roy Fedden really did try and do it the hard way with an air-cooled sleeve valve.

Whatever it's other faults the liquid-cooled Sabre did not have major cooling problems.

Fedden was related to Ricardo by blood and they discused the issue which is one reason Fedden had Bristol go that way. My understanding is that Fedden was eased out of Bristol by 1943 over conflicts with the family that owned Bristol.

I don't think there was anything wrong with the Sabre by the time the Tempest entered service. Napiers quality control and poor manufacturing had been the cause of half the problems rather than the design. I believe it did end the war as the worlds fastest piston engines fighter at low altitude. It went the way of the P-51H, P-47N etc.

Bristol Poppet valve engines were pretty good. Siemens-Bramo which was merged into BMW turned Feddens 1918 developed Juiper into a 1200hp machine, the Bramo 323 Displacement: 26.82 litres (1,636.8 in³) that powered the Fw 200 Condor.

A straight H block engine with poppet valves may have done much better.

 

[fastmongrel]

A straight H block engine with poppet valves may have done much better.

I believe with big capacity poppet valve H engines you would run into problems with the inlet and exhaust passages being too cramped for good breathing.

 

[Shortround6]

As a general rule that's right: However provided the following criteria are met

• At least two aircraft are involved, and for the sake of the argument: There will be two aircraft
• Both aircraft have the same armament: Say 4 x 20mm with 150 rpg
• Both aircraft have the same weight when armed, full fuel and oil: For some reason, I'll pick 10,000 pounds
• Both designs have the same wing-area (310 ft^2), same aspect ratio (6), same L/D ratio (13.5%)
• Both designs have the exact same horsepower engine (1875 hp) same critical altitude (24000 feet with ram)
• Design A has a fuel fraction of 11.1% (185 gallons) and Design-B has a fuel fraction of 16.68% (278 gallons)
• Both aircraft are stressed for a g-load of 9g ultimate when fully loaded at takeoff.

The following will be the case:

• Aircraft 1 & 2 will perform exactly the same when on 100% internal fuel
• Aircraft 2 will have a longer range providing the same basic cruise settings, and combat settings & limits will be used even if both use drop-tanks of the same size and capacity.
• Aircraft 1 & 2's performance will more radically diverge as fuel load goes from 100% to 80%, and from 80% to 50% as the different percentages of fuel will correspond to a different amount of fuel relative to the plane's weight: At 100% weight, Aircraft & 2 have no difference; at 80% weight, Aircraft 2 will weigh 98.85% the weight of Aircraft 1; at 50%, is 97% the weight of Aircraft 1.
• Aircraft 1 & 2's power-loading, maximum g-load capability, and corner velocities will all more radically diverge.

Diverging by single digit percents, especially very low single digits makes very little difference in some aspects aircraft performance. You could vary the weight of a clean Mustang for example by hundreds of pounds (source claims 1000) and only vary the top speed at altitude by 3mph. Difference between a good paint job and a poor one is more than that. A change of 3% in wing loading while turning would be undetectable to a pilot while flying. You would need a very accurate recording "G" meter to find the difference.
I would note that in tests of a Hawker Hurricane adding 434lbs internally to it's 6316lb weight (6.9%) slowed it down from 316mph to 309mph at 20,000ft. Unfortunately they had to change the engine between tests on the same airframe so it's not 100% accurate.

Probably in climb you could tell the difference but probably not at any reasonable fighting altitude.

The wings, fuselage and tail of a P-51B/C went around 2000lbs. doesn't include landing gear or fuel tanks.

But horsepower went up considerably in that time....

It did but they still weren't using all the volume. ANd there were a number of restrictions on how the plane was to be loaded,flown and even taxied at those high weights. A biggie was that at high weights you HAD to fill the outer wing tanks (they didn't exist on the early planes) to spread out the span loading, you couldn't fill the bombbay with AP bombs, hang a pair of 4000lbs under the plane and just fill the inner wing tanks, you had a good chance of bending/breaking the wing even if the gross weights were the same.

Honestly, I'm surprised they never thought of fitting the R-2600 to it.

If you are referring to the B-15 I am sure some bright spark thought of it. It just wasn't worth the effort. There was only one plane, they were never going to build anymore. They had learned a lot about structural design and aerodynamics in the meantime.
Look at the B-19, took so long to build (Douglas didn't even want to finish it) that it was obsolete on the day it first flew. Army wanted it for analyzing/evaluating structural design. Not because they thought it was a viable warplane.

 

[Koopernic]

Tye

Pictures taken from the 'Motoren und Strahltriebwerke' book. The 801R was supposed to be outfitted with 2-stage, 4-speed S/C. The 801Q was, IMO, close sibling to the 801E. 801J is noted as 801D-2 with turbocharger.
The graph notes that 801E was to be core for the P. 8035 engine project. Please note the gear speed change at 8.5 km for the 801J, the turbo engine.
'Luftschraubenwelle + Luefter' = 'prop shaft + fan'.

View attachment 474086 View attachment 474087

The Data quoted in Gersdorf looks like R+D data. Almost all of those engines look like minor variations of the
BMW801D except for the BMW 801E or F, the others were changes in the superchargers, accessories, propeller or gearbox.

The BMW801R had a 2 stage supercharger with 4 speeds. In fact each stage had independent drive so it was more like a 2 x 2 speed supercharger.
It also had a massive integral inter cooler. It was actually a little longer than the Jumo 213 and DB603 engine without their radiator.

The BMW factory testing and developing the 801R was bombed and the damage set back the project so far it could not be recovered in a sensible time.

It was intended for a version of the Ta 152C and obviously could have incorporated 801F technology to work its way to 2400/2600hp but its key characteristic was its high power at high altitude. It could maintain sea level power to 5200m/17,060 and 1760hp to 8600m/28220ft.

Noteworthy the length of these units
801D and 801S = 2006mm
801F = 2256mm
801R = 2550mm

So what was the F doing with the extra 250mm length? I have read on other forums that it was supposed to have a 3 speed supercharger? Could they have introduced a moderate amount of inter-cooling.

To achieve 2400hp or 2600hp requires 2.0 ata so this engine would have substantial jet thrust. It would also do well with good supercharger matching.

For fun Ive attached the airflow paths of the BMW 801TJ

 

[Koopernic]

I believe with big capacity poppet valve H engines you would run into problems with the inlet and exhaust passages being too cramped for good breathing.

The Fairy Monarch was a H block engine. One problem was that the intake passages had been cast into the block making it difficult to improve its performance without changing castings. I don't know if this was a critical thing but it did restrict its development.

Argus made a H Block called the AS412 and there was an AS413 H24 project that used Jumo 213 pistons. Don't know how they planed to get around the problem of cramped gas flow passages.

 

[fastmongrel]

Don't know how they planed to get around the problem of cramped gas flow passages.

Possibly run a higher boost and a lot more valve overlap than a V or X of the same size would need.

 

[Koopernic]

Concur!

Does anyone know if there are less moving parts to this set up than the "standard" valve layout? Also, it seems to have a weakness if hit in the "watch" mechanism it could take out the entire motor (sort of like liquid cooled engines and their ever so required liquid coolant).

Cheers,
Biff

There is a cutaway
https://www.youtube.com/watch?v=EyPvpdy4dggt away of a R-2800 working on this you tube video. If you go to 1.30 you can see the cam pushrod mechanism working.

I'd suspect the Sleeve Valve radials valve drive mechanism would be deeply buried behind the gearbox and parts of the crankcase thereby being heavily protected.

Some WW2 aircraft had cutoff valves to isolate a leaking coolant flow to a radiator, I imagine it would be more useful in a twin radiator design but as the oil cooler was still keeping the the temperature down it might be quite useful to have coolant flow without a radiator just to even out heat in the engine.

Isolation valves were fitted to the Me 109F and Me 109K but skipped most of the Me 109G series, they were well regarded and maintenance crews would race to crash sites to recover the precious valves before salvage crews or even other maintenance teams could get there.

It wouldn't be too hard to make a coolant flow 'fuse'. You would have two flow meters: one measuring outward flow of coolant and ther other the return flow. If there was a major difference the radiator could be bypassed. Several ways of doing it. A simple orifice plate that produces a pressure differential or maybe a vane type flow meter that was subtracted by a differential gear mechanism.

I don't know how coolant loss was detected. It apparently looked white (steam?)

I would have thought a good 'limp home mode' would have simply blocked of fuel to one bank of cylinders for a few seconds alternately so that it cooled. The engine would be running at 50% power but it would get the pilot home.

Coolant flow on a Jumo 213E was 9.5L/second so on a Merlin there must be about 2 US gallons coolant flow per second. That's why I suggested an coolant leak fuse, one would have to react quickly.

The problem with water cooled engines is that a single bullet could end the mission. Not so bad over your own country because you could get back to base. The engines didn't stop straight away.

Russian and German troops didn't dive for cover when aircraft came, they started shooting. The smaller 10-12 man squads the Germans were organised into at the end of the war could probably get 8 rifles firing of which 2 might be semiautomatic Gw 43. The Machine gunner might get his MG42 into operation if he had a stand for it. The Corporal, NCO or Sargent that ran things would have had a sub machine gun.

 

[Shortround6]

Simply space the cylinder blocks a decent width apart.

more photos and story/data here: Arsenal 24H and 24H Tandem Aircraft Engines

engine went about 4000lbs so you need a good sized plane to carry it

 

[Shortround6]

There is another thread here that I don't want to look for at the moment is which I took 3-4 posts to cover all the supposed advantages for the sleeve valve engine as listed by Roy Fedden in a talk given in England.
Not trying to take away anything from Fedden but he was comparing his sleeve valve engines to his own poppet valve engines which were 10-20 years older in basic design. He was NOT comparing his sleeve valve engines to P &W or Wright radial engines. Maybe he didn't consider the upstart Americans to be competition?
For instance he claimed the Sleeve valve was more oil tight and didn't leak as much oil into the engine compartment or onto the airplane. I am not claiming the American engines were oil tight but the Bristol poppet valve radial's valve gear was lubricated by grease gun and the valve springs, rockers and pivots were all flapping around in the breeze. NOT enclosed.
Photos were published of all the parts needed for ONE cylinder of both types of engine. Of course Bristol poppet valve cylinders used four valves so there are a lot more parts than a 2 valve cylinder would show. The photo also doesn't show all the gears in the crankcase vs the gears and cam ring/s needed for a single row radial.

The Mercury and Pegasus didn't get a whole lot of development during the 30s as most of the R&D was going into the sleeve valve engines.

 

[Koopernic]

Simply space the cylinder blocks a decent width apart.
View attachment 474135
View attachment 474136
View attachment 474137

more photos and story/data here: Arsenal 24H and 24H Tandem Aircraft Engines

engine went about 4000lbs so you need a good sized plane to carry it

The dimensions of the Sabre according to Wikipedia were

• Width: 40 in (1,016 mm)
• Height: 46 in (1,168 mm)

The dimensions of the Arsenal H24 were
3.94 ft (1.20 m) wide, 4.92 ft (1.50 m) tall,

• Width: 47.5 in (1.2m)
• Height: 59 in (1,5m)

The Sabre had 1/2 the swept volume so taking the cube root to scale the engine to the size of the Sabre

• Width: 1.067m
• Height: 1.144m

The dimensions are almost exactly the same though the Sabre is horizontal and the Arsenal H24 vertical.

The Sabre's pistons are over square whereas the Arsenal H24 like the Jumo 213 it was based on under-square. The horizontal arrangment should have some advantages in increasing RPM.

The Arsenal H24 like the Jumo 213 it was based on ran at 3250 rpm. The Jumo 213J was supposed to reach 3700 RPM. It was on the bench in Jan 1945 so the Arsenal H24 could probably have been pushed to 3700RPM and its boost level from 1.7 bar to 2.0 bar as Jumo were doing this.

 

[Peter Gunn]

The theory is that they are moving but not reciprocating like a valve against a spring, howver I would say the friction between the surfaces, even with oil lubrication probably means very little gain.

Watched a couple more videos and I agree, there's a lot of metal on metal friction there.

 

[pbehn]

Watched a couple more videos and I agree, there's a lot of metal on metal friction there.

There is a motorised cutaway at the Yorkshire Air museum, it makes your brain hurt watching it. Since it runs very slowly but dry you can see the effect of the constant movement, nothing is able to "bed in" against anything else, and the scoring is visible.. Also for a brief period the block machined surface is exposed to the exhaust gas through the sleeve port, this must have seriously roasted the oil.

 

[Juha2]

...Russian and German troops didn't dive for cover when aircraft came, they started shooting. The smaller 10-12 man squads the Germans were organised into could probably get 8 rifles firing of which 2 might be semiautomatic Gw 43. The Machine gunner might get his MG42 into operation if he had a stand for it. The NCO or Sargent that ran things would have had a sub machine gun...

I doubt that, I would say sometimes they fired back but usually they took cover, based on Allied gun camera films and by the fact that He 111s attacked Soviet troops with their nose mg even during winter 41/42.
When I did my military service with the Finnish Army in mid-70s, inspirated by the NVPA experiences during the Vietnam War, we were trained for anti-a/c and anti-helicopter firing but even with 7 assault-rifles and one LMG per squad it was stressed to us squad leaders never to open up against incoming planes only against those attacking your neighbours. Odds were heavily against you if you engaged incoming plane. And never disturb Mi-24 Hinds!

 

[Denniss]

Tomo, really sad this book use improper engine power and not proper shaft power as per BMW's documentation. Several power levels are wrong as a fair comparison is only given at sea level, not max power at various alts depending on SC setup.
The 801Q description is wrong unless this was a project code later reused (as 801Q-2). FW docs clearly state the Q-2 as D-2 with the mentioned oil system/amor changes which are in the 801S too.
Per BMW/Fw docs the U = TU is wrong as it contained the Q-2 engine
The 801J had the D-2 as base engine, this is correct. The further description is wrong as it was used in the Ju 388 (as TJ-1)
The 801S had a slightly improved supercharger for better alt performance, the 801F may have gotten an even further improved one taking more space + the external air intakes, stronger fan and probably larger oil tank/cooler may have caused a length increase

 

[tomo pauk]

The biggest quarrel I have with book is it's jet engine half - I'm not interested in those German engines of ww2 deserve far better books of greater number of pages. Unfortunately, the book about Junkers engines is too pricy for me. Hoperfully, Calum, our member, will have his book soon in print.

 

[pbehn]

There is another thread here that I don't want to look for at the moment is which I took 3-4 posts to cover all the supposed advantages for the sleeve valve engine as listed by Roy Fedden in a talk given in England.
Not trying to take away anything from Fedden but he was comparing his sleeve valve engines to his own poppet valve engines which were 10-20 years older in basic design. He was NOT comparing his sleeve valve engines to P &W or Wright radial engines. Maybe he didn't consider the upstart Americans to be competition?
For instance he claimed the Sleeve valve was more oil tight and didn't leak as much oil into the engine compartment or onto the airplane. I am not claiming the American engines were oil tight but the Bristol poppet valve radial's valve gear was lubricated by grease gun and the valve springs, rockers and pivots were all flapping around in the breeze. NOT enclosed.
Photos were published of all the parts needed for ONE cylinder of both types of engine. Of course Bristol poppet valve cylinders used four valves so there are a lot more parts than a 2 valve cylinder would show. The photo also doesn't show all the gears in the crankcase vs the gears and cam ring/s needed for a single row radial.
View attachment 474138
The Mercury and Pegasus didn't get a whole lot of development during the 30s as most of the R&D was going into the sleeve valve engines.

In my working life I met some very highly qualified and experienced people. On their professional subject they were masters of the subject. One man in particular was fundamental in getting his company as one of the world leaders in two products. If you asked him any question on metallurgy he would give you an answer but if the answer affected how his product was viewed he would omit, evade and at times tell outrageous lies, and he was not alone. Some of these lies were not only put in writing they were actually put in the company publicity brochure and prospectus.

When it comes to engines, when I was racing, all the tuners were completely disparaging about all the other tuners efforts. Some would allege cheating (over bored cylinders, bored out carbs, doped fuels etc). but only for favoured riders inmplying that other tuners were not selling what they advertised, on the other side there were dark mutterings about pistons collapsing. As a racer using one tuner racing against guys using all the others it was complete borrocks, there was only one guy I knew who cheated with a 350cc engine in the 250cc class. The man who won the national championship, that year in my class (Mick Crick) had a bike slower than mine and most other competitors, the rider makes the difference.

In the early 1980s there was a guy called Joe Ehrlich who was famous for engine tuning, he had some success in the 1950s and made a "come back" winning some races with his machines like the Junior TT. He was interviewed as Heddon was, and he declared that he was working on a new engine, a variation on the stepped piston which was so revolutionary it would make normal two strokes obsolete. He was so persuasive I you could imagine people buying one off the drawing board. The engine and the bike never materialised and probably never existed.

I cant say that Heddon was being dismissive of upstart Americans or not but to me he was just being like any other corporate engineer. From what I have read about the intrigues of the USA motor and defence industries they are pretty much the same, I hold up Curtiss and its War Hawk ads as evidence.

 

[tomo pauk]

From what I have read about the intrigues of the USA motor and defence industries they are pretty much the same, I hold up Curtiss and its War Hawk ads as evidence.

Hmm - what was so wrong about the Warhawk?

 

[pbehn]

Hmm - what was so wrong about the Warhawk?

Nothing wrong with it as a plane but it was no longer the world beater they continued to say it was until production stopped.

 

[Shortround6]

According to some sources (which could very well be wrong) Fedden came close to bankrupting Bristol in the quest to get the sleeve valve into large scale production (both basic engine development and production engineering).
Bristol had been, at the end of the 20s THE premier aircraft engine maker in Britain if not the world. Bristol engines (mainly the Jupiter) had powered over 260 types of aircraft world wide (many of them prototypes to be sure) and was being produced under licence in 16 countries. With the coming of the Great Depression of 1929 world wide demand for aircraft engines fell sharply, and then Fedden took off on the sleeve valve tangent. The Mercury and Pegasus (a modernized Jupiter) did get some improvements but it seems the Sleeve Valve projects sucked up most of the research and development.
Wright during the same time period (roughly) introduced the R-1750 in 1927 (with forged tungsten steel intake valves and hollow exhaust valves of forged tungsten-cobalt-chrome steel with metal salts inside for cooling) at 525hp. Introduced the R-1820E (0.125in larger bore) in July 1930, moved to the "F" in 1932, followed by the F50 (with a new crankshaft) and while production continued, the G series was introduced in April of 1935 with 2800 sq in of cooling fin per cylinder, automatic valve gear lubrication with no external tubes, stellite faced exhaust valves and stainless steel exhaust valve seats and more.
Jan 1937 sees the introduction of the G-100 with forged steel crankcase among other major changes and in March of 1939 the G-200 shows up with a new different forged steel crankcase. Wright continued on but that gives the idea. The Cyclone went from 525hp at 1900rpm in 1927 to 1200hp in 1939 at 2500rpm.

There were many more changes than the ones listed above.

While Bristol may have gone through lean times in the early 30s due to the world financial situation the market had changed in mid to late 30s with Both Wright and P&W intruding on the world stage. R-1830 Twin Wasps going into commercial service at the end of 1936 and the older Hornet making enough of an impression than BMW was building them under Licence since about 1928.
Development of the Sleeve valve engines dragged on and on and..........on. After spending so much time and money on the project could Fedde admit (even to himself?) that it might have been a mistake?
There is no doubt that the final model/series Hercules and Centaurus engines were very fine engines indeed, very long lasting and reliable while putting out competitive power. But the world wide market had disappeared.

 

[pbehn]

Heddon may or may not have been correct we will never know, if Bristol had made the investment then who knows what would have panned out. It is a bit like the Betamax and VHS debate or early AppleMac v Windows, in both cases the most popular VHS and Windows have people saying the other product was technically superior. I remember when Honda decided to take on the two strokes in 500cc motorcycles, they produced a 4 cylinder eight valve per cylinder oval piston engine, it didn't win but earned some points in races, with enough investment almost anything can be forced to work somehow but as you say the market had gone. By the end of WW2 big poppet radials had a huge track record on American bombers and the early passenger planes were based on them, a big consideration when people board a plane.