Best Japanese fighter

[tyrodtom]

Poor surface quality in an intake passage will rob some power, but not much. Poor quality exahust passages will not affect power much if any. It certainly would in a normally aspirated engine, but all these engines were supercharged. Boosted engines are completely different from normally asiprated engines as far as flow goes. Of much more concern would be the main bearing clearances.

Tight clearances will overheat at higher rpm and cause premature failure and loose clearances will not allow the engine to develop full oil pressure and will limit manifold pressure unless you want engine failure. So the pilot would be able to pull only so much power before the oil pressure gets to the limit. Any more power would be a personal gamble.

One place where quality would really affect performance would be if the props were of poor quality finish.

Naturally, the airframe fit and finish is high on the list, too. Leaky gaps are very draggy. You want to prevent air from entering anywhere it is not supposed to enter.

In my 20+ years of racing, and some experience with supercharged tow trucks, I've found what you said about the importance of intake and exhaust passages to be totally wrong, most of the improvements in horsepower come from increases in intake flow, AND exhaust flow, thru improvements to intake and exhaust passages and camshaft timing and duration. You can't get fuel/air mixture in if you can't get it out.

And a supercharger isn't as effective with bad intake and exhaust passages. All the same rules apply, supercharged or normally aspirated.
And I can't believe that drastically changes if the engine is at 10,000 or 20,000 feet in a aircraft.

 

[Ivan1GFP]

Talked with Steve Hinton and Ed Maloney today about the Ki-84 we used to have. He said our example was barely flyable and had a very slightly bent prop and they could not get rid of the vibration ... and the gear up locks were nonfunctional. So, although it was quite pleasant to fly, they never got into the higher-performance part of its envelope. At the time, there was simply no demand for it on the airshow circuit and they had NO spare parts so, when the offer came in from Japan, they jumped at it.

The only people who flew it were Don Lycans and Bud Mahurin.

I have seen photographs of this aeroplane in flight in old books.

It was a beautiful machine but had lots of obvious problems.
Your comment about the landing gear up locks is interesting because from the photographs I have seen, the gear never retracted all the way. The inner gear doors closed before the gear was up so the main gear was trapped outside. Also, the tail wheel must have had some serious issues because this plane looked like it had a new strut fabricated which could not possibly retract.

From what I have read elsewhere, this aeroplane's flying days are finished, but perhaps it can still be used as a template to make new ones. To save space in storage, apparently someone decided to saw through the wing spars.

- Ivan.

 

[Ivan1GFP]

I am coming to the view that the post-war American trial performance of the Ki-84 giving a maximum speed of 427 mph is misleading. My reasoning starts from the engine data given at q‹ó‹@ƒGƒ"ƒWƒ"ˆê——E"ú–{ŒR, which gives powers:

Ha-45 11 – Take off 1800/2900 – Military 1650/2000/2900 and 1460/5700/2900
Ha-45 22 – Take off 2000/3000 – Military 1890/1800/3000 and 1750/6450/3000

It also gives for the Mitsubishi's M9K

Ha 43-11 – Take off 2200/2900 – Military 2070/1000/2800 and 1930/5000/2800

Now the Navy's A7M1 Reppu was powered by a Nakajima Homare 22 and is quoted as only having a maximum speed of 310 Knots at 6,190 m (357 mph). The A7M2 was powered by the M9K and had a speed of 339 kt at 6,600 m (390 mph). If we take those speeds and cube them, we can guess the ratio of the horsepower as around 1.303 assuming the weight, drag and propeller efficiency is the same. In fact the M9K was heavier (980 kg vs. 830 kg.) and wider (1230 mm vs. 1180 mm) than the Homare and propellers get less efficient at higher speed. Thus the 1.3 ratio is too low. However, dividing the 1930 hp of the M9K by 1.303 gives 1480 hp. Thus we can be fairly sure that a Homare 22 as supplied to Mitsubishi in Japan gave less than 1480 hp at around 5000 m.

Now it is unlikely that Nakajima was simply telling lies about the performance of the Homare 22. A prototype engine would almost certainly give 1750 hp at 6450m at 3000 rpm. However, the production engines could not match this. Rinkol http://www.ww2aircraft.net/forum/aviation/kawanishi-n1k1-j-kasei-37359.html#post1023836 suggests "The early Homare engines suffered from production quality control problems affecting the intake passages" and that is the best explanation that I have found. Thus the only Homare 22 engines that Nakajima could supply to Mitsubishi had much worse performance than Nakajima prototypes.

Now if we go to the post-war American tests, everything makes sense if they managed to get the quoted performance out of the Ha-45. It is possible that the test aircraft actually had a hand made prototype engine from Nakajima. Alternatively, it is possible that the highly skilled American technicians had a translated manual quoting the expected power, realised that their engine was underperforming and set to work with their well equipped machine shop to remake any defective parts until they brought their engine up to its quoted performance.

The point is that most Homares in either the Ki-84 or the Shidens did not give their quoted performance and that the high speeds quoted could only be achieved with specially manufactured engines.

ps. On reflection, another possibility is that Nakajima actually solved the manufacturing problem very late in the war so that some Ha-45 engines were mass produced in 1945 with the theoretical performance and that one of those went to America for the Ki-84 test.

Hello Cherry Blossom,

For the most part, I believe you are correct. Not that it makes much difference but there is a difference in altitude of 400 meters which might swing the results by a couple percent.

The early models of he Ha-45 had a high pressure fuel injection system. It didn't work reliably so they eventually switched to a low pressure system. That apparently didn't work reliably either. The Hayate's Constant Speed Propeller (CSU as described in tests) was constantly failing. The spark plugs apparenly fouled badly so the US service people eventually went with a hotter US plug (I wish I could find the report of all the repairs done). IIRC, the engine itself wasn't bad, but a lot of the supporting equipment was not reliable and needed work.

Another issue to take into account is how the power ratings of the engine were achieved.
As an example, for the Ki-61-I with a Ha-40 engine:
Take Off: 1175-1180 HP @ 2500 RPM at +330 mm Hg (1.4765 ATA, 42.9 inches Hg)
Military Power: 1080 HP @ 2400 RPM at +240 mm Hg (1.346 ATA, 39.37 inches Hg) (4200 Meters altitude)

Japanese testing would have used the Military rating. US Testing would have used the War Emergency rating which in this case is probably pretty close to the Take Off rating.

420-something mph isn't unreasonable for an aircraft like this. The FW 190 is quoted at 408 mph at around 20,000 feet on only about 1500 HP or so and that was probably a bit of an understatement because US testing of a ground attack version (G Model?) achieved a bit over 420 mph.

- Ivan.

 

[JtD]

In my 20+ years of racing, and some experience with supercharged tow trucks, I've found what you said about the importance of intake and exhaust passages to be totally wrong, most of the improvements in horsepower come from increases in intake flow, AND exhaust flow, thru improvements to intake and exhaust passages and camshaft timing and duration. You can't get fuel/air mixture in if you can't get it out.

And a supercharger isn't as effective with bad intake and exhaust passages. All the same rules apply, supercharged or normally aspirated.
And I can't believe that drastically changes if the engine is at 10,000 or 20,000 feet in a aircraft.

In an aircraft, it would usually effect full throttle altitude, and power above full throttle altitude.

 

[Shortround6]

Aircraft engines were pretty high performance engines for their day. They were assembled using a lot of "tricks/techniques" used in racing engines of the time. Especially as the war went on. Cranks and rods were shot peened and polished. Piston sets were matched for weight ( perhaps not as well as a modern Hi-Po engine but then they didn't have digital scales) to an acceptable tolerance, not just any set of pistons of the right number would do. Intake and exhaust ports may already be polished, at least to some extent.
I don't know about Japanese or German engines but US and British engines were run on test stands for several hours to break them in and provide a final quality assurance, they were run on dynamometers and failure to met a specified power meant the engine was torn down to find out why. Once the the engine reached the airframe factory or repair shops it was up to the factory or shop workers to install the engine correctly. Engine maker was seldom responsible for the exhaust manifold/pipes. But most of those were fairly short compared to car manifolds/pipes and no baffles.
Intakes could be tricky, being a number of feet of ducting at times leading to the actual engine intake.

By WW II mistakes in the size of the intake were pretty rare as that had been worked out during the 20s and 30s. Most installations of high powered engines offered the ability of "RAM" air, forward motion of plane created higher than atmospheric pressure in the intake duct at the engine intake. Leaks in the duct could cause a loss of this pressure but this would only really affect the engine power at speed and at altitude.

 

[tyrodtom]

I realize aircraft engines are already high performance engines, but to state defects in the intake and exhaust passages can't have much affect on performance because they're supercharged on very high boost is just wrong.

 

[razor1uk]

Any and all imperfections are more noticeable as the oxygen thins and the power drops at altitude, boosting the intake via supercharging helps with reduce that, but the imperfections and manufacturing differences in castings, parts, machining etc, tyrodtom, are still there at altitude.

So you have worked on tuned high torque engines, so have you been many places and different heights above sea level, or stayed in a similar geological and geographical area with said engines.

It is truer, when chasing torque, that imperfections are less worry-some, and that in some parts of the intake or exhaust that some slight roughness* can be useful for encouraging some boundary layer turbulence and thus fuel atomisation with reduced fuel pooling, assuming the bore/diameter to flow rate ratios are already acceptable to the evisaged usual speed range of that engine.

When chasing power though, and in my mind and most others too, its just the same for torque as well, that just dependant on how you tune and set the engine up.
Power or torque; any misalignemts, imperfections, improper finishes inside the intake and exhausts reduce what gains you have per cylinder, and you really don't want each cylinder having its own character, fuelling, heat soaking and balaning issues, as that would result in a very compromised engine settings to try and get any sort of smooth engine.

Why shoot your self in the foot running a higher boost, to loose a few percent of it with higher heat soakings and fuel usage when a proper 'blueprinting' for example would produce the same power at few percent less boost with the option to have more boost when needed.

Ahem, blueprinting is a modern term for hand made/balanced/fitted to specs or to a certain standard. In WW2 and industry prior to then and in some places still, this largely didn't exist, so long as the parts were in tolerances, they'd just be slung together so to speak, and hence in a worsening home war situation, some worked well, others were dire.

*There are recent studies into the micro flow of fluids (both liquid and air) around 'dentines' like surface - aka akin to the dentines that form the skin of a shark, increasing the flow smoothness and lowering a sharks (or skates) energy consumption.

When it comes to speed or war, you want all you can get when you need it, and anything less than suitable just isn't right and somebody should get a rollocking for it.

 

[fastmongrel]

Would a poor intake/exhaust design have more effect the higher the altitude. For example I believe Bristol Hercules had a poor design of intake compared to the Hooker redesigned Merlin intake but as the majority of Hercules engined aircraft were used at low and medium altitude this restriction didnt effect performance.

 

[Shortround6]

It affects performance but the effect was known. That is to say that a Hercules engine performed as the factory said it would, not hundreds of HP less than the factory claimed. Now Bristol may have wondered WHY the RR engines performed better at high altitudes

These engines DEPENDED upon their superchargers for performance. A poor supercharger hurt performance everywhere but was more noticeable at higher altitudes. The supercharger itself was made up of several components. A poor inlet restricted airflow to the impeller. At sea level this may not be important as the supercharger can provide more boost than the engine can use anyway. BUT the supercharger has to work a little harder, Superchargers at the time were around 65-75% efficient which means that a poor supercharger was using 60-70% of the input power to actually compress the air. A good supercharger was over 70% most of the time. The EXTRA 30-40% was turned into heat (or all but 1-2% that went into friction in the gears and such) and this was the bugaboo. A hotter charge is more likely to detonate, a hotter charge is less dense so even at the same pressure it has fewer pounds of air per minute running though the engine and thus less power.

Now if hypothetical engine "X" needs 60 hp to compress the air it needs at sea level and the supercharger is 60% efficient you need 100hp going into the input shaft ( forget friction) and 40 hp is heating the intake charge over and above the heating that would take place from compression alone. If you improve the intake entry of the super charger ( and fiddle with the impeller?) and get the efficiency to 70% you only need about 86hp going into the shaft and have ONLY 26HP turning into extra heat. Only 65% as much power going into excess heat.

Now with a 1000hp engine at sea level the difference of 14hp going to the supercharger is not going to be real noticeable but perhaps the cooler, denser charge will be?

Most, if not all, of these supercharger intakes were castings and many had the carburetor mounted on them so making trial versions was difficult (expensive) until somebody else showed that better performance was possible. Hooker had a number of claims to fame, one of which was realizing that some of the formulas used to design/evaluate superchargers were wrong.

And until better than 87 octane fuel came along engines were limited to 6lbs or under of boost (many 1930s engines used 2-3lbs) most any supercharger could provide the performance required.

And again, the intake of the supercharger is made by the engine maker while the ducting leading to the engine is made by the airframe maker ( in co-operation with the engine maker at times, R-R having an airfield where trials were carried out to try to solve installation problems).

 

[fastmongrel]

Nice one SR6 but you really dont belong on the internet with all these clear concise posts that even a bonehead like me can understand

 

[GregP]

Hi Tyrodtom,

I am assuming, possibly incorrectly, that most of your supercharger experience has been either with automotive engines or possibly general aviation engines of relatively low displacement. Say, 580 cubic inches or less. If I am wrong, I apologize right now. In those engines, the finish of the intake and exhaust manifolds is important for best performance given the relatively small superchargers and displacement involved. That doesn't mean they can't produce great power, but it means the tricks are different. An NHRA Top Fuel engine makes more power than any Reno racer, but wouldn't last for the takeoff run much less for the climbout and race.

In the case of boosted large-displacement aero Vee engines, such as a Merlin (smallest), Allison (middle), or a Daimler-Benz 6 (biggest), the supercharger is moving enough air at enough pressure that the finish is barely noceable in performance. You might see a very slight improvement, but not enough to matter to anyone. In the case of big radials, the supercharger doesn't have to move the air very far and the pressure will make the finish of the intake and head passages irrelevant.

Now it is not entirely immaterial, just largely. If you are racing, every little bit helps and yes, they do it all for racing. But if you are flying a restored Bearcat ... then the difference between 2,135 HP and 2,150 HP is nothing, relatively speaking. If the boost on your R-2800 is 60 to 75 inches of Mercury (15 to 22 pounds for you British boost users), then you won't notice a performance difference after polishing the intakes or exhaust manifold passages in the head.

If I were doing an engine for myself, I'd probably smooth all the head passages ... just for the sake of doing it. But I'd not see any more power than one left essentially stock. The intake and exhaust passages in the head came from the factory smooth enough to make rated power, and are still smooth enough to do that today. When racing at Reno, the extra power comes from a lower compression ratio, higher boost, and higher rpm, not from smoothing the head passages. "Porting and polishing" are meaningless to a Merlin or Allison (balancing is another story). The 3,850 HP in a top racing Super-Merlin comes from taking the 65 inches of Mercury manifold pressure to 125+ inches and from raising the rpm from 3,000 to 3,400 or even slightly higher. An engine is really just an air pump. Move more air and you make more power.

Of course, it had better be running shot-peened Allison G-series rods and had better have a good ADI system and a spray bar or it won't live for very long at full power. Might not even GET there before blowing up. Stock Merlin rods aren't designed with enough of a safety factor to handle more than about 2,500 HP. That level was sufficient for all wartime Merlin use and it still suffices for all warbird use other than racing at the top level in Reno.

Most of the Bronze and Silver class birds don't need anything more than that. Only if you are running in the top 3 in Unlimited Gold will you need a Super-Merlin or Super R-3350. We'd love to put a racing Allison in one, too, but someone has to want to do it. Also, they don't even use anywhere NEAR full power until the final race. Nobody wants to blow his super-Merlin in a heat race! During race week they start flying about Tuesday or Wednesday, and the only full-power runs happen in the final race, when it matters. All the rest of the heats and races are just to get to the final race.

Then the dirty little secret of Reno Unlimited air racing surfaces. The strategy at the highest level is to win the Unlimited Gold Final at the lowest possible speed, thus saving the engine for next year. Typically, someone will jump out to a good lead and then slowly back off every few laps until they are running somethin like 85 - 90 inches on the last lap, and they only push the go-fast lever in again if somone is catching them. If they could get away with it, they'd win at 300 mph! Only competition makes them go faster than cruise power. By way of example, Steven Hinton Jr. can turn laps in the 515 mph range at Reno (on the old course), but won last year at 477.5 mph. So he probably flew the last lap at 460 mph or so since he qualified at 493.3 mph on the new course.

I notice this IS off-topic and if you want to discuss it further, maybe we should revisit a Reno thread or start a new one. I am not up on the engine hot-rod tricks for the jet class, but I'm sure FlyboyJ is since he is a team member on a jet team. I doubt if he'll tell you all the tricks in any case since they are still racing! Naturally, they do all the drag-elimination things they can to start with. But more thrust and a good smooth racing line are probably the real secrets beyond the aerodynamic clean-up.

 

[tyrodtom]

But the question is how refined was the intake and exhaust passages of Japanese engines as they were produced.
American and other allied engine manufactors usually had the time to optimize their engines, then quality control that insured that the majority of the engines produced met acceptable standards.
But did the Japanese ?

I'll bow to your expertise when it comes to large aircraft engines Greg, because I have zero experience when it comes to anything over 500 ci.

 

[GregP]

The ONLY experience I have with a WWII Japanese Aero engine is with the Planes of Fame A6M5 Model 52 Zero. It is a Sakae 21 and the quality of the parts matches anything I have seen on an American or British engines of similar vintage. That is to say, excellent.

If they DID suffer quality issues, it was probably with the assembly and bearing clearances or perhaps with the metal alloys themselves. But the fit and finish of the Japanese engines I have seen is excellent. Even the engine in our Mitsubishi Radien is quite good, though unrestored.

I would think the main possible quality issues would be with bearing clearances, possibly ring end gaps, fitting the rings upsides down maybe (causes excessive oil consumption and fouls plugs), and other issues of assembly. It is also possible the rubber parts were substandard and cracked easily, causing leaks in either intake or exhaust paths.

I do NOT know that the fit and finish were ever bad.

Another issue might be quality control in the carburetors. A great engine with a bad carb comes across as a bad engine.

Wish I knew more about the so-called Japanese quality issues with their engines later in the war. Seems like a good thing to go look for ...

 

[Shortround6]

On some of their engines the cooling fins were finished by hand on the heads. They were used to hand ( or hand held) finishing grinders. Unless they were incredibly sloppy there shouldn't be enough difference in the port passages to make a several hundred horsepower difference. Cadillac made parts for Allison like connecting rods an aircraft rods had around twice as many inspections as Cadillac car connecting rods and Cadillac was a quality car maker in the 1930s. Aircraft engines, even small, low powered ones, were in another class of manufacture than 99% of the worlds car engines in the 1930s. I am not saying bad parts were not made but it takes an awful lot of people looking the other way to actually get them into an engine. It takes an awful lot of screwing up to have an engine make around 75% of it's rated power. not finishing the intake or exhaust ports isn't going to make that much difference. Aircraft heads do not have the amount of "extra meat" that car heads often do. It is not a question of opening them up for extra flow but just finishing them off, de-burring if you will. Quality may very well have slipped near the end of the war but loose pistons (or bad rings) might make more sense than poor airflow through the ports.

 

[GregP]

I agree Shortround. The parts in the WWII Aero engines I have seen "up close and personal" are universally of very good quality.

I HAVE seen great parts poorly assembled to produce an engine that needed help rather quickly once it started running. I have seen overhauls last for an entire 5 - 10 hours! Nothing wrong with the parts, but the clearances were simply ignored and the engine self-destructed in normal operation. Perhaps the Japanese experienced a lot of poorly-trained operators at the engine plants later in the war? As I said, makes me want to look into it .... but where to start ....

 

[bobbysocks]

the beauty and magic of the old norton engines was how well they were ported...which was damn near perfect. i know a guy who races triumph motorcycles....he claims he can gain me 5+ hp with his port job on my bonneville. the better an engine breathes supercharged or not has to have some effect on performance.

 

[GregP]

Normally aspirated, yes. Once you get boost in a big engine, it is realtively meaningless. I am, of course, assuming decent surfaces from the factory. As I said, in WWII the factory surfaces were sufficient to obtain full power. That is, I'm not talking about very rough finishes to start with. Rather, factory smooth passages.

In a small, normally aspirated engine, the air enters at only atmospheric pressure, 29.92 inches of Mercury on a standard day, and smoothing can help quite a bit. Hence "porting and polishing" are a good thing, and I've done a few myself. I had a particularly good result on a Suzuki GS 1000 once.

In a big supercharged engine (the bigger the better), the air is being forced in and a decent surface is all that is needed ... and they all had that straight from the factory. Once you start boosting, the power comes from moving more air. Higher manifold pressure and rpm does it best, and does WAY more than smoothing an already decently smooth head passage.

 

[fastmongrel]

the beauty and magic of the old norton engines was how well they were ported...which was damn near perfect. i know a guy who races triumph motorcycles....he claims he can gain me 5+ hp with his port job on my bonneville. the better an engine breathes supercharged or not has to have some effect on performance.

A port job will certainly help your Bonnie breathe better but 5+hp gain. Mmm not sure about that not without some other work. My BSA Thunderbolt has been tweaked over the years with a balanced crank, ported head, CV carb conversion, electronic ignition, oil cooler and a belt primary drive and I reckon all that has raised power by about (Holds finger and thumb up) that much. Where porting does work is in the mid range on Brit twins, the Thud was always a torquey beast but now it can shame a Jap 600 up to about 60mph. Of course at 60mph the rice boiler goes past like I have hit the brakes

I used to race classic bikes and it was all about getting it to rev and not spew oil and engine parts all over the track not about peak power on a dyno. The original Fastmongrel was an Aermacchi 350 flat single that was bored out to 408cc and wrapped in a Drixton frame. It was fast for a pushrod aircooled motor and you could go hunting TZ250s on a twisty track but the engine quite often went home in a bucket. I detuned it with a new piston and a new cam and it became a beauty to ride but was never in danger of catching a 2 stroke on the straight.

 

[Shortround6]

Porting can help set the engine up for different things. Ford got carried away at times with huge ports on both the 351 Cleveland engine and some special 427 heads. Great for power at 6000 rpm and above but they actually hurt the engine/s down low. For street use the smaller port heads were actually better. The huge ports had too much volume and reduced the velocity of the incoming gas at low rpm ( low gas flow).
This is not the same as just smoothing things but often a good port job also changes the angle or curve inside the passage.

And again, what works very well on a rough, small cast iron port may have less benefit on a port that is already about the right size and somewhat smooth (less improvement since you are already halfway there).

Un-supercharged engines have less than atmospheric pressure in the intake manifold/ports. Gauges hooked up to the manifold of such engines are vacuum gauges and show how many inches less than normal the pressure is.

The downward moving piston is creating a low pressure that sucks the air through the engine instead using a pump to force the air in. Smooth passages, gentile turns and large cross sections can help but the force feed engine does have different conditions.
One of which is that it already can flow more air than the engine can use until it hits full throttle height.

 

[cherry blossom]

The consensus here seems to be that it would be hard for the likely "quality control" problems to reduce the power of the Homare 22 from 1750 hp at altitude to an estimated 1300 hp. The founder of Nakajima was a politician Chikuhei Nakajima - Wikipedia, the free encyclopedia and I suspect that they did not undersell their products. I have zero evidence but is it possible that the Homare was redesigned over 1944-5 along the lines of Hooker's improvement of the Merlin over 1940-1 crucially without changing the model designation? A possible motivation for such apparently bizarre behaviour might be that the high power had already been falsely claimed for the initial models 21 and 22. Thus calling the redesigned models by the same names and blaming quality control problems could allow Nakajima to avoid accusations of fraud.

ps. If the Homare 21 of 1944 gave only 1300 hp at altitude, it makes it clear why the lighter Ha-112 II giving 1250 hp. at 5800 m. at 2600 rpm could replace it in the Ki-116 and give similar performance to a standard Ki-84 (but presumably not matching the later American tests).