- Thread starter
- #61
N1K2-J Shiden-Kai Performance
- Thread starter Laurelix
- Start date
Ad: This forum contains affiliate links to products on Amazon and eBay. More information in Terms and rules
AerialTorpedoDude69
Airman 1st Class
- 180
- Mar 1, 2022
Laurelix
the 363 and 369 numbers are all very odd to me because the only source I could find was published by Japanese original sources in the 1950s.But going off some of the combat reports of Hellcat pilots, it seems that the N1K1-J was almost the same speed as a F6F-3 Hellcat, which was rated at 371-MPH. USN pilots described only "very slowly" being able to overtake a fleeing Shiden.
Japanese wikipedia has the N1K1-J at 363 MPH and the N1K2-J at 380 MPH. This seems to be more accurate than Francillon's numbers, which were derived from (likely) original sources that were published in the 1950s.
Some sources on Japanese wikipedia list the Shiden-Kai prototype's maximum speed as being 620.4 kph. Another source also lists the Reppu A7M1's max speed as being 624 to 640 kph with the second prototype having a top speed that was even 10-15 kph faster, due to individual thrust stacks (around 634 to 655 kph).
I believe the Reppu data is also from a pilot's, Tomokazu Kasai, interview on the Shiden-Kai. The reason ranges are given is that Kasai (and another pilot) were having to remember decades old data. So while this is a first-hand account, it cannot be accepted as completely accurate on its own, although I assign a lot of credibility to it.
- Thread starter
- #63
Yes
But going off some of the combat reports of Hellcat pilots, it seems that the N1K1-J was almost the same speed as a F6F-3 Hellcat, which was rated at 371-MPH. USN pilots described only "very slowly" being able to overtake a fleeing Shiden.
Japanese wikipedia has the N1K1-J at 363 MPH and the N1K2-J at 380 MPH. This seems to be more accurate than Francillon's numbers, which were derived from (likely) original sources that were published in the 1950s.
Some sources on Japanese wikipedia list the Shiden-Kai prototype's maximum speed as being 620.4 kph. Another source also lists the Reppu A7M1's max speed as being 624 to 640 kph with the second prototype having a top speed that was even 10-15 kph faster, due to individual thrust stacks (around 634 to 655 kph).
I believe the Reppu data is also from a pilot's, Tomokazu Kasai, interview on the Shiden-Kai. The reason ranges are given is that Kasai (and another pilot) were having to remember decades old data. So while this is a first-hand account, it cannot be accepted as completely accurate on its own, although I assign a lot of credibility to it.
However… whenever the 369mph is mentioned for the N1K2-J, the rate of climb is also shown as completely abysmal for an aircraft with such good power to weight.
The only explanation is can give is that the 369mph (595km/h) max speed was achieved with the huge underbelly fuel tank attached, flying at military power which would add significant weight and drag and would explain why the climb is much worse.
In the IJN performance table
It claims 321knots (595km/h) at 5600m
It also claims 7:22 to 6000m at military power
Now….
According to U.S. the Ha-45-21 has 1765hp at Military Power
N1K2-J loaded weight is 3800-4000kg
That's power to weight ratio of 0.44-0.464hp per kg
7:22 to 6000m
According to this
P-51D at 4360kg (9611lb) running at 61Hg military power (1490 horsepower)
Which is 0.34hp per kg, being 35% worse than N1K2-J is rated to climb to 6000m in only 7:00, which beats the 7:22 for N1K2-J
Then there's the A6M3 Mod 22
1010 horsepower at military power
2678kg loaded weight
0.377hp per 1kg
So again… worse than N1K2-J
Yet it climbs faster to 6000m, 7:19
—-
N1K2-J with the additional 400L under belly fuel tank at overload weight is literally 4860kg according to the IJN table and it would also suffer from extra drag
1765hp / 4860kg = 0.36hp per kg
This would put its power to weight ratio similar to that of those planes that also need 7 minutes to 6000m
Last edited:
MIflyer
1st Lieutenant
Well, here is the page from the Profile on the airplane.
AerialTorpedoDude69
Airman 1st Class
- 180
- Mar 1, 2022
I missed your comment when you first posted it last year, but according to Japanese wikipedia, the gondolas reduced speed by something like 15 kph (IIRC). I looked a little more closely into the matter and found that the original N1K1-J prototype only had two 20mm cannon in underwing gondolas. Kawanishi had rushed out the N1K1-J so quickly that they hadn't even had time to match the fuselage to the new engine. I believe that the "wart" that you're referring to is the "oil cooler" or at least what some refer to as the "oil cooler" but it looks more like a fuel cooler.
The oil cooler was a supplemental add-on that appeared later in the design phase of the N1K1-J because they didn't even bother to redesign the cowling to add a duct for additional oil cooling. There are other problems too, like the Shiden Kai's high level of wing-fuselage fillet which supposedly helped with takeoff but probably impaired level flight.
------>
Laurelix
that is something you mentioned earlier and it's an interesting hypothesis. But how often did anyone record aircraft speeds using overload/ferry performance and not military performance? Wouldn't they have recorded the speed using cruise values if it were an overload/ferry test flight? It's a good guess because the numbers seem about right, but as far as I'm aware, it's not standard practice to write down overload/ferry as military performance.We know that Japanese sources always use the military non-WEP maximum engine setting to determine their top speeds, unless specifically stated otherwise. So, the 620.4 kph was probably higher. Also it's a very specific number so this seems to be something that Kasai remembered with clarity.
My personal belief is that there's some kind of translation error or typo because the 363 figure for the Shiden 11 seems accurate given wartime descriptions of the Shiden 11 from Hellcat pilots. But a cleaned up and lightened aircraft should probably be more than six MPH faster. Removing the wing gondolas themselves increased the speed of the N1K1-J by -around 15 KPH (according to Japanese wikipedia IIRC).
In fact, that number is suspiciously similar to the N1K1-J Otsu which was around 9 MPH faster than the N1K1-J. In other words, it seems that translators may have swapped the values for the N1K1-J Otsu with the N1K2-J.
- Thread starter
- #66
The engine in N1K1 has 1280hp at 6000m
It achieved 490km/h at 5700-6000m right
Now this plane has shit ton of drag obviously due to all these external floats.
N1K2-J has 1625hp at 6000m with Ha-45-21
If we use same drag coefficient of N1K1 with those floats and calculate what speed it would have with 1625hp… we would get this;
1625 / 1280 = 1.2695
Cube Rooted = 1.0828
490 x 1.0828 = 531km/h
Now the question is… do those huge floats only reduce max speed by 64km/h or 104km/h
That bottom central main float might as well be another fuselage cuz its that big
- Thread starter
- #67
In this F6F gun can footage
View: https://youtu.be/413k38v_oqs
The N1K2-J is seen in combat with the underbelly tank
If we look at pictures of N1K2-J fighters, that underbelly tanks seems like it's a standard operational loadout.
View: https://youtu.be/413k38v_oqs
The N1K2-J is seen in combat with the underbelly tank
If we look at pictures of N1K2-J fighters, that underbelly tanks seems like it's a standard operational loadout.
Last edited:
MIflyer
1st Lieutenant
MIflyer
1st Lieutenant
Depends a lot on altitude. F6F had two stage supercharging, not nearly as good as an R-2800 with a turbo at altitude, but still better than single stage.
MIflyer
1st Lieutenant
Here is something for comparison. Slowed the Wildcat down to 240 mph.
- Thread starter
- #71
F4F-3 has 331mph vs 240.5mph with floats
90.5mph difference, that's 145-146km/h speed difference
- Thread starter
- #72
So yeah…. Back to this…
It is not going to be a mere 60-65km/h speed difference
AerialTorpedoDude69
Airman 1st Class
- 180
- Mar 1, 2022
Your hypothesis is plausible, but we need evidence to show that testing with a drop tank was standard practice and it was not. The prototype testing of the N1K1-J was done without a drop tank attached, affording to these images.
But you're right, when flying any long-range mission they flew with drop tanks. In Genda's Blade, unless it was an interception mission, they were using external tanks. CAP. Escort. You name it, they were probably using drop tanks for it. So I imagine that they did do testing at cruise settings, with a drop tank attached. But, again, that's for cruise, not max power. If they were using max power for long range missions, it wouldn't give usable data for calculating the distance required for long range missions. Did any air force use VMAX with drop tanks attached for any reason?
In other words, if Japanese sources did list the "max performance" of the Shiden Kai with drop tanks attached, it would have been an error. But errors are common in the literature of the era due to translation and sourcing issues.
I think an alternative to your line of reasoning is this:
Francillon's is taken from General View of Japanese Military Aircraft in the Pacific War, which was published in the 1950s. This book was patched together from a variety of sources, from engineers to ground crew. One error it introduced to the historical record is top speed. General View uses the first Ki-84 prototype's speed as the production model's maximum performance. The error is that it verifiably doesn't use overboost settings for max speed, it doesn't include airframe modifications that were made to production versions, and the prototype used an 1,800 HP engine. Something similar probably happened in regard to their reporting of the Shiden's top speed. It probably used the early prototype's top speed, which would have lacked individual thrust stacks, WEP, and a Homare 21.
As far as we know, General View was the first time the N1K1-J and N1K2-J was assigned a top speed in Western sources and it's the top speed used by Francillon. In other words, there is evidence that this same source was used to inform Francillon and all subsequent Western sources on the N1K2-J. And because it uses a patchwork of sources, it's loaded with inconsistencies and errors.
- Thread starter
- #75
What we know for sure is that in the N1K2 manual… the prototype definitely didn't fly with the fuel tank achieving 611km/h with derated Ha-45-21
What's also worth noting is that my estimate is that N1K2-J is 658km/h vs 672km/h max speed relative to Ki-84
(Derated Ha-45-21 having 611 with 1440-1450hp, which gives 635 with 1625hp for full power Ha-45-21)
Now whilst N1K2 had bigger wings at 23.5m2 compared to Ki-84's which were 21.00m2… the wings of Ki-84 are standard high lift wings with rather high CL_Max of 1.46 wing lift coefficient according to its manual. So high lift wing which also means more draggy. N1K2-J however actually has low drag, Laminar Flow wings for speed, so around 1.3 CL_Max wing lift.
This is another reason why I don't believe N1K2-J would be so much slower compared to Ki-84. Simply wouldn't make sense. Also the climb rate included. N1K isn't that much heavier than Ki-84 and both use same engine. The climb rate shouldn't be so much worse
Hence why I completely disregard the 369mph max speed and 7:22 to 6000m sources.
The N1K2 prototype official test flight performance figures in its manual, to me is the highest authority, it's not second hand source but a full legitimate primary source and gives the most expected, realistic performance figures.
What's also worth noting is that my estimate is that N1K2-J is 658km/h vs 672km/h max speed relative to Ki-84
(Derated Ha-45-21 having 611 with 1440-1450hp, which gives 635 with 1625hp for full power Ha-45-21)
Now whilst N1K2 had bigger wings at 23.5m2 compared to Ki-84's which were 21.00m2… the wings of Ki-84 are standard high lift wings with rather high CL_Max of 1.46 wing lift coefficient according to its manual. So high lift wing which also means more draggy. N1K2-J however actually has low drag, Laminar Flow wings for speed, so around 1.3 CL_Max wing lift.
This is another reason why I don't believe N1K2-J would be so much slower compared to Ki-84. Simply wouldn't make sense. Also the climb rate included. N1K isn't that much heavier than Ki-84 and both use same engine. The climb rate shouldn't be so much worse
Hence why I completely disregard the 369mph max speed and 7:22 to 6000m sources.
The N1K2 prototype official test flight performance figures in its manual, to me is the highest authority, it's not second hand source but a full legitimate primary source and gives the most expected, realistic performance figures.
Last edited:
AerialTorpedoDude69
Airman 1st Class
- 180
- Mar 1, 2022
Yes, that's true. But does the manual have the overboost performance? Tomokazu Kasai's estimate of 620.4 KPH is about in line with a weak WEP on a machine that can make 611 KPH.
I wish we had just one performance rating that was with a fully rated Ha-45-21 at its critical altitude with WEP because then we could reasonably have the ZLDC. But like you say, it was definitely lower than the Ki-84's ZLDC. Although the fuselage-wing fillet (according to Japanese Wikipedia) supposedly helped with takeoff, it was a high-drag feature that cost the design some speed, although how much, we don't know. For some reason, Japanese military planners with obsessed with runway length and the Shiden had the most complex flaps out of all WW2 aircraft. I'd guess that they combined split with fowlers because there was a military requirement to have a short takeoff and landing and laminar flow airfoils tended to dramatically increase takeoff and landing rolls. Like I mentioned earlier, it's also probably why it had such a huge fillet angle. The fillet might have generated lift at a high angle of attack.
But the 611 KPH + 4% performance (roughly WEP) would get 635 KPH, which is close to your numbers. Another 4% gets it within your lower boundary for performance. So we're really speaking of a relatively trivial performance difference.
- Thread starter
- #77
The Japanese were expecting the Ha-45-21 to achieve 1700hp at 6000m hence why they estimated the full power N1K2 would achieve 644km/h (348 knots) at 6000m at military power without W/M boost and 5:15 to 6000m
However the engine only did 1625hp at 6000m instead
As far as I'm aware the 1700hp at 6000m wouldn't be achieved until Ha-45-25
AerialTorpedoDude69
Airman 1st Class
- 180
- Mar 1, 2022
Laurelix
I had read your post earlier a while back and then forgot that you had reverse-engineered the ZLDC from the Shiden-Kai manual. I tried to also find the ZLDC but my numbers were not correct. Anyway, my understanding of your work is that you have a few assumptions in it:- That the Shiden Kai which the manual references had a de-rated Ha-45-21.
- That this de-rated engine produced 1,800 horsepower at sea level and not the 1970 HP.
- Your max speed calculation adds 200 HP to get the WEP rating rather than the military rating.
The Ha-45-23 (which was installed on the finalized but not produced N1K3 and N1K4 series IIRC) solved that issue but switching to fuel injection. The direct fuel injection technology may have caused other issues with total output that was not solved until the -25. My source for this is a Polish aviation book which oddly had more information on it than any English source.
But the -23 would have been a vastly superior engine at altitude compared to the -21 as the carburetor caused engine cutouts, fuel economy problems, and problems developing power at altitude. It's strange that Nakajima didn't move to the -23 entirely.
- Thread starter
- #79
Ha-45-11 / derated 21
Did I understand it correctly? My thoughts on this are that you would be correct but only if the assumptions above are true. I think the third assumption is completely correct and backed by the evidence. But I couldn't find whether the Shiden Kai from the manual used a de-rated engine or another de-tuned version of the Homare.
- That the Shiden Kai which the manual references had a de-rated Ha-45-21.
- That this de-rated engine produced 1,800 horsepower at sea level and not the 1970 HP.
- Your max speed calculation adds 200 HP to get the WEP rating rather than the military rating.
The Ha-45-23 (which was installed on the finalized but not produced N1K3 and N1K4 series IIRC) solved that issue but switching to fuel injection. The direct fuel injection technology may have caused other issues with total output that was not solved until the -25. My source for this is a Polish aviation book which oddly had more information on it than any English source.
But the -23 would have been a vastly superior engine at altitude compared to the -21 as the carburetor caused engine cutouts, fuel economy problems, and problems developing power at altitude. It's strange that Nakajima didn't move to the -23 entirely.
Has 1460hp at 5700m at military power
Full power ha-45-21 ended up being
1625hp at 6000m at military power
1625 / 1450 = 1.121
Cube rooted = 1.0387
611 x 1.0387 = 635
The Japanese speed of 611 and the calculated speed of 635 is all at military power
AerialTorpedoDude69
Airman 1st Class
- 180
- Mar 1, 2022
Thank you, that makes perfect sense. My best attempt to calculate how an 175 HP increase impacts the Shiden Kai's speed matches your calculation as well. As far as I can tell you are correct about the continuous military speed of the N1K2-J and I have to compliment your math skills.
But just so others can follow, I've done my best to show my work below:
1. Military speed of Shiden Kai with full-rated Ha-45-21 = Speed of derated Ha-45-21 *(1625 HP/1475 HP)(to the 1/3rd power)
2. Military speed of Shiden Kai with full-rated Ha-45-21 = 610 KPH * (1.11 to the 1/3rd power = 1.03228011)
3. Military speed of Shiden Kai with Full-rated Ha-45-21 = 630 KPH
I rounded off a lot so my numbers are not as precise as yours. Overall, your estimates are the best calculated top speed for the fully rated Shiden Kai available, although I think there are some assumptions made in your WEP calculation that might need some additional research in order to validate. For example, the WEP setting for the Ha-45-21 is 500mm of mercury at 3,000 RPMs. But given that the Homare was already using MW50 injection at military power, then I don't know if it could get another 200 HP because MW50 already added 200 HP according to Japanese records.
But that calculation itself is probably better than anything else out there. However, we could make a better estimate if we knew the Sumitomo-VDM prop efficiency used in the Shiden-Kai. It's also worth noting that their props got better throughout the war. Although US bombing eventually destroyed the Sumitomo Water factories, they did manage to relocate to a basement beneath a department store and continued production.
By the way, I went ahead and uploaded the Shiden-Kai manual to a free filehosting solution, which anyone can download here. You already have it but in case someone else wants to download it, I hope it helps them.
EDIT: It was deleted
At nearly 400 pages it will take forever to translate, and the scan quality is bad enough that even AI OCR cannot do an adequate job at translating it automatically. There are a few options available to me but none are good enough to do a comprehensive translation. This AI seems to be one of the better free options out there.
I have a few questions:
First, are we sure the Shiden Kai in the manual has a derated Ha-45-21? I could not find what engine they are referring to in the engine section. The kanji is almost unreadable.
Second, are we sure that the Homare 21 overboost setting was capable of adding 200 horsepower? It was probably weaker than that given that it used MW50 injection at its military power setting.
Third, if would be helpful if we knew the four-blade VDM propeller efficiency in order to determine how an increase in horsepower would impact top speed. Although this isn't necessary, the fact that their props got better over time means there could have been a speed bump later in the war that's not accounted for.
Last edited:
