Aces High Bulletin Board
General Forums => Aircraft and Vehicles => Topic started by: F4UDOA on November 19, 2006, 10:35:10 PM
-
HT/Pyro
I was running through the new F4U FM's and I noticed the low alt climb rates seemed a little low fuel levels so I ran a few climb test.
Climb to 20,000FT
180 Gallons of fuel= 1080LBS (50% in F4U-1/1A and 75% in -1D)
Fuel burn = 1
Test method takeoff level on runway, accelerate to 150MPH engage WEP and auto climb, start timer. stop timer at 20K. WEP expires after 5 minutes do NOT re-engage.
1. F4U-1(Birdcage) with Water 50% fuel= 8 minutes 4 seconds
2. F4U-1A with Water 50% fuel= 7 Minutes 19 seconds
3. F4U-1D With Water 75% fuel= 7 minues 21 seconds
4. F4U-1D With Water 100% fuel= 7 minutes 45 seconds<==about 45 seconds slow.
The F4U-1A/1D should reach 20K in roughly 7 minutes flat. They appear to be an average of a couple of hundred FPM off. The F4U-1 is also slow considering I tested it about 400lbs light and it is still slower than listed time at it's reduced weight.
I'm guessing there is some hidden weight attached to the new FM's or the climb FM of all the F4U's has changed from the listed climb rates from AH2.
I think it is the former as it affects all F4U's the same.
-
Another possibility could be that simply your experimentation had flaws.
This is not a direct criticism pointed at your own methods, but rather an explanation on why it might be problematic to compare in-game climb performance with real-life performance by engaging "auto-climb".
For one thing, auto-climb is not known to be the most efficient method of climbing in Aces High. Despite popular beliefs "auto-climb" is actually "auto-speed" - the plane will stabilize its general heading and its balance in the roll axis, while adjusting its pitch to keep the plane stable at a certain set speed.. and therefore, it does not take into account a number of factors that might effect climbing efficiency. The optimal speeds for climbing might change throughout the range of altitudes a plane travels - which, is clearly more than the "auto climb" feature might handle.
Therefore, my guess is that the auto-climb can provide an objective reference of climbing capabilities only when it is directly compared with other in-game planes also using auto-climb.
ie) When one compares the Bf109 with the F4U, the difference in magnitutde of climbing capabilities between the two planes will be presented at a relative scale: one will first engage the Bf109 in auto-climb, and then compare it with the F4U engaging in auto-climb. It is highly probable that both the 109 and the F4U will not match real-life figures per se, but rather only the relative difference between the two planes engaged in auto-climb, will match that of the relative differences between real-life figures.... for instance, if real life climb figures show a 30% advantage in climb in favor of the 109, the same 30% ratio will show between the auto-climb results compared, despite the ultimate "time to reach X altitude" figure might not match that of real-life.
The only way to actually try and match real-life climbing figures with that of the game, IMO, would be to look up historical documents and see what sort of methods the actual planes used, to perform climb tests. One particularly necessity would be try to find out the average climb speed the plane used in actual real-life testing, and adjusting AH "auto-climb" speed to match that.
If there's one thing I've found out during my own series of comparative tests in turn performance, it is that coming up with an objective and mechanical testing method is incredibly difficult. My suggestion is that you should try to consult these boards and have a discussion in which testing method might provide the results you are seeking... before any testing is actually done in the first place.
To the extent of my knowledge, every piece of information we have for AH planes were actually tested by different people with different test methods - which seriously questions the objectivity of it all.
As a side note, I myself do not trust other people's results in turn performance testings, for example, because I find serious faults in the objectivity of their testing methods - whereas my own tests were devised to set each of the plane's to their mechanical limits and leave out as much human factor as possible, others were rarely so thorough.
A stark example of why a totally objective method of testing is required can be observed from one of my own test notes, linked in my sig below; "The AH Compendium of...."
-
Kweassa,
I would say that test methods are always in question however the differance in climb was immediately noticable in low speed maneuvers at low alts. This led me to initiate the test in the first place.
I generally try to test in more "real world settings" than flight simm settings however the auto climb does a very good job in maintaining a consistant IAS that matches the best climb speed of the F4U (approx 155MPH IAS). give or take almost +/- 10MPH will not affect climb rates or times very much so climb method has some slop built in to it. However the loss of time is roughly 45 seconds over 20,000FT which may not sound like that much but it is fairly significant in average climb rate.
The F4U-1D is now reaching 20k at 7.45 which is an average of 2580FPM which is up from 7.00 which is 2857FPM.
This is almost 300FPM a slow climbing aircraft like the F4U does not have to spare.
-
Originally posted by Kweassa
As a side note, I myself do not trust other people's results in turn performance testings,
Nearly choked on my coffee when I read that :)
Badboy
-
on the hogs has changed with the new release. In some aspects it appears to be a bit better but its performance as a pure angles fighter has been affected more than a little. If and how the changes affect climb to alt I dont know....
I do know that the hog has lost a bit of its "zoom". I noticed the nikki for example retains E comparatively better vs the hog then it did before (or at leat seems to)....
-
Nearly choked on my coffee when I read that
It doesn't necessarily mean I dismiss other tests altogether. Rather, if there may ever be serious disparities in certain figures, I'd have an unwaivering faith in the faultlessness of my own figures, than believe someone else's, which were gained from questionable test methods, according to my own standards.
I generally try to test in more "real world settings" than flight simm settings however the auto climb does a very good job in maintaining a consistant IAS that matches the best climb speed of the F4U (approx 155MPH IAS). give or take almost +/- 10MPH will not affect climb rates or times very much so climb method has some slop built in to it. However the loss of time is roughly 45 seconds over 20,000FT which may not sound like that much but it is fairly significant in average climb rate.
It could be due to a variety of factors..
For instance, how much trust do we give AH's FM when the altitude ranges go over 'non conventional' ranges seen in gameplay? Over 20k? 25k? 30k? Frankly, I'm not so sure about that part.
Another could be a variety of real-life factors that are not introduced in-game, such as specific cowl settings, radiator flap settings, etc etc.. For instance, in IL2/FB a certain level ofaccurate testing is possible because the game allows the players to set specific supercharger/mixture settings and radiator cowl/flap settings as well. This became extremely handy when a certain band of Luftwaffe fans actually compared in-game tests with the plane in the exact configuration of engine settings and radiator settings, as proposed in real life test documents. They found some numbers did not match, and the fact that the FM had problems became unquestionable since the testing configuration matched that of real life configuration almost exactly. Can we have the same amount of confidence in our own tests with Aces High?
Thirdly, (a personal question), what was the reasoning behind engaging ADI(wep) right after takeoff? Could the results be different if you perhaps engaged it somewhere around 17~18k, when the planes starts to reach higher altitudes? Again, this calls for a separate discussion in what kind of testing method we should use, to hold AH's numbers against real life figures.
My 2cents.
-
If you're not loading up any external ordnance, there shouldn't be any weight or performance difference from the previous version. I would try the same test with the previous version.
-
Originally posted by Squire
Whats the empty weight difference between the F4U-1A and F4U-1D? They arent identical.
1a has wing tanks, 1d doesn't.
Bronk
Edit: squire pulled his post.
-
Ya I took off the post when I realised the debate isnt about the differences in the 1A/1D its about the F4Us climb to 20k compared to version 2.08.
-
1 thing i noticed in the -1 and -1a hog is fuel load out are not the same. When u take 75% fuel, it gives u 75% in wings and 75% in main. Is this correct as I thought the -1 and -1a are same plane all except for paddle prop and canopy. The -1 hog loads 25% in wings tanks and 100% in main if u take a 75% fuel load out.
Anybody got any info on this?
-
The load is the same in gallons, 272, its just pumped in differently, but thats no big deal.
-
As far as the climb rate goes, I also have to wonder wether the 7 minutes to 20k that was done in real life was done with ADI the entire time. I find it hard to beleive the pilot would shut it off after 5 minutes, when he's trying to get the best time to alt time. The AH tests above are done when the WEP is off after 5 minutes, and we all know the 5 min limit was nothing more than a suggested limit in real life, and not always adhered to, unlike AH where its shut off for you?
-
Squire,
You know that is a great question. I have to find where exactly but I do remember reading that the climb test were done at WEP for 5 minutes then at mil power. BTW the internal fuel load on the -1/-1A is 361 gallons. If you select 25% you will get 90 gallons.
Pyro,
Thanks for looking into it. My guess was extra weight was in there somewhere. I kept checking my gear flaps and ordinance but I had none.
I have a chart from the POH for the F4U-1D (It is a pretty common chart). It shows the F4U at Mil power climb times to 20K at weights from 9,000lbs on up. Basically every 1,000lbs equals 1 minute to 20,000FT. That is roughly 350FPM. When I tried that on the -1A it was much slower than I expected. That is what got me started in the first place.
Here is a F4U-1A climb chart with paddle prop. The mil power is significantly different.
(http://mywebpages.comcast.net/markw4/air1777climb.jpg)
-
Is that a chart of an actual test? or just a calculation?
Because if its an actual test it would indicate the pilot did an ADI climb to 38,000 feet without letting up. Might not have been great for the engine, but they are trying to establish parameters.
I realised after giving it a think, that its really academic anyways, the chart shows to do 20,000 feet in 7 minutes, you gave to use ADI the entire way, wether its a calculation or not. If it was a 5 minute ADI climb, the chart would indicate that, but it doesnt.
-
Originally posted by Kweassa
It doesn't necessarily mean I dismiss other tests altogether.
Phew, after analyzing flight models now for around 15 years that's a relief :)
Badboy
-
Squire,
It is an actual flight test, I have the entire report of 7 pages non of which makes reference to using ADI any longer than 5 minutes.
The actual duration of ADI in the aircraft was 8.5 minutes (10 in the Hellcat) not the 5 minutes that is rrequently referred to. Which is more than enough to get to 20K.
The climb at 20K in the text of the report states the climb to be over 2500FPM. The climb chart indicates 2300FPM which would seem to show a reduction in power at that altitude.
(http://mywebpages.comcast.net/markw4/air1777pg1.jpg)
-
Perhaps testing them to 16k (within 5 minutes) would be a better test, at least that would indicate with no doubt how the two compare.
Of course, as with all the climb tests, I have to ask as to method in real life. From brakes release? or after airborne? or? it can be the difference of a minute sometimes.
It can get pretty messy, as we have seen in many other debates on climb rates in the past.
The F4U-1A seems to climb according to the AH chart (for the F4U-1D), it holds @2500 fpm untill 17500 ft, then starts to drop off.
...I will let Pyro sort it all out I guess. ;)
-
Pyro,
I have run this many times in prior version at different weights and loading. It is unquestionably slower than prior versions and even when removing weight via fuel the problem does not correct.
Under 10K I would say that it is more noticable.
-
I tested the F4U-1C thinking it may have been unchanged.
8 Minutes 7 seconds to 20K at WEP until it expired and then mil power.
Definitely a pronounced change from the previous version.
-
So it climbs slower than a Spit I?
-
Slower than the Hindenburg towing a B-52.
Slowwwww.
-
Originally posted by F4UDOA
Slower than the Hindenburg towing a B-52.
Slowwwww.
:rofl
-
When the drag model was revised the first time, I tested many of the aircraft for changes in climb rate and compared the data to the previous modeling. There was no difference other than normal testing variations.
Today, I tested several aircraft using the same method.
This method begins by flying at 50 feet ASL at 300 mph TAS. I then engage auto-climb and engage WEP. I measure the time required to reach 10,000 feet. This method introduces the zoom-climb factor, and thus, does not measure pure climbing ability alone. A similar method was used by the USAAF to help define what they called "combat climb". Only between 4,000 and 10,000 feet does the aircraft climb at its steady-state rate.
Whether or not this is a fair measurement of climb is not important, because the data is repeatable, and therefore perfectly acceptable to determine if the flight model changed with a new update.
Previous data/current data
F4U-1D: 2:24.47/2:27.96
F4U-4: 1:55.67/1:57.04
F4U-1C: 2:32.74/2:33.35
F6F-5: 2:15.67/2:15.63
Spit14: 1:47.13/1:46.53
Spit16: 1:39.90/1:40.25
190D-9: 2:03.35/2:01.63
109K-4: 1:45.69/1:44.25
Tempest: 2:05.38/2:03.86
As you can see, the differences are within what could be considered Normal Test Variation (less than 2% on average).
My regards,
Widewing
-
You see, the old one had the cowl flaps CLOSED... but you can clearly see they're OPEN now...
That's what's slowing you down!! :aok :rofl :cry
-
Widewing, - I cannot but admire your dedication on the subject.
I've done some testing, but not to that amount on a single subject.
I did some testings with friends in the training arena. Not so easy to clock like this (how do you clock anyway, - stopwatch?), I was doing turn-tests like 109F vs Spit IX with various loads etc. What I got out was enough for a little frame of truth, such as that a lightly loaded 109F would turn on a pair with a Spit IX with full tanks etc.
I did some climb tests, and found the Spit XIV too slow to 20K, the 109G too fast, and in comparison, the Spit XVI is probably climbing too well, - well, it's a rather tight ballpark anyway.
-
Originally posted by Angus
Widewing, - I cannot but admire your dedication on the subject.
I've done some testing, but not to that amount on a single subject.
I did some testings with friends in the training arena. Not so easy to clock like this (how do you clock anyway, - stopwatch?), I was doing turn-tests like 109F vs Spit IX with various loads etc. What I got out was enough for a little frame of truth, such as that a lightly loaded 109F would turn on a pair with a Spit IX with full tanks etc.
I did some climb tests, and found the Spit XIV too slow to 20K, the 109G too fast, and in comparison, the Spit XVI is probably climbing too well, - well, it's a rather tight ballpark anyway.
I prefer to test offline, with fuel burn set at zero. This eliminates one variable.
Here's a general outline of the tests I perform.
Acceleration at SL, 5,000 ft, 10,000 ft, 20,000 ft and where applicable, 30k as well. I measure 150 mph to 250 mph and 250 mph to 350 mph or max attainable speed.
Max speed at SL, 10k, 16k, 20k, 25k and at best altitude. Sometimes done in increments of 1k to establish speed curve. Consumes huge chunks of time.
Turn radius and rate at SL, with full flaps, 25% fuel.
Turn radius and rate at 10k, with full flaps, 25% fuel
Turn radius and rate at 20k, with full flaps, 25% fuel
Turn radius and rate at SL, clean, 25% and 100% fuel
Turn radius and rate at 10k, clean, 25% and 100% fuel
Turn radius and rate at 20k, clean, 25% and 100% fuel
Rate of climb; combat climb, 300 mph TAS from SL to 10k, 25% fuel.
Rate of climb; from stop on runway to 10k, 25% fuel.
Rate of climb; from stop on runway to 20k, 25% fuel.
Dive speed from 25k to 10k, max speed recorded.
Dive acceleration, from 25k; Bunt over at 300 mph TAS, record time to accelerate to 500 mph TAS.
Roll rate at 150 mph, 250 mph, 300 mph and 350 mph. Testing done at 1,000 ft ASL.
Max range, recorded from E6B at best RPM/MAP for max range.
As you can imagine, this testing takes 8 to 10 hours per aircraft....
My regards,
Widewing
-
What is deck speed for a F4U-1A ?
I had lot of trouble shacking one from the tail of my yak9U the F4U-1A looked to be faster than the yak.
-
Originally posted by straffo
What is deck speed for a F4U-1A ?
I had lot of trouble shacking one from the tail of my yak9U the F4U-1A looked to be faster than the yak.
once up to speed its a very fast plane... it just has sucky acceleration.
-
Originally posted by straffo
What is deck speed for a F4U-1A ?
I had lot of trouble shacking one from the tail of my yak9U the F4U-1A looked to be faster than the yak.
If you fly it clean, it will reach 366 mph at sea level. If you have taken a bomb or drop tank, the pylon/shackles cannot be dropped and it will reduce your max speed at sea level to 360 mph.
My regards,
Widewing
-
So it's faster than a yak.
-
Originally posted by straffo
So it's faster than a yak.
Yes, but it's more complicated than that.
First we need to examine acceleration. I measured acceleration for the F4U-1A, Yak-9U and P-51D. All were loaded with 25% fuel. Testing was done at 50 feet ASL. Recorded time required to accelerate from 150 mph to 250 mph and from 150 to 350 mph.
150 mph to 250 mph-
F4U-1A: 23.37 seconds
Yak-9U: 22.50 seconds
P-51D: 24.32 seconds
150 mph to 350 mph-
F4U-1A: 1:37.06
Yak-9U: 1:47.04
P-51D: 1:38.56
So, the Yak has a very slight advantage to 250 mph. However, at 270 mph its rate of acceleration has already fallen behind that of the F4U. Where the Yak shows its greatest edge is between 150 and 230 mph. Over the whole speed range from 150 mph to 350 mph, the Corsair gets there first, with the P-51D close behind and the Yak trailing well back.
Now, we know that the Yak doesn't have a WEP setting, so it can run at max speed for as long as it has fuel. On the other hand, both the F4U-1A and P-51D have just 5 minutes of WEP, whereupon they are limited to MIL power. So, I measured max MIL power speeds and compared them to the Yak's max speed. Tests were done at 50 feet ASL.
F4U-1A: 352 mph MIL power
Yak-9U: 357 mph
P-51D: 355 mph MIL power
Let's assume that all three are in a tail chase, with the Mustang and Yak chasing the F4U.
Since the F4U-1A accelerates best (especially over 300 mph), it will pull away. Even when the Corsair peaks at 366 mph, the P-51D cannot close on it having the same max speed. Meanwhile, the Yak is falling steadily behind. After 5 minutes both the F4U and Mustang run out of WEP. The P-51D cannot close on the F4U until the Corsairs speed bleeds below 355 mph. That will take longer for the F4U, than for the Mustang or Yak (I'll discuss speed bleed later). It will take the P-51D quite some time to chase down the F4U, assuming WEP is not used again by either. Our Yak, which has fallen way behind might as well turn around. Its 5 mph advantage will require a long chase to catch up. Probably too long to bother.
Personally, if I were flying the F4U, I'd drag it away from the Yak and kill it... Then I'd go after the Yak, which is generally outclassed by the Corsair.
Speed bleed... A measurement of how fast the airplane sheds E when power is pulled off to idle. This can be a factor in a tail chase. I measured the time required to bleed down from 350 mph to 150 mph. Altitude was 300 feet using auto-level. This should be a function of drag, but moreso of mass.
F4U-1A: 46.12 seconds
Yak-9U: 27.93 seconds
P-51D: 37.69 seconds
So, the F4U-1A bleeds off speed slower than the P-51D and much slower than the Yak-9U. This hints toward the reason that the F4U zoom climbs better than many other fighters... Greater potential energy for any given speed.
To balance this, I decided to measure E bleed over a 180 degree 5g flat turn, beginning at 350 mph, max power. All three fighters began at 350 mph, speed at the end of the turn is recorded.
F4U-1A: 308 mph
Yak-9U: 302 mph
P-51D: 306 mph
Not a huge difference, but it tends to show that the Yak can pull a tighter turn from high speed as it bleeds more E in the process, which tightens the turning circle. Of course, simply pulling off some power can change the entire relationship.
My regards,
Widewing
-
Thank for you post , it confirm my intuition .
Originally posted by Widewing
F4U-1A: 46.12 seconds
Yak-9U: 27.93 seconds
P-51D: 37.69 seconds
I'm pretty surprised byt the E bleed of the Yak (even if again I expected this from my AH experience) what can justify this ?
Aerodynamic differences ?
Mass differences ?
A La7 is a comparable to the yak in size/weight should it bleed E like a Yak ?
-
Originally posted by straffo
Thank for you post , it confirm my intuition .
I'm pretty surprised byt the E bleed of the Yak (even if again I expected this from my AH experience) what can justify this ?
Aerodynamic differences ?
Mass differences ?
A La7 is a comparable to the yak in size/weight should it bleed E like a Yak ?
My inclination is to believe that mass is the primary factor.
I will test the La-7 and compare it to the Yak. If the La-7 holds E much longer than the Yak, it would be something to discuss with Dale and Pyro.
My regards,
Widewing
-
Originally posted by Widewing
F4U-1A: 46.12 seconds
Yak-9U: 27.93 seconds
P-51D: 37.69 seconds
This sounds awfully wrong. The P-51D has an empty weight of 7,635 lbs, the F4U-1a 8,982 lbs, and the Yak 5,526 lbs. With internal fuel the P-51 and F4U is rather even in weight, difference being only a couple of percentages. The P-51D is obviously the most aerodynamic plane of the three with the Yak a close second, both having inline engines with the Merlin offering 220 hp more than the Klimov. The F4U with its radial engine producing in excess of 2000 hp (2250 with water) is obviously the most draggy, only able to match the 1720 hp P-51D in speed. Being close in weight, but more draggy, the F4U should decelerate quicker than the P-51D, probably closer to the Yak.
-
Ok, I tested the La-7 for speed bleed. Here's how they match up, speed 350 mph TAS, pull off power to idle, time to bleed down to 150 mph is recorded.
F4U-1A: 46.12 seconds
Yak-9U: 27.93 seconds
P-51D: 37.69 seconds
La-7: 39.41 seconds
For the La-7, it cannot be correct. It weighs just 7,300 lbs compared to the Yak which weighs 7,050 lbs. Moreover, the Yak certainly has a lower drag coefficient than the air cooled Lavochkin. So, I would believe that they should bleed speed at a similar rate.
However, that is not the case. The La-7 holds its E better than the much heavier and extremely clean P-51D. There's no way that the radial-engined La-7 has a lower Drag coefficient than the .0176 of the Mustang.
I think we have partly discovered why the La-7 is so uber in Aces High.. It has the E retention of a much heavier aircraft without actually suffering the associated weight penalty. Obviously it doesn't require perking, just fixing.
My regards,
Widewing
-
Originally posted by Viking
This sounds awfully wrong. The P-51D has an empty weight of 7,635 lbs, the F4U-1a 8,982 lbs, and the Yak 5,526 lbs. With internal fuel the P-51 and F4U is rather even in weight, difference being only a couple of percentages. The P-51D is obviously the most aerodynamic plane of the three with the Yak a close second, both having inline engines with the Merlin offering 220 hp more than the Klimov. The F4U with its radial engine producing in excess of 2000 hp (2250 with water) is obviously the most draggy, only able to match the 1720 hp P-51D in speed. Being close in weight, but more draggy, the F4U should decelerate quicker than the P-51D, probably closer to the Yak.
I read somewhere that the La-7 had a CDo of .0223, and taken with it's wing area of about 190 sq/ft, that equates to a flat plate area of 4.24 sq/ft. Compare that to the P-51D with a flat plate area of 4.10 sq/ft. Now add in the far greater weight of the P-51D and it should be no contest. Yet, as modeled, the La-7 bleeds speed SLOWER than the P-51D... Has to be a modeling issue.
My regards,
Widewing
-
Sounds like there is more than one plane that is affected by this anomaly. For all we know the entire FM needs remodeling.
Makes me wonder if HTC do any testing themselves?
-
Widewing: In your deaccel test, you have not taking the prop into account at all.
Other than that Ill take a look at the numbers.
-
Originally posted by hitech
Widewing: In your deaccel test, you have not taking the prop into account at all.
Other than that Ill take a look at the numbers.
I appreciate you having a look.
I realized that the prop would have some effect, but since there's no way to eliminate it in game...
Another round of tests could be done by simply shutting off the engine, but I don't know if that would make much, if any difference. I will try that later this afternoon and compare the results.
Again, thanks.
My regards,
Widewing
-
Its interesting to throw in the Mosquito in there...with 100 percent fuel "clean" (and an a/c of low drag), its @ 20,000 lbs, and I tested it using the above method and it decelerates from 350 to 150 in 40 seconds. Almost the same as an LA-7 of 1/3 the weight.
Spit VIII at 25 gas does it in 30 sec, im assuming thats a function of the much reduced inertia from a lighter a/c.
-
Some additional data concern speed bleed.
I tested the original four fighters again, this time shutting off the engine. There was no difference in bleed-down time, all were within 1/10th of a second of the original numbers. There was one exception, and that was the P-38L. When you shut down the engines on the P-38s, the props go into feather, greatly reducing drag.
Here's the shut down bleed times... From 350 mph to 150 mph, 25% fuel (except where noted) and zero burn. Altitude was 300 feet ASL.
F4U-1A: 46.08 seconds
Yak-9U: 27.88 seconds
P-51D: 37.78 seconds
La-7: 39.34 seconds
Here's some additional types tested.
190D-9: 36.46 seconds
109K-4: 35.72 seconds
SpitVIII: 30.91 seconds
N1K2-J: 35.53 seconds
P-38L: 33.41 seconds (85.47 seconds if props feathered)
Tempest: 30.28 seconds
Typhoon: 30.21 seconds
P-47D-40: 49.84 seconds (54.25 seconds with 100% fuel)
I find it odd that the Brit fighters bleed E faster than the others. I expected than the P-47 would bleed slower by adding weight (full fuel) and it did. The Typhoon and Tempest were virtually identical, but the Tempest did have a lower drag wing design... I expected a difference. The N1K2-J and 109K-4 bleed speed at similar rates. I suspect that the 109's smaller flat plate area was offset by the N1K2-J's greater weight. Looking at the three American fighters, they tend to bleed speed according to the relationship of their relative weights. Again, this is what I expected.
On the other hand, the lightweight La-7 retains speed like a much heavier fighter would without the weight penalty. However, the Spitfire Mk.VIII, with a similar weight to the La-7 and 109K-4, bleeds speed like crazy. This is despite having a very low drag wing design.
I also do not understand why the Tempest and Typhoon bleed speed as rapidly as they do. Especially when compared to the 2,000 lb lighter 190D-9. Can the prop diameter/design make that much difference?
My regards,
Widewing
-
Yes, despite the 100,000,000 posts from the "Spit Haters Club" (tm) concerning it "not bleeding E", its an interesting result. ;)
-
The thing with the Spit isn't that it doesn't bleed E, it's that it can pretty much instantaneously get it BACK.
Could the F4U's wing configuration have anything to do with its substantially increased E-retention over the P-51? It WAS determined that the gull design reduced drag on the airframe.
-
Well, post a film of it getting it back "instantaneously" and post it. I will look at. I have been in AH for almost 5 years, and have never come across such a thing yet.
...and the criticisms were precisely that: "wouldn't bleed E" (supposedly).
Trouble was, when asked to test and show a result, it was never forthcoming, just hot air.
I dont like the over use of the Spits in the MA either, and thats really what the problem is, not the flight model.
...In any case, I wont hijack the thread (every AH thread turns into a Spit thread at some point, its like a scientific constant in AH BB), but I would also like PYRO to look at the relationship with weight and drag with some of the types mentioned above.
Later. :cool:
-
Right when HTC changed the airflow code, folks mentioned the 109s, saying "I used to never have to slow down, now I have to sideslip, use rudder and flaps, all just to land!" so that might be a side-effect or bug from the code change.
Widewing: The prop is a huge brake. It's still spinning and the wind is pushing it, so you're slowing down a lot. When you're in AH and you shut the engine off, you "coast" a lot longer if you reduce RPMs as far as they will go. This lessens the bite of the wind, and might show better deaccel times. I don't know if a feathered prop on the P38 has the same drag as a reduced RPM prop on other planes, though. I suspect the 38 would still have the advantage.
-
Originally posted by Widewing
I read somewhere that the La-7 had a CDo of .0223, and taken with it's wing area of about 190 sq/ft, that equates to a flat plate area of 4.24 sq/ft. Compare that to the P-51D with a flat plate area of 4.10 sq/ft. Now add in the far greater weight of the P-51D and it should be no contest. Yet, as modeled, the La-7 bleeds speed SLOWER than the P-51D... Has to be a modeling issue.
I know it is tempting to use aerodynamic data, but you can't reach conclusions about deceleration by comparing Cdo values alone. The reason is that deceleration comes from two terms, the prop drag divided by the mass of the aircraft and the total aerodynamic drag divided by the mass of the aircraft. The aerodynamic drag is made up of two components, induced drag and profile drag. At high speed, where you would normally be able to assume that the induced drag was negligible, and you might therefore be able to compare the zero lift drag coefficients, you still can't, because unless the prop can be fully feathered the prop drag will have a very significant impact on deceleration. At lower speeds, this is compounded as the induced drag increases, so regardless of where you are in the envelope, a comparison of Cdo isn't a reliable indication of deceleration.
Badboy
-
Originally posted by Badboy
I know it is tempting to use aerodynamic data, but you can't reach conclusions about deceleration by comparing Cdo values alone. The reason is that deceleration comes from two terms, the prop drag divided by the mass of the aircraft and the total aerodynamic drag divided by the mass of the aircraft. The aerodynamic drag is made up of two components, induced drag and profile drag. At high speed, where you would normally be able to assume that the induced drag was negligible, and you might therefore be able to compare the zero lift drag coefficients, you still can't, because unless the prop can be fully feathered the prop drag will have a very significant impact on deceleration. At lower speeds, this is compounded as the induced drag increases, so regardless of where you are in the envelope, a comparison of Cdo isn't a reliable indication of deceleration.
Badboy
I understand. I also understand the relationship of drag and mass to deceleration. Added weight should increase induced drag at lower speeds. If for no other reason than needing a higher AoA to maintain level flight. If that is factored into the modeling, it's completely masked (in terms of test data) by the increase in mass. You can observe the effect of added weight visually while testing. At low fuel, with Combat Trim on in Auto-Level, cut power and watch as the aircraft is trimmed ever nose high to maintain level flight. Add weight and this occurs at a higher speed and the angle of attack is eventually greater. This would certainly increase induced drag, to my thinking at least.
By increasing the fuel weight of the P-47D-40 by 1,372 lbs, it decreased the speed bleed time by almost 9%. So, a 9.5% increase in mass resulted in a 9% decrease in deceleration.
I tested the La-7 with full tanks, or 122 gallons. This added an additional 549 lbs to the plane's mass. This was roughly a 7.5% increase in mass over the 25% fuel weight of the plane. When tested, this yielded a 7.2% decrease in deceleration.
This tells me that same rules apply to both aircraft. The problem as I see it is that the La-7's baseline mass is either too high or the total drag is too low (or some combination thereof). Can we assume that the La-7's total drag is greater than that of the P-51D? I think it's probably a safe bet. Even if these were identical, there's no getting around the difference in mass. There is no doubt that the P-51D has a far greater mass (about 30% more). You don't need a full-blown analysis to see that something is unusual here. It's the old dead skunk analogy; you really don't need to see a dead skunk to know it's there. ;)
I'm sure Hitech and Pyro will be looking at the equations to see why this exists. I'm also curious to know why the Spitfire decelerates like it's tied to a figurative tree. Must have the draggiest propeller in the history of aviation. :)
Seriously tho, sometimes simple testing can reveal things never thought to be unusual. In hindsight, I always thought that the La-7 retained E far better than I would have imagined prior to flying it online. I never expected it to be anything more than my own perception.
My regards,
Widewing
-
Originally posted by Widewing
Can we assume that the La-7's total drag is greater than that of the P-51D? I think it's probably a safe bet.
I don't think we can. The La-7 has only 180 hp more than the P-51D to make those 380 mph at sea level, so I think they are very close in parasitic drag. The La-7 is a small plane after all. When the prop is factored in I think that huge four-bladed prop on the P-51 is a much more effective air-brake the the La's three-bladed high-speed prop.
(http://upload.wikimedia.org/wikipedia/en/thumb/8/88/P-51_at_Oshkosh.JPG/250px-P-)
(http://upload.wikimedia.org/wikipedia/commons/2/28/LavochkinLa-7.jpg)
-
They should also look at the hurricanes flight model to. It does amazing things for a wooden plane at incredible speeds,lol.
-
Something is fishy :
http://www.netaces.org/ahplanes/comparisons/spdretover.htm
http://www.netaces.org/ahplanes/comparisons/deceloverview.htm
I know it's not exactly the test you have made WideWing but HAMMER result are too different IMO.
-
Could just be the different propellers like Hitech says, but it does smell a little ... marine life, yes.
-
Those results are from many revisions ago. AH1 even.
-
Originally posted by hammer
Those results are from many revisions ago. AH1 even.
Certainly but I expect the difference between the planes to be similar.
As far as I can remember there was no drastic change on the F4U-1D and the Yak9U for example.
This evening I'll try to redo your test with the actual version and check if there is a lot of differences
-
Widewing: Retest the p51d once using 2500 rpm instead of 2700
-
This thread has long ago been jacked.
However if you go to Hammer's website you can see that the climb of the F4U-1D was faster in AH1 than it is icurrently even when flown from a standing start on the runway.
BTW, it matched the climb of the NAVAIR chart almost exactly prior to the latest revision. It is almost 1 minute behind the NAVAIR chart in it's current format regardless of who does the testing.
-
F4UDOA, this thread has not been hijacked. They are discussing possible reasons for a possible bug that would lead to a slower rate of climb. It's related.
-
Hi Widewing,
I've just taken another look at your test data, and I think the evidence points to a different explanation. Your initial reaction to the test data is:
Originally posted by Widewing
The problem as I see it is that the La-7's baseline mass is either too high or the total drag is too low (or some combination thereof). Can we assume that the La-7's total drag is greater than that of the P-51D? I think it's probably a safe bet. Even if these were identical, there's no getting around the difference in mass. There is no doubt that the P-51D has a far greater mass (about 30% more).
But your following results indicate that there might be a better explanation:
By increasing the fuel weight of the P-47D-40 by 1,372 lbs, it decreased the speed bleed time by almost 9%. So, a 9.5% increase in mass resulted in a 9% decrease in deceleration.
I tested the La-7 with full tanks, or 122 gallons. This added an additional 549 lbs to the plane's mass. This was roughly a 7.5% increase in mass over the 25% fuel weight of the plane. When tested, this yielded a 7.2% decrease in deceleration.
This tells me that same rules apply to both aircraft.
This is fairly strong evidence that the mass and drag are behaving as you would expect. So the only other thing that could produce the results you are seeing is prop drag. That the prop drag is a significant component can be seen by looking at your P-38 results and the difference when fully feathered.
P-38L: 33.41 seconds (85.47 seconds if props feathered)
This would suggest that the difference in deceleration in your results could be due to the prop modelling, and not necessarily an error. Different propellers feathering to different pitch settings could explain the results you are seeing, and in view of the fact your earlier tests seem to show that the mass and drag are behaving as expected, I think this is the more likely explanation.
If that is the case, it would mean that there wasn't anything wrong with the flight model at all, and it might just be that some props are feathering more than others at low power, or power off.
Regards...
Badboy
-
Originally posted by hitech
Widewing: Retest the p51d once using 2500 rpm instead of 2700
Normal RPM is 2,900 for the P-51D. However, I tested at 2,700 RPM and 2,500 RPM. Results are as follows, same test parameters as before.
2,700 RPM: 41.47 seconds
2,500 RPM: 45.22 seconds
I also tested the La-7 (normally 2,500 RPM in MIL power, 2,600 RPM in WEP)
La-7 @ 2,300 RPM: 42.34 seconds
In addition, I tested the La-5FN for reference sake (at normal 2,500 RPM)
La-5FN: 36.75 seconds
My regards,
Widewing
-
Originally posted by Badboy
This would suggest that the difference in deceleration in your results could be due to the prop modelling, and not necessarily an error. Different propellers feathering to different pitch settings could explain the results you are seeing, and in view of the fact your earlier tests seem to show that the mass and drag are behaving as expected, I think this is the more likely explanation.
If that is the case, it would mean that there wasn't anything wrong with the flight model at all, and it might just be that some props are feathering more than others at low power, or power off.
Heya Badboy,
One problem with this explanation is that the props don't feather. Upon shutting down the engine, there is no drop in RPM, all the way down to 150 mph and lower. Any feathering would be reflected in reduced RPM. In all of the single-engine fighters I tested, props remain in full low pitch when the engine is shut off. Only the P-38 feathers. All others tested windmill at full RPM.
My regards,
Widewing
-
Originally posted by Widewing
One problem with this explanation is that the props don't feather. All others tested windmill at full RPM.
Ahh, ok, but your other tests do make it look as though the weight and drag are ok. So it still looks to me as though the different props may be producing significantly different drag regardless that they don't feather. Considering the variety of different prop/reductiongear/engine combinations that doesn't surprise me. Also when I look at the difference in the P-38 feathered test, I can see how those differences might produce the test results you posted.
I have a feeling that if you could feather all the props you might get test results more in line with those suggested by the weight and drag characteristics of the aircraft.
Regards...
Badboy
-
It is the prop that is making the difference. At minimal power settings the prop drag is providing so much drag, that any extrapolation about drag co's of the airframe is totally meaningless.
HiTech
-
Originally posted by hitech
It is the prop that is making the difference. At minimal power settings the prop drag is providing so much drag, that any extrapolation about drag co's of the airframe is totally meaningless.
HiTech
Which prop is making the difference, that of the P-51 or the La-7?
I can't find much info on the prop used on the La-7, but looking at photos it certainly does not have what Hamilton Standard called "high activity" blades. The prop appears to be about 10.5 feet in diameter. I don't see any visual difference between the the prop on La-7s and La-5FNs. They look identical.
That said, the AH2 La-5FN bleeds speed considerably faster than the La-7 does. Granted, the La-7 incorporated some aerodynamic improvements to increase speed. But, the La-5FN was a tad heavier. As it is, the AH2 La-5FN bleeds off speed 7% faster than the La-7 does. Whereas the P-51D bleeds speed 3.7% faster than the La-7.
Note that the C.205 is a bit heavier than the La-7, and swings a prop of similar diameter and blade chord. Yet, it bleeds speed at a much greater rate than the La-7.
This is where I'm confused. If the P-51 has a much higher drag propeller (apparently high enough to more than offset a 33% difference in mass), that causes it to bleed speed faster than the La-7, then why is there a significant speed bleed difference between the La-7 and La-5FN? After all, they have nearly identical weight and the same propeller. Something isn't adding up for me.
Is there something that I'm missing?
My regards,
Widewing
-
Widewing: What is your point.
-
Originally posted by Widewing
then why is there a significant speed bleed difference between the La-7 and La-5FN? After all, they have nearly identical weight and the same propeller.
They did not have the same propeller. The la7 had a new propeller. I will dig out details when I get home.
The key (and unusual) aspect of the La7 development was that there was no engine upgrade or additional thrust component (other than any efficiencies gained from a new propeller).
Apart from any benefits aquired from the new prop the La7 was purely an exercise in pure drag reduction plus some saving thru weight (induced drag) reduction.
Some theorise that the La7 had a greater applicable period of WEP 10 mins over the La5FN's 2 minutes. This (La5FN @ 2 mins) is not a categorical record however.
Edit
Having said all that I would have thought that in comparison with eg a Yak9U the La7 should lose speed more quickly in a high speed glide. I will check what prop data I have for both.
-
What Tilt said: They took the LA-5FN and reduced its drag to create the LA-7. Same engine, but with aerodynamic refinements. They should not be the same E retention at the same weight.
Tilt is the resident LA expert here, no question. ;)
-
I think WW's point is that with all the other factors, how can THAT much difference (a full 6 seconds with the La-5 at full RPM, and the La-7 operating at REDUCED power!) between the La-5 and La-7 be solely a difference in the prop and some aerodynamic cleanup, if E retention is as much a product of mass (as in the F4U, which btw Viking you GREATLY underestimated. She weighs in at roughly 12,000lbs under combat loads, not quite but close to twice that of the P-51).
-
Saxman:
p51d 2,500 RPM: 45.22 seconds
La-7:2,400 RPM 39.41 seconds
And we are not talking about reduced power, all test are done at min throttle. What we are talking about is more prop drag do to higher RPM.
-
Originally posted by Saxman
… if E retention is as much a product of mass (as in the F4U, which btw Viking you GREATLY underestimated. She weighs in at roughly 12,000lbs under combat loads, not quite but close to twice that of the P-51).
While I may have somewhat underestimated the F4U's mass, it is nowhere near twice that of the P-51D which had a normal loaded weight of 9,500 lbs and a max. load of 12,000+ lbs.
-
They key difference between the Vish 105 prop (la5fn) and the Vish 105-v4 La7
The V4 was a so called "constant speed" prop assembly with additional "anti flap" properties
Both were 3.1m in dia.
Both weighed 141 kg.
Whilst I realise the debate re prop drag has arisen to show that simple glide tests do or do not indicate the drag characturistics accurately...................
Once you have established there is an influence of prop drag then I would have thought that rpm etc becomes meaningless as the angle, blade area and back geared engine resistance would vary from ac to ac. at any given rpm.
Take a 41.2 litre 14 cylinder engine with a compression ratio of 7:1 (back geared 1:1.45)and compare turning it to a 30 litre V 12 with a compression ratio of 6:1 (back geared 1:2.1). We can all argue over which would be easier for the prop to turn but I doubt they would be the same.
I have noted that since the aerodynamics were universally modified that some ac do dive away from the La7 far more effectively than before. The P51 being one of them. It is a matter of record that a 109 G4 would eventually obtain a higher rate of acceleration in dive than an La7.
BTW WEP power as we know it on the Ash82FN was 2500 rpm (1200mm mg) being 100 rpm above the 2400 mil power setting (1020mm mg). The engine had a rating of upto 2600rpm but this was only ever used so fleetingly limited to a duration of 30 seconds presumably at lower altitudes.
-
Originally posted by hitech
Widewing: What is your point.
I'm not trying to make a point, I was trying to understand why there are odd looking variations in E retention.
Here's some test data, in descending order in terms of slowest bleed to fastest. 350 mph to 150 mph, power at idle. Measured in seconds/Max Takeoff Weight
P-47N: 49.91/16,300
P-47D-40: 49.84/14,500
P-47D-11: 48.34/13,582
F4U-1A: 45.91/12,050
190A-8: 45.65/9,682
F6F-5: 43.97/12,300
F4U-4: 43.65/12,400
190A-5: 42.94/8,583
Ta 152: 42.00/11,500
La-7: 39.41/7,300
Ki-61: 38.73/7,650
P-51D: 37.96/12,100
P-51B: 36.72/11,200
190D-9: 36.46/9,840
La-5FN: 36.75/7,379
109K-4: 35.72/7,440
SpitXIV: 35.69/8,500
C.205: 35.61/7,498
N1K2-J: 35.53/9,040
P-38L: 33.41/17,500
109F-4: 32.06/6,393
FM-2: 31.31/7,431
SpitVIII: 30.91/7,875
Ki-84: 30.69/7,965
Tempest: 30.28/11,400
Typhoon: 30.21/11,400
SpitXVI: 29.82/8,500
P-40E: 29.48/8,400
Yak-9U: 27.78/7,050
SpitV: 27.15/6,785
A6M5: 24.46/5,920
My regards,
Widewing
-
Originally posted by hitech
Saxman:
p51d 2,500 RPM: 45.22 seconds
La-7:2,400 RPM 39.41 seconds
And we are not talking about reduced power, all test are done at min throttle. What we are talking about is more prop drag do to higher RPM.
By the way Saxman, propeller RPM for these two would be:
P-51D: 1,293 RPM
La-7: 1,142 RPM
Add to this that the P-51D swings a 11'2" prop, while the La-7 swings a 10'2" prop. I believe that the P-51's prop generates more drag than that of the La-7 when windmilling at higher RPM. The unknown seems to be; is that drag so much that it overcomes the 3,000 pound weight differential? It does seem that in this particular case, it does (at least as its modeled) as testing at similar prop RPM gives the P-51 a very large advantage in E bleed.
My regards,
Widewing
-
I'm not trying to make a point, I was trying to understand why there are odd looking variations in E retention.
As I said in the beginning. It is do to prop drag, also understand that prop drag in AH is a close approximation, and isn't really a detail that we spend a lot of time on with each plane.
HiTech
-
Originally posted by hitech
... also understand that prop drag in AH is a close approximation, and isn't really a detail that we spend a lot of time on with each plane.
HiTech
(http://www.lordpanzer.com/images/outrage.jpg)
;)