Isa Speed Testing Of All Fighters In Game

[II./JG77_Manu*]

I am starting a new thread here, because the other one got dragged way off topic, so i didn't wanna see such an important matter disappear beyond a personal discussion.

Please stay on topic this time, and only discuss things, which have to do with the aircraft performances.

 

Finished my testing with all the 6 fighters in ISA conditions, once again i show you how i did it, for the ones who didn't read my first testing about this matter:

Altitudes 0 [did it on 70m, because didn't wanna fly into the next tree or hill], 3650, and 6000m.

My procedure was to take off, go to altitude,  and stay there till top speed is reached. Then i began to measure the speed. I used the Grids to do it, using 40km, and measuring the time the plane needed to go those 40km. In my opinion by far the most accurate method, even if your measuring error would be one second, that is still less then 0,5% error. It can also be replicated easily by everyone, who wanna doublecheck my results. I set the radiators this way, that oil doesn't exceed 110°C, and water doesn't exceed 100°C [within a long time]. The exact setting can be seen in my tableau. I also put the real life numbers next to my results, and the difference in between.

Here it is:

 

 

now we got the whole picture. What can we see? Let's draw some conclusions:

- Yak and F4 overperforming in general

- all planes but the 190 overperforming at higher altitudes, with PF-engined Yak and Lagg the most significant.

- all planes but the G2 and the Lagg overperforming at zero alt.

 

I think the Yak and the F4 have the most significant errors, and should definitely be fixed. Will compile a report about those two planes, and send it to Han. The other ones have "acceptable" errors, for now, in my opinion, because they mostly cancel each other out.

[MK_RED13]

Wow.. Thank you very much Cel!!!

[GOAT-ACEOFACES]

Got Tracks?

[JtD]

Thanks for testing. Biggest error imho is not with individual aircraft, but the fact that all overperform above full throttle altitude. In relative performance, aircraft with low full throttle altitudes, i.e. M105PF engine, i.e. Yak and LaGG, benefit the most.

 

What altitude did you change gear at with the La-5?

 

Could you please let me know what reference you are using for the 535 sea level maximum of the Bf109G-2? In direct comparisons between DB601E and DB605A, the DB601E was tested faster down low, and most figures I know are around 525 for G-1/G-2.

[II./JG77_Manu*]

Thanks for testing. Biggest error imho is not with individual aircraft, but the fact that all overperform above full throttle altitude. In relative performance, aircraft with low full throttle altitudes, i.e. M105PF engine, i.e. Yak and LaGG, benefit the most.

I thought the same, but La5 and G2 proved me wrong a little, because they even overperform below their full throttle altitude at higher alt (while not overperforming at lower alt of course). And i think F4 and Yak performance are really to far off.

 

 

 

What altitude did you change gear at with the La-5?

I think somewhere above 2000. Aren't you supposed to change it at 2k? Normally i try it a couple of times, and as soon as i see a performance increase on the instruments, i stay in second gear. As for the test at 3650 - definitely second gear

 

 

 

Could you please let me know what reference you are using for the 535 sea level maximum of the Bf109G-2? In direct comparisons between DB601E and DB605A, the DB601E was tested faster down low, and most figures I know are around 525 for G-1/G-2.

Rechlin tests (which say 525) used the G2/G1 with non retractable tailwheel. Same goes for the direct comparisons between the engines. The G2 with retractable tailwheel however had a performance increase of about 10-12kph at ground level, it managed 535, according to the official datasheet from Messerschmitt. I got it in my book, but you can definitely find it somewhere on the internet, too. And i thought we have the one with retractable tail wheel in game, don't we?! (never checked it in the ingame model to be honest). If we have the one with fixed tailwheel i will correct my tableau immediately.

Edited March 12, 2015 by Celestiale

[JtD]

2km is too low for change of gear on the La-5, best altitude is somewhere between 3000 and 4000m. There shouldn't be much of a difference at 3650 between the gears.

 

I've read up on the G-2 speed figures, apparently, 535 is a calculated figure and corresponding tests are often not clear about the tail wheel, so it is a little bit of guessing. Of the tests naming the type of tail wheel, only the Finnish tests with a non-retractable tail wheel appear to support the 535km/h, other tests are somewhat slower (down to 505 at SL). 535 is probably equivalent to the Yak's 510 - optimistic, but possible with a good aircraft. Tail wheel in game is retractable, so no need for corrections.

 

Does your book explicitly say the Rechlin test was done with an aircraft without retractable tail wheel?

Edited March 12, 2015 by JtD

[II./JG77_Manu*]

 

 

Does your book explicitly say the Rechlin test was done with an aircraft without retractable tail wheel?

Yes it does. It also says, there have been even lower speeds achieved with other 109G1s, with a "bigger fixed tailwheel" and worse coating. But that haven't been the one's regularly used in battle. 

[unreasonable]

Nicely tabulated: may I suggest that you add an extra column and show the Game-RL speed as a percentage of the RL?

 

The devs have stated elsewhere that they are within 2% of their target speeds (IIRC): highlighting what look like the results outside this range may get their attention.

[Tektolnes]

Quick % comparison attached.

[Phant0m]

It would be interesting to see the speeds obtainable without manual radiators (for the 109s) and additionally with AI controlled radiators (noting what temperature the AI tries to maintain).

 

I have a strong feeling this will account for the observed discrepancies.

 

Great work testing this celestial - numbers are better than off-topic arguments.

[[BTEAM]_Shifty_]

Am I the only one bothered that even under ISA some planes can fly with fully closed radiators? Is that realistic in any way? I mean winter ok, but +15C?

[DD_Arthur]

 

Great work testing this celestial - numbers are better than off-topic arguments.

 

+1.  Yes, well done Celestiale.  Very interesting results.  I note that between sea level and approx. 3700m all planes are just about within tolerence and only at high altitudes this over-performance becomes questionable for some planes. 

 

Also, in view of the 109F4's slight over-performance at all altitudes, can we now forget about any perceived bias in the way the devs have modelled these 'planes and accept that they need to do a small amount of tweaking to their high altitude model?

 

Once again; thanks for taking the time to do this Celestiale

[NachtJaeger110]

Thank you Celestiale, this is the right stuff  

Please do send it to Han, I'm curious for the answer. However I don't think that they'll have time to do something about it soon...

 

Alltogether, the errors are less serious than I expected.

I would be happy if they could only tune those two errors over 6%.

[GOAT-ACEOFACES]

The devs have stated elsewhere that they are within 2% of their target speeds (IIRC)

Problem is we don't know what the 'target' is.. 1C/777 has never posted the values the settled on.. We don't which set of real world data they used, or, if they averaged more than one set together. So, showing a difference greater than 2% could say one of two things..

 

1) The dev's missed the target, assuming they are using the same data you used.

2) The dev's hit the target, your just not using the same data they used.

[Kurfurst]

In direct comparisons between DB601E and DB605A, the DB601E was tested faster down low, and most figures I know are around 525 for G-1/G-2.

 

Werknummer 14 026 again... 

[NachtJaeger110]

Problem is we don't know what the 'target' is.. 1C/777 has never posted the values the settled on.. We don't which set of real world data they used, or, if they averaged more than one set together. So, showing a difference greater than 2% could say one of two things..

 

1) The dev's missed the target, assuming they are using the same data you used.

2) The dev's hit the target, your just not using the same data they used.

 

Even the two above 6% errors? The mistakes in either data would be quite large. Or is this within the errors of 1940s flight testing?

[GOAT-ACEOFACES]

Even the two above 6% errors?

IMHO yes..

 

First and formost, there is no way to validate the in-game test, Second, as noted, we don't know if the real world data used in the 6% error is the real world data that 1C/777 is using as their 'target'

 

The mistakes in either data would be quite large. Or is this within the errors of 1940s flight testing?

Well,

 

keep in mind that up till now, +/-5% errors has been the generally accepted rule-of-thumb acceptable error.. But, it looks like 1C/777 is raising the bar to 2% error.. Which IMHO does start to get into the 'noise' of WWII test flight errors, in English, is about as good as it can get!

[6./ZG26_5tuka]

Here we go again. People cried for test data and now that they got they make up reasons to discredit their value of proving FM errors...gosh. Let the devs do their job, you dont need to evaluate this data (unless the test appeares obviously faulty).

 

Anyway nice test series Celestiale. Keep it up and let us know what devs reply incase you send it to them

Edited March 13, 2015 by Stab/JG26_5tuka

[MK_RED13]

_{... and what about speed with closed rads and duration till engines damage ???}

Edited March 13, 2015 by MK_RED13

[NachtJaeger110]

so I guess there was no response from the devs?

[II./JG77_Manu*]

so I guess there was no response from the devs?

 

I am in the middle of changing my residence right now, so i didn't have the time to compile a proper report so far. 

Sorry, might last two or three further weeks to get sufficient time to do so.

[NachtJaeger110]

I am in the middle of changing my residence right now, so i didn't have the time to compile a proper report so far. 

Sorry, might last two or three further weeks to get sufficient time to do so.

 

Thank you for the info Celestiale, no reason to be sorry! please take as much time for it as you need. this will just make the report better. Keep up the good work

[[BTEAM]_Shifty_]

One response was that devs will look into 109 cooling. Something was fishy about 109 being able to fly full throttle with closed radiators even in ISA.

So expect top speeds to drop slightly for those.

[303_Kwiatek]

Good,  wonder if is any response for Yak-1/Lagg-3 overperforming at high alts?

[II./JG77_Manu*]

Shifty_' timestamp='1427454502' post='246893'] Something was fishy about 109 being able to fly full throttle with closed radiators even in ISA.

There is nothing fishy, they were able to fly with closed rads in real life (at ISA) as well. In fact, all the speed tests have been executed with radiator at 0. They flew a very long test track (from southern Germany to northern border of Italy) with 0 radiator. Climbing with 0 radiator would probably be not such a good idea. 

 

 

 

Good,  wonder if is any response for Yak-1/Lagg-3 overperforming at high alts?

 

Didn't still have the time for a report yet, but the Devs definitely saw this topic. If that's enough, that's another question. Right now i don't care to much about it, to be honest. I am not playing the game right now, the sound bug stops me from it, that's a far bigger issue at the moment, in my opinion.

Edited March 28, 2015 by Celestiale

[bivalov]

again?

 

ok, AGAIN.......

 

 

speed of Yak-1 №1569 with M-105PF (apparently, it is correct for all planes between s69 and 87, and with old cooling system) = 510 kph at sl. serial Yak-1 №2029 with M-105PA showed 487 kph at sl + 20-25 = 507-512. and serial Yak-1 №2898, apparently, with retractable tail wheel, showed 516 kph at sl.

 

 

best speed of Yak-1 s69 in game, with fully closed radiators = 560 kph at sl. with fully opened radiators = 532 kph at sl. max. permissible temperatures = 100 degrees of oil/water, and personally i got 551-552 kph at sl (i.e. 522) with 100/100 degrees.

 

so, 560 - 30 kph - 15-20 kph = 510-515.

 

 

test in mission with +15 degrees - 511 kph at sl with 100/100 degrees.................

 

and quick test with +15, and 910 mm hg + 2700 rpm (M-105PA nominal settings) - 486 kph at sl.

 

 

La-5 type 37 s8 (that plane could be a bit better that very early s4, according to a bit better quality in end'42) with nominal power, and with flap of oil cooler "по потоку", showed maximum ~540 kph at sl - 30 = 510..............

[[BTEAM]_Shifty_]

Indeed, oil at 110 or 100 is considered overheating. Every manual states to keep in-going oil below 80 or 75. Max speed run or not.

[Original_Uwe]

Funny that the only aircraft within the 2% limits is the 190-the aircraft that has received the most scrutiny!

 

Well done OP. Seems there might be something wrong with the games high altitude model?

 

I hope you can conduct the same tests for climb.

[Venturi]

Well done.

[Deadly_Dolphin]

Did the top speed of the F4, not increase to 660ish Kph, when the restrictions on the Db601E was lifted by february '42 ?

Edited April 13, 2015 by Deadly_Dolphin

[303_Kwiatek]

I think rather about 650 kph.  660 kph is for G-2 at 1.3 Ata with tail wheel retractable.

[Original_Uwe]

Celestiale, would you mind sharing the test mission you have built?

Ive been trying to duplicate your results but simply cant.

For instance my test of the 109G2 is:

Stalingrad single mission

No other units

No wind or turbulance

100% fuel

No unlocks (standard configuration)

6000m

AI controls aircraft heading and altitude, player controls throttle and rads.

at 1.28ata (highest I could get indicated on dial) I made 493kph.

using these converters (http://www.csgnetwork.com/tasinfocalc.html for IAS to TAS  and http://www.csgnetwork.com/knotstatuekmcalc.html for knots to kph conversion) I got 677kph TAS which is a bit faster than yours.

[CaK_Rumcajs]

Celestiale, would you mind sharing the test mission you have built?

Ive been trying to duplicate your results but simply cant.

For instance my test of the 109G2 is:

Stalingrad single mission

No other units

No wind or turbulance

100% fuel

No unlocks (standard configuration)

6000m

AI controls aircraft heading and altitude, player controls throttle and rads.

at 1.28ata (highest I could get indicated on dial) I made 493kph.

using these converters (http://www.csgnetwork.com/tasinfocalc.html for IAS to TAS  and http://www.csgnetwork.com/knotstatuekmcalc.html for knots to kph conversion) I got 677kph TAS which is a bit faster than yours.

That calculation of TAS is simplified. Do not rely on it if you want to determine TAS. To avoid any problems with calculations and conversions and the need to feed the formulas with variables people use the map grid. Create a scenario with desired air conditions, set the airplane heading 0 or 90 or 180 or 270 degrees, turn on level autopilot and record time to cross two squares. Two squares = 20km, time comes out of your stopwatch, calculate speed. That way you get ground speed by definition and if wind is set to zero it's equal to TAS.

[Original_Uwe]

Of course, but we don't have a test track as yet, and I have tried using the mission editor to create one but find my skills wanting.

That might be a good project though. What all would need to be included in the c re at ion of a test bed mission by which we can standardize these tests?

[Original_Uwe]

I finished my testing of the German fighters just now.

Methodology was similar to the OP-a 40 kilometer straight track run at highest combat power rating, with a 30km run up prior to the track to get up to speed.

AI maintained level control of the aircraft with one exception (noted below). Player controlled throttle input.

Mission was assembled in FMB and is identical for each of the aircraft, only changing in altitude. Each aircraft was timed in its 40 kilometer run at 200 meters, 3000 meters and 6000 meters, all ASL as far as I know.

Conditions are set to game default, with zero wind and turbulence.

Fighters are fueled 100%, and all operational systems are left in their automatic default positions. No unlocks are used, even down to the game default skins.

Time was taken via a handheld stopwatch (My G-Shock 9400 Rangeman) at the cues given in game (see tracks)

 

AIRCRAFT@Powersetting

 altitude  time  airspeed

 

BF-109F-4@1.3ata

 200m    4:33.88   525 kph

 3000m  4:08.89   578 kph

 6000m  3:37.49   662 kph

 

BF-109G-2@1.3ata

 200m    4:34.21   525 kph

 3000m  4:06.16   584 kph NOTE: Aircraft would not maintain level flight under AI control, player took control of aircraft to best of abilities to keep level.

 6000m  3:42.11   648 kph

 

FW-190A-3@1.32ata 2450rpm

 200m    4:30.11   533 kph

 3000m  4:16.55   561 kph

 6000m  3:50.61   624 kph

 

Sources to measure by:

Unfortunatly OP didnt post sources, but I believe Ive seen them at Kurfurst.net. As this is my source I will post what I can. Of course these are all to be taken with grains of salt and Im  sure  documents countermanding them will be easily and quickly produced as our community is prone to do.

 

BF-109F-4

As usual there are some discrepancies in the reports, but at sea level in game we are dead on. Otherwise it seems to be in between the two sheets, closer at 3000m to the IV/78/42 calculations than the uncorrected test flight, but closer to the flight without compressability correction at 6000m. At 3000m our aircraft is right at the 3% performance margin garaunteed by Messerschmitt compared to IV/78/42, but we are at or above 5% over at 6000m. 

 

http://kurfurst.org/Performance_tests/109F4_Datenblatts/109F4_dblatt_calculated.html

 

BF-109G-1 

Reichlein tests with a G-1 (no use in tests) that is surmised to have a fixed tail wheel unlike our ride in game. Regardless the numbers are nearly identical between the test aircraft and in game.

 

http://kurfurst.org/Performance_tests/109G_Rechlinkennblatt/rechlin_G1_blatt.html

 

FW-190A-3

Its not precise, but it sure looks like were in the ball park.

Quick conversions make our test aircraft 331 at S.L., 348 at 9842ft, and 387 at 19685ft.

I think the FW is the closest, its really shocking.

 

Comments are welcome, I will post the missions and tracks in one second...

[Original_Uwe]

Here are the zipped files containing my missions for testing. Simply drop entire webmissions folder into your data/missions.

Tracks just drop the whole file into your root/data folder.

http://www.mediafire.com/download/hkg5mubth0i6f60/webmissions.7z

http://www.mediafire.com/download/g8awfe9wwkgoln8/Tracks.7z

[Sgt_Joch]

I have been reviewing Gordon, Khazanov's book and the figures for the Yak-1 in game seem pretty much right on.

 

The OP recorded max speeds of 520 kmh at sea level and 588 kmh at 3.685 km.

 

Now it is true that the june 42 test of Yak no. 1569 with a M-105PF engine (weight: 2917kg) did record a max speed of 510 kmh at sea level and 571 kmh at 3.6 km, but that test was not run at full power. No. 1569 had the new enlarged oil cooler and there was worry about overheating so the rpms were limited to 2550 at "low altitude" rather than the normal 2700 RPM limit. # 1569 recorded a climb time of 6.4 minutes to 5 km even though an earlier test of a M-105PA engined Yak-1 (weight: 2883kg) had recorded a faster time of 5.9 minutes.

 

There were two later tests of Yak-1s with M-105PF engines, one in late 42 (weight: 2900kg) which recorded a max speed of 523 kmh at sea level, 590 kmh at 3.8km and a climb time of 5.6 minutes to 5 km; and one in early 43 (weight: 2884kg) which recorded a max speed of 531 kmh at sea level, 592 kmh at 4.1 km and a climb time of 5.4 minutes to 5 km.

 

The later tests are probably more representative of what a M-105PF Yak-1 could achieve at 100% power.

 

[JtD]

Actually power output and prop efficiency on the Klimov as in the Yak are better at 2550 rpm then at 2700 rpm, unless above full throttle altitude, where the supercharger performance is better with higher rpm. Therefore, in real life, higher speeds were obtained with 2550 than with 2700 rpm at low altitudes. Also, the June 1942 test is with a series 69 aircraft which is what we've got in game.

[Sgt_Joch]

Interesting, do you have a source for that?

 

This Yak-1 pilot manual states that max. speed at all altitudes is achieved at 2650-2700 RPM:

 

74. To achieve the maximum speed (when you met with the enemy, to make pursuit,for air combat or out of it) should do following steps:

 

a) close the mixture;

 

b) at all altitudes set the revs at 2650-2700 rpm; if the engine expels oil, reduce rpm to 2,550 rpm;

 

c) close the water and oil radiators, fixing in position "with the flow";

 

d) flying above 3,000 m with M-105Pa engine and above 2000m with M-105PF engine, put the second stage supercharger.

 

 

http://www.docdroid.net/gmjm/yak-manual.pdf.html

 

Gordon/Khazanov's book on Soviet aircraft is not the most clearly written, but its does state that in the first version of the M-105PF, i.e. the field mod, RPM had to be reduced from 2700 down to 2400-2500 to control overheating which negated all the performance advantages of the increased boost.

 

Talking about the M-105PF in Yak #1569 (june 42 test), it had improved cooling, but temperature control was still critical so RPM was reduced to 2550 rpm for "optimum performance", but the way the text is written strongly implies that this was done to control the temperature, not to achieve best performance.

[MK_RED13]

Is it also applied in this game ??? 

 

....from the manual mentioned above

 

IV. FLYING Turn

 

84. Before starting the turn, stabilize the aircraft in level flight at a speed of 320-340 km / h . This same speed must be maintained when making a turn with twist 65-70° . Making a left turn the plane tends to lower the nose, while with the right turn tends to lift. The plane is stable making turns and passes easily from one side to another. If you are in a turn and you pull the stick in excess, and the speed of 270-280km / h is reached, the aircraft becomes unstable; if you keep pulling the stick, the aircraft wing drops. When you pull the stick in excess and lose speed (the characteristic symptom is when the plane starts to shake) should push the stick forward and put the aircraft in level flight. If the plane goes into a spin, is carried out according to standard procedure.

 

Combat turn

 

85. To perform the combat turn must be given full throttle, increase the speed to maximum, then create a roll up 50° simultaneously lifting the tail. The exit of the combat turn is effected at 230-240 km / h at full engine power.

 

During combat turn, the plane wins till 1000m altitude.

 

The forces applied to the rudders are normal