Jump to content

Holtzauge

Members
  • Content Count

    1095
  • Joined

  • Last visited

Everything posted by Holtzauge

  1. Interesting info about flap usage @[TLC]MasterPooner. However, the use of flaps in sims bugs me: I know this has been discussed a lot before but I think there is a reason that flaps are so seldom mentioned in combat reports: IRL they are not as good as they have been modeled in many games and while many swear they would never use them in-game so it's a non-issue I have seen the same persons post tracks in which they used them anyway so I guess they are not as "bad" in-game as some would have you believe........
  2. At 25,000 ft the Me-109K4 is far better than the P-51 when it comes to stationary turn rate: If fact I'm surprised no one has commented on that yet since it's pretty obvious in one of the old charts I posted years ago and that @[TLC]MasterPooner posted above. The curves for 8 km there is pretty much the same altitude (25,000ft = 7.6 Km) and as can be seen the K4 is way better at this altitude. Don't know how that is reflected in the game though? Maybe time for some high altitude testing?
  3. I was not planning to participate any more in this thread but since there seems to be so much confusion about this measurement: The figure below shows that there is nothing wrong with the in-game modeling because of the same SL turn times at 280 Km/h IAS because that was just what was expected: Follow the 280 Km/h line up to where it intersects the turn rate curves in the figure below and lo and behold: The Me-109K4 has the SAME turn rate at this speed irrespective if the power setting is 1.8 or 1.98 ata. All propeller planes depending on power output have a "sweet spot" turn speed at which they achieve the best turn rate. As you can see it's slightly different depending on power setting and that is why in-game just as IRL you have to get to know your plane so you know at what speed this is depending on your power setting. I tried to explain this in this post and hopefully this added example above will help illustrate how this works.
  4. Honestly, I think von Helsing should have driven a stake through this thread long ago but just let me add this: You can't use a jet chart to argue propeller plane turn characteristics: While not totally true as a first order approximation this holds true: One is a constant power producer the other a constant thrust producer which means an apples and pears comparison. Finally: As far as I can see most here seem to agree that there is nothing wrong in IL2 at the moment: The P-51D's best sustained turn rate is a tad slower with the same fuel load as the Me-109K4 just as expected. Now if someone thinks what we have is wrong the onus is on THEM to prove their case not the other way around. As far as I can tell some here are trying to make it sound like we have to prove that part span slats (Yes, apparently the word PART SPAN needs to be highlighted in red) are not some kind of Wunderwaffe and that the P-51 doesn't have a terribly low Clmax. That is not how it works and I'm sure that by now the developers are either running out of popcorn, have dozed off in boredom, or simply switched channel.
  5. My understanding of how this works: For any plane keeping all other things the same the speed for best turn rate will increase if you add more power moving the right hand (power limited) curve of the doghouse chart up to the right. Conversely, if you keep the power the same and add wing area or increase Clmax this will move the left hand part of the doghouse chart curve up to the left, decreasing the speed for best turn rate. In my simulations the speed for best turn rate for the Me-109K4 goes up by about 13 km/h when going from 1.8 ata to 1.98 ata.
  6. Never put someone on ignore before but since you via PM added clear malicious intent to discredit on top of the slander accusation you've earned it.
  7. No, what muddies the waters is unsupported speculation and generalizations about how laminar profiles work and Skip Whatsisname quotes and people not reading reports: I can quote Erwin Leykauf for you saying he can out turn Spitfires in his 109 so now the 109 out turns Spitfires as well? I already explained why I use the Clmax data I do for the Mustang: I use two NACA reports to lean on: NACA Report 829 and NACA report TN 1044 that BOTH give a Clmax in the order of 1.4 for the Mustang in service condition which is also what I use. Excuse me a while. I have to go watch the BRIO flic again.
  8. My apologies @Aurora_Stealth: Parallel to this Me-109K4 issue I’m modeling the Pfalz D.VIII and unfortunately I did the latest run of the Me-109K4 in the wrong C++ module which has different propeller characteristics modeled (the K4 ported but not tuned in that module) so in fact my estimates in the correct simulator model are lower than the ones you measured: 19.8 s with full fuel load, 1.8 ata at 1 Km and 18.5 s at SL. In fact I already published the 19.8 s turn time for the Me-109K4 with full fuel load in this post: 360/18.2=19.8 s so I should have reacted myself when I posted the longer turn time estimates. To excuse myself, the C++ compilator I use has a tab GUI so all it required was to choose the wrong tab. That being said, I still think this is a damn good simulator: However, sometimes things need to be tuned that’s all. However, the jury still seems to be out on that one because as far as I can see you, @HomicideHank and @[TLC]MasterPooner seem to be getting somewhat different results measuring in-game so maybe more testing is needed. However, as you pointed out above, theirs (and my simulations) are done with autumn conditions and yours were I believe done on a summer map and that does impact results.
  9. Thanks for doing the additional low speed turn rate testing @Aurora_Stealth. That the Me-109K4 was under performing at very low speeds was just a subjective feeling on my part based on how I perceive it to behave in-game but comparing your measured numbers to what I was expecting based on my simulations it does not look like there actually is anything off in-game at speeds ”behind the curve” because I don’t get any shorter turn times at speeds below the speed for best turn rate for the K4. TBH, I’m really impressed how well the planes in BOX are modeled given that everything has to be done in real time while in my simulations I have no such constraints and while turn rate estimates just take a few seconds for me to run, time to climb estimates take minutes. So we can be very happy with what we have and that the developers are doing such a great job continuously developing this fine simulator.
  10. First of all thanks for asking this question at all. Very nice that you take the time to collect an analyze what your customers think about spotting and are looking into the subject to improve this fantastic simulator even more. I don't fly that much without icons or tweak the system for better spotting but even with icons on I think it is very difficult to pick out the outline of an aircraft against the ground, which you need to set up the correct deflection in a head on pass for example. Probably a lot of this has to do with depth perception not being present on my 2D screen but compared to how I perceive aircraft against a background in real life I can think of two improvements that could be done: One is to compensate for the lack of depth perception by introducing a more hazy coloring of the background compared to the plane which would then appear to have more saturated coloring and consequently be easier to pick up against the background. In addition, I don't really think this would be unrealistic at all since in many times (other than after a cold front passage) there can be quite a lot of haze giving the background a bluish tinge. The other idea I think was mentioned before already: Introduce more pronounced "glittering" to emulate canopy reflections and reflections from various surfaces of the plane. To make this fair, it would of course have to be some random system so that everyone who is flying rolls the dice at the same time intervals and consequently run the same risk of generating a reflection.
  11. That car would complete that circle much faster if BRIO put slats on it.
  12. Well you see this is exactly what gets you into problems: Either you simply did not understands what I wrote when I said e was a function of Cl or you are willfully misrepresenting the situation when you say we agree about e. About proving you wrong I don't need to do that since you are doing an admirable job of that yourself. I just point out the inconsistencies that's all. Being wrong is bad enough but assuming a superior attitude and asking people like you did earlier if they have formal education in aerodynamics when they air different opinions than yourself just digs you deeper down that credibility hole. Putting you on ignore would probably be the best course of action but since you insist on framing your hypothesis and conjectures as facts this could do untold damage. That being said don't expect any more replies to me. I will just intervene if it's absolutely necessary.
  13. You are quite right: It may very be well be a Cdi related problem and it really creates a lot of friction when people state things as facts when in reality it is simply their opinion and conviction and not a fact at all. In addition, it does not help when they are plain wrong: While it is true that Cdi is coupled to Cl, there is also in the denominator a term ”e” sometimes referred to as the Oswald factor. This factor denotes how well aligned the lift distribution over the wing is to an elliptical pattern which is defined as e=1. However, wings are usually not elliptical in shape and even those that are usually has washout built in (e.g. like the Spitfire). So ”e” is usually also in IRL less than 1 and is commonly set to a fixed value, say 0.8 which gives good approximations in most cases. However, what happens at high angles of attack is that the flow partly breaks down which then means that the lift distribution over the wing departs even further from elliptical meaning that the factor ”e” goes down even further. This is why in my simulation e is a function of Cl meaning you have to have an idea of how e is connected to Cl. So in many cases my low Cl e is around 0.8 but then at higher Cl e goes down substantially. This means that if I model this optimistically or pessimistically I can totally different results meaning there is every possibility that a model of the induced drag can be right or wrong. So if someone says it can’t be a Cdi issue as the Cdi is directly coupled to Cl they are plain wrong even though the statement is presented as a fact. Now about the thrust being off I think when it comes to the best turn rate issue evidence as I see it points to thrust being correctly modeled at those speeds after all since if it was related to the thrust modeling, then the climb rates would be off as well which as far as I know they are not. Rationale: If the thrust model for turning at 300 Km/h IAS was off then so would the climb rates at 300 Km/h IAS. In addition, as I see it all in-game measurements posted so far for both the P-51D and Me-109K4 align well with my simulations meaning I don’t see a problem but that of course does not rule out that both my simulations and the modeling in IL2 is off but so far I have seen nothing that convinces me of this. However, as I said before, there may be an issue with the thrust or drag modeling below speeds for best turn and climb rate but in that case I think all bets are off if it’s a thrust or drag issue because you can’t use the climb rates anymore to check the thrust modeling. One way to test this though would be to test acceleration from speeds below best climb rate (basically prop hanging) and then see how long time it takes to reach say 400 km/h IAS. I can do acceleration estimates in my simulator but I’m not going to post too much here since I nurture a vain hope to publish a book on aircraft performance some day and if I reveal everything now no one will bother to get it.
  14. First part essay 500 words. Second part multiple choices of the type "What is the maximum lift coefficient of the Me-109 K4?" (1=below 1.1, 2=About 1.4, 3=Well in excess of 5) which is why I suspect some will fail miserably.
  15. I just realized that this whole discussion about Clmax and turn rates is actually a bit of a groundhog day since it was all covered exhaustively already back in 2016 here. In addition, I just spoke to the Headmaster and everyone is now tasked with reading through all 11 pages, exam covering Clmax effects on turn rates will be tomorrow, no exceptions. If you fail you will NOT be allowed any sim time next week. Understood?
  16. Thanks for reminding me about TN1044: Had it on my drive and looking at the figure you posted that ties in nicely with the NACA report 829 figure of Clmax=1.4 for low Mach in-service and 1.44 for pristine polished then. In addition, I did a calculation to what is being used in-game and I came up with Clmax=1.34 for the Mustang and 1.39 for the K4 assuming that the lower quoted stall speed figure is for 10% fuel no ammo which I believe @unreasonable has concluded from his studies of the correlation between the stall speed numbers and weights in the in-game tech specs. So it seems that my simulations and BOX power off Clmax assumptions are pretty much on the same page here and in addition, I believe those are also quite close to the numbers used in DCS as well so hopefully this can put a cap on the Clmax discussion at least. When it comes to Clmax at power on conditions I have not dug up so much information about that since the way I set up the simulator I use Clmax power off as input. I did post results for the Spitfire in power on condition though earlier in this thread. I think I said Clmax=1.8 there based on looking at the graphs but it turns out this was also listed in the text and a more accurate number is 1.89.
  17. @Aurora_Stealth and @[TLC]MasterPooner: Thanks for doing the tests and as far as I can tell your results for the Me-109K4 don't really contradict each other or my simulations: Beginning with the tests @Aurora_Stealth did at as I understand it 1 Km altitude summer conditions with 100% fuel 1.8 ata: At 300 Km/h IAS I get 22.1 s turn time in the simulator while the measured result in-game is 21.7 s which is a good fit. Also at 320 Km/h the results tab well: Simulator 21.0 s and in-game 21.8 s. However, the simulations are at 15 C and the in-game results from summer conditions which I believe is 20 C which if anything, suggests that the in-game results are slightly optimistic. However, this was also what I said earlier on: My impression is not that the best sustained turn times are off but that it may be so that the induced drag/propeller efficiency modeled at low speeds is a bit pessimistic which was why I suggested testing at speeds lower than 300 Km/h IAS as well. The 19.5 s turn time result posted by @[TLC]MasterPooner for the the Me-109K4 with 100% fuel 1.8 ata were as I understand it for SL? If so there is good agreement between simulations and in-game results here as well: In the simulator I get 19.8 s while in-game it is 19.5 s. In this case results are directly comparable since they were done on the 15 C Autumn map.
  18. The C++ program I use is not CFD but a parameter based simulation and an adaptation of a program I originally wrote for my master’s thesis to study the effect of external stores on jet aircraft but about 15 years ago I adapted it to also handle prop aircraft. As an example, Clmax is a function of Mach as is Cdo so I can capture subsonic drag rise and compressibility effects. Propeller efficiency is a function of blade loading, advance ratio and blade tip Mach number. Oswald factor is a function of Cl etc. For the P-51 I currently assume 1.4 as low Mach Clmax based on NACA report 829 which gives Clmax 1.44 for the Mustang (Airplane 1 in the report) in pristine sealed conditions and 1.4 as Clmax with in-service roughness. Then as I mentioned above, Clmax goes down with Mach in the code and for example, at M=0.5 I’m down to a bit over 1 in Clmax. When it comes to exactly how the simulator works I’m not going to divulge that here since I’ve spent years collecting the data I use and I’m working on a book project so I’m not ready to undercut my market just yet . On a more serious note: I have been posting results for more than 10 years in various forums and you have to judge the results I produce in exactly the same way you do flight sims: No one posts trade secrets and you have to look at the results for tested/simulated scenarios, compare that to historic data or what you think the results should be, and then make up your own mind if you think the simulation is any good. That being said I for sure want to discuss the assumptions going in (like Clmax) so if you have an interest in this and have collected data as well that sounds good. OK, for 100% fuel 400 Km/h IAS at 1 Km 1.8 ata I get 22.5 s but as I said earlier, I think the main issue could be the low speed induced drag/propeller efficiency combination starting from say 300 Km/h IAS and below. At that weight I get 21.3 s at SL and 26.5 s at 3 Km with Steig&Kampfleistung so I’m still thinking that there may be a typo in the IL2 spec sheet. OTOH this should be easy to test in game: In the current release, if you test under these conditions do you come close to the numbers I posted or do you come closer to 24.2 and 32.3 s ?
  19. Well if the 50% fuel in the chart above means 50% for both aircraft that means (in-game) 510 l for the P-51 and only 200 l for the Me-109K4 and in that case the difference should be bigger: Looks like they are about even in-game under those conditions while my simulation indicates about a 1 s advantage for the Me-109K4 with 290 l (i.e. 73% fuel) and the P-51D also with 290 l (i.e. 28% fuel). Don't have the time to simulate this right now but obviously I will get an advantage bigger than 1 s for the K4 in this case since the P-51 is hampered by a bigger fuel load. OTOH, the simulation I did earlier was assuming 1.8 ata for the K4 and 67" for the P-51 so it's not a 1:1 mapping then between in-game test and the simulations. Sounds good! If you do could you do it with 290 liters of fuel in both aircraft and 1.8 ata for the K4 and 67" boost for the P-51? Would simplify the comparison to the simulations. Edit: I saw the IL2 specs you just added for the K4 and that 24 s at SL number has got to be a typo: I remembered there was something like that before as well for another aircraft in the aircraft descriptions and they updated it. However 24 s is the right number for the K4 at 3 Km altitude so that could have been it.
  20. Well if the Me-109K4 with 290 l fuel 1.8 ata is not out-turning the P-51D with 290 l fuel 67" boost in-game by about a 1 s then of course I'm on board: I just wrote that. However, as far as I understand that is being contested from both camps and IMHO a fair distribution of work could be that I do the simulations and maybe someone else who has a horse in this race could chip in and do some in-game flight trials? Honestly, there are a lot of opinions and statements being made in this thread but I have yet to see someone posting in-game turn rate numbers and tracks. However, I see another issue with the current Me-109 FM: My impression is that it suffers too much drag when you go slow or alternatively when you pull high angles of attack and IMHO it seems like the Me-109 suffers more from this than it should but I have yet to figure out a good way to show that. To me it seems that if you go to slow in the Me-109 in-game and get "behind the curve" as in a low speed high angle of attack state, it takes more time and effort to get back energy than it should. I have discussed this with @ZachariasX (has from the Spitfire TR-9) who has the same opinion. Edit: Just thought of a way to test this: Simply fly sustained turns at 1 Km with a full fuel load in the Me-109K4 starting at speed for best turn rate starting at say 300 Km/h IAS and then reduce in increments of 20 Km/h until you can't do a circle at all. Post the results here and I will do a chart comparing that to the C++ simulations. In this way we can see how induced drag has been modeled in-game when you get "behind the curve". It will of course take some time doing the in-game tests because I want to see at least 3-5 consistent turns holding altitude and speed for each turn speed to get some level of statistical sampling. If this is to much work then by all means, continue the discussion without data. But if some of you could chip in and do the in-game flight tests then I will do the simulations and do the comparison.
  21. Frankly @Panthera, how do you come up with Clmax=2 power on for Me-109? The way you framed it now looks more like a value you would like to see and something simply pulled out of a hat because I don’t see anything supporting that number at all. In addition, if I were to use your higher estimates on Clmax for the Me-109K4, I would get better than historical numbers not only for the K4 but for all Me-109 that I have modeled so no sale. Also cherry picking the highest number available for the Me-109 from the Chalais Meudon wind tunnel tests does not exactly strengthen your case: The results were 1.42, 1.45, 1.44, 1.42, 1.44, 1.48, 1.43 and 1.45. Taking the average of these we get instead 1.43 not 1.48 so picking the highest available number puts a dent in your credibility TBH. Thinking back I remember I have had this exact discussion back in 2014 in the DCS forum: A guy using the handle Hummingbird made a remarkably similar case to yours in this DCS thread. One nice aspect of that discussion in DCS though, was that the DCS FM developer Yo-Yo participated in the debate and I would say that Yo-Yo and I landed very much on the same page regarding Hummingbird’s and Crumpp’s wish to jack up the Clmax for the Fw-190D9 and Me-109K4, as in no sale. Attached below is a chart from that discussion which was done more to compare the Fw-190D9 and P-51D carrying the same fuel load so the Me-109K4 looks a bit better in that since it’s carrying a smaller fuel load. So in order to compare apples with apples, I did another C++ simulation just now for SL conditions assuming the P-51 and Me-109K4 carry the same fuel load 290 l and for that I get the exact numbers [TLC]MasterPooner mentioned above: For the 109 K4 18.48 s at 1.8 ata and for the P-51D 19.50 s at 67" boost. And that about concludes it for me: This dead horse has suffered enough and while it is true that the P-51 is a lot worse than a Me-109 in a slow speed knife fight when hampered by a full or even half fuel load, my conclusion is that it is only a shade worse when compared on similar terms, i.e. the same fuel load.
  22. The links you provide are for full span slats @Panthera and nobody as far as I know has contested that slats allow you to go to higher in alfa and get higher lift in that case. However, the Me-109 has only part span slats that if you look closer coincide with the location of the ailerons and the problem I’m having is that you are making sweeping statements about the benefits of full span slats for Clmax and then applying that to the Me-109 like that was the intention of the designers when they put the slats on the Me-109 evidence for which is still missing as far as I can see. In addition, with the qualifier you now added ”this has always been the point of slats, to increase the CLmax of the covered area” then we are back to where we started and which no one is contesting. Also, If you look real close the drag is actually a little bit higher with slats out until the profile stalls but the effect is very small. But yes, that effect is minute and for sure I agree that the drag of slats is insignificant in this context and that was actually what I said earlier: the slat drag is irrelevant compared to the induced drag in a tight turn. Returning to the Clmax at power on conditions: The Spitfire also gains a lot in Clmax in power on conditions in straight flight. There is actually a good report on Spitfire turn capability that quantifies this: Notes on the turning performance of the Spitfire as affected by flaps, RM 2349 by Morgan and Morris from 1941. In figure 10 of that report you can see that the Spitfire gets Clmax 1.8 in 1g full power conditions with no flap but in accelerated conditions (e.g. 3g) like in a turn it drops down almost to the same as in power off conditions. The main problem as I see it with all this is that you are claiming that the slats mean that the Me-109 should be a whole lot better than the Spitfire and I think we could go on arguing about this for some time so why not instead try to put a number on how much better you think the Me-109 should be in terms of Clmax? The Spitfire has a power off Clmax of 1.36 and according to RM 2349 a Clmax of around 1.8 power on with no flaps at 1g. The Me-109 has a power off Clmax of slightly over 1.4 and given that the Me-109 has slats on the outboard portions of the wing how much better should it in your opinion be than the Spitfire’s Clmax in power on conditions? In BoX, what should be modeled as Clmax power on for the Me-109 at 1g, 2g, 3g and 4 g?
  23. Can you explain why you think they don't add drag due to "the circulation effect"? I would be surprised if they did not come with any drag penalty so I would appreciate if you could reference that statement. In addition, most aircraft have washout to get decent stall charcateristics and the Me-109 does not, so slats are very much needed on the latter. I have never seen any reference to that the slats "are merely there to increase the amount of lift available for the same given wing area"? Source?
  24. According to theory Handley-Page type slats like on the 109 do not come out with a bang but creep out gradually as the lift coefficient (Cl) goes up. There have been tests done on the Me-109 slats as well and IIRC then they start to open up at Cl=0.9 and are fully out at around 1.3. You can find this info in Hoerner’s book on lift. In addition to what Mark Hanna says I can add some new info I just got from Mikael Carlson who flies Hangar10’s Me-106G-6 and he says he does not even notice when the slats come out: No snatching, no disturbance in roll, no nothing which ties in nicely with theory because that is just how you expect them to work on paper. In addition I have a German WW2 report stashed away somewhere about Me-109 slat kinematics that says the same. Given the weight of evidence it’s simply amazing that people still insist that as a rule the Me-109 slats came out with a bang upsetting roll angles just because Eric Brown says so based on flying one captured specimen. I have no doubt they did that on the one example he flew but what is more likely, the aircraft he flew was not properly rigged or maintained or this is right and Mark Hanna, Mikael Carlson and theory is wrong? Second, the drag added by slats in a tight turn is insignificant: A ballpark figure for the wings contribution to zero lift drag (Cdo) is about 35% and only a limited portion of the wing has slats so even if that goes up a significant amount it only has a limited impact on Cdo. Now at best L/D Cdo is roughly the same as the induced drag (Cdi) but at higher lift coefficients (Cl) when the slats come out on a Me-109 at 0.9 and above the Cdi is totally dominant and Cdi is affected by span loading and not by slats so if the slats add some zero lift drag that does not matter because in a turn when the slats are out induced drag is everything.
  25. Yes, I think the info from Mikael about the light controls on the Dr.1 was especially interesting and that the ailerons on the Pfalz D.VIII were so heavy. As you say, while Javier Arango does not get into specifics about actual control forces he is quite critical of the aileron arrangement on the Camel and from looking at the pictures a ballpark guess would be that the Camel is probably then more close to the D.VIII than the Dr.1 when it comes to aileron forces. The Dr.1: The Fw-190 of WW1! Regarding the Pfalz, Mikael said that he is proceeding testing it cautiously which seems prudent. Especially given he said he want to approach stall carefully since early impressions seem to be it does not have the same benign stall as the Fokkers. But even so I think you are right about the future: The D.VIII is probably more of an energy than angles aircraft and it will be very interesting to see what happens when he expands the envelope but that thin wing profile with the sharp leading edge should theoretically at least put a cap on what's possible in terms of hard manouvering. About the Pfalz cockpit video: On the left is IIRC the oil pressure gauge just like in the original: Mikael made it on a lathe himself based on measurements taken of the real thing in a museum and convinced some company specializing in glass to blow that part for him. Talk about attention to detail! Mikael also did an early turn trial during the eight test flight and I'll PM you about that as well.
×
×
  • Create New...