Jump to content
JtD

Another look at turn times

Recommended Posts

I would not take tacview g loads as accurate without confirmation. I will try to do it (I know that speed is not captured from game as IAS).

Share this post


Link to post
Share on other sites

AnPetrovich has shown/proved on Russian forum that Tacview is not reliable with G in BoS

 

Cheers, Dakpilot

  • Upvote 1

Share this post


Link to post
Share on other sites

Can you link please? I have neither. The 109 G manuals I have show no stall speeds at all.      Alternately summarize in a table - throwing the odd number into a post does not help to sort out what it is you are highlighting as evidence of a problem at all. 

 

The figures in my tables are all the clean configuration speeds - ie stall at Vmin with flaps and gear up.   Using landing speeds is going to cause all sorts of problems - stalling speed with F+G up is completely different to with F+G down.  So CL with flaps and gear down is radically different - much higher - to that in clean configuration, which is the whole point of flaps, and not going to help at all in understanding issues in normal flight CLmax, if indeed there are any., since we cannot compare the results with the data on the airfoil used in the wing.

 

I brought up the stall speeds in landing configuration to show that there was an issue there as the aircraft certainly shouldn't stall above or even at the recommended landing speed, and thus there might very well be a similar issue with the clean configuration stall speeds as well, the G2 & G4 figures indicating this.

 

If you look in the F & E manuals you see landing speeds in IAS & TAS listed. I don't have any G manual so I can't tell you wether it's listed there.

 

From the Friedrich manual:

 

post-12047-0-51464500-1397408609.gif

 

 

Landing speed is listed as 140 km/h IAS and 150 km/h TAS, which will be atleast 10 km/h above stalling speed. 

 

For comparison's sake the indicated landing speed for the noticably heavier K-4 is listed as 150 km/h, whilst the indicated landing speed of the Fw190A5 & P-51D are 165 km/h and 172 km/h respectively.

Edited by Panthera

Share this post


Link to post
Share on other sites

So i have a question to the sheet you just posted.

 

It says that the values are regarding "Start and Notleistung" (Take off and Emergency Power) @ 1.42 ata. Further down it says "Start and Notleistung" is limited to 3 mins. Anyone ever notice that? Currently we get 1 minute...

That is for the Bf109 F1 and 2. And that thing had the engine of the Bf109 E (DB601) which had a limit of 3 minutes. We see this ingame as well.

Share this post


Link to post
Share on other sites

I am aware that it is for the F2, right now the lint on1.42 ata is 1 minute for the F2

Yeah. I just looked that up. It is stated in the manua for some odd reason. ingame however the engine is running longer. Measured 2 min 20 sec in only one test. So the engine is running longer as the stated 1min. testing again to confirm.

 

Tested.

 

2,5 min

3 min

2.5 min

 

Why the manual states one and why the engine runs not full 3 mins has to be answered.

Edited by DerSheriff

Share this post


Link to post
Share on other sites

I got it. The "Start and Notleistung" in the sheet is refering to 2600 RPM. But full power with automated pitch is 2800 RPM. So to get your rated 3mins you have to fly with manual pitch and 2600RPM like its stated in the sheet.

  • Upvote 1

Share this post


Link to post
Share on other sites

Currently our F-2 has 1.35 ata at 2600 RPM as Emergency Power for 3 mins, and then Boosted Mode as 1.42 ata at 2800 RPM for 1 min.

Edited by -=PHX=-SuperEtendard

Share this post


Link to post
Share on other sites

depends what the definition of start and notleistung IRL and Game is. That has to be clearified and more sources would be nice.

Share this post


Link to post
Share on other sites

Sorry I just saw that it had been discussed before... moderators please move or delete!

where? I like to read that

Share this post


Link to post
Share on other sites

Thinking some more about the Me-109 turn rate problem, I don’t think the in-game propeller efficiency is the culprit: The reason is that the best turn rate and climb rate occur at pretty much the same speed for many aircraft and the Me-109G2 is no exception. So this means if the Me-109G2 propeller is 20% wrong due to it being modelled based on some other than VDM data, then this would hit climb performance as well. However, if anything the Me-109G2 climbs better than it should at 21 m/s at SL which is means that the problem is not the propeller efficiency but the L/D ratio at high Cl, or more specifically the Cd at close to Clmax.

 

Rationale: The propeller efficiency e can be expressed as e=(T*v)/P where  T is thrust, v TAS speed and P the power. So if we fly at the same speed v with the same power P then in order for e to change this means that T will changes as well. Now since the Me-109G2 climbs extremely well at 21 m/s in-game (Better than it should IMHO, but that is covered in another thread), this means that the thrust in-game is on a historical or even better level and since the speed at best turn rate and climb rate is about the same (circa 280 Km/h IAS) this means that there is no problem with the propeller efficiency at speed for best turn rate.

 

So I don’t think VDM propeller data is needed to fix this: The problem is probably instead the drag at high lift coefficients. This would also explain why the Me-109G in-game has such poor performance once you slow down below speed for best turn rate.

  • Upvote 1

Share this post


Link to post
Share on other sites

Thinking some more about the Me-109 turn rate problem, I don’t think the in-game propeller efficiency is the culprit: The reason is that the best turn rate and climb rate occur at pretty much the same speed for many aircraft and the Me-109G2 is no exception. So this means if the Me-109G2 propeller is 20% wrong due to it being modelled based on some other than VDM data, then this would hit climb performance as well. However, if anything the Me-109G2 climbs better than it should at 21 m/s at SL which is means that the problem is not the propeller efficiency but the L/D ratio at high Cl, or more specifically the Cd at close to Clmax.

 

Rationale: The propeller efficiency e can be expressed as e=(T*v)/P where  T is thrust, v TAS speed and P the power. So if we fly at the same speed v with the same power P then in order for e to change this means that T will changes as well. Now since the Me-109G2 climbs extremely well at 21 m/s in-game (Better than it should IMHO, but that is covered in another thread), this means that the thrust in-game is on a historical or even better level and since the speed at best turn rate and climb rate is about the same (circa 280 Km/h IAS) this means that there is no problem with the propeller efficiency at speed for best turn rate.

 

So I don’t think VDM propeller data is needed to fix this: The problem is probably instead the drag at high lift coefficients. This would also explain why the Me-109G in-game has such poor performance once you slow down below speed for best turn rate.

 

That is indeed a valid theory Holtzauge.

 

Btw, you're climb rate of 18.7 m/s for the G2 at 1.3 ata fits pretty well with the tested climb rate of the G2/R2 at SL which was 18.2 m/s @ 3,130 kg :)

 

Erla tested a 3,070 kg G1 to 19.2 m/s @ 1.3 ata at SL, and 20 m/s at 2 km. 

 

So a climb rate of around 19-20 m/s is probably the most realistic for the G2 at 0-2 km, and definitely not 21-24 m/s  :)

Edited by Panthera

Share this post


Link to post
Share on other sites

I brought up the stall speeds in landing configuration to show that there was an issue there as the aircraft certainly shouldn't stall above or even at the recommended landing speed, and thus there might very well be a similar issue with the clean configuration stall speeds as well, the G2 & G4 figures indicating this.

 

If you look in the F & E manuals you see landing speeds in IAS & TAS listed. I don't have any G manual so I can't tell you wether it's listed there.

 

From the Friedrich manual:

 

 

 

 

Landing speed is listed as 140 km/h IAS and 150 km/h TAS, which will be atleast 10 km/h above stalling speed. 

 

For comparison's sake the indicated landing speed for the noticably heavier K-4 is listed as 150 km/h, whilst the indicated landing speed of the Fw190A5 & P-51D are 165 km/h and 172 km/h respectively.

 

Thanks for posting that  - although I have difficulty reading it and cannot find a better version online. Gothic in low res - ugh.

 

The overall problem I have is that deducing CLmax from stall speeds is incredibly sensitive to the number for speed that is used, since v in the lift formula is squared.

 

When you are using a landing speed as a base you are making two adjustments. Firstly from IAS to TAS, then from landing speed to stalling speed.

 

Both of these are a problem: in the IAS case since there is the specific high AoA error that is not accounted for in standard PEC tables that never show speeds near the stall. In the F-2 table you show, I would be interested to know how the difference was estimated (if that is indeed what it is). In this case it is very close to using cosAoA as an adjustment factor, which seems reasonable as a guide, but the RAE and US tests suggest under measures the error at stall speeds. ie stall speeds are a fair bit higher than that adjustment would suggest.

 

In the second case - landing vs stall speed - I can find on line quotes from experienced pilots of tail draggers saying that landing speed should be almost identical to stall speed - effectively you stall the aircraft when the wheels are a few inches above the ground so that lift is gone as you make contact. So a 10 kph adjustment in this case could be far too much - and given that we are talking about 10kph adjustment to a 150kph speed, this makes a huge difference. 

 

So, 140 is 93% of 150, but 140^2 is 87% of 150^2.  87% of 1.40 = 1.22   

 

In a nutshell, the sensitivity of the calculations to small variations of Vmin makes this kind of back of the envelope estimation from numbers not explicitly measured for the purpose extremely unreliable.

 

I do agree that Holtzauge's theory could be a valid explanation in general for the turn time issues.  In the case of the odd G-4 number for min stall speed at min weight, I have not actually tested that myself: just taking the tech specs number. It is possible that there is a simple data entry error either in the original table of results or transcribing into the tech specs page. (Easy enough to do after a long shift staring at columns of numbers, as anyone who has had the misfortune to do this kind of work will know :()  

Share this post


Link to post
Share on other sites

Well if it’s a drag or lift problem or a combination I guess remains to be seen but to me it looks like it should be possible to fix the Me-109 turn rate issue without VDM propeller data since the problem seems to lie elsewhere. In addition I would not mind if they reduced the Me-109G climb rates as well since they seem too optimistic as well as Panthera pointed out above.

Share this post


Link to post
Share on other sites

Just had a quick question. So basically you are saying L/D is too high at the moment. Right?

Woldn´t it make sense for the devs to ask about prop efficiency then though?

If we assume they lowered L/D and would meet both climb and turn rate (historically), then as a result the max speed would also go up right?

However they probably derive max speed from a fixed input on hp that goes into their model.

As a result the only thing left for them to adjust to meet their max speed again, would be the prop efficiency which would have to differ at different speeds...

Wouldn´t that make sense in this context?

Edited by =EXPEND=SchwarzeDreizehn

Share this post


Link to post
Share on other sites

Well, prop efficiency will effect both climb and turn, because it happens at the same speed. While turn is in the low region, climb is not. So improving turn to historical levels by increasing prop efficiency will also increase climb to unhistorical levels.

 

L/D, however, is taken at different states of flight for turn, climb and high speed. Turn happens at high angle of attack, climb at medium and high speed at low ones. So you could just change L/D at high angles of attack, and not change anything for climbing or high speed flight.

 

Personally, I don't get the prop statement by the devs, but then I don't know the details of the flight model. I'd play with L/D.

 

How's VDM coming along, btw.? L3Pl4K had asked me about contacting VDM originally, but after being hesitant for a while I lost my computer, so I didn't do it.

Share this post


Link to post
Share on other sites

Yes, I think it looks more like it's the L/D at high aoa that may be the problem. At least that is what comes to my mind since it's probably not the propeller efficiency. Can only guess at the reason the developers asked for the VDM propeller data but maybe they have data for other planes but are missing it for the Me-109 and used some other data instead. In that case it makes sense to think of the prop data as the first thing to look into but since the Me-109 climbs as well as it does then it's probably something else and then the L/D ratio and Clmax comes to mind. OTOH without knowing the details of how the BoX FM works they may have other knobs to turn so it may be something else but for the moment the prop efficiency does not seem to be it.

 

However, that being said, if someone can come up with VDM prop data for the Me-109 I'm sure it would always be welcome as input to fine tuning and I for one would love to see it so by all means continue the quest if you can! :good:

 

By the responses so far in this thread it seems I'm not alone in thinking it's not a prop issue so then it makes sense to alert the developers about this and see what their take on this is I think. I'm planning to report it when I have put together some more data and structured the argumentation a bit but then it looks like it may be worth while to do that anyways. Any ideas about what to put in the report would be welcome.

Share this post


Link to post
Share on other sites

 

 

i contacted vdm, but they still have not answered...

Maybe it will make sense that 777/Petrovich communicate with VDM. From Company to Company, signed with Senior Engineer could be impressive enough to react.

Share this post


Link to post
Share on other sites

If someone sent a mail asking for info out of my companies archives, the central mail address folks would certainly not know what to do and try to sit it out.

Share this post


Link to post
Share on other sites

If someone sent a mail asking for info out of my companies archives, the central mail address folks would certainly not know what to do and try to sit it out.

Yes, I know. That is why I asked them if there are any archives like this at all.

Of course I did not ask them for info right away...

Share this post


Link to post
Share on other sites

Well, prop efficiency will effect both climb and turn, because it happens at the same speed. While turn is in the low region, climb is not. So improving turn to historical levels by increasing prop efficiency will also increase climb to unhistorical levels.

 

L/D, however, is taken at different states of flight for turn, climb and high speed. Turn happens at high angle of attack, climb at medium and high speed at low ones. So you could just change L/D at high angles of attack, and not change anything for climbing or high speed flight.

 

Personally, I don't get the prop statement by the devs, but then I don't know the details of the flight model. I'd play with L/D.

 

How's VDM coming along, btw.? L3Pl4K had asked me about contacting VDM originally, but after being hesitant for a while I lost my computer, so I didn't do it.

 

The answer is probably that they already have the correct L/D ratios and do not want to change it. That is why they ask for VDM propeller efficiency curves, or how to jerry-rig the F/M with guesswork. In the old Il-2 they had these odd solutions to reach to historical turn rates on the G-2 but it only seem to have worked if the planes was modelled several hundred kg lighter otherwise - needless to say, this immediately effected a other aspects of the FM that less visible or measurable things than speed, climb and turn. Solutions like this are bound to broke more things than they fix, not to mention the FM is then based on what, pure fantasies and theorycrafting?

 

Obviously they already know that and probably tried, since the FM is developed by professionals they know how things should be done. 

Edited by VO101Kurfurst

Share this post


Link to post
Share on other sites

I wonder if they might have added an artificial drag increase upon slat opening, incorrectly thinking the slats would increase drag and thus arriving at way too low an L/D ratio?

  • Upvote 1

Share this post


Link to post
Share on other sites

The answer is probably that they already have the correct L/D ratios and do not want to change it.

No, they said they are waiting for prop data to make a complete overhaul, they are not just waiting for it so they can only tune the prop.

 

And to go into a bit of details, just changing prop efficiency in order to chop off a couple of seconds from the turn time, you'd need about 20% more thrust. This is just not the margin you have there, unless the prop right now is completely porked. Which I can't imagine it to be. Anyway, a side effect of just improving prop efficiency would be another ~5m/s in climb, which is unhistorically high. So they cannot fix the FM by just adjusting the prop model.

Share this post


Link to post
Share on other sites

I wonder if they might have added an artificial drag increase upon slat opening, incorrectly thinking the slats would increase drag and thus arriving at way too low an L/D ratio?

 

Well since modeling viscous flow is extremely costly from a processing perspective my guess is that drag in the BoX FM is added based on a calculated lift distribution using inviscid flow analysis methods and in that sense basically all drag is artificially added and just like you say it may well be that the drag added when slats open is currently on the high side. However, I think the main takeaway from the recent discussion is that it should be perfectly possible to fix the Me-109 turn rate issue without specific VDM data since the problem of the missing 20% is not related to propeller efficiency.

  • Upvote 2

Share this post


Link to post
Share on other sites

Well since modeling viscous flow is extremely costly from a processing perspective my guess is that drag in the BoX FM is added based on a calculated lift distribution using inviscid flow analysis methods and in that sense basically all drag is artificially added and just like you say it may well be that the drag added when slats open is currently on the high side. However, I think the main takeaway from the recent discussion is that it should be perfectly possible to fix the Me-109 turn rate issue without specific VDM data since the problem of the missing 20% is not related to propeller efficiency.

 

Yes I was ofcourse trying to say there should be no additional (or at least only an extremely minute amount) parasitic drag as the slats deploy, as ofcourse the increase in lift will always be accompanied by an increase in induced drag. Thus I wonder(ed) wether they may have added some parasitic drag that really shouldn't be there :)

 

And yeah, I really don't believe in the propeller efficiency being the issue either, if that were the case there would have to be serious flaws in other parts of the FM if at the same time real life level speeds are achieved.

Share this post


Link to post
Share on other sites

However, I think the main takeaway from the recent discussion is that it should be perfectly possible to fix the Me-109 turn rate issue without specific VDM data since the problem of the missing 20% is not related to propeller efficiency.

 

Because reasons.

Share this post


Link to post
Share on other sites

I see he did the comparison at 1 km there, I wonder what CLmax he used for both aircraft back then. TN 1044 suggests a rather large difference in lift coefficient at 25,000 ft between the two, so I'd expect something similar down low.

 

My own estimates puts the Fw190D9 ahead in both sustained & instantaneous rate, but again it's for 25,000 ft and not ~3,300 ft as I lack the mach specific Clmax of the 190's wing at low altitude. 

Edited by Panthera

Share this post


Link to post
Share on other sites

I dont remember mate so I wont speak for him, but I only recall that he arrived on a number that was deemed a bit too optimistic by Yoyo - their lead FM developer, in one of the other discussions: https://forums.eagle.ru/showthread.php?t=199139&page=18

Yo-yo posted this many years ago: https://forums.eagle.ru/showpost.php?p=2369933&postcount=105 

As I said elsewhere ... we have the same discussions every year or two ... with same or almost the same people. Should be celebrated almost like a groundhog day or something. 

 

Personally I am more interested in seeing P-47, 190 A-8 and ... Me-262 lol 

Share this post


Link to post
Share on other sites

yes, this is vaguely familiar:

 

 

As P-51 has lower wing loading and almost the same maximal lift as Dora it has advantage in turn rate and turn radius in low speed range. Its thrust limited steady turn is worse than D9 turn at higher speed.
So, if Dora begins to play on Mustang's field it won't win turnfight.

 

https://forums.eagle.ru/showpost.php?p=2369933&postcount=105%C2%A0

 

bit early to be discussing it though, BOK is'nt even out yet.

Edited by Sgt_Joch

Share this post


Link to post
Share on other sites

So it seems that in DCS the P-51 has almost the same lift coefficient as the Fw190? I wonder why that is and what their basis is for that considering that flight testing suggests the opposite, and in a rather significant fashion even.

Share this post


Link to post
Share on other sites

The same report you posted. Yo-yo made a number of points on that there. But they also have close cooperation with Fighter Collection and said in some videos they used Miss Velma to confirm flight behavior and characteristics. I just find different perspectives interesting. 

Share this post


Link to post
Share on other sites

They used the same report? Really? In that case I'm really surprised how they could arrive at that conclusion as the differences are quite marked in the report  :blink:

 

I wonder if they only used the P-51's mach specific Clmax at low altitude from that report, and then completely disregarded the data available for Fw190's wing/airfoil type ? If so I would definitely be sceptical about their FMs on a comparative scale.

Edited by Panthera

Share this post


Link to post
Share on other sites

@Panthera: Have you looked through the whole thread Hiro linked to? There was actually a lot of interesting discussions earlier on in that thread about both the Clmax of many planes, the Fw-190, P-51 and Me-109 included and it was nice that the DCS developer Yo-Yo participated like he did.

 

Beginning with the stall speed Clmax, there is another report (NACA 829) that you should look at that has a full scale P-51 test which gives Clmax 1.4 for the Mustang. The thread is also interesting since there are some wild ideas put forward that the no flap Clmax for the Fw-190 should be as high as 1.58 and the Me-109 a whooping 1.7!

 

When it comes to the Clmax at higher than stall speed, I don’t have the figures in front of me but IIRC then at M=0.3 -0.4 the Dora and Pony are both modelled at around 1.2 to 1.25 in the C++ code. However, I don’t think the low speed Clmax is that big an issue since as you can see in the figures I posted, the “roof” of the doghouse charts are pretty much on the same level so if the stall boundary is moved to the right a bit on both planes due to a more conservative Clmax estimate that is not a big deal: They are still pretty much on par when it comes to stationary turn performance.

 

The above is however with the following caveats:

 

1) I have assumed the same fuel load for both the Pony and Dora: I don’t see why the Pony should be penalized for good endurance so I compare at a state with a similar endurance capability not with max internal fuel.

 

2) I have assumed the Dora with B4 fuel and MW-50 boost and the Pony at two WEP power states: one being 67” boost and the other 75” boost.

 

Under these conditions it turns out the difference was less than I anticipated and to me it looks like they are pretty evenly matched. The Clmax at moderate to higher Mach, say 0.4 to 0.6 (affected by buffeting etc.) is of course especially interesting for the instantaneous turn rate and there the report TN 1044 is really interesting. Another interesting report is NACA WR L6110 which has more input on the P-51 Clmax as affected by Mach.

 

@=LD=Hiromachi: About the other BoBp aircraft: I have a number of them modeled in C++ already since many of them are after all iconic and were a part of the earlier Il-2 plane set as well. Will be interesting to see how they are modeled in BoBp when the time comes.

Share this post


Link to post
Share on other sites

Thats great to hear man! You're fantastic!

 

And yes, it certainly will. We will not avoid comparisons to DCS ww2 aircraft in this regard, it's unavoidable. I cant wait to have those warbirds released. 

Share this post


Link to post
Share on other sites

@Panthera: Have you looked through the whole thread Hiro linked to? There was actually a lot of interesting discussions earlier on in that thread about both the Clmax of many planes, the Fw-190, P-51 and Me-109 included and it was nice that the DCS developer Yo-Yo participated like he did.

 

<snip>

 

 

Just about half way through it now and trying to follow the arguments - thinking that the fellow called Pilum was making some sharp points. ;)

 

Reminds me of certain threads we have had here. 

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×
×
  • Create New...