In-game Nieuport 28.C1 & Albatros D.Va sustained turn rate comparison to C++ simulations

[Holtzauge]

Below is a figure showing the relative turn performance of the Nieuport 28.C1 and Albatros D.Va at an altitude of 6000 ft based on C++ simulations.

 

The figure also gives some idea about the type of data that will be included in the book “WW1 Aircraft Performance” which will soon be published.

 

I have tried in-game to replicate the turn times indicated in the attached figure, but I get the reversed results there: The Albatros has a turn time in-game that is 2-3 s faster than the Nieuport at 6000 ft in-game as far as I can tell?

 

The C++ simulation OTOH (See figure below) indicates the reversed result with the Nieuport needing 15-16 s to do a 360 degree turn at 6000 ft while the Albatros needs 17-18 s.

 

This of course raises the question of what figures we should expect IRL and I would appreciate any input on this. However, I believe the C++ simulation is solid, and more info about the C++ model itself and how it has been validated can be found at this site:

 

www.militaryaircraftperformance.com

 

The site is still work in progress but it still gives some idea about the type of data that will be included in the book which is due to be published before the end of the year.

 

I hope you find the simulation results interesting and any constructive input as to which numbers are right and why would be most welcome!

 

PS: Please be patient for the initial loading of the page! I think there is some kind of server issue at my host since it can sometimes take 10-15 s for the initial page to load!

 

 

[No.23_Starling]

Can’t wait to read the book! Am veeeeeeery interested in what you find about the SE5a vs Dva

[Zooropa_Fly]

I think we should wait for Plank to comment before we jump to any hasty conclusions. 

[Guest deleted@83466]

Looks very similar to John Boyd’s E/M diagrams(of ‘Fighter Mafia’ fame). Muy interesante.

 

 

 

Edited October 17, 2022 by SeaSerpent

[Holtzauge]

Well, when it comes to how the S.E.5a fares against the Albatros D.Va, I will not reveal any details here to undercut my marked: Hell, I spent 2 years writing a book so I want to sell a few copies anyway! But let me say this much: I don’t think @ST_Catchovwill be disappointed when he sees the results!

 

About Boyd: A very interesting character indeed and I thoroughly enjoyed reading the book by Robert Coram about him. A fascinating story that comes highly recommended.

 

Regarding Knalp’s ideas about N28 turns, those are above my paygrade so I’ll have to pass on commenting those! 

[No.23_Starling]

1 hour ago, Holtzauge said:

Well, when it comes to how the S.E.5a fares against the Albatros D.Va, I will not reveal any details here to undercut my marked: Hell, I spent 2 years writing a book so I want to sell a few copies anyway! But let me say this much: I don’t think @ST_Catchovwill be disappointed when he sees the results!

 

About Boyd: A very interesting character indeed and I thoroughly enjoyed reading the book by Robert Coram about him. A fascinating story that comes highly recommended.

 

Regarding Knalp’s ideas about N28 turns, those are above my paygrade so I’ll have to pass on commenting those! 

The other one i would love to see is dh2 vs dii. Leon Bennett’s book calculates the pusher as having a much tighter turn but a far slower top speed and climb. This matches contemporary descriptions from pilots too. It will also give us more insight into Hawker’s last flight vs MvR. There’s a narrative I want to challenge that the DH2 was inferior in all respects to the Dii.

 

Lastly the Eindekker vs DH2 would be fascinating given the match up in Rise of Flight.

[Holtzauge]

1 minute ago, US103_Rummell said:

The other one i would love to see is dh2 vs dii. Leon Bennett’s book calculates the pusher as having a much tighter turn but a far slower top speed and climb. This matches contemporary descriptions from pilots too. It will also give us more insight into Hawker’s last flight vs MvR. There’s a narrative I want to challenge that the DH2 was inferior in all respects to the Dii.

 

Lastly the Eindekker vs DH2 would be fascinating given the match up in Rise of Flight.

 

Well if I get sufficient interest in the WW1 aircraft performance book which covers the mid- to late WW1 period, I may instead of finishing the book about WW2 aircraft performance do early WW1 fighters and keep the WW2 book on the back burner. In that case the scouts you mentioned seem like natural candidates. But it all depends on the kind of response the first book gets.

 

BTW: Does anyone fancy doing a similar turn rate/speed chart for the in-game planes? I just did a quick test myself yesterday trying to keep roughly the same speeds as indicated in my simulations but I have no idea where the "sweet spot" in terms of speed to turn is in the in-game N28 & D.Va? In addition, does the "doghouse" chart for the in-game planes look the same as in the simulations or do they keep/lose turn rate differently as a function of speed?

[BMA_Hellbender]

Tha

2 hours ago, Holtzauge said:

BTW: Does anyone fancy doing a similar turn rate/speed chart for the in-game planes? I just did a quick test myself yesterday trying to keep roughly the same speeds as indicated in my simulations but I have no idea where the "sweet spot" in terms of speed to turn is in the in-game N28 & D.Va? In addition, does the "doghouse" chart for the in-game planes look the same as in the simulations or do they keep/lose turn rate differently as a function of speed?

As reported almost a year ago (directly quoting your work):

 

Edited October 18, 2022 by =IRFC=Hellbender

[ST_Catchov]

3 hours ago, Holtzauge said:

Well, when it comes to how the S.E.5a fares against the Albatros D.Va, I will not reveal any details here to undercut my marked: Hell, I spent 2 years writing a book so I want to sell a few copies anyway! But let me say this much: I don’t think @ST_Catchovwill be disappointed when he sees the results!

 

I don't think @US103_Rummell will be disappointed either.  

[No.23_Starling]

2 hours ago, =IRFC=Hellbender said:

Tha

As reported almost a year ago (directly quoting your work):

 

Interesting. So the N28 is closer to the model but the Dva is wayyy out

Are you going to send a copy to 1C? I’ll buy one for them if you like

[Holtzauge]

21 hours ago, =IRFC=Hellbender said:

Tha

As reported almost a year ago (directly quoting your work):

 

 

Good God man! How do you expect me, a pensioner, to remember what I posted more than a year ago? 

 

On a more serious note, thanks for digging that up! It does highlight the differences in a good way I think. First of all the in-game time for the Albatros at 13 s seems very optimistic. Secondly, the simulations indicate that the roles are reverse with the Nieuport being the better turner.

 

Another interesting observation is that the IAS speed used in the in-game trial at 105 Km/h is very close to the 114 Km/h TAS in the C++ simulations for the Albatros but for the Nieuport the TAS speed for best turn rate in the simulations is at 133 Km/h TAS which is about 122 Km/h IAS at 6000 ft, which is then quite a bit higher.

 

So while the Albatros's speed in the test is in line with what the simulations indicate, the simulations hint that for the Nieuport the speed for best turn rate is closer to 122 Km/h IAS at this altitude. So does it turn better at this speed in-game as well or is the speed for best turn rate in-game 100-110 Km/h IAS for both aircraft? I’m guessing the latter since you guys who know the in-game planes well would have picked up on this otherwise.

 

21 hours ago, ST_Catchov said:

 

I don't think @US103_Rummell will be disappointed either.  

 

Absolutely! The reason your name came to mind is that you are more, how shall I put it, vociferous in your opinion on the S.E.5a. 

 

19 hours ago, US103_Rummell said:

Interesting. So the N28 is closer to the model but the Dva is wayyy out

Are you going to send a copy to 1C? I’ll buy one for them if you like

 

That is a very generous offer and thanks for that @US103_Rummell! However, if you get a copy of your own and spread the word if you like it, I would be content with that. I was thinking about sending Jason a copy of the book when I’m done with the hopes that he could channel it to the right developers. I think that could be to our mutual benefit: They obviously have a lot of knowledge and data on these aircraft that could help me fine tune my models, and I could help them out with simulation input from my perspective.

 

Edited October 19, 2022 by Holtzauge
spelling......

[No.23_Starling]

48 minutes ago, Holtzauge said:

 

Good God man! How do you expect me, a pensioner, to remember what I posted more than a year ago? 

 

On a more serious note, thanks for digging that up! It does highlight the differences in a good way I think. First of all the in-game time for the Albatros at 13 s seems very optimistic. Secondly, the simulations indicate that the roles are reverse with the Nieuport being the better turner.

 

Another interesting observation is that the IAS speed used in the in-game trial at 105 Km/h is very close to the 114 Km/h TAS in the C++ simulations for the Albatros but for the Nieuport the TAS speed for best turn rate in the simulations is at 133 Km/h TAS which is about 122 Km/h IAS at 6000 ft, which is then quite a bit higher.

 

So while the Albatros's speed in the test is in line with what the simulations indicate, the simulations hint that for the Nieuport the speed for best turn rate is closer to 122 Km/h IAS at this altitude. So does it turn better at this speed in-game as well or is the speed for best turn rate in-game 100-110 Km/h IAS for both aircraft? I’m guessing the latter since you guys who know the in-game planes well would have picked up on this otherwise.

 

 

Absolutely! The reason your name came to mind is that you are more, how shall I put it, vociferous in your opinion on the S.E.5a. 

 

 

That is a very generous offer and thanks for that @US103_Rummell! However, if you get a copy of your own and spread the work if you like it, I would be content with that. I was thinking about sending Jason a copy of the book when I’m done with the hopes that he could channel it to the right developers. I think that could be to our mutual benefit: They obviously have a lot of knowledge and data on these aircraft that could help me fine tune my models, and I could help them out with simulation input from my perspective.

 

I’ll preorder one as soon as you have it ready. Will get one for the squad too.

 

A huge thank you for doing this work.

[No.23_Starling]

 

I’d add that this is probably the best way to support the developers in refining flight models where we don’t have real test data. There are surviving examples of some of our birds and excellent replicas with original engines, but the former can’t be tested to the extremes. I have performance data from Shuttleworth for their SE5a, but obviously it doesn’t include min turn rate etc.

 

Im not sure if you’re going to hand your C++ model over to 1C so their engineers can assess it; either way this is a much better approach to petitioning for FM changes than “my super plane should do X”. Even better if it does align with multiple primary sources.

[HagarTheHorrible]

Sod the SE5a ?,  what about the Dolphin ?  While I concede it might be lacking in the reversing a turn stakes, it should be pretty much unmatched (by aircraft that actually have a ‘modicum’ of directional stability) in a sustained turn. Providing, that is, that it actually ‘has’ wings to start with ?.

[Chill31]

Here is a little something to fuel your WWI day dreams!

 

 

[ST_Catchov]

6 hours ago, US103_Rummell said:

I have performance data from Shuttleworth for their SE5a, but obviously it doesn’t include min turn rate etc.

 

Does the data shed any light on the FC "bleeding revs" issue?  Like, it's all bollocks!

 

6 hours ago, US103_Rummell said:

Im not sure if you’re going to hand your C++ model over to 1C so their engineers can assess it; either way this is a much better approach to petitioning for FM changes than “my super plane should do X”. Even better if it does align with multiple primary sources.

 

It sounds like a good idea but I'm not convinced 1C would do anything with it. Doesn't seem to be their style .... if you know what I mean.  

[Holtzauge]

17 hours ago, US103_Rummell said:

Im not sure if you’re going to hand your C++ model over to 1C so their engineers can assess it; either way this is a much better approach to petitioning for FM changes than “my super plane should do X”. Even better if it does align with multiple primary sources.

 

10 hours ago, ST_Catchov said:

It sounds like a good idea but I'm not convinced 1C would do anything with it. Doesn't seem to be their style .... if you know what I mean.  

 

I’m not planning to give 1C my source code for the same reason 1C won’t give access to theirs but I am willing to share results. In addition, they don’t need to see the source code: Proof of the simulator’s ability lies in the results it gives, just like when we do tests in-game. Find a historical number you think is good, test it against your model and compare. If those numbers line up well then you can be pretty sure that the others (I mean the ones calculated by your simulation) are as well because then it’s just a matter of physics: There is no magic involved: The aircraft performance will be determined by the weight, power, wing profile, wing area, span etc.

 

As to the possibility to get changes done in-game, 1C has done modifications of the flight models on a number of occasions when presented with evidence that they find compelling. From my end that begins with publishing my book and letting them have a look at it and taking it from there. 

[No.23_Starling]

4 hours ago, Holtzauge said:

 

 

I’m not planning to give 1C my source code for the same reason 1C won’t give access to theirs but I am willing to share results. In addition, they don’t need to see the source code: Proof of the simulator’s ability lies in the results it gives, just like when we do tests in-game. Find a historical number you think is good, test it against your model and compare. If those numbers line up well then you can be pretty sure that the others (I mean the ones calculated by your simulation) are as well because then it’s just a matter of physics: There is no magic involved: The aircraft performance will be determined by the weight, power, wing profile, wing area, span etc.

 

As to the possibility to get changes done in-game, 1C has done modifications of the flight models on a number of occasions when presented with evidence that they find compelling. From my end that begins with publishing my book and letting them have a look at it and taking it from there. 

I get you can’t give the source code but I’m guessing you can explain the methodology and factors included?

[Holtzauge]

3 hours ago, US103_Rummell said:

I get you can’t give the source code but I’m guessing you can explain the methodology and factors included?

 

Yes, that would work. In addition, I think that would be the best way since I think we have very different ways of doing the modeling. Anyway, I think the first step is to agree on what to target. And in the case of the N28 versus the D.Va, I think my model gives the correct relationship between the two: The N28 should be better at sustained turns due to two fundamental advantages: One is a better power loading, i.e. less weight per hp, and the other is something called the span loading which is basically the weight of the airplane divided by the span. Note that this is not the physical span, but something called the effective span which is dependent on the bi- or triplane configuration and the vertical separation between the wings.  This is explained in more detail in the book, but given that the N28 bests the D.Va on both counts, it simply should be better according to the laws of flight mechanics.

 

Edited October 19, 2022 by Holtzauge

[J2_Trupobaw]

How do you account for the bracing density in all that? I remember reading that the wires added insane amount of drag (+50% is figure I remember, it can't be right?), and light bracing was one of few clear  advantages of German single-seaters.

[Holtzauge]

37 minutes ago, J2_Trupobaw said:

How do you account for the bracing density in all that? I remember reading that the wires added insane amount of drag (+50% is figure I remember, it can't be right?), and light bracing was one of few clear  advantages of German single-seaters.

 

Yes, I have an atmospheric model: It's the ICAO standard 15 deg C model so as the aircraft go up, the density and temperature of the air changes. And you are right: Bracing wire is roughly circular and circular objects create insane amount of drag as you say. Behind a circular object you get a so-called von Karman vortex street which causes a lot of parasitic drag. As a ballpark estimate a round wire causes the same amount of drag as a symmetrical wing profile ten times as thick!

[No.23_Triggers]

Fantastic stuff. Should the devs ever be so-inclined, this is the kind of thing that could really aid them in turning FC into its own thing, rather than RoF-in-a-different-engine, which is sorely needed IMHO. The current FMs are long-outdated. 

[J2_Trupobaw]

13 hours ago, Holtzauge said:

 

Yes, I have an atmospheric model: It's the ICAO standard 15 deg C model so as the aircraft go up, the density and temperature of the air changes. And you are right: Bracing wire is roughly circular and circular objects create insane amount of drag as you say. Behind a circular object you get a so-called von Karman vortex street which causes a lot of parasitic drag. As a ballpark estimate a round wire causes the same amount of drag as a symmetrical wing profile ten times as thick!

Entente actually developed wires with streamlined (oval?) profile to reduce that, but used more wires than Germans  .

[No.23_Starling]

13 hours ago, Holtzauge said:

 

Yes, I have an atmospheric model: It's the ICAO standard 15 deg C model so as the aircraft go up, the density and temperature of the air changes. And you are right: Bracing wire is roughly circular and circular objects create insane amount of drag as you say. Behind a circular object you get a so-called von Karman vortex street which causes a lot of parasitic drag. As a ballpark estimate a round wire causes the same amount of drag as a symmetrical wing profile ten times as thick!

Are you also modelling prop pitch and type of prop, eg four blade vs two? Is this a substantial issue with ww1 birds?

[Holtzauge]

1 hour ago, J2_Trupobaw said:

Entente actually developed wires with streamlined (oval?) profile to reduce that, but used more wires than Germans  .

 

Yes, and they were called "RAF wires" even if they were one piece solid hammered/drawn to an oval shape! I write a bit about this in the chapter devoted to drag in the book.

 

1 hour ago, US103_Rummell said:

Are you also modelling prop pitch and type of prop, eg four blade vs two? Is this a substantial issue with ww1 birds?

 

Yes, I model the propellers thrust depending on the rpm, diameter, pitch and so-called "solidity". The solidity is basically how much blade area you have (could be distributed over 2 or 4 blades) which in turn determines how "loaded" the individual propeller blades get, which in turn determines it's efficiency. This (the modeling of propeller efficiency) is covered in a short chapter in the book and also discussed in the dive chapter where it is explained how the propeller goes from producing thrust to causing extra drag by windmilling which of course also needs to be modeled.

 

Edited October 20, 2022 by Holtzauge

[No.23_Starling]

29 minutes ago, Holtzauge said:

 

Yes, and they were called "RAF wires" even if they were one piece solid hammered/drawn to an oval shape! I write a bit about this in the chapter devoted to drag in the book.

 

 

Yes, I model the propellers thrust depending on the rpm, diameter, pitch and so-called "solidity". The solidity is basically how much blade area you have (could be distributed over 2 or 4 blades) which in turn determines how "loaded" the individual propeller blades get, which in turn determines it's efficiency. This (the modeling of propeller efficiency) is covered in a short chapter in the book and also discussed in the dive chapter where it is explained how the propeller goes from producing thrust to causing extra drag by windmilling which of course also needs to be modeled.

 

This sounds amazing. I’m going to learn a lot here. Did you say you worked in engineering??

[Holtzauge]

21 minutes ago, US103_Rummell said:

This sounds amazing. I’m going to learn a lot here. Did you say you worked in engineering??

 

Not any more but I used to. 

[No.23_Starling]

2 hours ago, Holtzauge said:

 

Not any more but I used to. 

This CV is the real deal and worth highlighting for the context of your work. This is not the work of an armchair general who “read in an old book not translated yet” but that of a highly skilled and experienced engineer. We are going to struggle to find anyone better qualified to help supply this kind of objective performance data. 

[HagarTheHorrible]

Any evidence of ‘engine braking ‘ for inline engines ?  By which I mean, as the engine speeds up it increasingly struggles to draw in ‘enough’ air, deliberately or otherwise, into the cylinder heads, creating a partial vacuum, and therefore impeding cylinder stroke efficiency.  Or, does propeller drag alone impede engine speed sufficiently to reduce possible over speeds ?

 

If I have a problem with purely ‘theoretical’ engineering, to ascertain ‘hidden’ truths it is that it doesn’t allow for the likes of the Me 210, which I assume worked ‘very’ well on paper.  WW1 aircraft were even more ‘quirky’, how those ‘traits’ are accounted for, I do not know.

 

Edited October 20, 2022 by HagarTheHorrible

[Holtzauge]

27 minutes ago, US103_Rummell said:

This CV is the real deal and worth highlighting for the context of your work. This is not the work of an armchair general who “read in an old book not translated yet” but that of a highly skilled and experienced engineer. We are going to struggle to find anyone better qualified to help supply this kind of objective performance data. 

 

You make me blush Sir!

 

25 minutes ago, HagarTheHorrible said:

Any evidence of ‘engine braking ‘ for inline engines ?  By which I mean, as the engine speeds up it increasingly struggles to draw in ‘enough’ air, deliberately or otherwise, into the cylinder heads, creating a partial vacuum, and therefore impeding cylinder stroke efficiency.  Or, does propeller drag alone impede engine speed sufficiently to reduce possible over speeds ?

 

If I have a problem with purely ‘theoretical’ engineering, to ascertain ‘hidden’ truths it is that it doesn’t allow for the likes of the Me 210, which I assume worked ‘very’ well on paper.  WW1 aircraft were even more ‘quirky’, how those ‘traits’ are accounted for, I do not know.

 

 

Well the propeller efficiency for a fixed pitch propeller is a bit bell shaped with speed, and depending on if the propeller is climb or speed optimized, it will either stop or continue to supply thrust a bit longer as you go faster. Again, this is covered in the book (and in the C++ modeling of course!) and the Entente scouts tend to have props with coarser pitch so they keep producing thrust to a higher dive speed. However, in the end the local angle of attack on the propeller blade goes down as the speed goes up and eventually it gets negative. It is at this point it starts windmilling and producing drag both aerodynamically and also due to the internal friction resulting from cranking the engine which would otherwise be at idle rpms. More details in the book!

[HagarTheHorrible]

Book, when, price and availability in UK ?

 

Vagaries or approximations are fine.

 

Just need to know if it’s a possibility for my Santa list ?

[Holtzauge]

17 minutes ago, HagarTheHorrible said:

Book, when, price and availability in UK ?

 

Vagaries or approximations are fine.

 

Just need to know if it’s a possibility for my Santa list ?

 

The book should be available before the end of the year globally via Amazon and directly from the publisher Lulu Inc. Release date and price not yet settled but unfortunately it seems that Amazon will take a rather hefty charge to sell it through their sites. 

[HagarTheHorrible]

Cheers.  
 

 

[J2_Trupobaw]

The Lozenge Book ?

[Holtzauge]

19 minutes ago, J2_Trupobaw said:

The Lozenge Book ?

 

Well, I always thought that nothing is as cool and good looking on a plane as doped "Flugzeugstoff" which is why I chose the same covering for my book as well! 

[1PL-Husar-1Esk]

15 hours ago, Holtzauge said:

 

The book should be available before the end of the year globally via Amazon and directly from the publisher Lulu Inc. Release date and price not yet settled but unfortunately it seems that Amazon will take a rather hefty charge to sell it through their sites. 

Kindle version would be available ?

[Holtzauge]

25 minutes ago, 1PL-Husar-1Esk said:

Kindle version would be available ?

 

No, to begin with only hardcover. I have to admit to being a bit old fashioned in this regard: I put a lot of work into it want it to be a book in the classical sense that you can hold in your hand. But maybe later!

[No.23_Starling]

8 hours ago, Holtzauge said:

 

No, to begin with only hardcover. I have to admit to being a bit old fashioned in this regard: I put a lot of work into it want it to be a book in the classical sense that you can hold in your hand. But maybe later!

Have you ever done any work on damage modelling? Asking for a friend

Edited October 21, 2022 by US103_Rummell

[Zooropa_Fly]

2 hours ago, US103_Rummell said:

"Asking for a friend" 

 

Nudge nudge, Wink wink, Say No More ! 

[TheSaxman]

On 10/17/2022 at 2:17 PM, Holtzauge said:

Below is a figure showing the relative turn performance of the Nieuport 28.C1 and Albatros D.Va at an altitude of 6000 ft based on C++ simulations.

 

The figure also gives some idea about the type of data that will be included in the book “WW1 Aircraft Performance” which will soon be published.

 

I have tried in-game to replicate the turn times indicated in the attached figure, but I get the reversed results there: The Albatros has a turn time in-game that is 2-3 s faster than the Nieuport at 6000 ft in-game as far as I can tell?

 

The C++ simulation OTOH (See figure below) indicates the reversed result with the Nieuport needing 15-16 s to do a 360 degree turn at 6000 ft while the Albatros needs 17-18 s.

 

This of course raises the question of what figures we should expect IRL and I would appreciate any input on this. However, I believe the C++ simulation is solid, and more info about the C++ model itself and how it has been validated can be found at this site:

 

www.militaryaircraftperformance.com

 

The site is still work in progress but it still gives some idea about the type of data that will be included in the book which is due to be published before the end of the year.

 

I hope you find the simulation results interesting and any constructive input as to which numbers are right and why would be most welcome!

 

PS: Please be patient for the initial loading of the page! I think there is some kind of server issue at my host since it can sometimes take 10-15 s for the initial page to load!

 

 

 

Nice to see some hard maths applied to show how wrong the Nieuport's flight modelling is. Now if only the devs would do something with it...