109 it fly wrong

[HoHun]

Hi Blogs,

>The first and third charts seem to imply a supercharger tax on the engine between 2 and 4 km.

The FAF speed chart is not the result of mere morphing of the DB605A power chart due to ram effect. The FAF speed chart requries the minimum power in the variable-speed altitude band to drop 6% below the minimum power indicated by the DB605A power chart I'm using for reference (relative to respective maximum power).

>In this case it really does matter what speed-altitude combination you calibrate from.

I'm calibrating from speed at critical altitude. (I could use any data point below 2.4 km to get a virtually identical result, so this calibration is fairly robust.)

Regards,

Henning (HoHun)

[Badboy]

Originally posted by HoHun
You probably wouldn't learn much - it's like a one-dimensional flight simulation incapable of real-time operation, and I think you're already one or two steps ahead of that with your own project ;-)
Henning (HoHun)

Hi HoHun, nice thread.

However, I’m going to disagree on a couple of points, at least if we are talking about MS Excel, and probably with other spread sheets too if we checked. You can if you want, run a six degree of freedom, non linear aerodynamics model, with coupled total force equations of motion, solved using 4th order methods (for greater accuracy than you get in most sim’s) and all run in real time. You just need to use the high level language that comes built in. It is also quite possible to include graphics and I think early versions of excel had a fairly basic flight simulation Easter egg. It’s probably still there if anyone knows how to access it.

Why would you want to? Well if you want a state of the art simulation for game purposes, online or otherwise, you wouldn’t, but spread sheets are a very powerful tool, particularly if you are more interested in processing the output data for engineering purposes. Running a simulation and interrogating it in real time to produce graphs and other analysis can be one major advantage of building a simulation that runs in a spreadsheet. Another advantage is that you can build and run a flight model that way, fairly quickly, without investing the time required for a full blown simulation project. There are many other pros and cons of course, but I’m also fairly sure that HiTech is just pulling your leg, judging by the performance graphs on the HTC web site, I’d say he knows as much about the use of spreadsheets as we do

Badboy

[HoHun]

Hi Badboy,

>However, I’m going to disagree on a couple of points, at least if we are talking about MS Excel, and probably with other spread sheets too if we checked. You can if you want, run a six degree of freedom, non linear aerodynamics model, with coupled total force equations of motion, solved using 4th order methods (for greater accuracy than you get in most sim’s) and all run in real time.

Agreed, but it would not be my first choice for the job :-)

However, if you know anything about the tools the professionals use, that would be highly interesting!

Regards,

Henning (HoHun)

[gripen]

blogs,
The throttle valve of the DB 605 is located between impeller and intake manifolds so below FTH there is more pressure between impeller and the throttle valve (gebläsedruck) than in intake manifolds (ladedruck). The MAP itself and control of the throttle valve is a fully automatic system ie pilot just selects wanted rpm and MAP combination and the system keeps these in wanted levels below FTH. The controll of the 2nd supercharger oil pump is simply by barometric valve which starts pump at around 2km.

I quess the best way to understand how MAP and pressure between impeller and throttle valve are connected, is to look power curve of the DB 605 and compare it to both pressure curves. The peaks of the power curve are located in same places where the relative difference between these pressures is smallest ie around 2,5km and FTH and above. Between these points pressure difference is higher and therefore relative efficiency decreases. Due to RAM effect this pressure difference increases at high speed flight, therefore efficiency is worse than in lower speed.

Overall it's up to you what you want to believe; documented stuff or something which might sound cool but is pretty much undocumented.

gripen

[mw]

"Now that could be used for making some aerodynamic comparisons... "

I've found that Daimler Benz chart useful for making comparisons.

Check the Spit V with Merlin 45 against this data set.  Very good agreement.

Check the "Me 109 Serie" curve against the Mtt AG curves shown here.  Again, very good agreement.

Compare the "Me 109 Serie" curve against some speculative estimates posted here.  Hmmm, not so good.

Gripen, thanks for explaining Gebläsedruck.  I couldn't figure that out.  Its fortunate we have the Finnish trials data.  Its interesting and valuable. I don't suppose I place quite as much weight on it as perhaps you do (shrug).  I've admired your knowledge of the material, logic and clear presentation of facts  in this thread. Nice job.

[HoHun]

Hi Mw,

>Check the "Me 109 Serie" curve against the Mtt AG curves shown here.  Again, very good agreement.

Interesting data :-)

Is it correct that Werknr. 14026 achieved 364 mph @ 33000 ft?

Regards,

Henning (HoHun)

[joeblogs]

There are known problems with using excel for ordinary least squares let alone nonlinear least squares.

-blogs

Originally posted by Badboy
Hi HoHun, nice thread.

However, I’m going to disagree on a couple of points, at least if we are talking about MS Excel, and probably with other spread sheets too if we checked. You can if you want, run a six degree of freedom, non linear aerodynamics model, with coupled total force equations of motion, solved using 4th order methods (for greater accuracy than you get in most sim’s) and all run in real time. You just need to use the high level language that comes built in. It is also quite possible to include graphics and I think early versions of excel had a fairly basic flight simulation Easter egg. It’s probably still there if anyone knows how to access it.

Why would you want to? Well if you want a state of the art simulation for game purposes, online or otherwise, you wouldn’t, but spread sheets are a very powerful tool, particularly if you are more interested in processing the output data for engineering purposes. Running a simulation and interrogating it in real time to produce graphs and other analysis can be one major advantage of building a simulation that runs in a spreadsheet. Another advantage is that you can build and run a flight model that way, fairly quickly, without investing the time required for a full blown simulation project. There are many other pros and cons of course, but I’m also fairly sure that HiTech is just pulling your leg, judging by the performance graphs on the HTC web site, I’d say he knows as much about the use of spreadsheets as we do

Badboy

[VO101_Isegrim]

Originally posted by HoHun
Hi Mw,

>Check the "Me 109 Serie" curve against the Mtt AG curves shown here.  Again, very good agreement.

Interesting data :-)

Is it correct that Werknr. 14026 achieved 364 mph @ 33000 ft?

Regards,

Henning (HoHun)

Yep, Werknr. 14026 in the DB601E/DB605A swapped engines test did 586 km/h at 10000m, farily low for the type, since the same plane when fitted with the much less powerful 601E engine did 580 km/h at the same altitude.

I guess there may be some problem`s with the 605A engine supercharger, given that it could attain the rated altitude in level flight, tested plane reached level speed VDJ height at 6100m instead of 7000m, and climb was even worser, VDH being only 5300m instead of 5800m.

Still, even this low performing Gustav at reduced boost (1310 HP) compares surprisingly well with the standard Spitfire IX at 150 octane fuel (1940 HP), their speeds at 20k feet being 626 km/h (Bf 109) and 624 km/h (Spitfire LF IX). This might give some idea for high speed aerodynamic effiency comparisons.

[HoHun]

Hi Blogs,

>There are known problems with using excel for ordinary least squares let alone nonlinear least squares.

Interesting - I tried to use the "solver" to fit a curve through the scattered FAF data points using 3rd order polynom least squares, and it gave me a strictly linear graph. I decided not to believe that :-/

Regards,

Henning (HoHun)

[HoHun]

Hi Isegrim,

>Yep, Werknr. 14026 in the DB601E/DB605A swapped engines test did 586 km/h at 10000m

Rather funny! :-)

That's exactly the value I predicted for the FAF aircraft, and which Gripen found so impossible to believe that he accused me of data manipulation and forgery.

Regards,

Henning (HoHun)

[VO101_Isegrim]

Hi HoHun,

That's exactly the value I predicted for the FAF aircraft, and which Gripen found so impossible to believe that he accused me of data manipulation and forgery.

Well, what can I say, typical of him. :/

Anyway, why not check your e-mail ?

[joeblogs]

Better to use Stata or Matlab.

-blogs

Originally posted by HoHun
Hi Blogs,

>There are known problems with using excel for ordinary least squares let alone nonlinear least squares.

Interesting - I tried to use the "solver" to fit a curve through the scattered FAF data points using 3rd order polynom least squares, and it gave me a strictly linear graph. I decided not to believe that :-/

Regards,

Henning (HoHun)

[Dweeb]

Originally posted by joeblogs
There are known problems with using excel for ordinary least squares let alone nonlinear least squares.
-blogs

That's interesting because I've recently done both linear and non-linear curve fitting and used excel and done the calculations manually, and the results came out the same in every case. What are the known problems exactly?

Dweeb

[Dweeb]

Originally posted by HoHun
Interesting - I tried to use the "solver" to fit a curve through the scattered FAF data points using 3rd order polynom least squares, and it gave me a strictly linear graph. I decided not to believe that :-/

Have you tried the "trendline" option for curve fitting? It seems to work very well.

Dweeb

PS, yes, I am a FSforum survivor from the late eighties, early nineties

[joeblogs]

I referred to excel's stat features not to its other functions. The primary issue is numerical precision, which can be a problem when inverting matrices.  I use excel all the time for other stuff, but not for regression work.

-blogs

Originally posted by Dweeb
That's interesting because I've recently done both linear and non-linear curve fitting and used excel and done the calculations manually, and the results came out the same in every case. What are the known problems exactly?

Dweeb