Author Topic: K4 vs Spit 16 climb  (Read 3614 times)

Offline Stoney

  • Gold Member
  • *****
  • Posts: 3482
Re: K4 vs Spit 16 climb
« Reply #45 on: April 16, 2011, 09:14:24 AM »
What specs/parameters do you need to know to calculate excess power?

Power Available = Thrust X Velocity (TAS)

Power Required = Drag X Velocity

Excess power = Pa - Pr

So, you need the ability to accurately predict thrust from stall speed to maximum speed (a tricky proposition with propellor driven aircraft, especially since thrust changes with altitude, either through changes in engine power or propeller efficiency).  You also need to do an accurate model of total drag from stall speed to maximum speed.  This is easier to do, but you need to include zero-lift drag, and drag due to lift (and compressibility drag if it rears its ugly head).  Drag also changes with altitude due to changes in dynamic pressure, and the parasitic drag and induced drag components change with speed.

So, even if you're merely doing a comparison in level flight, you can imagine how complex this estimation can be.  Introduce changing altitudes, and it gets even more complex, since you have to factor in some way to account for changes due to differences in dynamic pressure and engine power and do it repeatedly over some sort of time span.

This is why a lot of people fall victim to over-simplifying by using such metrics as wing-loading and power-loading--because to do the actual analysis required is very time consuming, complex, and sometimes beyond the grasp of most folks aerodynamics knowledge.  The best model I've seen on these boards was from DTango, and his response to that epic Brewster thread last year.  He mathematically estimated the zoom climb of the Brewster using a Pr/Pa model and posted an Excel graph in the thread.  He actually was kind enough to send me the actual spreadsheet used to generate the graph, and it was stunning.  Huge spreadsheet involving multiple variables broken down into each second of about a two minute period.  Even as detailed as it was, Tango himself will tell you it was a mere estimation--an accurate estimation, but an estimation.  I myself have generated a spreadsheet that I can use that will tell me Cdo and Cdi at a given condition for each aircraft in-game, but its dependent on determining an accurate stall speed for each aircraft in that specific configuration (i.e. 25% fuel, 50% fuel, etc.) and does not yet correct for compressibility drag.  This was not easy to do, still needs some refinement, and the testing time in-game to determine the stall speed is very time consuming.  To do something like this on paper alone (i.e. analyzing an aircraft based purely on its aerodynamic metrics without the ability to test in-game) would take even longer.  Just creating thrust curves for each aircraft is very daunting, since propeller efficiency can be such a huge obstacle to model.

This is why everyone on the board should be thankful when guys like Badboy or Tango take the time to actually graph this stuff out, or provide us with tools like the Bootstrap Calculator to do some casual analysis with.  Its also why I am continuously impressed with HTCs ability to model all of these very complex calculations in-game.  And finally, its the only true reality check you can get to combat all the anecdotal crap that gets slung about on these boards, and from other sources.

Ultimately, there's not one simple equation that takes all of these variables into consideration--its a combination of a lot of formulas and will still only give you an estimation.  I don't know if that answers your question or not...
"Can we be incorrect at times, absolutely, but I do believe 15 years of experience does deserve a little more credence and respect than you have given from your very first post."

HiTech

Offline moot

  • Plutonium Member
  • *******
  • Posts: 16333
      • http://www.dasmuppets.com
Re: K4 vs Spit 16 climb
« Reply #46 on: April 16, 2011, 04:10:13 PM »
I can see a nice 3D graph (speed, alt, excess power) for comparing any two planes, if I could manage the calcs.

Is Badboy's 'bootstrap' method for calculating stall speed not accurate enough?

I do remember that prop calcs are a major effort though.  The last time I'd tried to calculate a plane's performance in this much detail, I gave up when I got to prop efficiency.. I couldn't even find the prop data.
Hello ant
running very fast
I squish you

Offline Stoney

  • Gold Member
  • *****
  • Posts: 3482
Re: K4 vs Spit 16 climb
« Reply #47 on: April 16, 2011, 04:35:31 PM »
I can see a nice 3D graph (speed, alt, excess power) for comparing any two planes, if I could manage the calcs.

Is Badboy's 'bootstrap' method for calculating stall speed not accurate enough?

I do remember that prop calcs are a major effort though.  The last time I'd tried to calculate a plane's performance in this much detail, I gave up when I got to prop efficiency.. I couldn't even find the prop data.

Badboy's method is very accurate, it just takes time to fly the profile precisely so you can trust the numbers.  I generated some of the aircraft stall speeds using it.  For example, the F6F numbers I generated looked good, but I don't trust the 190 numbers I got because the resulting Clmax was higher than I thought it should be.  The 190 is a tough plane to determine stall speed for, because it gets so unstable at slow speed, high alpha conditions.  But, it'll take maybe 10-15 mins per aircraft per configuration to get numbers you can trust, at least in my experience.  I got far enough along to get good numbers for the F6F and generate this:



For thrust calcs, it is tough.  However, there are some close approximations that can be made as long as you stay away from the static condition.  Theoretically, with drag numbers, you can determine prop efficiency.  Badboy has posted some methodology on that.  I'll have to climb back into my books and see if I can find something.
"Can we be incorrect at times, absolutely, but I do believe 15 years of experience does deserve a little more credence and respect than you have given from your very first post."

HiTech

Offline dtango

  • Silver Member
  • ****
  • Posts: 1702
Re: K4 vs Spit 16 climb
« Reply #48 on: April 16, 2011, 05:47:35 PM »
Just a comment on Badboy's fabulous Bootstrap tool.  It's doesn't model an aircraft per se i.e., aero coeffs, thrust, drag, prop, eff. etc.  It assumes whatever is modeled in the AH virtual world for a given aircraft for these variables and then calculates turn performance basis whatever the virtual AH flight test data you plug in are.  There are gigantic reasons for this - one being Badboy want's the output to be representative of what you will find in the AH virtual world.  Badboy is not attempting to remodel the aerodynamics with Bootstrap; he's given a tool to calculate turn performance of AH aircraft based on the aerodynamics modeled in AH itself embedded in the flight test data you plug in.
Tango / Tango412 412th FS Braunco Mustangs
"At times it seems like people think they can chuck bunch of anecdotes into some converter which comes up with the flight model." (Wmaker)

Offline dtango

  • Silver Member
  • ****
  • Posts: 1702
Re: K4 vs Spit 16 climb
« Reply #49 on: April 16, 2011, 05:51:45 PM »
This is why a lot of people fall victim to over-simplifying by using such metrics as wing-loading and power-loading--

This needs to be stickied :).
Tango / Tango412 412th FS Braunco Mustangs
"At times it seems like people think they can chuck bunch of anecdotes into some converter which comes up with the flight model." (Wmaker)

Offline moot

  • Plutonium Member
  • *******
  • Posts: 16333
      • http://www.dasmuppets.com
Re: K4 vs Spit 16 climb
« Reply #50 on: April 16, 2011, 06:09:42 PM »
Gotcha.   What I'd want to do is compare such an altitude/speed/excess power graph for each of the AH twins, to a crude powerloading/wingloading/level speed/etc table of these twins, to see just how accurate/inaccurate such a table of vulgar specs is.  If you make speed/altitude the horizontal axes, and vertical axis the excess power metric, it would make for very intuitive illustration of a plane's performance.

I'm also curious...  How different would such a 3D plot for a plane's excess power in level flight be, from one of Pe while maneuvering (some standard like 2 G)?  Could you get a good enough (for general MA dogfighting purposes) idea of the plane's performance/character looking at the level flight 3D plot, or would there be for some/all planes such a change in character/performance (whether taken on their own, or compared to other planes - eg a P38's excess power looking great in level flight but looking barely competitive compared to a Spit's maneuvering excess thrust) that you'd have to see this 3D plot in context with all other maneuvering plots (i.e. a range of G if that's a good metric)... e.g. such a 3D plot with G as fourth dimension - time-varying for G.
« Last Edit: April 16, 2011, 06:12:35 PM by moot »
Hello ant
running very fast
I squish you

Offline TheDude

  • Zinc Member
  • *
  • Posts: 16
Re: K4 vs Spit 16 climb
« Reply #51 on: April 16, 2011, 11:12:56 PM »
where u get of saying that...numbers dont translate into a pilots skill...funny i climb with both spit 16's and k4s.. and have no trouble in a vert fight vs either
just my 2 cents
TheDude

Offline Stoney

  • Gold Member
  • *****
  • Posts: 3482
Re: K4 vs Spit 16 climb
« Reply #52 on: April 16, 2011, 11:25:23 PM »
Gotcha.   What I'd want to do is compare such an altitude/speed/excess power graph for each of the AH twins, to a crude powerloading/wingloading/level speed/etc table of these twins, to see just how accurate/inaccurate such a table of vulgar specs is.  If you make speed/altitude the horizontal axes, and vertical axis the excess power metric, it would make for very intuitive illustration of a plane's performance.

I'm also curious...  How different would such a 3D plot for a plane's excess power in level flight be, from one of Pe while maneuvering (some standard like 2 G)?  Could you get a good enough (for general MA dogfighting purposes) idea of the plane's performance/character looking at the level flight 3D plot, or would there be for some/all planes such a change in character/performance (whether taken on their own, or compared to other planes - eg a P38's excess power looking great in level flight but looking barely competitive compared to a Spit's maneuvering excess thrust) that you'd have to see this 3D plot in context with all other maneuvering plots (i.e. a range of G if that's a good metric)... e.g. such a 3D plot with G as fourth dimension - time-varying for G.

There are various ways to graph it for comparison.  The term "envelope" actually comes from a type of excess power graph where you can see graphically, the speeds needed to keep the aircraft "inside" the envelope.  Some of them can get pretty complicated, but some are more simple.  Let me get some things cooking this week and I'll see if I can put together a full blown power curve.  Basically, my chart is just missing the power available plot and a curve for power required.  I'll see if I can't finish it up and post it.  I'll use the F6F data I already have, then post it just for an example.  Then perhaps we can treat your specific comparison.  I'll move it to a new thread to keep from mucking this one up...
"Can we be incorrect at times, absolutely, but I do believe 15 years of experience does deserve a little more credence and respect than you have given from your very first post."

HiTech

Offline moot

  • Plutonium Member
  • *******
  • Posts: 16333
      • http://www.dasmuppets.com
Re: K4 vs Spit 16 climb
« Reply #53 on: April 16, 2011, 11:54:02 PM »
Alright.  I'll make a quick & dirty sketch of what I had in mind, you can tell me if it's realistic in that other thread where I'll post it.
Hello ant
running very fast
I squish you

Offline dtango

  • Silver Member
  • ****
  • Posts: 1702
Re: K4 vs Spit 16 climb
« Reply #54 on: April 17, 2011, 12:24:41 AM »
Gotcha.   What I'd want to do is compare such an altitude/speed/excess power graph for each of the AH twins, to a crude powerloading/wingloading/level speed/etc table of these twins, to see just how accurate/inaccurate such a table of vulgar specs is.  If you make speed/altitude the horizontal axes, and vertical axis the excess power metric, it would make for very intuitive illustration of a plane's performance.

I'm also curious...  How different would such a 3D plot for a plane's excess power in level flight be, from one of Pe while maneuvering (some standard like 2 G)?  Could you get a good enough (for general MA dogfighting purposes) idea of the plane's performance/character looking at the level flight 3D plot, or would there be for some/all planes such a change in character/performance (whether taken on their own, or compared to other planes - eg a P38's excess power looking great in level flight but looking barely competitive compared to a Spit's maneuvering excess thrust) that you'd have to see this 3D plot in context with all other maneuvering plots (i.e. a range of G if that's a good metric)... e.g. such a 3D plot with G as fourth dimension - time-varying for G.
Several things to comment on:

1) Keep in mind power-loading does not equal excess power (nor specific excess power).  Not the same things.

2) As to usefulness of power-loading or wing-loading they are really simplistic figures of merit that need to be used carefully.  For issues with power-loading see this post:
Power-to-Weight Ratio Pitfalls

Let's take a quick look at wing-loading.  For Wing-Loading we can arrange the lift equation with wing-loading (defined as variable Ws where Ws = Weight/wing_surface_area) as:

Vstall1g^2 = 2*g*Ws / air_dens * Cl

Notice that the smaller Ws is, the lower the stall speed.  The lower the stall speed the lower the turn radius, & also the better the corner performance.  That's why wing-loading gets mentioned a lot because it gives us a thumb-in-the-air check of this.  Here's the catch & it's a big one.  Just looking at wing-loading ignores a huge variable, lift-coefficient which is a function of the shape of the wing airfoil.  Looking at the equation for Vstall you'll note that Cl is in the denominator.  Ignoring Cl means we would be ignoring any impact the design of the shape of the wing has which is significant.

3) If you plotted power-loading or wing-loading vs airspeed you wouldn't get anything meaningful.  They would each be just a flat straight line (straight with respect to the velocity axis).  Most of aerodynamics is quite curvy :) with respect to airspeed, or most any other variable.  You can surmise the usefulness of such a crude thing :).

4) Excess power curves would look something like the drag curves Stoney has created above.  Multiply drag by velocity and you get power required.  Of course you need power available as well.  It would look like the following graph labeled as Pxs:  (also Pa=power available, Pr=power required)


Image Source: FIU

My recommendation for doing any analysis is to stick with aero conventions (e.g. like the drag curves Stoney has posted).  You'll learn a lot in the process too.  These things have been thought about by lots of brilliant people for over 100 years.  If there were good simplifications for analysis we've already been using them.

If you're interested in some other interesting charts for performance comparisons you can start here:
http://www.vmihosting.com/MWS/Documents/FTM108-chapter6.pdf

Cheers
« Last Edit: April 17, 2011, 12:28:02 AM by dtango »
Tango / Tango412 412th FS Braunco Mustangs
"At times it seems like people think they can chuck bunch of anecdotes into some converter which comes up with the flight model." (Wmaker)

Offline Stoney

  • Gold Member
  • *****
  • Posts: 3482
Re: K4 vs Spit 16 climb
« Reply #55 on: April 17, 2011, 07:07:37 AM »

If you're interested in some other interesting charts for performance comparisons you can start here:
http://www.vmihosting.com/MWS/Documents/FTM108-chapter6.pdf

Cheers

Moot, this document Tango linked is Chapter 6 of the Navy Fixed Wing Performance Flight Test Manual.  I found a site where you can download the entire document in .pdf --can't remember the site, but do a Google Search and you can probably find it.  Its a very helpful reference, especially for performance equations.  It doesn't give you a "Reader's Digest Version" that's easier to understand, but everything is in there...
"Can we be incorrect at times, absolutely, but I do believe 15 years of experience does deserve a little more credence and respect than you have given from your very first post."

HiTech

Offline drgondog

  • Copper Member
  • **
  • Posts: 326
Re: K4 vs Spit 16 climb
« Reply #56 on: April 17, 2011, 10:32:41 AM »
Stoney - if you download the document it automatically converts to .pdf.

As I went through the document I found only two areas where the analysis (in my opinion) would break down as a foundation for a stand-alone dynamic model (vs Comparative).  Dtango - thanks for the link.

1.) For sustained Ps=0 turn, constant altitude turn, there is both a Delta drag and a Delta thrust component at low speed which is not fully incorporated when using a level flight Drag Polar.

2.) CLmax for a level flight condition is different from CLmax for sustained turn.

The 'root cause' for the difference (in my opinion) is due to the Elevator as a contribution to "lift"
a. In low speed level flight the elevator deflections and therefore lift contributions due to elevator deflection is relatively low... the 'negative lift' contribution due to incidence required for pitch equilibrium at near stall/level flight is low both because the effective angle of attack and absolute lift force (of the elevator-horizontal stabilizer combination is relatively small to the Wing.

b. In high G sustained turn, the opposite is true. The negative lift contribution due to up elevator on the horizontal stabilizer is of much higher (relative) magnitude to maintain the high AoA of the Wing as it approaches CLmax...thereby increasing the required amount of lift which must be supplied from the wing (beyond that required to offset nW). The 'Net' effect is a slight increase (virtual) to the Weight vector to supply the negative lift effect of the tail.

Having said this, the rudder deflection 'incremental positive' lift offsets somewhat the negative contribution from the elevator.

The net effect is that the resultant contributions of lift vector from the tail and wing are in opposite directions - implying a stall at a higher velocity than predicted using CLmax at level stall for turn modelling? The eternal question is 'how much'

The Delta drag is also increased due to both the trim drag of a higher lift [absolute value] tail (and rudder) but also the miscellaneous viscous drag contributions due to the high AoA.
I need to ponder this a little more.
Nicholas Boileau "Honor is like an island, rugged and without shores; once we have left it, we can never return"

Offline dtango

  • Silver Member
  • ****
  • Posts: 1702
Re: K4 vs Spit 16 climb
« Reply #57 on: April 17, 2011, 10:53:43 AM »
The full manual can be found here at the US Naval Flight Test Alumni pages.  FTM108 is the fixed-wing performance manual.

US Naval Flight Test Pilot Alumni Library
Tango / Tango412 412th FS Braunco Mustangs
"At times it seems like people think they can chuck bunch of anecdotes into some converter which comes up with the flight model." (Wmaker)

Offline moot

  • Plutonium Member
  • *******
  • Posts: 16333
      • http://www.dasmuppets.com
Re: K4 vs Spit 16 climb
« Reply #58 on: April 17, 2011, 06:36:25 PM »
Thanks guys I saved these posts..  I'd grabbed ch5 FTM108 already from a previous discussion where it seemed excess power was a better sense of performance than power/wing-loading, but haven't gotten around to figuring out the math yet.  I guess ch6 is closer to the heart of how excess power works?

I'll get back to this in about a week once I've got other stuff finished.
Hello ant
running very fast
I squish you