[ACE-OF-ACES]

Sorry for taking so long, we got a lot of snow last week and I didn't have a lot of free time between shoveling snow and the power going out.

But as promised I did a series of climbs at different climb speeds to determine the best climb speed (BCS). I did 5 tests, starting at an IAS of 250mph up to 450mph at 50mph increments.

As part of this test I also wanted to see if there was a climb speed that would produce the 10,000fpm ROC that IL2Compare reported.

First lets take a look at the 5 different time to climbs (TTC) graphs



Looking at the graph you can see the 300mph BCS reaches 20kft sooner (in less time) than the other BCS speeds.

Now lets take a look at the 5 different rate of climb (ROC) graphs



Looking at the graph you can see it agrees with the TTC graph, in that the 300mph BCS produces the largest ROC. At least it does up to 20kft. Note at 20kft the 250mph BCS starts to overtake the 300mph BCS wrt ROC. In English, if you were flying, you would want to climb at ~300mph up to 20kft than switch to a BCS of ~250mph as you continue to climb.

As you look at the graphs, keep in mind that in the real world the raw data will have a best fit curve drawn threw the data points (see next) to account for the settling time of the pilot. Also note, these ROC graphs show a bit of a curve (concave) to them. Thus a straight line fit of these data points is not the best method, but it is the simplest. With that said I applied the straight line fit to the middle of the data, which means the end points (i.e. Sea Level (SL) and 20kft) are going to be a little less than they actually are. But, looking at the graph you should be able to imagine the 'curve' beyond the 'straight line' fit at SL and 20kft.

Now lets take a look at the best fit, straight line fit of the ROC data @ 300mph BCS.



As you can see, the straight line fit matches the slope of the data best at the middle, where the straight line fit is going to report lower than actual ROC values at the end points.

Now lets graph the best fit curve to the real world data (RWD)



As you can see, as predicted by IL2Comp, the in-game F-86A has a much larger ROC than the RWD! Granted it is not the 10,000fpm @ SL that IL2Comp predicted, the straight line fit only shows 9,400fpm @ SL. But keep in mind the straight line fit is a little less than the actual ROC at the end points. Looking at the graph you can imagine the actual ROC is ~10,000fpm.

So in summary, the theory that Freddys' modified Java class might be hiding something from IL2Comp does not appear to be the case. In that even if you only use the straight line results you can see the % error is around 25%.

Now that I have the BCS per altitudes I can do a full up ROC test where I adj the BCS with altitude, for example at 20kft I can switch from 300mph to 250mph, with the goal of obtaining the best TTC values. Which I can than compare to the real world TTC data to find the % error. In that an error in ROC only affects the ROC at that alt, but, an error in ROC has an cumulative effect on the TTC values. Which result in a plane being able to reach a specific altitude much sooner than it could in real life.

Here is the RWD I used, it is from the F-86A Flight Manual


Here are the track files of each test
ROC @ 250mph BCS
ROC @ 300mph BCS
ROC @ 350mph BCS
ROC @ 400mph BCS
ROC @ 450mph BCS

Now I have to go shovel some more snow! ;(