Talking a planes inertia.

[bomber]

Within the mass balance section of an FDM are the values of planes inertia..


<mass_balance>
<ixx unit="SLUG*FT2"> 539.57 </ixx>
<iyy unit="SLUG*FT2"> 524.36 </iyy>
<izz unit="SLUG*FT2"> 987.49 </izz>
</mass_balance>

Can anyone tell me how these are worked out, because I'd love to know.

Simon

[jwocky]

Well, I had the -160 up and aside of the torque, she felt good to me. Didn't do any math on it because ... well, she felt already good.

[bomber]

But how do you know that these values are correct ?

How can we be sure that something else within the flight model hasn't had to be adjusted to make it feel right because these numbers are wrong ?

Where's the proof that these numbers are correct ?

How does it stand up to a challenge ?

[jwocky]

Well, technically, it's the same formulas we discussed with propellers, only used on the plane itself.

for ixx, it should be sufficient precise to say at least whether she is in the right ball park, for iyy, it's more of a guess, I assume, nobody calculated that plane in parts (and whether we have enough data to do so) and izz is the same problem. By all means, rotations around y and around z have always the problem, you have on oane and a plane nose, on the other end a tail, differently shaped and with different mass.

My problem is, I can't sit down right now and try to figure some smart math because something else is going on outside of FG and I actually need to take a few days off.

[bomber]

My point is that these values are very important to how a plane handles.

Now after all these years of fg development, you'd expect some sort of process for determining these values for planes without manufacturers data.

So why isn't there.... in the years I've been at FG I've not seen anyone explain it.

My fear is that it's just 'borrowed ' from another plane which if we look closely is also borrowed from another plane adinfinitum.

Simon

[IAHM-COL]

Sometimes I feel like going positivist about the whole thing, and make a trivariate experiment, moving those values on a ranged manner, and reading outputs from FG, seeing how the inertia can per example change the lifting coefficients or something.

Problem is that the output seems not to easy to read.

Maybe I just need to pipe to console and collect via a pipe ( cat | ) or something.

That one, I could graph correlates.

----Thinkin' outloud--- IH-COL

PS: Actually I tried it once. I was trying to adjust fuel consumption speed on the HondaJet to better match published specs. I was expecting that fuel consumption to be a variable on weight (which reduces on the plane as more fuel was consummed). That is. the longer I fly, on a same engine power output, the less fuel would be needed to keep the same engine output. After doing this experiment (annoying one cause I needed to place the plane of cruise for every data point, and do the NONO of increase the sim rate to collect data more efficiently), my data seemed to suggest that fuel consumption in JSBSim was not reacting as a function of aircraft weight, and it was maintained as a constant being a function of throtle levels. At the end of the day, I could not figure out how to better manage the fuel consumption from the BFSC, and I had to resort to thrust tables in the FDM to obtain a significant change in range.--and a subsequent dissapointment in my abilities as FDM builder.

[bomber]

Over at outerra uriah set me something similar so as I could see thrust over time and a variety of other jsbsim property values for the hellfire missile flight model I was making for him...

Output was in excel .

I'll go look it up

[jwocky]

Ooookay,

thinking out loud:

1.) We have in many cases not enough data, so we borrow from a similar plane for which we have data and tweak till "it feels right". Which can run an engineer nutz, engineers don't feel, and if they do, they don't trust their feelings, so they can only relax after it's calculated. Bomber, I so see you point, but I have currently no time to write a Java toolbox. (we could take dimensions form the model and enter the weights into input fields or something)

2.) Israel, I am very sure, you played with TSFC and the thrust tables, not BSFC on the HondaJet. Now, if I remember right, the thrust tables were just changed to get you more power on higher altitudes so you can make it up there and don't need to fly all the time full throttle. Which of course indirectly changes your fuel consumption as well.
For a jet, fuel consumption in JSB depends mainly on throttle and on drag. The weight, one expects to be so important, doesn't play the big role in many cases simply because the relation between (empty weight + payload)/MTOW is in many cases so near to one, it produces not a really measurable difference. It does on extreme long distance planes, for example 747s. They use at the end of the flight about 25-30% less and it does definitively on long range propeller planes (who carry more fuel in relation to the total weight) like the Condor or the Koenigsadler.

3.) Gosh, we so need a toolbox for all those calculations ... *cry - no time to play*

[IAHM-COL]

Over at outerra uriah set me something similar so as I could see thrust over time and a variety of other jsbsim property values for the hellfire missile flight model I was making for him...

Output was in excel .

I'll go look it up

cool stuff. This is what I feel sometimes.
I feel filling JSBsim simulation with very "real-life" values does not produce the more accurate simulation outputs.
So, if that case would be true, I'd suggest fitting JSBsim with whatever unrealistic value is necessary to produce the better/more realistic output out of it.

that, short of just fixing JSBsim itself.

[IAHM-COL]

Ooookay,


2.) Israel, I am very sure, you played with TSFC and the thrust tables, not BSFC on the HondaJet.

Myself, I tweaked both. But only the Thrust tables did output anything of use. The BSFC of the engine was very limited on its ability to give me a good fuel consumption range... and very effective to give me developers' rage.