Airspeed change

[pub00515]

Hi!

While trying to program an IAS Hold autopilot sub-system, I stumbled upon the following problem:

for the mode to work reliably, I need to have the first derivative of the airspeed, namely the acceleration. The only variable I found was the longitudinal acceleration relative to the world axis.

Unforunately, this isn't the acceleration I am looking for as it shows the change in GS (ie total kinetic energy) so I need a different variable.

One method would be to measure the variable inside my module but that would create quite a big lag.

Does anyone know a workaround over this?

Thanks!

Mark

[Pete Dowson]

... I need to have the first derivative of the airspeed, namely the acceleration. The only variable I found was the longitudinal acceleration relative to the world axis.

Unforunately, this isn't the acceleration I am looking for as it shows the change in GS (ie total kinetic energy) so I need a different variable.

Did you search thoroughly? What about 3070, the longitudinal acceleration relative to the body axes?

Pete

[pub00515]

To tell you the truth, I didn't try that.

Can you confirm that the longitudinal acceleration (relative to the body axis) will be 0 as long as the IAS is constant? (even if GS is changing due to wind or when climbing/descending)

[Pete Dowson]

To tell you the truth, I didn't try that.

Can you confirm that the longitudinal acceleration (relative to the body axis) will be 0 as long as the IAS is constant? (even if GS is changing due to wind or when climbing/descending)

No idea, but it doesn't sound right. It seems to me that you can't keep both the IAS constant AND the acceleration zero if the wind velocity is changing. Also, though I don't know whether it is modelled in FS, I would have thought that a vertical or lateral movement would have some (small) effect on the pressure in the pitot tube even though it is aligned with the longitudinal axis.

The FS simulation engine is dynamically recalculating all these things all of the time. Results like the IAS would likely be a derived value, not one resulting from any trend or prediction, so why would there be a specific IAS "acceleration" for you to read?

Incidentally, the TAS is more what you would want in any case, as some changes in the IAS will also be due to pressure and temperature changes, which will certainly be occurring during climbs and descents as well as more slowly in level flight.

Why not just use FSUIPC's Monitoring (on the Logging page) to display it dynamically on the FS screen (adv display option checked). You can see what it is doing then.

[LATER]

If all you want is some trend indication in order to adjust thrust, or display the trend on the airspeed tape, then I don't think you are going to get anything better than the body-axis acceleration. Maybe you would need to "smooth" that a little too, to remove spikes due to wind gusts or turbulence, but it is worth looking at in any case.

Regards,

Pete

[pub00515]

Hi Pete!

I've been experimenting with both values but I can't seem to get what I want (nor do I understand what the difference between the two longitudinal accelerations is)

Anyway, I want to be able to control pitch to keep a set speed constant (ie IAS Hold or FLCH or open climb/descent) but I need a value for IAS change over time as my controller needs rate information :-(

[Pete Dowson]

Anyway, I want to be able to control pitch to keep a set speed constant (ie IAS Hold or FLCH or open climb/descent) but I need a value for IAS change over time as my controller needs rate information :-(

I think the only way you'll get a value for IAS changes over time is to read it and record it at known intervals. Read the millisecond counter at the same time so you can keep it correct with respect to varying intervals (due to Windows message processing, process switching and varied FS graphics, sound and disk loading).

The best timer to read is probably the millisecond one provided by FSUIPC itself, at offset 3374, as this will almost precisely reflect the actual time it met your request.

Regards,

Pete