YASim next testing (henning/jsb changes)

I mentioned in a previous post that Henning/jsb is reworking YASim, and I (possibly stupidly) volunteered on the dev list to do testing

From inspecting the merge request which I have exported into patches it's all refactoring the wing code so that he can rework the mstab code & update the CG/MAC stuff later - mostly structural & renaming of functions

While he's added a new version check (2018.1), none of the changes are wrapped in a version check

for reference the valid versions to go in the yasim xml header are:

YASIM_VERSION_ORIGINAL
YASIM_VERSION_32
2017.2
2018.1 (inserted by this patch)
YASIM_VERSION_CURRENT (defaults to latest above)

What I'm seeing is that the merge/patch set does create slight changes in the solver - creating small changes in drag coefficient, lift ratio, cruise aoa, tail incidence, & approach elevator

there is no change to inertia tensor or CG

As the changes only affect the wing element, they don't appear to affect helicopters (they might if they had a wing element - I've only looked at the lynx)

I want to present the current changes to James & Henning in order to give them an idea as to how it affects aircraft developers

What I'm asking is for a suggested list of FGUK craft to test against - preferably a small list that represents a range of aircraft & helicopter types that represents the FGUK hangar

any suggestions?

F-15 Active, typhoon, Jag, L39 for Jets with differing uses of M/sab + wing configurations.
Grob + commander, Vb Spitfire

the dev version of the AH-64 uses Msatb to represent its wings, so you might want to try that one.

Just grabbing the release versions of all of these to test against

The dev AH64 FdM is the one using the Mstabs.

Just out of interest. How do you test them?

I'm starting with the command line solver, looking at it's output tables, then moving on to simple flight tests - takeoff, landing, cruise as specified in the xml, and max speed at/near sea level

I'm creating a spreadsheet on google docs, working on the Aero Commander 1st as I'm more familiar with it out of all the craft the list.

I've done a simple test (command line solver, then takeoff from EGOV, flight at max speed at 2000ft to EGOD, and then land - all in core high pressure weather) of the Aero Commander

The results are here:

docs.google.com/spreadsheets/d/1ta30HRhq...n88/edit?usp=sharing

It's in the Tools directory of the Gdrive

What's evident from the solver output (I've included the lift/drag & TAS/drag tables) is that while the solver nearly matches the approach & cruise settings with slightly different drag & lift settings (less than 1% difference), how it gets there is a bit different, and what it produces outside of small aoa is quite different

As to flying, where you're mostly in small AoA (at least in the AC), it's not noticeable until landing - I just couldn't match the settings of the previous flight at all (both landings were at higher weight than in the xml & with full flaps.

I found that at cruise (at the settings found in the xml) that it was identical, and climb seemed the same as well, but low altitude (<1000ft) max speed was ~15kts higher in the older code

I need to check I've not missed something in my patching, and then check a couple more aircraft

Henning has updated his changes based on my testing & found the bug (a missing parameter), so the current merge doesn't cause any changes

I have some time off work, so will be working on the friction stuff