Kudos for fixing the long-standing issue of not taking into account payload effects on MOI. However, the SDK description on how to enable this using the “empty_inertia_tensor” parameter in the online flighting SDK appears to be missing.
2 Likes
Any idea how and what value we should use?
Found it is a vector of three floats, but didn’t know what values are “normal”
Hello,
This part of the documentation is “work in progress”.
It’s being worked on as we speak, and will be available before the SU3 release.
Regards,
Sylvain
2 Likes
Too beta test this new feature, it would be great to know how it worked. A hint would be helpfull.
1 Like
I understand but it’s rather complex (~4 pages of documentation) and some technical aspects of the implementation are still being discussed.
Unless @An.Petrovich feels comfortable giving you a glimpse, I’d rather wait for at least a preliminary version of the public documentation.
Regards,
Sylvain
2 Likes
Hello,
Indeed, the SDK will provide complete documentation,
but just to give you a very brief idea of the main points:
empty_inertia_tensor = Jxx, Jyy, Jzz, Jyz, Jxz, Jxy
It is a comma-separated array of 3 to 6 doubles represents the full inertia tensor of an empty aircraft: Jxx (pitch MOI), Jyy (yaw MOI), Jzz (roll MOI), Jyz (coupled MOI), Jxz, Jxy
Jxx, Jyy, Jzz - are mandatory, Jyz, Jxz, Jxy - are optional.
Units: [slug*ft^2]
Sign convention for centrifugal MOI: standard (not alternative)
It is relative to the CG of the empty aircraft (not to the origin or datum point),
and in the MSFS/DirectX basis (X - lateral axis, Y - vertical axis, Z - longitudinal axis)
This new param overrides params: empty_weight_pitch_MOI, empty_weight_roll_MOI, empty_weight_yaw_MOI, and empty_weight_coupled_MOI BUT only if it passes the validity check for the tensor - you can observe errors/warnings in the console, if any.
If it is present and valid it activates the following fixes:
- The addition of payload stations now correctly affects the aircraft’s moments of inertia by updating the full inertia tensor matrix accordingly.
Previously, all payload stations were merged into a single equivalent mass before being added to the aircraft, which led to incorrect accounting for the effect of spatially separated masses on the moments of inertia (a long-standing issue present since FS 2004 at least).
NOTE:
If you previously compensated for this by increasing the empty aircraft MOIs, re-adjustment may now be necessary for the new parameter values.
-
Fixed incorrect recalculation of moments of inertia during simulation when the aircraft’s weight changes (e.g., fuel addition/consumption, payload release, etc.). The previous algorithm computed the additional terms to moments of inertia relative to the CG of the empty aircraft, instead of relative to the actual CG, which shifts as the mass distribution changes.
-
The full 3×3 inertia tensor is now used for aircraft rotation modeling. Previously, only the three main moments of inertia (calculated for the loaded aircraft, i.e. including additional weight) were taken into account, while the computed centrifugal (product of inertia) terms were ignored as if they were zero.
-
Fixed several issues in rotational dynamics modeling that could result in physically incorrect aircraft rotation, especially at high angular velocities.
The laws of conservation and evolution of angular momentum and rotational kinetic energy are now satisfied with high accuracy. -
Increased limits for angular velocity (up to 10 revolutions per second) and angular acceleration (up to 10 revolutions per second within 0.1 sec) in critical simulation scenarios caused by incorrect flight model settings. These limits serve solely as safeguards to prevent the sim from crashing and are not expected to be triggered under normal conditions in any gameplay scenario.
Hope it helps!
S!
7 Likes
Thank you got it to work 
2 Likes
Hi everyone,
Please have a look at the announcement we just made regarding the new “empty_inertia_tensor” being postponed until after SU3.
Best regards,
Eric / Asobo
1 Like
I understand these aren’t the complete docs. BTW the note mentions the ‘Payload’ stations affecting the calculated MOI values, but not the systems.cfg [Water Ballast System] Tank.N entries. I believe (unlike the current payload stations) these ballast entries already affect at least the roll MOI - FYI a 300kg empty glider may carry an additional 200kg of dumpable ballast mostly in the wings.
I’d like to request the final docs do summarise which aircraft parameters are affecting the MOI.
Gracias.
2 Likes
Can someone please clarify the definition and calculation of the coupled moi. This has been a mistery to me for a long time and none of the developers I talked to was able to give me an answer. Since we´re now getting new moi parameters it would be a good point in time to finally get some clarity. Additionally the coupled moi value did not seem to have any effect on my helicopter, no matter how extreme I set it up.
2 Likes
Hello b1e1n7e,
You’ve raised a completely valid question, and it’s no surprise that this topic has been confusing. The inertia tensor and its coupled terms are non-trivial, and can be difficult to explain clearly.
Unless you’re using a specialized tool to compute the full inertia tensor based on the aircraft’s mass distribution in 3D space, I would not recommend trying to define the coupled moments of inertia manually. These off-diagonal components (like Ixy, Ixz, Iyz) only have physical meaning in the context of a full volumetric mass/inertia computation — something typically done with CAD or simulation software.
Until you dive into that level of detail, it’s perfectly fine to stick with the three principal moments of inertia (Ixx, Iyy, Izz). In most cases, that’s sufficient for a reasonably realistic simulation of the aircraft’s inertial behavior, especially for airplanes and simpler helicopters.
If you’re curious to dig deeper, here’s a solid reference on the concept of the inertia tensor and what the coupled components represent:
Hope it helps!
Andrey / Asobo
1 Like
Hello B21,
Thanks for pointing this out - you’re absolutely right to expect that all changes in mass distribution, including water ballast systems, should be accounted for.
As mentioned in point #2 above, the simulation will now correctly recalculate the moments of inertia whenever the aircraft’s mass changes during flight (fuel consumption, payload drops, etc.).
So yes, any system that contributes to total mass - including water ballast system, tank.N entries - will affect the MOI as expected. This behavior is part of the general improvement.
Best regards,
Andrey / Asobo
2 Likes
According to the SDK, an entry for the empty weight coupled MOI is required, yet one cannot use the weights debug screen for a recommended value as you can for the other MOIs. What would you suggest be used since it is too difficult to calculate manually?
1 Like
Hello AwarePlot117729,
I’ve checked the SDK, and you’re absolutely right - the “empty_weight_coupled_MOI” parameter is currently marked as required, although in practice it’s optional. I’ll ask our SDK team to correct the documentation, thank you for pointing it out!
In the meantime, if you’re unsure what value to provide for this parameter (i.e., if you haven’t calculated it), it’s safest to simply leave it as zero.
In any case, please note that in the current FM, the “empty_weight_coupled_MOI” parameter is not yet used in the simulation of the aircraft’s rotational dynamics. This will be fixed in SU4, along with the introduction of the new “empty_inertia_tensor” parameter.
Best regards,
Andrey / Asobo
2 Likes
Has the documentation for this now been update with the SU4 release?
I am trying to understand what to fully do to implement this new way of doing inertia.
Best,
Raul
1 Like
@SimbolFSReborn No, the documentation is missing, which has been reported here:
I found this is now fully documented actually after inspecting the latest release notes:
And also on weight and balance:
Is all there now, will try to set it up and see how my flight model gets affected 
R.
Please report back!