Units issue in drag equations

[jake-asquith-spire]

Hello there,

Is there a units problem in the equations for calculating drag force (e.g. code)?

From looking at other force calculations in the repo that use the same ForceModel trait, I can see that they are returning 10^-3 * Newtons. E.g. for solar pressure it looks like the equation has (10^-3 * m^2 * N * m^-2)

Similarly for the gravitational acceleration if we multiply by mass (I suppose kg, as that seems conventional to other parts of the repo), we derive the units (kg * km * s^-2) => (kg * 10^-3 * m * s^-2) => (10^-3 * N).

For the aforementioned drag equation though, it looks like the units of the equation in existing code are (kg * m^-3 * m^2 * km^2 * s^-2), which resolves to ((kg * km^2 * s^-2) / m^1), which doesn't equal the others. It seems like it might be appropriate to multiply the equation out by 10^3 here, which would give ((kg * km^2 * s^-2) / (m^1 * 10^-3)) => (kg * km * s^-2) => (10^-3 * N), which is consistent with the other forces.

Thoughts? Happy to open a PR to change this.

[ChristopherRabotin]

Hi Jake,

I think you're absolutely correct. The drag models have not been validated as the other models have, in part because I have worked only on cislunar missions for 7-8 years now, but also because these models are much simpler than the ones I could find in other trustworthy software. If you have a verification case you could add, that would be even better, but if you don't , I totally understand.

Also note that #86 will significantly improve the gravity models, and I'm actively working on this (although https://github.com/nyx-space/hifitime/milestone/14 has been taking a lot of my free time recently).

Thanks

[jake-asquith-spire]

No problem!

Shall I open a PR to put in the missing factors, or will the units change be encompassed in #86 (I see some mention about drag changes in there)?

Thanks for linking that milestone, I had a look at the nyx-space ones too.

[ChristopherRabotin]

Yes, please open a PR with the unit changes. I expect that you'll need to update the associated regression test. Ticket #86 still needs quite some work, so I'll be sure to update that branch to whatever is on master before merging. That branch adds a lot of breaking changes though, and I'll be happy to guide you through updating your code base once it's merged.