Don't forget that we are discussing the Type 517M radar here, not the Nebo-M. The former has just 4 antennas in a row, while the Nebo-M has 22 antennas. The Nebo-M is credited with an impressive 0.5 degree accuracy, although it is unclear exactly how it achieves this as its mainlobe beam is going to be around 5 degrees wide. If it uses monopulse, then its power-aperture will drop to 25% significantly impairing its detection range. But for the sake of argument, let's assume that's still sufficient to detect the LRASM at 30km range. Scaling to 517M, azimuth accuracy will be 22/4 * 0.5 = 2.75 degrees. Can you redo your calculation with this figure?
The number and size of antennas should just affect the total power output, number of beams and 3D tracking.
I don't see how it should affect the angular accuracy of each individual beam in the vertical plane.
How did you get 7 degrees? In you example above, if the error is 838m doesn't that mean that you should draw a circle of radius 838m, as the area of uncertainty?
Yes, the LRASM could be anywhere in a circle with a radius of 419m. But since you know it is a sea skimmer, I'm assuming a line across the horizon.
It would help if you specify the speed that you used for HHQ-9. Given the short range of the engagement, it will likely be close to its top speed: 2000 m/s?
Mach 4.2 as listed.
No, you can't do that. First you need to figure out what it is that you detected. That's called an OODA loop.
It's a simplified assumption assuming zero time required.
But by all means, go ahead and add some OODA loop.
You still end up with 3-4 engagements before the 13km engagement start on CMO.
When you redo your calculations, don't forget to take into account that the Type 517M is a rotating radar and that your signal updates are some 6s apart. In those 6s, a maneuvering LRASM can cover at least 1,5-1,8km.
The simplest thing would be to make the VHF radar directional for terminal tracking, like we see with the E-2D
It's a software change after all.