SM-3 repeatedly misses ballistic missiles by more than 3 miles

[fmachanda]

I attempted to defend a CG Ticonderoga (Surface Ship #2862) with an SM-3 Blk IIA (Weapon #3256) from a Chinese DF-26 ASBM. However, despite having an active datalink with the CG as well a high fidelity ABM radar track, the missiles miss the target with almost 4 miles of lateral error, with almost no apparent mid-course corrections. I replicated this behavior multiple times with various engagement ranges, multiple ship types (all AEGIS using SM-3), and both manual and automatic firing.

Using Command Open Beta v1.05.1328.7 (Steam) on Windows with DB v501

sm3 failure.zip

(35.72 KiB) Downloaded 10 times

[DWReese]

For the Devs:

I agree with the OP. The SM-3 (and SM-6) worked better with 1307,14. Now, with 1307.19 (and apparently the beta zest version as well) the SM-3 often flies right on by. The bottom line is, they don't work as well as they did.

[Transient]

I suspect there may be something not quite right with the ballistic missile interception mechanics at the moment.

Unlike air breathing targets or HGVs, a ballistic missile in its ballistic phase is fully predictable throughout its ballistic trajectory. As such, it should be a binary yes or no whether the SM-3 can or cannot get into position to intercept the ballistic missile, if the intercept is to occur in the ballistic missiles’s ballistic phase.

Of course there should be a prob of intercept still for whether the KV actually impacts the ballistic missile.

Currently we are seeing too many cases of the SM-3s not even getting in the vicinity of the ballistic missile for a possible intercept, even with near perfect tracking information throughout the ballistic missile flight path.

[DWReese]

As I recall, this worked fairly well about 3 months ago. Shortly afterward, someone reported a problem with the interception, and some tweaks were made. Right after the tweaks were made, the intercept sequence worsened, in my opinion.

I have run a test scenario on this for years, so it was easy to see when the changes were applied. I didn't bother reporting this new issue because the Devs' efforts seem to be aimed at the 64-bit beta, and I am not participating in that.

Good luck, and I hope that they address and fix this soon.

[blu3s]

Hi guys, thanks for the report, it is being investigated because there seem to be some strange issues with the new model.

Logged: 0015222

[SteveMcClaire]

sm3 failure.zip

Hi fmachanda,

Thanks for the report -- I am in the process of investigating it. Do you perhaps have an earlier version of this saved game? Being able to debug the engagement from the beginning would be very helpful. If I delete the missiles in flight, reload the DF-21, and shoot again I can't reproduce the lateral error that the save file shows.

The reason there is no mid-course guidance for the SM-3 is because the Tico does not have the incoming ballisitc missiles on its own radar. The incoming missiles are so high that they are outside the maximum range of the Tico's own radar and the SM-3s are firing based the contact data from the SBX-1.

[DWReese]

The reason there is no mid-course guidance for the SM-3 is because the Tico does not have the incoming ballisitc missiles on its own radar. The incoming missiles are so high that they are outside the maximum range of the Tico's own radar and the SM-3s are firing based the contact data from the SBX-1.
Top

i agree. If the SM-3 (e-version) is fired at great distance, then it will miss every time because it can't see it to be guided via mid course. But, by changing the range to 200 miles seems to work fine. The SBX-1 can track it originally, and when the Tico finally sees it itself when it is within 200 miles, the midcourse guidance takes over.

The long range of the e-version sort of seems like a complete waste because nothing will be able to track the oncoming missiles for that distance if they are in space. You would need a bunch of SBX-1s along the path (or something like that) to get the full distance out of the missile. That's not going to happen. So, it seems as though the best use of all versions of the SM-3 is geared to be used at 200 miles or less.

That's my take on it.

[fmachanda]

Thanks for the help and discussion! I reset the simulation to start right at missile launch for further investigation. I also turned off the SBX and T-AGM radars to only use the Tico's sensors (which makes interception almost impossible except at extremely short ranges in the original engagement).


Also, for further testing, I added another instance of the engagement but at a much closer range to the North (this one is 150 miles instead of 500 miles). I am not sure if any of this information will be useful, but here are some behaviors of the missiles I found in this closer situation:

1) The Tico detected the launch of the DF-26s with its SPY-1 and tracked their ascent. SM-3s were launched at a range of ~150 miles.

2) The SPY-1 lost the track of the DF-26s as they climbed through 196.2 km, but the SM-3s continued to ascend while connected to datalink.

3) Roughly 4 minutes later, a 'Weapon Issue' message appeared in the log, as the SM-3s decided that the datalink information was not firm.

4) The SM-3s immediately pitched down at this point and began to descend. It appears they were attempting to hit the position of the last contact to the DF-26 (5 minutes old at that point) rather than the expected position on the trajectory.

5) The SM-3s continued to the last known position, crossing this area with an altitude of approximately 196.2 km (the real missile was of course much closer and much higher at this point). Most of the SM-3s still had a large (>3 nm) lateral deviation, but 1 or 2 of them occasionally passed the expired track with no error. All missed and no 'Weapon Endgame' report was sent in the message log.

6) The DF-26 continued untracked by the Tico until reentry. Occasionally, the radar would never even pick up on the reentering targets (only IR and visual sensors), which meant that an 'Insufficient data for BMD engagement' message prevented SM-2/3/6 to launch. However, this short range terminal behavior was far less consistent that the other behaviors described above.



NOTE: I ran the simulation again but with the SBX and T-AGM radars on (just turn on at save file start). The US could see the missiles up to 1412.3 km on ascent this time. Everything was the same as described above except...

4) The odd pitch down behavior at datalink loss was not replicated this time, even if the Tico could not see the missiles with its own sensors. The SM-3s continued to climb instead of suddenly descending.

5) The SM-3s reached the DF-26 with the same approximate altitude of the real-time missile track this time, but all missed. All but 1 or 2 still had large lateral deviations, and a 'Weapon Endgame' message was never sent for any of the missiles.

6) Terminal engagement behavior was not valid because of presence of two cruisers (situation design error, not game error).


These behaviors appeared on all the tests of this situation on my computer. Please let me know if you get similar results or if you have more ideas for testing. Thanks for any help!

sm3 test.zip

(40.95 KiB) Downloaded 15 times