SA-2 LN hitbox issue

[LetMePickThat]

Hi,

 

The hitbox for the SA-2 launcher (SM-90) seems to be way too big, extending both on the sides and above the 3D model. This can be seen when firing HARM at the central radar, using the default SA-2 template: most missiles collide with the nearest launcher's hitbox, despite clearing it by several meters visually. Check these screenshots, as well as the saved track below, showing the issue in detail.

 

https://imgur.com/Y1z64Mm

https://imgur.com/Y1z64Mm

 

 

SA-2_hitbox_issue.trk

Edited March 5, 2021 by LetMePickThat

[LetMePickThat]

Another example:

 

https://imgur.com/PN5hRtu

 

https://imgur.com/4pK7kJo

 

https://imgur.com/8rro4sB

[Flappie]

Hi. The SA-2 hitbox is perfectly OK (see attached track).

 

It is the force of a powerful explosion next to it that ignites it.

SA2_hitbox_is_OK.trk

[LetMePickThat]

On 3/5/2021 at 11:26 PM, Flappie said:

Hi. The SA-2 hitbox is perfectly OK (see attached track).

 

It is the force of a powerful explosion next to it that ignites it.

SA2_hitbox_is_OK.trk 123.1 kB · 2 downloads

Thank you for your answer. Do you have an idea about what is happening in these clips? The HARM aren't touching anything, yet they explode when close to the SA-2 launchers for seemingly no reason. Removing the launchers makes the HARM go all the way to the targeted radar as intended. With the launchers in place, the radars suffered no damage. Both missiles were heading for their targets at the moment of impact/detonation.

 

 

 

Edited March 7, 2021 by LetMePickThat

[Jarlerus]

Could be an issue with the HARM code as well?
Maybe it has some kind of proximity-sensor, and explodes when it passes over the launcher?

Try to replicate the trajectory with, say, a Maverick - and see if it also destroys the launcher instead of the radar.

Edited March 7, 2021 by Jarlerus

[Wdigman]

I noticed this on the SA6 as well. The HARM detonates by proximity. However even though no damage model appeared the SAM system did go down. I don't even think there was a percent damaged either.

 

I will not pretend to know how a HARM really works but not everything needs to be a direct hit to disable a system.

 

In real life will a HARM try to hit the radar dish/ Dome?

[Flappie]

27 minutes ago, Jarlerus said:

Could be an issue with the HARM code as well?
Maybe it has some kind of proximity-sensor, and explodes when it passes over the launcher?

 

It would be weird if the HARM would have a proximity sensor when A/A missiles don't, right? Yet, the video is pretty explicit. I'll make some tests next week.

[Northstar98]

The HARM IRL has a FMU-111/B laser proximity fuse - I'm not sure what the range is but in DCS it seems to be set at 7 metres. Looking at the model, the aperture for the laser faces downwards (there looks to be a downward facing window in roughly the missile's mid-section, can't see anything else).

 

In this video it shows the HARM coming it at much steeper angles to hit targets, perhaps alleviating the issue. (Timestamped at where the proximity fuse is mentioned).

 

 

Spoiler

 

I think the problems related to the HARM is maybe how DCS models warheads - DCS doesn't have any fragmentation modelling at all from what I can find, what it does do is add a modifier to warheads to approximate fragmentation - in missiles with large fusing distances DCS' inflates the warhead size to something 5x IIRC to achieve something desireable.

 

The other issue is the damage modelling and how the HARM guides in DCS. Firstly, ground units do not have component or subsytem level damage modelling; the HARM has a blast fragmentation warhead which should be very good at shredding antennae, even if the unit it's mounted to isn't totally destroyed. What probably happened in @Wdigman  example is that the HP of the unit dropped below a threshold when the sensors of a unit stop operating, even though the unit appeared as if it wasn't damaged (DCS also only has 2 states for graphical damage of ground units - fully functional or completely destroyed - unlike aircraft which typically have multiple stages depending on severity).

 

As for HARM guidance, it doesn't actually home in on antennae themselves, it instead just goes for the geometric centre of whatever unit it's targeting. This is almost certainly down to DCS' lacklustre implementation of AI RADARs, which I could go on for ages about. This is particularly apparent when fired at ships (and yes, in the video above, targets on the water are sometimes used). 

 

 

Edited March 7, 2021 by Northstar98

[LetMePickThat]

15 minutes ago, Northstar98 said:

The HARM IRL has a FMU-111/B laser proximity fuse - I'm not sure what the range is but in DCS it seems to be set at 7 metres. Looking at the model, the aperture for the laser faces downwards (there looks to be a downward facing window in roughly the missile's mid-section, can't see anything else).

 

In this video it shows the HARM coming it at much steeper angles to hit targets, perhaps alleviating the issue. (Timestamped at where the proximity fuse is mentioned).

 

 

  Reveal hidden contents

 

I think the problems related to the HARM is maybe how DCS models warheads - DCS doesn't have any fragmentation modelling at all from what I can find, what it does do is add a modifier to warheads to approximate fragmentation - in missiles with large fusing distances DCS' inflates the warhead size to something 5x IIRC to achieve something desireable.

 

The other issue is the damage modelling and how the HARM guides in DCS. Firstly, ground units do not have component or subsytem level damage modelling; the HARM has a blast fragmentation warhead which should be very good at shredding antennae, even if the unit it's mounted to isn't totally destroyed. What probably happened in @Wdigman  example is that the HP of the unit dropped below a threshold when the sensors of a unit stop operating, even though the unit appeared as if it wasn't damaged (DCS also only has 2 states for graphical damage of ground units - fully functional or completely destroyed - unlike aircraft which typically have multiple stages depending on severity).

 

As for HARM guidance, it doesn't actually home in on antennae themselves, it instead just goes for the geometric centre of whatever unit it's targeting. This is almost certainly down to DCS' lacklustre implementation of AI RADARs, which I could go on for ages about. This is particularly apparent when fired at ships (and yes, in the video above, targets on the water are sometimes used). 

 

 

 

 

Removing the launchers in DCS makes the HARM go all the way to the radar, with no pre-detonation, so it might be a proximity fuse issue.

I'll try to mod the KillDistance value for the AGM-88 from 7.0 to 1.0 and report on the changes.

[Northstar98]

12 minutes ago, LetMePickThat said:

Removing the launchers in DCS makes the HARM go all the way to the radar, with no pre-detonation, so it might be a proximity fuse issue.

I'll try to mod the KillDistance value for the AGM-88 from 7.0 to 1.0 and report on the changes.

 

In my limited tested, it seems that the HARM is likely to detonate on something unintended when coming in at a shallow angle - trees, building etc, even if it doesn't directly hit them - so yes, I think it's the proximity fuse being triggered.

 

EDIT: Think I've found another issue, DCS models the fuze as being spherical around the missile; anything within 7m will trigger the fuse. However, it looks like the HARM (just going by the in-game model) only has a window for the laser on the bottom, around the missile's mid-section, so it should only be measuring distances along essentially a line coming out of the bottom of the missile. I imagine that proper implementation of the fuze would probably necessitate something like raycasting. 

Edited March 7, 2021 by Northstar98