can not reproduce GBU38 misses targets

[tsgucci]

I'm playing the "persian gulf freedom" capmaign by ground pounder sim. Already with 2 missions I experience that JDAM's (GBU38) misses its target by several hundred feets.

Wind data is entered into the LASTE (CBU97 drops accurately). THe bombs does not fall short, they fall usually port side of the target. Bombs are dropped when MAN REL mode is open.

 

Any idea?

Edited August 1, 2021 by tsgucci

[BIGNEWY]

Please include a short track replay showing the problem. 

 

thank you

[Foka]

JDAM's autopilot is messed up for months, ED keep asking "send track" and do nothing with it.

[Jafferson]

LASTE Page bug was reported a couple of years ago for the first a10c module. Never noticed they ever changed something.

Ignore the LASTE page and your bombs should have good hits.

 

Maybe it's not a bug? it's a long time ago but I remember a statement, there is no need for enter wind data, cause the CDU is taking data during the flight. Manual enter for backup.

But I'm not sure about that, please correct me if I'm wrong.

Edited August 2, 2021 by Jafferson

[BIGNEWY]

2 hours ago, Foka said:

JDAM's autopilot is messed up for months, ED keep asking "send track" and do nothing with it.

 

GBU-38 Was fixed, which is why I am asking for a track. Please keep the report posting useful information. 

 

I know we say it a lot but unless we see a track replay example it is difficult to tell what is going on, with a track we can see the same circumstance as you and take control. 

 

I have tested GBU-38 I am getting hits with no problem. 

 

thanks

[jaylw314]

19 hours ago, tsgucci said:

I'm playing the "persian gulf freedom" capmaign by ground pounder sim. Already with 2 missions I experience that JDAM's (GBU38) misses its target by several hundred feets.

Wind data is entered into the LASTE (CBU97 drops accurately). THe bombs does not fall short, they fall usually port side of the target. Bombs are dropped when MAN REL mode is open.

 

Any idea?

 

Make sure the appropriate sensor is SPI.  If the SPI is still STPT, they will target the steerpoint instead of TGP

The LASTE shouldn't affect JDAMs, they have their own GPS and will aim for the point on the ground.  proper LASTE setup might give you a more accurate max range display by a little bit, though, since throwing the bombs with some wind correction saves them a small amount of energy during guidance.

[tsgucci]

I'm sending a track and a tacview file. Persian gulf map, 4 targets. SW BMP was targeted. GBU38 missed by 10-15 meters.

I have made several runs. The target miss happens not all the time. I have seen also perfect hits. But something is really off, because it used to be much more accurate. Wind does not seem to be have an influence, LASTE neither.

 

 

I have made short video about it as well:

 

 

 

The bug I discovered when I was playing with the Persian Freedom Campaign. At mission 8 I try to drop it. I've made a video about it. The first GBU misses the target by 5-10 meters. This you cannot see in the video, but you can see the black crater next to the target. When droping the 2nd GBU, it misses 100-150 meters.

 

 

Thanks for the support!

 

GBU38 bug big miss.trk Tacview-20210806-184739-DCS-GBU38 bug.zip.acmi

Edited August 6, 2021 by tsgucci

[Yurgon]

12 minutes ago, tsgucci said:

I'm sending a track and a tacview file.

 

Excellent, thanks!

 

Looking at the track, everything looks good to me and the GBU-38 should have hit the target dead in the middle.

 

I ran the track again and took control to drop the bomb a few seconds after coming into range, and it missed by almost exactly the same margin. There used to be a problem where JDAMs released right at max range would miss, but this doesn't seem to be the case here. And in any case, in the track as well as in the first video, the bomb overshot the target, so it doesn't look like a range or energy issue.

[EasyEB]

19 hours ago, Yurgon said:

 

Excellent, thanks!

 

Looking at the track, everything looks good to me and the GBU-38 should have hit the target dead in the middle.

 

I ran the track again and took control to drop the bomb a few seconds after coming into range, and it missed by almost exactly the same margin. There used to be a problem where JDAMs released right at max range would miss, but this doesn't seem to be the case here. And in any case, in the track as well as in the first video, the bomb overshot the target, so it doesn't look like a range or energy issue.

Any wind?

[Yurgon]

10 minutes ago, EasyEB said:

Any wind?

 

If only there was a track file that would allow you to answer that question...

 

Edited August 7, 2021 by Yurgon

[tsgucci]

I do not know if this is relevant but I have realised the different behavior of the GBU38 already on the hornet when the GBU24 was introduced. Since that GBU38 has a tendency to come relatively high over the target and in the terminal phase it pitches down agressively, it looks like the bomb falls completely vertical perpendicular to the target, like it would be a bunker buster bomb. Earlier it always had a shallower terminal phase and the target was hit by the side. I thought with the GBU24 that was also changed so I thought it is normal.

And on the hornet 1 out of 20 cases the bomb falls completely off target, usually falls short. I use to take 8xGBU38, and drop it one-by-one in the same run-in (search for target, SCS designate, drop one, then repeat). Usually one bomb misses the target but completely wrong place.

[Hammer1-1]

On 8/2/2021 at 8:47 AM, BIGNEWY said:

 

GBU-38 Was fixed, which is why I am asking for a track. Please keep the report posting useful information. 

 

I know we say it a lot but unless we see a track replay example it is difficult to tell what is going on, with a track we can see the same circumstance as you and take control. 

 

I have tested GBU-38 I am getting hits with no problem. 

 

thanks

and half the time the tracks dont even play right which is a big part of the overall problem. when are they going to cut that stalactite down?

Edited August 7, 2021 by Hammer1-1

[BIGNEWY]

1 hour ago, Hammer1-1 said:

and half the time the tracks dont even play right which is a big part of the overall problem. when are they going to cut that stalactite down?

 

 

Lets not go off topic please, but I deal with track replays all day everyday, most of them work for bug reporting if they are kept short, dont have user mods or loads of scripts. Its important to remember the track replay system only replays inputs, nothing else, so AI may react differently, or turbulence if enabled my be slightly different.

 

 

[BIGNEWY]

On 8/6/2021 at 6:03 PM, tsgucci said:

I'm sending a track and a tacview file. Persian gulf map, 4 targets. SW BMP was targeted. GBU38 missed by 10-15 meters.

I have made several runs. The target miss happens not all the time. I have seen also perfect hits. But something is really off, because it used to be much more accurate. Wind does not seem to be have an influence, LASTE neither.

 

 

I have made short video about it as well:

 

 

 

The bug I discovered when I was playing with the Persian Freedom Campaign. At mission 8 I try to drop it. I've made a video about it. The first GBU misses the target by 5-10 meters. This you cannot see in the video, but you can see the black crater next to the target. When droping the 2nd GBU, it misses 100-150 meters.

 

 

Thanks for the support!

 

GBU38 bug big miss.trk 848.06 kB · 2 downloads Tacview-20210806-184739-DCS-GBU38 bug.zip.acmi 35.3 kB · 1 download

 

 

When you target try using point track and not area track and let me know if you see a difference. I have tested and all seems ok to me. 

[Yurgon]

1 hour ago, BIGNEWY said:

When you target try using point track and not area track and let me know if you see a difference. I have tested and all seems ok to me. 

 

I took control from OP's track a bunch of times and set the TGP to point track. Every single time the GBU-38 goes long. Most of the time, it'll at least kill the southern most BTR-80, but in one out of 3 attempts, it only damaged it (rather severely, but it was not a k-kill); new track attached.

GBU38 bug big miss 2.trk

[Frederf]

Point track is going to have a target location which is the intersection of the TGP LOS and the DTSAS surface (assuming that's what's providing elevation/slant data). I expect the target to be slightly long as a function of the height of the target center above terrain and the grazing angle.

 

If you're inside 8nm and laser rangefinder is latched on during JDAM release does accuracy improve?

[Yurgon]

19 hours ago, Frederf said:

Point track is going to have a target location which is the intersection of the TGP LOS and the DTSAS surface (assuming that's what's providing elevation/slant data).

 

Right. That sort of implies that AREA track does something different in this regard, but that's not the case, is it? I mean, whether the TGP is using POINT track or AREA track, it's always going to use either laser ranging when available, or LOS DTSAS ground intersection, right?

 

19 hours ago, Frederf said:

I expect the target to be slightly long as a function of the height of the target center above terrain and the grazing angle.

 

I don't think the slant angle of OP's setup (and my flying from his track) would account for the bomb falling so long that it's outside the killzone for a BTR-80. If the bomb was consistently one or two meters long, that would be a perfect explanation. But unless the DTSAS database has wrong elevation data, I see no reason for the bomb to consistently go as long as it does in this setup.

 

19 hours ago, Frederf said:

If you're inside 8nm and laser rangefinder is latched on during JDAM release does accuracy improve?

 

Haven't tested that.

Edited August 8, 2021 by Yurgon
Typo

[jaylw314]

21 hours ago, Yurgon said:

 

I took control from OP's track a bunch of times and set the TGP to point track. Every single time the GBU-38 goes long. Most of the time, it'll at least kill the southern most BTR-80, but in one out of 3 attempts, it only damaged it (rather severely, but it was not a k-kill); new track attached.

GBU38 bug big miss 2.trk 668.44 kB · 1 download

 

Wait, how do you take control from a track file???

[Frederf]

AREA and POINT should operate on the same logic and should come to the same conclusion for target point when using the same LOS. This is easy to test by making a mark point in both modes and comparing the numbers. I have a tendency to area track the ground underneath the target due to knowing the process.

 

I've played with JDAMs and found that in freeze with infinite ammo and dropping several bombs found that there's a rather wide natural dispersion. Even with identical target and release parameters the bomb has a rather large CEP, 8m or so.

 

Press the escape key and there is a "TAKE CONTROL" option in the menu. You can also go to manual time acceleration or view control by adjusting either during playback.

[Foka]

16 hours ago, jaylw314 said:

 

Wait, how do you take control from a track file???

ESC -> take control

[Tippis]

On 8/7/2021 at 9:48 PM, tsgucci said:

I do not know if this is relevant but I have realised the different behavior of the GBU38 already on the hornet when the GBU24 was introduced. Since that GBU38 has a tendency to come relatively high over the target and in the terminal phase it pitches down agressively, it looks like the bomb falls completely vertical perpendicular to the target, like it would be a bunker buster bomb. Earlier it always had a shallower terminal phase and the target was hit by the side. I thought with the GBU24 that was also changed so I thought it is normal.

 

This sounds an awful lot like it's trying follow some pretty warped interpretation of the terminal guidance parameters. There have been these on-again-off-again issues where you've not been able to clearly set INS-guided weapons to just go straight in — they act as if you always (well… “always,” but sporadically, and for no clear reason) have some pre-set azimuth and final attack angle even when those settings are not touched.

 

Maybe there's some logic loop in there that just goes off the rails and suddenly decides that the default no az/no fall angle actually means “90° drop, attack from the north” and then goes bananas due to the trigonometry not working properly for that combination of extremes?

[Hammer1-1]

Supposedly theres an argument the A-10C is able to select the profile the GBU38 has upon impact, and one of those profiles is a high energy nose down profile, vs a flat trajectory profile. Im not 100% sure of this, but the -18 doesnt have the ability (at the moment at least) to select a different profile like the A-10C can.

[tsgucci]

I have re-tried the drop with point track (tbh never had different experience in point or area track). It missed the target with approx 25 meters. I am attaching the track.

 

 

GBU38 bug point track 25m miss.trk

[Frederf]

Bomb simply runs out of energy. It's going 250KTAS with a left miss until endgame. At a certain distance it switches into precision mode and can't make the turn. It misses more to the left than long/short. The JDAM guidance is just being poopy or the LAR is not conservative enough. It has a good target location info it just can't hit it when it's at the edge of kinematic range. That's definitely a problem with modeling, releasing exactly at max range should have normal accuracy.

 

Attached is same situation but waiting until about 4.0nm.

GBU38 bug point track 25m hit.trk

[Khallimero]

Maybe GBU-54s in GPS mode suffer the same issue. (sneaky topic update )

But as far as I tested, GBU-38s seemed to work fine once you changed the default 0° impact angle. Unlike GBU-54s.