Jump to content

Can't get INS bombing right


mdee

Recommended Posts

Hey

 

Thought I'd ask for help here as I can't get INS bombing right - always off 10-20 meters and I am not sure what causes it.

 

Here is what I do

1. Aircraft ready to go on a runaway, started up.

2. Check theta/rho offset from designated IP waypoint. I've tried both to measure it on F10 map and in mission editor. The differences are minor.

3. Target is about 12 miles north from IP point.

4. Alt difference same as in #2 (waypoint alt - target alt)

5. Stick the data into INS as BAD waypoint in reference to IP waypoint. I am positive I am entering the data correctly

6. Set weapon release parameters etc. Take off.

8. About 3 miles to IP waypoint, Master Arm on, P1 selected (RS etc. is on too, the aircraft is configured correctly), INS update diamond visible on the HUD. "X" waypoint visible as well

9. Point the diamond at the landmark, press "target designate" button on hotas tu update INS. THe "X" on the HUD doesn't move. "VAL" lights up (I am not pressing it - going by training mission), "REC" lights up, but not "BAD" (according to training mission it should?).

10. After overflying IP I get CCRP wings guiding me to the target.

11. "X" on the HUD is usually about 20-30+ meters off and not lined up with the target

12. The bombs miss, consistently.

 

I am bit puzzled. This is supposed to be a precise method. My theory is that measurements in the editor aren't precise enough - if this is the case how can I measure the offset better?

 

Any other ideas?


Edited by mdee
Link to comment
Share on other sites

Can't help you on that one, I'll give it a go myself and see if I have better luck. I do remember a pilot of a different platform saying that this method of coordinate-based bombing isn't exactly loved in real applications because unless you have someone giving you an exact target position to the last metre, the bombs won't be accurate, and the fact that it uses two systems that are prone to errors (CCRP and INS) makes it impractical unless you're saturating a large area or target.

 

 

EDIT: Tried it in the Mirage 2000, I think I cracked it. The INS we use is accurate to a pretty decent extent, and it accepts decimal inputs for for distance and heading. However, our blessed little Mission Editor doesn't do decimals for degrees and the INS only takes two decimals for distance! I flew four attacks, two with Mk 82 and two with Snakeyes. In all of them, the target according to the INS was within 1000ft to either side of the intended target, but that's way too far for small bombs to still have an effect. Not sure if there's a way to get the decimals out of the heading properly. Has anyone tried it with coordinates instead of rho/theta?


Edited by Lucas_From_Hell
Link to comment
Share on other sites

Can't help you on that one, I'll give it a go myself and see if I have better luck. I do remember a pilot of a different platform saying that this method of coordinate-based bombing isn't exactly loved in real applications because unless you have someone giving you an exact target position to the last metre, the bombs won't be accurate, and the fact that it uses two systems that are prone to errors (CCRP and INS) makes it impractical unless you're saturating a large area or target.

 

Yeah.. after 50+ runs I am coming to the same conclusion.

 

A couple of things I've found

 

- VAL pressed like normal update works much better

- Toss the bombs (30 pull about 15 km to target)

- Have a big target and long interval :)

- Use a JTAC


Edited by mdee
Link to comment
Share on other sites

Example mission, Beslan-Terek-Nalchik bombing.

 

IP) Terek telephone pole on square bend in road

X-00123008

Z+00801061

817'

 

TGT) Nalchick control tower NW of midfield

X-00124614

Z+00760429

1411'

 

IP-to-TGT relative

dX -1606 (S)

dZ -40,632 (W)

dY +594'

 

BAD entry S01.606, W40.632, +594'

Theoretically my theta-rho are 267.74° 21.940nm but for example an error of 0.05° results in a 35m displacement at 40km range. It's not nearly as precise simply due to the limited number of digits the system will allow.

 

With drift disabled (special options, main menu) fine crosshair accurate to within 10m.

 

With drift I noticed that my IP point was not displayed on HUD where the IP was either before or after the temporary PI INU update (~100m). However despite not updating the IP visually, going on to the target did have the target point over the target to good precision <10m.

 

What is probably significant is that INU updates in Mirage module are not happening in the real way. They erase true position error instead of applying the update transformation honestly. This can be demonstrated by giving IP wrong coordinates, say 0.1' too much north. In real airplane if you fly over and designate properly this error doesn't matter as long as your offset is accurate. But in DCS the offset is also shifted 0.1'. DCS acted like your INU update diamond was exactly on coordinates and not wherever they actually were.

 

All in all, it's workable with the following suggestions:

1. Be as accurate as possible for your IP and realize that coordinate precision is a grid. Your landmark won't be exactly on these grid lines. There is a maximum precision available by manual entry. Dragging a waypoint over IP in mission planner will get precision that's simply impossible from cockpit entry. It's not possible for automatic BAD values so the limits for cockpit entry apply here always.

 

2. Use X-Z coordinates to find dLat/dLong in meters. It's much more precise than theta-rho by ruler.

 

3. For now, INU update quality doesn't matter as long as it's a valid permanent or temporary one but you should do one pretty soon before attack since drift is quite high per minute. The PI temporary update seems to work fine but do a permanent one as well (or validate the temporary) if you want.

 

4. Crosshair HUD positioning is pretty terrible. It doesn't display the center of the cross on the point but more like the bottom of the top segment. If it looks like the point is underground and parallax-ing, that's just how it is for now.

 

5. "Wing" HUD guidance is pretty terrible. Good luck with actual release.

 

6. 30m accuracy is excellent for this form of attack historically


Edited by Frederf
Link to comment
Share on other sites

CCRP release is not working properly currently using PI, so bombs are released when you press the trigger instead of being relased at calculated released point, making things worse when not using GBUs.

 

Not sure it's the cause of your problem though

 

I'm primarly using PI+Toss bombing with GBU 12 so it works pretty well

 

Also IRL, seems that RS isn't selected with TAS as it causes accuracy issue. But i'm pretty sure this is not implemented in DCS

Link to comment
Share on other sites

 

What is probably significant is that INU updates in Mirage module are not happening in the real way. They erase true position error instead of applying the update transformation honestly. This can be demonstrated by giving IP wrong coordinates, say 0.1' too much north. In real airplane if you fly over and designate properly this error doesn't matter as long as your offset is accurate. But in DCS the offset is also shifted 0.1'. DCS acted like your INU update diamond was exactly on coordinates and not wherever they actually were.

 

 

Was working fine initially, but indeed it's broken since a while now. That's a known bug.

Link to comment
Share on other sites

Frederf - thank you so much for this information! And, yes, INS updates issues - offset effect you described - spot on, noticed that too.

 

I've switched over to offset coords instead of polars and have much better rate of success now. I hope little quirks/bugs the system has now are going to be fixed. It's a very fun thing to do, especially when simulating low visibility conditions.


Edited by mdee
Link to comment
Share on other sites

For polar coordinates you can have a look here:

https://forums.eagle.ru/showthread.php?p=3481491#post3481491

 

There is a software used to calculate the values based on IP and target coordinates.

Only Caucasus so far.

 

40km seems to me too far for IP.

I would rather choose one around 10Nm...

Mirage fanatic !

I7-7700K/ MSI RTX3080/ RAM 64 Go/ SSD / TM Hornet stick-Virpil WarBRD + Virpil CM3 Throttle + MFG Crosswind + Reverb G2.

Flickr gallery: https://www.flickr.com/gp/71068385@N02/728Hbi

Link to comment
Share on other sites

Nothing wrong with 40km run in. At 0.5nm/h drift the time it takes to fly 40km is on order 20m drift. OK it's not 20km run in with ~10m drift but no big deal.

 

 

Don't trust Earth-to-grid coordinate transforms for DCS because DCS world isn't Earth shaped. I measure from 47N45E to 47N28E is 1287km with one of those online calculators but in DCS it is 1293km. X-Z coordinates

 

 

 

In XZ that's 00277381, 00812290 to 00219193,-00479611

Deltas 58188, 1291901

Distance 1293.2km

 

 

Best coordinate transform is whatever coordinate to XYZ grid (using the unique formulae for DCS) and then normal flat trigonometry.

Link to comment
Share on other sites

Nothing wrong with 40km run in. At 0.5nm/h drift the time it takes to fly 40km is on order 20m drift. OK it's not 20km run in with ~10m drift but no big deal.

 

 

Don't trust Earth-to-grid coordinate transforms for DCS because DCS world isn't Earth shaped. I measure from 47N45E to 47N28E is 1287km with one of those online calculators but in DCS it is 1293km. X-Z coordinates

 

 

 

In XZ that's 00277381, 00812290 to 00219193,-00479611

Deltas 58188, 1291901

Distance 1293.2km

 

 

Best coordinate transform is whatever coordinate to XYZ grid (using the unique formulae for DCS) and then normal flat trigonometry.

 

At least try the software before rejecting it.

 

It is purposely designed for DCS and takes into account DCS earth geometry (flat terrain).

 

I tried it with good results in polar coordinates mode.

Mirage fanatic !

I7-7700K/ MSI RTX3080/ RAM 64 Go/ SSD / TM Hornet stick-Virpil WarBRD + Virpil CM3 Throttle + MFG Crosswind + Reverb G2.

Flickr gallery: https://www.flickr.com/gp/71068385@N02/728Hbi

Link to comment
Share on other sites

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...