[NO BUG] M-61 Vulcan and Gau-8 Avenger dispersion values - ED Forums
 


Notices

Reply
 
Thread Tools Display Modes
Old 09-05-2019, 08:56 AM   #1
nighthawk2174
Member
 
nighthawk2174's Avatar
 
Join Date: Mar 2014
Location: USA
Posts: 408
Default [NO BUG] M-61 Vulcan and Gau-8 Avenger dispersion values

This thread is all about all the information I’ve been able to gather that supports that the dispersion of the Gau8 and M61 in DCS are WAY WAY WAY too high.

Current values in DCS (Da0):

- 22 milliradians
- 17 milliradians

M61

Video 1:

https://www.military.com/video/guns/.../3732679073001

In this video a firing test in an aircraft resulted in dispersion of 95-99% falling just over 4milliradians

Video 2:

This video is of a test of an M61 vulcan

https://youtu.be/Ma4pa3DqA1s?t=4




As you can see the grouping is quite small but in order to get usable data out of this, we’ll have to do some math. Now from this still image we see that a bullet is about ~5x5px. Meaning that assuming (since this is cardboard) that the bullet hole is pretty much the size of the round we get 4mm/px or 0.1575 in/ px. Making the grouping 65x80px or 10.24x12.6in across.

From here we can now find the range, knowing that the grouping is just about 12.6 inches or 1.005ft across we can do some basic geometry and trig to get the range. Using the barrel diameter .6343ft and the approximation that if our camera was right behind the gun it would be 7.5ft away we can find the range to the target. This ends up resulting in a range around 100-115ft, which is quite reasonable here. This would mean a dispersion of 4.35-5.00 milliradians. Now this number makes the 5 milliradian values from the documentation bellow more than likely to be representative of active duty guns.


We get the range to be ~100ft leading to a dispersion value of ~4.35 milliradians.



Documentation:


From paper “Project Vulcan” tests of active duty M61 Vulcans









Of note Ship based Vulcan's have additional clamping to significantly reduce their base dispersion values. This is done as various studies had identified as 1 milliradian or less of dispersion to be the required value for gun based CIWS.




-These are from a paper disusing how to improve dispersion of the gun, and as far as I can tell the recommendations of the paper were put into place.

--
Breifly the
GAU-8]

Now knowing the approximate distance to the target from the M61 (100-115ft) we can give this gun a go. The grouping is x px. With a single bullet being 11x11 px meaning our scale is .1074 px/in. Taking this scale, we get the main grouping to be 115x82px or 12.3x8.8inches. Now that we know the size of the grouping and using some trig we get the dispersion to be: 4.45-5.12millirad for the largest dimension of the grouping.

Documentation:

https://ia802700.us.archive.org/6/it...e80007stol.pdf

In this doc we get several runs on M48 tanks, looking at the hits on the tanks, the firing ranges, and grouping on the tank would support the above dispersion values.

Last edited by nighthawk2174; 09-05-2019 at 08:59 AM.
nighthawk2174 is offline   Reply With Quote
Old 09-05-2019, 09:07 AM   #2
BIGNEWY
ED Community Manager
 
BIGNEWY's Avatar
 
Join Date: Aug 2011
Location: UK
Posts: 17,388
Default

Thanks for the information we will take a look.
__________________
Windows 10 Pro x64, NVIDIA MSI RTX 2080Ti VENTUS GP, INTEL i7 4790K @4.4 GHz( Cooled by H100i ), 32GB DDR3 @1866 , Asus Z97-AR, TM Warthog, Jet provost rudder pedals, VIVE Cosmos
BIGNEWY is offline   Reply With Quote
Old 09-05-2019, 11:09 AM   #4
toutenglisse
Member
 
toutenglisse's Avatar
 
Join Date: Jun 2016
Location: French Alps
Posts: 764
Default

From my little understanding :

The values you provide here are for a gun, or a plane+gun, that is attached to a Bench, locked.
The values in DCS (22 milliradians IF "Gau-8 : 0.0022" = 22 milliradians ?) are for an airborne gun, which can't be the same.

The Bench absorbs recoil and vibrations, not what an airborne plane can do. More obvious, if the stabilization system on A10c doesn't work the nose goes up when firing, giving a catastrophic dispersion.

For a better comparison you should provide airborne dispersion values from real tests in first place, which seems to me not possible ? (it's not tested this way for real ?) Or the comparison results would be aberrant ?

Click image for larger version

Name:	grouping.jpg
Views:	163
Size:	61.3 KB
ID:	216735

Nice grouping !? no ?


EDIT : Ok Ramsay - your post and the link you provide are clear, I was thinking wrong
__________________
Hardware : I7 7700K - Z270 gaming M5 - 32GB ( 4x8 ) ddr4 3200Mhz - 1080Ti gaming X - SSD M2 - TM Twcs / TM Rgt yaw pedals / MS sidewinder FF2 + TM Warthog - Oculus Rift CV1.

Last edited by toutenglisse; 09-05-2019 at 01:52 PM. Reason: in post
toutenglisse is offline   Reply With Quote
Old 09-05-2019, 11:42 AM   #5
Ramsay
Senior Member
 
Join Date: Aug 2012
Location: UK
Posts: 1,974
Default

Quote:
Originally Posted by toutenglisse View Post
From my little understanding :

The values you provide here are for a gun, or a plane+gun, that is attached to a Bench, locked.
The values in DCS (22 milliradians IF "Gau-8 : 0.0022" = 22 milliradians ?) are for an airborne gun, which can't be the same.
Planes in DCS are in motion, there is no need to add additional dispersion to account for effects that are modelled in DCS.

Also ED's DCS values for "some" ammunition match published specs., while others do not.

Quote:
IF "Gau-8 : 0.0022" = 22 milliradians ?
AFAIK,

• Da0 = 0.0022 does not equal 22 mil.

Dispersion in mils = Da0 (from shell table.lua) * 8 * 1000, so

• Da0 = 0.0022 = 0.0022 * 8 * 1000 mil = 17.6 mil

Source: https://forums.eagle.ru/showthread.p...53#post3956553

IRL 80% of the GAU-8's bullets should fall within a 5 mil circle.

For Info:

The 1962 M-61 test paper (page 24, December/January) lists 80%/5 mil => 100%/10 mil

• Da0 = 10 mil = (10 / 8 ) / 1000 = 0.00125

Last edited by Ramsay; 09-05-2019 at 12:14 PM.
Ramsay is offline   Reply With Quote
Old 09-05-2019, 12:10 PM   #6
TLTeo
Member
 
TLTeo's Avatar
 
Join Date: Mar 2018
Posts: 689
Default

The M-39 guns on the F-5 and M3 on the F-86 also show similar behavior iirc.
TLTeo is offline   Reply With Quote
Old 09-05-2019, 05:24 PM   #7
nighthawk2174
Member
 
nighthawk2174's Avatar
 
Join Date: Mar 2014
Location: USA
Posts: 408
Default

Quote:
Originally Posted by Ramsay View Post
Planes in DCS are in motion, there is no need to add additional dispersion to account for effects that are modelled in DCS.

Also ED's DCS values for "some" ammunition match published specs., while others do not.



AFAIK,

• Da0 = 0.0022 does not equal 22 mil.

Dispersion in mils = Da0 (from shell table.lua) * 8 * 1000, so

• Da0 = 0.0022 = 0.0022 * 8 * 1000 mil = 17.6 mil

Source: https://forums.eagle.ru/showthread.p...53#post3956553

IRL 80% of the GAU-8's bullets should fall within a 5 mil circle.

For Info:

The 1962 M-61 test paper (page 24, December/January) lists 80%/5 mil => 100%/10 mil

• Da0 = 10 mil = (10 / 8 ) / 1000 = 0.00125
First, there are no additional dispersion effects for planes in motion in DCS there is a person in another thread on this that compared the dispersion of a jet on the ground and in the air and they were identical. Additionally I highly doubt this has any impact at all on actual dispersion values. The amount of time an actual shell is in the barrel while accelerating and then the amount of time it would spend in the air flowing over the jet is just so unimaginably small that it has practically no impact. The reason for dispersion, as pointed out by the paper I linked, is the actual warping of the barrel due to the torque being applied to it. It causes the barell to both twist and warp up/down a very very very small amount but enough to imapct dispersion.
Me quoting these values as the actual miliradian values comes from in game tests:

https://www.dropbox.com/s/d1w8d8lsvq....zip.acmi?dl=0

In this ACMI you can see that I got my A-10 right at 2007 ft from the target. The resulting dispersion circle has a radius of 1.5 to 2 btr lenghs (25.7ft = length of a btr). It is harder to tell in the ACMI so I will see if I can either A) replicate the scenario and take screenshots or B) see if I took any and update this post latter. Additionally the GSH301 and GSH23 have a value set of .0005 and .0007 respectively. There is a whole fourm thread where testing was done to prove this I just haven't found it yet.



Last edited by nighthawk2174; 09-06-2019 at 05:58 AM.
nighthawk2174 is offline   Reply With Quote
Old 09-06-2019, 08:00 AM   #8
Yo-Yo
ED Team
 
Yo-Yo's Avatar
 
Join Date: Apr 2003
Posts: 14,916
Default

Quote:
Originally Posted by toutenglisse View Post
From my little understanding :

The values you provide here are for a gun, or a plane+gun, that is attached to a Bench, locked.
The values in DCS (22 milliradians IF "Gau-8 : 0.0022" = 22 milliradians ?) are for an airborne gun, which can't be the same.

The Bench absorbs recoil and vibrations, not what an airborne plane can do. More obvious, if the stabilization system on A10c doesn't work the nose goes up when firing, giving a catastrophic dispersion.

For a better comparison you should provide airborne dispersion values from real tests in first place, which seems to me not possible ? (it's not tested this way for real ?) Or the comparison results would be aberrant ?

Attachment 216735

Nice grouping !? no ?


EDIT : Ok Ramsay - your post and the link you provide are clear, I was thinking wrong

Before starting the discussion one needs to have an agreement about dispersion unit - what kind of units you are operating: 100% circle, 80% circle, 50% circle, standard deviation, median deviation and recall the ratios between them.

You are absolute right: the documents that were mentioned, as far as I can see, is for factory test bed or naval heavy duty mounts. Airframe mount can not be so rigid, so the overall dispersion indeed will be noticably more.
__________________
Ніщо так сильно не ранить мозок, як уламки скла від розбитих рожевих окулярів
There is nothing so hurtful for the brain as splinters of broken rose-coloured spectacles.
Ничто так сильно не ранит мозг, как осколки стекла от разбитых розовых очков (С) Me
Yo-Yo is offline   Reply With Quote
Old 09-06-2019, 01:46 PM   #9
nighthawk2174
Member
 
nighthawk2174's Avatar
 
Join Date: Mar 2014
Location: USA
Posts: 408
Default

Currently in DCS the value is for a 100% circle but all the rounds are evenly spread within that cricle. Its not like rlf where the 100% circle is determined by a few outlier rounds. Now it is just a pure guess that an aircraft mount can't be as rigid as a test bed. In the first video the tests were conducted with the gun in the actual aircraft. Just to also fully disprove this point:



here you can see the mounting mechanism for the M61 in the F15. It is every bit as extensive as the test bed if not even more so as it actually has a clamp near the muzzle where the test bench does not.

Last edited by nighthawk2174; 09-10-2019 at 07:18 AM.
nighthawk2174 is offline   Reply With Quote
Old 09-06-2019, 05:50 PM   #10
Yo-Yo
ED Team
 
Yo-Yo's Avatar
 
Join Date: Apr 2003
Posts: 14,916
Default

Quote:
Originally Posted by nighthawk2174 View Post
Currently in DCS the value is for a 100% circle but all the rounds are evenly spread within that cricle.
Absolutely wrong statement. Absolutely. In both parts.
__________________
Ніщо так сильно не ранить мозок, як уламки скла від розбитих рожевих окулярів
There is nothing so hurtful for the brain as splinters of broken rose-coloured spectacles.
Ничто так сильно не ранит мозг, как осколки стекла от разбитых розовых очков (С) Me
Yo-Yo is offline   Reply With Quote
Reply

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump

All times are GMT. The time now is 06:07 AM. vBulletin Skin by ForumMonkeys. Powered by vBulletin®.
Copyright ©2000 - 2019, Jelsoft Enterprises Ltd.