WWII gun jams/misfires (research needed for W.I.P. mod)

[Avimimus]

Hello,

 

I'm working on a possible upcoming mod that would add the WWI misfire mechanic to WWII. While the misfire mechanic is pretty simplistic and data is hard to come by, it should be an improvement to simulate the fact that cannons were not completely reliable. However, I need help with research.

 

I can produce two effects:

- Average number of rounds between misfires

- Average number of seconds of firing prior to an increased misfire chance

 

Firing a long burst won't guarantee a failure (there will still be a chance of 'getting away with a long burst') - but keeping a burst below these lengths will make failures very unlikely.

 

Information so far:

- 50mm Bk-5 pilots were instructed to keep bursts below 5 rounds (so approximately four seconds)

- 57mm Molins pilots were instructed to keep bursts to four or less rounds (so approximately four seconds)

- 20mm Hispano II (70% probability of misfire with 30 round burst - so less than three seconds)

- 0.303 Browning (less than four seconds)

- 0.50 Browning (less than three seconds)

 

Feel free to correct the above if it is incorrect.

 

Reputation will be used when such information isn't available, for example:

Mk-103 more reliable than Mk-101 more reliable than Mk-108 (e.g. bursts of 5 seconds, 3 seconds, 2 seconds)

Mg-151/20 more reliable than MG-FF/M more reliable than MG-FF

Hispano Mk.V more reliable than Hispano Mk.II

ShKAS relatively unreliable

 

So feel free to suggest numbers even if you can't back them up.

 

Note: Unfortunately we cannot simulate things like g-forces disrupting feed mechanisms, but we can still introduce some unreliability. Also let me know if you want the guns to be able to be cleared in flight or remain disabled for the rest of the battle once jammed! I might be able to do this differently for different types of guns.

[Stonehouse]

3 hours ago, Avimimus said:

I can produce two effects:

- Average number of rounds between misfires

- Average number of seconds of firing prior to an increased misfire chance

 

Is that the bulletMisfireData and overheat limit parameters associated with barrel temperature or something else?  Nifty idea if you can get it working well. I noticed previously that it seems like those two lines seem identical across most WW2 fighter weapons. The 0.02 becomes 0.03 on some WW1 guns though.

 

eg mg151-20

BarrelTemperatureIncreasePerShot = 5.657
BarrelTemperatureDecreaseCoefficient = 0.045

BulletDispertionAngleData = 0.0, 0.6, 700.0, 3.5 //разброс зафиксированной пушки при 0К нулевой, разброс перегретой 0.6°
BulletSpeedReducingData = 0.01, 700.0, 0.25, 0.05, 3.5
BulletMisfireData= 0.0, 0.02, 280.0, 700.0, 2.0 // 280K - the temperature at which a techno-message appears (see OverheatLimit)

CartridgeCasesEjectionPosition = -1.320, -0.053, 0.0
CartridgeCasesEjectionSpeed = 0.0, -2.0, 0.0

OverheatLimit = 0.4f // 0.4=700/280

[Avimimus]

Yes. BulletMisfireData is the main value (although I was also going to adjust the Barrel temperature increase and decrease to allow building up and bleeding heat faster... to ensure that short bursts don't accumulate enough to lead to a failure).

 

BulletMisfireData= probability per round at heat level 1; probability per round at heat level 2; heat level 1; heat level 2;

 

There is also a fifth value on that line (2.0 in the example you chose) which is commented as "четвёртое число - степень зависимости угла рассеивания пуль (снарядов) относительно оси ствола стрелкового оружия от абсолютной температуры ствола стрелкового оружия, [безразм], (>=1)"

 

Make of that what you will - and then let us all know what it means  I've been puzzled by it.

 

P.S. OverheatLimit might be the maximum possible heat level, a hard cap - but I haven't tested that. The comment suggests it is hand calculated from a ratio of temperature level 1 and temperature level 2.

[Stonehouse]

1 hour ago, Avimimus said:

Yes. BulletMisfireData is the main value (although I was also going to adjust the Barrel temperature increase and decrease to allow building up and bleeding heat faster... to ensure that short bursts don't accumulate enough to lead to a failure).

You probably already know but don't forget that temps feed into dispersion. So if you change the temp increase per round and cooling rate you will end up impacting accuracy as well as misfires. Just my opinion only but I would try misfire rate first................

 

Interesting that translates to talking about the 4th number. Guessing you got it from somewhere like RoF as I don't see that comment in the files I have looked at in IL2?

 

Anyway, the best translation I get with a variety of engines is:

 

the fourth number - the degree of dependence of the scattering angle of bullets (shells) relative to the axis of the gun barrel on the absolute temperature of the gun barrel, [indiscriminate or iron not sure which is better. 1st may be talking in maths terms while iron may refer to the material of the barrel the formula the algorithm behind these parameters is based on], (>1)

 

So, I read that as the degree of dependency between dispersion angle relative to the barrel axis and barrel temp is a factor of 2 (in my example from the MG151/20).  There would be a formula behind the scenes that for a given barrel temp and that 5th factor derives the angle of dispersion.

Edited January 18, 2023 by Stonehouse

[Raptor341]

This should be standard in the simulation already - looking forward to seeing the mod in action. As far as I am aware unless you can re-cycle the action then the gun in question should remain jammed for the rest of the mission.

[Stonehouse]

Just to be clear I believe Il2 does not model jams but only misfires due to overheating

 

I think the overheatlimit = 0.4 may just be saying that once you get the barrel temp past the ratio of the two temps display the overheat message. So, once you get past temp1/temp2 of temp 2 display the warning. So in my example issue the message once past 280K. Maybe it was once a variable value instead of a hardcoded one or perhaps it is just better performance wise to not have to do the calc for each round fired (ie each temp change). Note the comment against it is upside down. 700/280 actually is 2.5.

 

Edited January 18, 2023 by Stonehouse

[kraut1]

Hi,

if you need an example for overheating, please check the zip for the Spit Ib and the BF109-F1, both to simulate unreliable cannons according to the real issues with the early Hispano and the MG-FF installed in the shaft of the engine.

 

[Stonehouse]

23 minutes ago, kraut1 said:

Hi,

if you need an example for overheating, please check the zip for the Spit Ib and the BF109-F1, both to simulate unreliable cannons according to the real issues with the early Hispano and the MG-FF installed in the shaft of the engine.

Ok you've made the temperature increase per shot fired extremely large with the original 5 point something altered to 300. You are right and this will cause misfires often, but it also makes the weapons almost immediately very inaccurate as the dispersion would go off the scale with perhaps only a few rounds fired. I honestly don't think it's the right approach and it would be better to increase the chance of misfires if that could be made to work. 

[kraut1]

2 hours ago, Stonehouse said:

Ok you've made the temperature increase per shot fired extremely large with the original 5 point something altered to 300. You are right and this will cause misfires often, but it also makes the weapons almost immediately very inaccurate as the dispersion would go off the scale with perhaps only a few rounds fired. I honestly don't think it's the right approach and it would be better to increase the chance of misfires if that could be made to work. 

Yes, this was in 1940 a very special situation with these both weapons. I agree with you that for a good working weapon the values are much too high.

 

[Avimimus]

13 hours ago, Stonehouse said:

Just to be clear I believe Il2 does not model jams but only misfires due to overheating

 

I think the overheatlimit = 0.4 may just be saying that once you get the barrel temp past the ratio of the two temps display the overheat message. So, once you get past temp1/temp2 of temp 2 display the warning. So in my example issue the message once past 280K. Maybe it was once a variable value instead of a hardcoded one or perhaps it is just better performance wise to not have to do the calc for each round fired (ie each temp change). Note the comment against it is upside down. 700/280 actually is 2.5.

 

 

Yes, indeed. Upside down - but I think the intent was 280/700? Maybe it is something like reverse Polish notation (or just the programmer being tired when they wrote the note)?

 

10 hours ago, kraut1 said:

Hi,

if you need an example for overheating, please check the zip for the Spit Ib and the BF109-F1, both to simulate unreliable cannons according to the real issues with the early Hispano and the MG-FF installed in the shaft of the engine.

 

I was actually going to cite you in the readme of the released mod - as you're mod tipped me off that this was possible.

 

That said I'll probably use a different technique - focussing on 'BulletMisfireData' rather than 'BarrelTemperatureIncreasePerShot' as focusing on the former should allow more precise adjusting of probabilities, and creating two probability (as baseline failure rate using the first and third columns, and an overheat/burst failure rate using the second and third columns).

 

 

13 hours ago, Raptor341 said:

This should be standard in the simulation already - looking forward to seeing the mod in action. As far as I am aware unless you can re-cycle the action then the gun in question should remain jammed for the rest of the mission.

 

Indeed - and as Stonehouse mentions the sim is only set up to handle misfires rather than proper jams.

 

Currently the player can recharge/cycle the guns by hitting 'r'. However, I may be able to disable this for some weapon types by adjusting the animation times in the file to very large values (e.g. ten minutes) ensuring that the weapon wouldn't re-activate while in the same fight at least. I could probably also make a gun jam mod for Flying Circus using a similar technique to replace misfires with more severe jams.

 

If anyone has data about which guns could be cycled while in flight - feel free to share that information as well!