Flash fire frequency

[brucesim2003]

I'm thinking that these happen too often. Just did a battle against an AI Chinese fleet. They had 4 battlecruisers with their main armament in a hexagonal layout ala the Nassau class. 3 ships had flash fires: 1 had one, 1 had 3, and the 3rd had 5 turrets cook off. Even accounting for the AI having thin belts (8" iirc), and minimal bulkheads, 5 out of 6 turrets is a bit much I think.

I've never had a flash fire on my designs, but the AI seems VERY prone to them.

[Cptbarney]

Ye, they seem to be more promenant than magazine explosions. Too be honest the main factors for flash fires should be a combo of ignited ammo, poor or obstructed damage control, fires both external and internal, and maybe additional shells and explosions.

I guess because everything is basic atm it will take some time before we get proper damage models, animations and the amrour rework most have talking about (myself included) on the forums.

[Norbert Sattler]

From my observations the propellant/explosives seem to play a big part in the flashfire chance. With Lyddite shells flash fires happen a lot more often than with more stable explosives.

And right now the AI seems to be very fond of Lyddite.

 

[Hussarball_PL]

As far as I have noticed, flash fires on DD's is nothing uncommon, in DD vs DD scenario. Well it might have something to do with AI ship design (some main guns and 3 lesser caliber guns broadside). But i'm not so sure if 3 inch gun ammunition would be able to send turret flying.

[Nick Thomadis]

This is going to be improved in the next update. Flash fires should happen more likely on ships with large main guns and ammo storages, while on DD they will not happen as often.

[DougToss]

18 minutes ago, Nick Thomadis said:

This is going to be improved in the next update. Flash fires should happen more likely on ships with large main guns and ammo storages, while on DD they will not happen as often.

Is that the difference between bagged and QF charges? Does mechanical versus handling play a role?

[madham82]

Need to keep in mind, flash fires are created by the same type of hit that would create a regular fire. The difference is what is hit. If a shell hit creates a "flash" in ammo storage area (i.e. the handling rooms under turrets and the magzines) it has the potential (based on the type of explosives/powder) to ignite all of it.

Without making it too complex, if a hit happens in handling rooms, that would likely destroy turret. If it happens in the magazine, you're looking at destroying the ship. Now that is oversimplified because there are lots of factors to determine how big the explosion is and damage, chaining to other magazines, etc...

Nick will have to provide more details as to how they are calculating it. DDs technically should not be any less susceptible to turret explosions from flash, but as Nick implied they don't carry as much ammo either. Also over-pens are more common which would not cause a flash directly. 

As for mitigation techs, propellant would be the biggest factor. Followed by armor scheme, barbette thickness, and the amounts of armor (specifically belt, turret top and sides, and deck). In reality, bulkheads should not impact this since we are talking flash fire caused by a pen. Now fire that spreads inside the ship is another story and bulkheads would matter for the ability to contain fires. 

[Wowzery]

Did a video and created the Weymouth class cruiser, it was like the fourth of July fireworks as my turrets flew off.

[Skoggatt]

On 5/21/2020 at 11:52 AM, madham82 said:

Need to keep in mind, flash fires are created by the same type of hit that would create a regular fire. The difference is what is hit. If a shell hit creates a "flash" in ammo storage area (i.e. the handling rooms under turrets and the magzines) it has the potential (based on the type of explosives/powder) to ignite all of it.

Without making it too complex, if a hit happens in handling rooms, that would likely destroy turret. If it happens in the magazine, you're looking at destroying the ship. Now that is oversimplified because there are lots of factors to determine how big the explosion is and damage, chaining to other magazines, etc...

Nick will have to provide more details as to how they are calculating it. DDs technically should not be any less susceptible to turret explosions from flash, but as Nick implied they don't carry as much ammo either. Also over-pens are more common which would not cause a flash directly. 

As for mitigation techs, propellant would be the biggest factor. Followed by armor scheme, barbette thickness, and the amounts of armor (specifically belt, turret top and sides, and deck). In reality, bulkheads should not impact this since we are talking flash fire caused by a pen. Now fire that spreads inside the ship is another story and bulkheads would matter for the ability to contain fires. 

While bulkheads in theory shouldn't reduce the chances of fires, in the game the bulkhead level represents the overall internal protection and damage control (for better or for worse). More advanced ships had better protection against flash fires, including both fireproof doors and valves to flood magazines should fire break out which should be represented either by the bulkheads or by another option in the ship constructor.

[Steeltrap]

The visual effect is clearly one of propellant charges igniting en masse, as is reckoned to have happened with HMS Hood, for example.

To my mind that sort of effect should NOT happen on any warship that doesn't have 2-part ammunition, as, unless I'm much mistaken, a brass round isn't typically vulnerable to that sort of flash effect.

Equally, seems silly for it to happen to an open mount hit.

The 'famous' flash fires came from main turret hits and the subsequent bursting charge making an explosion that travelled down the trunk of a main gun turret. Poor storage discipline can increase the chances.

A fire that reaches a magazine, however, ought to result in an explosion befitting the ammunition. For most ships, their main gun ammo magazine exploding would likely cause catastrophic loss of the vessel.

As it is, turrets flying left, right and centre on jets of blue flames like some sort of infernal champagne corks might look amusing, but they're pretty damn dopey ON A TRANSPORT, or indeed any small warship without propellant charges.

If it's a magazine explosion, it ought to be fatal. If it's a ship without propellant charges, there shouldn't BE such a thing other than a magazine explosion.

I commented elsewhere I thought putting this in while we/the AI have no means of mitigation seemed a pretty silly idea, and that creating effects that don't make much sense re a "propellent burn jet" on a ship without separate propellant isn't perhaps the greatest appeal to 'accuracy',  but what do I know.

[Jatzi]

I just thought it was high for testing purposes. Also the armor system isn't done. Barbettes and citadels and whatnot are getting reworked afaik so that'd probably help a lot with it, same with crew mechanics whenever that comes into play. I do agree it looks ridiculous in most situations and should be reworked a little. Also like have fatal magazine detonations just disappeared? I've seen secondary ammo dets still plus torpedo magazine hits but main gun magazine hits just seems to result in flash fires now and the turret blowing off

[DougToss]

There's a lot of granularity here that depends on improvements to the internal modelling of a ship and performance of her crew. There are so many interconnected systems, each of which plays a part in the attendant risks and rewards of the others. If turret and barbette compartmentalization and safe shell handling does not effect rate of fire, there is no reason for player and AI not to use them. Whereas, the British did not forgo safety procedures because they were stupid and reckless, but because they feared reduced ROF would itself lead to the loss of the ship.

Some detailed quotes: 

“Unfortunately ships did not carry enough ammunition for both a long-range engagement and rapid fire at shorter range. In half an hour a gun might well fire sixty rounds, but designed magazine capacity was only eighty. To provide enough for both long- and short-range firing, in 1913 Admiral Callaghan ordered his ships’ magazines overloaded. Later orders show that the dilemma was well understood. Orders issued in 1914–15 explicitly warned against wasting ammunition (for fear that not enough would be left for the decisive phase of the battle), but it must have been difficult both to urge gunners to open fire as soon as possible and not to waste ammunition.”

“By late 1914 the emphasis was shifting to rangefinder control and rapid fire at medium range – ie, within German torpedo range. In that scenario firing could not last long, perhaps only ten minutes or less. Much would depend on how quickly that firing could be achieved, which was dependent upon how rapidly shells and powder could be supplied to the gun turrets. Overloading magazines caused congestion at the bottoms of ammunition hoists. That might not matter if most ammunition was expended during a long-range phase when firing would be relatively slow, but given North Sea visibility, the battle might begin at short range. In that case, rapid fire would be needed at once. Without an initial long-range phase the extra cordite would still be congesting the bottoms of the hoists, making it difficult to supply the guns nearly fast enough.”

“Early war experience made rapid fire seem even more important. The Germans fired more rapidly than the British in the early battles. German survivors of the first major action, the Falklands (December 1914) said that slow British fire made German gunnery easier. Thus rapid fire was both a defensive and an offensive measure. If a German ship was smothered in splashes, her gunlayers might well fail to hit altogether. Orders issued early in 1915 emphasised the need for rapid fire.”

“At the Dogger Bank in January 1915, the British battlecruisers had the excess ammunition but were probably still following the usual magazine rules. In particular, charges were kept in their cases until needed. These cases, in particular, caused congestion at the bottoms of ammunition hoists, slowing the rate of fire. British gunnery at Dogger Bank seemed far more effective than the German, the implication being that somewhat faster fire, and more aggressive ship-handling, would have converted that battle into a decisive victory for the British. Probably following the battle, British gunners seem to have reached a very unfortunate (if officially unsanctioned) solution. Removing cordite bags from their

“At the Dogger Bank in January 1915, the British battlecruisers had the excess ammunition but were probably still following the usual magazine rules. In particular, charges were kept in their cases until needed. These cases, in particular, caused congestion at the bottoms of ammunition hoists, slowing the rate of fire. British gunnery at Dogger Bank seemed far more effective than the German, the implication being that somewhat faster fire, and more aggressive ship-handling, would have converted that battle into a decisive victory for the British. Probably following the battle, British gunners seem to have reached a very unfortunate (if officially unsanctioned) solution. Removing cordite bags from their protective (if clumsy) cases before they were needed helped relieve congestion at the bottoms of hoists. Loading was made even faster by using the working chamber and the handling room as ready-service magazines. Gun mounts were designed so that only charges on their way to a gun could be at these positions, to ensure against flash from the turret passing down into the magazine.”

“In fact the turrets were loaded so that a rich explosive train led directly from turret to magazine; a magazine explosion sufficient to destroy a ship was nearly inevitable after a penetrating turret hit. This became evident when the Admiralty investigated the losses during the autumn of 1916. HMS Lion was lucky. A turret hit blew the roof off, so that no explosion further down could build up enough pressure to breach her hull. Half an hour after the hit (and the flooding of the magazine) embers from the cordite in the working chamber were still smouldering. Ignited by a down draft, they fell down the hoist and ignited cordite still stacked in the handling room at the bottom. Despite considerable water in the magazine, the explosion in the adjacent handling room was severe enough to deform the bulkheads and doors leading to the magazine: this also in spite of the water load on the other side of those structures. The Grand Fleet battleships survived because their turrets were not penetrated.”

“The fleet could not be told that its own efforts to fire rapidly had been fatal. Soon after the battle it was claimed that plunging fire had destroyed the battlecruisers, and considerable deck armour was added, presumably as a way of convincing the Grand Fleet that its ships were safe. The DNC of the time (Sir Eustace Tennyson d’Eyncourt) pointed out that machinery spaces, which had much the same deck protection as the magazines, and which covered far more of the ships’ lengths, had not been penetrated, suggesting that magazines had not been penetrated, either.”

“Later it was admitted that turret hits, not plunging fire, were at fault. However, the losses were attributed to unsafe practices in the battlecruisers that had been adopted to achieve higher rates of fire: safety interlocks at the working chamber were deliberately disconnected. Without those interlocks in place, the flash of an explosion in a turret could propagate all the way down into the magazine. This account now seems a grossly inadequate cover-up. The flash explanation makes itself evident in later British attention to magazine regulations (which were extended to the closed hangars of British carriers) and to the anti-flash clothing adopted by the Royal Navy. Flash experiments were conducted on board the pre-dreadnought Prince of Wales in 1917. They demonstrated that the precautions in force before Jutland should have sufficed, had additional unsafe practices not been followed.”

“The choice both to overload magazines and to seek high rates of fire makes sense given the North Sea visibility problem and the choice not to invest in whatever would be needed to achieve high hitting rates at longer ranges. If the Royal Navy had really thought it possessed what was needed to hit at high rates at very long ranges, it would not have worried about wasting ammunition at those ranges, and it would also have worried less about evading German torpedoes, because it would have fought mainly outside torpedo range.”

Excerpt From: Norman Friedman. “Naval Firepower: Battleship Guns and Gunnery in the Dreadnought Era”. Apple Books. 

TLDR;

There was a reason safe handling practices were not followed, and there should be a design and campaign fleet management choice to be made between safety and rate of fire. Without any benefit to unsafe handling, this is a no-brainer and not very interesting - players and the AI would just wait for safety procedures and design techs to become available and then use them universally. When players are presented with having insufficient rate of fire or available ammunition to win engagements, the choice to operate unsafely is much more appealing. 

[Steeltrap]

Good info about the nature of the problem, and I was aware of the crux of it when I made my comment. (As an aside, I believe Jellicoe had in fact stipulated that ships were NOT to fire rapidly, but were to establish proper solutions as shown by straddles before shifting to full rate of fire. I seem to remember he expressed it as "deliberate" fire.)

Indeed it's part of the basis on which I made the point that this whole "flash fire" thing on weapons without two-part ammunition (projectile plus propellant charges) appears to make no sense.

Equally, it appears to make no sense on a weapon mount that cannot channel the blast/fire downwards to reach propellant storage/handling spaces.

It bothers me when we see mechanics introduced that appear NOT to be supported by evidence.

Flash fire from a turret penetration on a gun using two part munitions? Fine.

Not two part ammo, or not an enclosed gun mount with a propellant handling trunk as part of it? Sorry, not convinced that's at all valid unless someone can produce evidence of it.

A fire reaching a propellant store? Fine, but that's not a flash fire, it's a propellant ignition/fire, and to all intents and purposes it's as devastating as a straight magazine explosion. Look at the BCs at Jutland, and HMS Hood.

It's one thing to implement something before half the necessary surrounding mechanics are in place (not that I think that makes sense, but that's up to the devs to decide).

It's quite another to implement something that appears simply NOT TO BE CORRECT in several clearly verifiable respects.

That's what this current "flash fire" is. Nearly EVERY ONE of them I see are in fact fires reaching magazines regardless of whether the gun would have separate propellant, NOT turret penetrations.

Even IF it occurs less frequently, that doesn't address the question of whether it ought to occur at all in many instances.

Cheers

(Not that I think anyone's listening, LOL)

Edited May 26, 2020 by Steeltrap

[SiWi]

my experience is that yes they happen often but also often after the ship is already beaten up.

Aka often after a ammunition explosion a flash fire follows.

[Norbert Sattler]

I can't confirm that.

My "highlight" was a group of two enemy destroyers. My ship made a sinlge hit on one of them, which started a fire. After that my ship couldn't hit the enemy DD (not that I could see any hit at any rate and nothing showed up in the battle-log either) anymore, but the fire still caused each of the DDs' seven guns to go flying with just a couple of seconds between each.
That left the DD with roughly half it's structural integrity, full floatability and it's torpedoes fully intact, but all it's guns gone.
When my ship finally managed to sink this partially defanged DD, pretty much the same thing happened with it's sister ship, except this time I got a couple of hits in while the firework-chain went on.
Unsurprisingly the DDs were packing Lyddite I, which seems to be tailor made for fireworks. My ship was firing with white powder. I don't remember the exact date I put in, but it was something in the 1910 to 1920 period.
Both DDs were undamaged and only received a relatively light hit around 100 points of damage when the fireworks started.

I've also seen numerous flashfires happen without any ammo explosions. In fact I'd estimate that I've seen about five times as many flash-fires than I've seen ammo detonations in the current version.

Except on my own ships, where they both seem to happen equally often, but never paired up with one following the other. But then I barely ever use the HE and firechance buffing explosives and prefer the saver ones like white powder and TNT, as well as higher level barbettes, which seems to drastically reduce flash-fire chance.

Edited May 26, 2020 by Norbert Sattler

[DougToss]

I believe how RTW handles this is that flash fires are a small risk until 6" and grows with shell caliber after that. The reasoning was that smaller shells are QF and as such have brass casings and are handled manually, where larger projectiles have bagged charges and more complicated shell handling procedures, manual or not. 

There is a world of difference between 120mm shells in a ready locker and the complicated series of hoists, shell rooms and powder rooms. 

versus

Hitting a shielded 4.7" gun mount is a very bad day for all involved, but in and of itself will not cook off the shells in the locker.

The splinters from that hit may cut down the guys daisy-chaining shells from the locker to the gun, but it's unlikely they will cook off and is nothing like flash in a shell hoist. 

Edited May 26, 2020 by DougToss

[madham82]

12 hours ago, Steeltrap said:

A fire reaching a propellant store? Fine, but that's not a flash fire, it's a propellant ignition/fire, and to all intents and purposes it's as devastating as a straight magazine explosion. Look at the BCs at Jutland, and HMS Hood.

That's what this current "flash fire" is. Nearly EVERY ONE of them I see are in fact fires reaching magazines regardless of whether the gun would have separate propellant, NOT turret penetrations.

Your first sentence here is confusing me. Are you saying the BCs at Jutland and Hood are examples of flash fires or not? 

In regards to your the second sentence, there really should only be two cases of flash fire in the game, hits to turrets (or more precisely their stored shells) and hits to magazines. Both should be subject what has already been shared about the right type/amount of shells to be susceptible to flash.

Other factors I would say (not currently modeled) would be fire/shell protection measures (i.e. the bulkhead/fire doors/hoists/etc...) and the type of enclosures of the shell storage. More on this below.

3 hours ago, DougToss said:

I believe how RTW handles this is that flash fires are a small risk until 6" and grows with shell caliber after that. The reasoning was that smaller shells are QF and as such have brass casings and are handled manually, where larger projectiles have bagged charges and more complicated shell handling procedures, manual or not. 

There is a world of difference between 120mm shells in a ready locker and the complicated series of hoists, shell rooms and powder rooms. 

versus

Hitting a shielded 4.7" gun mount is a very bad day for all involved, but in and of itself will not cook off the shells in the locker.

The splinters from that hit may cut down the guys daisy-chaining shells from the locker to the gun, but it's unlikely they will cook off and is nothing like flash in a shell hoist. 

Great find on the diagram. I assume brass casings probably offer some fire/thermal protection, whereas the bags are designed to ignite and so are vulnerable to flash. If I thinking about it correctly, flash is not so much open flame as it is heat from a shell hit/explosion. So when you store propellant in bags designed to ignite, in a closed environment, the heat from these hits causes ignition. Hence flash fire, boom, explosion force pops the turret (all happening in short space of time). When this happens in a main magazine, the results are tenfold. 

You can look at many of the worlds modern Main Battle Tanks for an example of the dangers and how the designers ensure crew survival. It was quite common for tanks to blow their turrets off after a hit which started a fire and detonated their stored shells. Now tanks have armored doors separating them from the shells and panels on the roof designed to blow off during an explosion to vent the forces up, saving the tank and crew. We are dealing with the exact same issue here, just on a much larger and more destructive scale. 

The picture of the 4.7" also highlights my point about the enclosure being a factor. By being open, you remove the closed environment that really is key to flash being able to ignite the shells. You are more likely to detonate only a few of the shells this way, since the open space will allow the heat (flash) to dissipate quicker.

Edited May 26, 2020 by madham82

[DougToss]

As in all things, consulting the RTW wiki was helpful:

Flash Fires 

Guns may occasionally be the victim of flash fires, where an unlucky hit makes it to the gun's magazine, detonating it. This destroys the ship instantly. The mechanics on this are somewhat uncertain (see this thread, for example), but suspicion is that all main guns may flash-fire, and secondary guns either 6" and larger or 7" and larger. Many players prefer to use substantially more armour on guns above 6" calibre to avoid this effect. Great Britain has a national penalty that gives them increased vulnerability to flash fires as of the 1900 start date, to reflect their very high battlecruiser losses at Jutland (among others), so this is especially relevant for British designs.

Every turret flash fire reduces the subsequent risks of flash fires, to represents improvements in safety procedures in light of disastrous explosions. Note this only applies to turret flash fires and does not affect magazine explosions from penetrated magazines.

From Their Devs:

The 6" guns and lower caliber cannot blow if they are secondary or tertiary armament. However as main armament, they can blow after flash fire.

But this risk needs to be taken as armouring them enough against own guns is just too expensive.

 

5" guns can have flash fire which can blow ship too however it is very rare.

 

I'd speculate that 6" and lower being safe as secondary's is a reflection of the use of manual shell handling, ready-use lockers and brass casings. Although shells larger than 120mm, especially at the start of the period may not have been QF, or differed by nation, someone more knowledgeable might know. RTW also makes the distinction between open, shielded and turreted mounts. When used as a primary weapon, it would make sense that they would be connected to a hoist and magazine.

 

The increased level of detail in UA:D means that @Nick Thomadis can build on this and better differentiate between bags and QF, open mounts, shields and turrets (and casemates!), and hoists and lockers. In gameplay terms each offers a scale between cost vs weight/protection and rate of fire vs safety.

 

Better yet, this scale shifts as calibres increase. A 2" firing QF shells, on an open mount, with ammo coming from a locker has a great rate of fire, though vulnerable to splinters it would have no chance of flash fire. It would not be practical to have the same gun in a turret.

 

On the other hand, hand-working the bagged charges of a 10" is a slow process, it would be very slow indeed to get ammo to the mount without a hoist, and protecting a gun of that size by a turret against its own caliber would be important. As an enclosed gun, with bagged charges and a hoist, it would be vulnerable to flash fire but not to splinters. 

[Steeltrap]

On 5/27/2020 at 2:18 AM, DougToss said:

I believe how RTW handles this is that flash fires are a small risk until 6" and grows with shell caliber after that. The reasoning was that smaller shells are QF and as such have brass casings and are handled manually, where larger projectiles have bagged charges and more complicated shell handling procedures, manual or not. 

There is a world of difference between 120mm shells in a ready locker and the complicated series of hoists, shell rooms and powder rooms. 

versus

Hitting a shielded 4.7" gun mount is a very bad day for all involved, but in and of itself will not cook off the shells in the locker.

The splinters from that hit may cut down the guys daisy-chaining shells from the locker to the gun, but it's unlikely they will cook off and is nothing like flash in a shell hoist. 

That's a lovely way of demonstrating visually a significant part of what I'd written.

😁  😎

[Steeltrap]

On 5/27/2020 at 5:55 AM, madham82 said:

Your first sentence here is confusing me. Are you saying the BCs at Jutland and Hood are examples of flash fires or not? 

In regards to your the second sentence, there really should only be two cases of flash fire in the game, hits to turrets (or more precisely their stored shells) and hits to magazines. Both should be subject what has already been shared about the right type/amount of shells to be susceptible to flash.

If you take it within its complete context, the point I was making is the DESTRUCTION of the Jutland BCs or Hood were examples of their propellant charges going off en masse, which may or may not also have detonated adjacent projectile magazines. The famous "impression" of HMS Hood as drawn by an eye witness on HMS Prince Of Wales showed very clear evidence initially of a "propellant volcano", which is to say a huge inferno of propellant cooking off as intended but not within the tight confines of a weapon.

I didn't specifically classify them as flash fires or not as that wasn't the point I was making. I was discussing the RESULT, not the cause. In the case of the Jutland BCs, they clearly were flash fires from turret hits. HMS Hood was not, as the evidence points most strongly to suggest she suffered a direct propellant magazine penetration.

As for the rest, I believe we're saying much the same thing.

Cheers