Pages

Wednesday, August 1, 2012

Warship Design

The Navy no longer builds warships.  The Navy builds fragile weapon delivery platforms but not warships.  A warship is a vessel that can stand and fight;  one that can take damage and continue to fight;  a ship that is designed to mitigate the damage it receives.

In WWII the Navy built superb warships that could deliver immense amounts of damage, were built to resist damage, and were designed to continue fighting when damaged.  Consider the destroyers on the Pacific picket lines towards the end of the war.  There were numerous examples of these small ships absorbing multiple bombs and kamikaze hits and yet continuing to fight.  Or, what about the cruisers and destroyers involved in the battle for Guadalcanal?  They absorbed tremendous amounts of damage from torpedoes and shellfire while continuing to fight.  Yes, many eventually sank but only after extensive damage and while continuing to fight to the end.

Current Navy ships have almost none of the characteristics of their WWII predecessors.  Armor, other than small amounts of shrapnel protection around key areas, has been abandoned.  Manual/local fire control is largely absent though, to be fair, trying to shoot down missiles in manual control is probably a pointless exercise.  Torpedo protection in the form of blisters is non-existent.  Manning, the most important damage control factor there is, has been “optimized” to the point where damage control has been seriously compromised.  And so on …

Having just touched on some of the features that make for a good warship design, what are the characteristics that would make for a good modern warship design?  Here’s a good starter list.

Separation.  Key pieces of equipment should be physically separated to the extent possible.  The modern Burke, for instance, has all of its Aegis arrays mounted within about a 30 ft diameter or so.  A single hit there would likely take out every array.  The arrays should be separated as on the Ticonderogas which have two arrays on the forward superstructure and two on the aft.  Similarly, two of the three target illuminators are within ten feet of each other on the aft superstructure.  The Harpoon Mk141 rack launchers are always installed in side by side pairs.  A single hit would cost a ship its entire Harpoon loadout.

Note the Aegis Arrays Clustered on the Forward Superstructure


Redundancy.  Redundancy is almost non-existent.  Consider the modern destroyer as typified by the Burke class.  It has a single 5” gun versus the WWII Fletcher’s five 5” guns.  The modern Burke has no redundancy in gunfire.  The main radars of modern ships tend to be single installations with no significant backup. 

Armor.  Modern armor, such as it is, is limited to shrapnel suppression around a few key electronics areas.  There is no general hull or deck armor.  The modern gun mounts consist of fiberglass/composite shells  which provide no protection whatsoever as opposed to their WWII counterparts which were largely proof against shrapnel and ranged from a couple inches for 5” guns to the massive armor of the battleship turrets which could shrug off almost any hit.

Single Point of Failure.  Closely related to redundancy, single points of failure can incapacitate a system or multiple systems.  The example of the single 5” gun illustrates the single point of failure concept wherein an entire magazine of ammo is useless if the gun fails.  More commonly, single points of failure relate to ship’s utilities.  For example, electrical power for multiple systems routed through a common cable run represents a single point of failure.

Damage Control Requires Lots of Manpower
Manning.  As has been vividly demonstrated repeatedly since the start of WWII and through today, manning is one of the most important features of a warship.  Adequate manning allows battle casualties to be replaced.  Most importantly, successful damage control is a function of manpower more than any other factor as demonstrated even today by the Stark or the Cole incidents.

Steel.  WWII ships were built tough.  All ships were armored commensurate with their size.  Today’s ships have only isolated armor segments for protecting electronics from shrapnel.  Modern gun mounts are protected only by fiberglass type enclosures that offer no protection for the gun, whatsoever.  Even setting aside the issue of armor, per se, today’s ships are built with aluminum, composites, plywood cores, and very thin steel skins.  They are built to hold out the elements but that’s about it.  Simply using thicker and stronger steel would be a vast improvement as shown in the earlier post about the Cole.

In summary, if the Navy were serious about building WARships, we would not have the LCS, the Burkes, or for that matter, any of our current ships.  Given the enormous cost of modern ships, it’s puzzling why much greater survivability and toughness aren’t built in.  Ships represent far too big an investment not to make every attempt to ensure that they can fight hard and, if need be, fight hurt.



22 comments:

  1. I think you also take too hard a line on the systems and weapons on the Burke. By the late 1980’s the effectiveness of the 5” gun was waning. At that point large caliber guns hadn’t been considered effective against high-performance aircraft for over ten years. And the primary anti-ship weapons were the Harpoon and Tomahawk. The 5” gun was seen as a backup and a way to provide limited naval gunfire support if an Iowa wasn’t available. Looked at in this way, the Burke has triply redundant anti-surface ability (I know the Flight IIA have no Harpoons, but the weight and space is there for them).

    The Fletcher and Sumner destroyers had so many 5” guns precisely because that was the ship’s primary anti-air and land-attack. It also supplemented the torpedoes in anti-ship for redundancy.

    As to the placement of the SPY-1, I thought it was to improve performance of the radar. I read somewhere that one of the largest “blocks” in assembling a Burke DDG is the forward superstructure with the framing to support the radar arrays. That’s because the arrays have to be precisely aligned relative to one another in order to eliminate discrepancies between “hand-offs” from one array to the next one for fast-moving crossing targets. This implies that the Ticonderoga was a compromise design in that only two of the four arrays could be mounted close to each other. If you look at all the other ship designs with either Aegis or APAR this seems to be borne out.

    WireguidedMarine

    ReplyDelete
    Replies
    1. I hadn't heard that about the placement of the SPY-1 arrays. Let's assume it's true for sake of discussion. Logic says it doesn't make a huge difference or else the Ticos, the Navy's main AAW platform, wouldn't be retained as they are. They'd either be reconfigured or decommissioned. So, with that logic in mind, would you rather have a slightly better radar setup that is susceptible to being completely taken out by a single hit or a slightly inferior setup that retains 50% function if one of the two clusters is hit?

      These are the kind of "war"ship decisions that a navy has to make and the US Navy is consistently opting for convenience, cost, and ease of maintenance over warfighting and toughness. Those decisions are fine in peacetime but costly in war.

      Consider further the VLS systems on the Burke. VLS can be set up in groups of eight cells as the smallest unit or clustered in multiples of eight (hence, the groupings of 32 and 64 that are commonly seen). From an ease of maintenance perspective or convenience of utility supply, the larger the grouping, the better. From a damage mitigation perspective, would you want all your offensive (Tomahawk) and defensive (Standard/ESSM) weapons clustered in two compact locations? Again, an example of the tough decision that the Navy has opted for convenience.

      If one really examines the Burke from a pure warfighting perspective, it comes up wanting in many areas - not all but more than it should for a multi-billion dollar ship.

      Delete
    2. It’s not something I think is critical to the radar’s performance, but it improves the coverage and the handoffs from one array to the other. But if you look at the issues with AAW and how limited the fleet was with older, conventional systems, the Ticonderoga class couldn’t come soon enough. I think the Navy accepted the tiny penalty in performance and corrected it with the Burkes.

      That leads me to the vulnerability of the radar. Older, rotating antennas were much more vulnerable to damage than a fixed array. A lot of single-point failures like the waveguide, bearing mount, etc. In my opinion, the Aegis, even clustered like they are on the DDG, is still more reliable and more likely to survive battle damage than conventional radars like the SPS-49.

      The Navy agrees with you on the VLS placement long-term. The Zumwalt has the cells peripherally mounted on the ship. But this is weight and space inefficient, which alone drives up ship size (minimal stealth and two AGS also drive up the Zumwalt’s size). And keep in mind that just about every ship out there, from the Daring class to the older Kidd class, had their missile magazines concentrated for less volume.

      The issue of survivability is a long one. One example I know of is the Fletcher/Sumner armament. The famous 5”/38 caliber gun was initially not liked by the USN. It was a compromise between the short, manually handled 5”/25 caliber gun and the long 5”/51. The short one was felt to be good for AAW and needed no electrical power. The 51 caliber gun was good for flat trajectory firing and had more hitting power against ships. The 5”/38 was splitting the difference, and for armored mounts on WWII destroyers, needed electrical power for AAW or even rapid surface fire. And there were sailors who swore that the Sumner was much more vulnerable than the Fletcher because of this total dependency on electricity for their dual mounts. But the majority of the DDs that were hit off Okinawa by kamikazes were Sumners (or the Gearings based on them) that fought back and survived.

      Could the Burke be better? Absolutely, but compare it to other warships of similar design being made today. The only ones that have the sensors, weapons, capacity, and versatility are the Burke’s half sisters in Japan and South Korea. What Western Europe, Russia, and China make today still gets compared to the quarter-century design of the Burke, and they often come up short.

      Delete
    3. Just an admin note... Your posts are going to the Blogger spam folder rather than being directly posted. I don't know why and I don't know how to correct it. I can easily restore them as I'm doing but until/unless I figure out why, your posts may not be visible right away. Rest assured that I'm not editing or moderating them. Thanks for your patience!

      Delete
    4. I've heard varying opinions about the survivability of the SPY-1 panels. At the smallest scale, the individual emitter elements are certainly more survivable. Whether a lattice work radar is more susceptible to complete failure through damage than an array panel through loss of a critical threshold number of emitters is a debatable point.

      The other aspect to survivability is on-scene repair. In that, the rotating radars are far more likely to be repaired and far easier to maintain. One of the acknowledged problems in the fleet right now is that the Aegis systems, fleetwide, are operating a significantly less than optimum level due to a lack of highly trained techs and resultant inability to diagnose and repair the systems at sea. For all practical purposes, Aegis cannot be repaired at sea.

      With that in mind, I'd have to give serious consideration to choosing a rotating radar system (the -48/-49 combo) and NTU (the New Threat Upgrade program that was the alternative to Aegis) over Aegis.

      You raise the point that the Burkes, with whatever deficiencies, are better than the ships of other navies. My point is not whether the Burke is better than someone else's poor design but, rather, whether the Burke is the best design that the US Navy could produce. My contention is that it is not and that the Navy has abandoned warfighting design priciples in favor of peacetime design considerations. Something as simple as building ships entirely out of HY-80 or even HY-100 steel as a minimum should be a standard design spec and yet we use aluminum and very thin steel. Look at my previous post, Ship Construction and Naval Armor, 9-Jun to see graphic evidence of the effect that a simple choice in steel has.

      Further, my point is that the problem is fleetwide, not just the Burkes. Consider the LCS which has not a single battleworthy feature or the LPD-17 class which, even setting aside the myriad of construction problems, has been deemed unsuited for its intended purpose.

      Battleworthiness is not just a question of installing the very best technology. It's a balance of acceptable technology, toughness, repairability, hitting power, and so on. The Navy's balance is out of whack in their recent ship designs.

      Delete
    5. Thanks for checking the spam folder. The computer and the network I’m on, how do you say it, sucks?

      You won’t find any disagreement with me about LCS. If you have seen my posts on galhran’s ID I think it is an awful design that sacrifices far too much for top speed, including survivability. They are so bad that I believe a deep SLEP for the remaining Perrys is the only way to have something approaching a viable “low-end” past 2015-2020.

      I think the LPD-17 is a good platform; it’s just that the USN/shipyard did a horrible job constructing the ship. Misaligned stern gate hinges and engine mounts? How can that be missed?

      The Zumwalt could be a real winner, if the electric drive, the stealth aspects, and tumblehome all work. And if they can keep the price down so we can afford more than a handful. The PVLS and the AGS won’t be the problems. I hope the shipyard does a better job building it than the one that did the San Antonio!

      I never served on an Aegis ship. I did do a six month pump on the USS Nassau as part of the Marine BLT in 1997. I found talking to some of the sailors that the ship’s main radar (SPS- 48?) was down quite often. And the ship had a civilian tech-rep from the manufacturer aboard at one point. And this from a fully mature system!

      I believe the current issues the Aegis ships are having are due more to the Navy skimping on maintenance and “lean manning”. One sailor I talked to three years ago was a EW rating on a Burke and she said they were undermanned in her department, along with most departments onboard. If properly maintained a Aegis should have good reliability. Let’s face it, when the USS Bainbridge and her Burke sisters are chasing down Somali pirates they don’t need the SPY-1 or the bow sonar to be 100%. Unfortunately the Navy got carried away with that and now there are systemic problems fleet-wide.

      I understand your concern about a single hit knocking out the whole Aegis. But my point still stands in regards to a SPS-48/49 NTU: a single hit will also knock that system out. During the Vietnam War a frigate, I think it was the USS Worden, was hit with a Shrike ARM launched by an F-4 in a friendly fire incident. The Shrike’s small warhead crippled the ship’s AAW ability and she had to steam to Pearl or the West Coast for repairs. I think more than one radar system topside was destroyed. And I don’t think a single sailor was injured.

      My real problem with the SPS-48 or 49 is that they rotate. If supersonic sea-skimming missiles are a serious threat then a conventional radar will not be able to “refresh” the radar plot fast enough. I also think the Aegis-type radars perform better in a cluttered environment near land and with ECM. They are not perfect, which is why the SPY-1D(V) variant was introduced and the BMD upgrade is ongoing.

      I agree with you that the Navy has botched up the recent ship classes either in design or in constructing them. But I think with the Burke they got much of it right. It was designed during the Cold War, when admirals thought their warships might find themselves in a real shooting war. Could it have been designed with more guns and armor? Sure. But even a more armored Cole would have a hole blown into it in Yemen. When an IED that big goes off like that right next to the target it will do serious damage.

      Delete
    6. This comment has been removed by the author.

      Delete
    7. How do they achieve the separation fore and aft on the Tico's?

      Delete
    8. This comment has been removed by the author.

      Delete
    9. So, I'm guessing there's no technical reason why the D model couldn't have two xmitters and separate the arrays? I suppose it was reduced to one for weight/cost/space savings?

      If my assumptions are correct, that's possibly another example of poor decision making regarding warship design. Clustering arrays to save weight/cost/space at the expense of greater vulnerability to battle damage seems a poor trade.

      Delete
    10. This comment has been removed by the author.

      Delete
  2. I agree with you about the trend with ships being built much lighter and less protected than before. Beside the LCS there is the Mistral class, HMS Ocean, and the Absalons, which were all built to commercial rather than naval standards. I’m sure you could add to that list.

    My understanding as a naval enthusiast and not a naval architect is that warships built to naval standards are shock-hardened with stronger framing and mountings for the hull and individual components to withstand a near miss from a bomb or torpedo. There is also added strengthening for the ship to survive two or more watertight compartments being completely flooded. Redundancy is also made in the fire-fighting, electrical, and propulsion systems.

    Warships as a rule have larger complements than civilian ships of similar size to deal with damage control, which is often the single biggest factor in surviving heavy damage. As you pointed out, newer warships have smaller crews to cut life-cycle costs, but that could be put the ship at risk in combat. I believe that Western politicians and admirals think that peer to peer naval combat is unlikely but “police actions” where a ship patrols offshore and bombards and helos in support is more likely.

    In this sense the ships I mentioned above are very fragile in that none of these requirements was made during design or construction and are impossible to retrofit.

    I do feel differently than you about the Burke class. When it was designed in the 1980’s its design took seriously the ability to survive damage control more so than any other post-WWII destroyer or cruiser in the USN.

    The Burke has two completely separate fire mains for the fire-fighting aboard ship and each one I believe has a tank with something like 500-600 gallons of AFFF concentrate, that can be mixed with seawater as needed. Redundancy is similarly built into the electrical mains. The propulsion system can run on any one of the four LM2500 engines on either propeller shaft.

    The Burke class was the first surface combatant that paid more than lip service to NBC by having a sealed and over pressured “citadel” that all the crew could live within in the event of a WMD attack.

    Many Cold War era destroyers, frigates and cruisers were built with aluminum superstructures. Not the Burke. This was a direct lesson from the USS Belknap collision in the 1970’s; the picture of her after her collision says it all. The Burke is all-steel, with Kevlar around key areas. While classified I imagine the CIC and VLS are two areas wrapped in Kevlar.
    I believe if the attack on the Cole had happened on a Spruance or Ticonderoga the damage and loss of life would have been much worse. From what I have read of the attack the terrorists used over 400 pounds of explosive made into an improvised shaped charge. Even assuming the charge was molded poorly and the distance wasn’t optimum that is a lot of force being driven into the ship right at the waterline. While in the Marines I worked with shaped charges and explosives and know that even a misshapen charge can still focus much of the energy of the blast toward a given area.

    The big lesson in warship construction learned during WWII was that destroyers and cruisers with redundancies in the areas I mentioned before made a big difference between being sunk and limping home.

    The other lesson from WWII was that by the end of the war torpedoes with reliable magnetic detonators and air-launched guided bombs/missiles could do tremendous damage to even the toughest warships. The Italian battleship Roma in 1943 was sunk with just two German guided bombs. HMS Warspite, a few months later, was almost destroyed by another guided bomb which nearly set off her second 15” turret magazine. By 1945 guided bomb, missiles, and torpedoes were seen as the future. The only way to defend against that, while fielding meaningful numbers of warships, was to build bigger, structurally strong ships, with multiple compartments, that could take a large blast and still float.

    ReplyDelete
    Replies
    1. You make a good point about the power of torpedos and missiles having grown to the point that even battleships are vulnerable. No one is suggesting that an invulnerable ship can be built nor even that it should be attempted. Conversely, no one should suggest that just because powerful weapons exist we should throw up our hands and make no attempt to protect our ships - and, to be fair, I don't think you're suggesting that.

      What we should be doing is attempting to ensure that our ships are not susceptible to easy kills. A 5" gun on a Burke should not be able to be put out of action by simple, low velocity shrapnel. A ship's entire long range anti-ship capability should not be able to be wiped out by a single hit because the Harpoon racks are sitting out on an open deck in unarmored tubes. Radar should not be able to be totally wiped out by a single hit. Ships should be armored to deal with peer threats. A destroyer should be able to absorb damage from another destroyer. An LCS should be able to absorb damage from ... ah, that's a bad example since it can't take any damage.

      We can't make ships immune to damage but we can use better steel, armor, compartmentation, redundancy, separation, manning, etc. to mitigate the effects of damage and, hopefully, stay in the fight.

      Delete
  3. Not to belabor a point, but the Fletcher, Gearing, and Sumner were armed with ten 21" torpedoes with 800 lb warheads and 5"/38's. They also carried twenty or more depth charges each with 200 lbs of explosive on the stern. Yet they had at the most 1" of armor on the hull sides and 1/2" for the 5" turrets. The torpedo tubes were not armored at all. The 20mm and 40mm guns often lacked shields due to topside weight problems. No one, especially someone like Arleigh Burke, would have said their ship was armored enough to deal with destroyers, never mind bigger ships. Does this sound familiar?

    Those ships could certainly take damage, yet on paper before the war they were called derisively "tin cans" because of what many expected in combat to happen to them. While some took horrific damage and kept floating, they usually weren't in any condition to continue with the mission. Only the great logistical effort forward deployed in WWII kept some DDs from sinking on the way home.

    The redundancy in fire-fighting, manning, and compartmentation are there in the Burke. A smaller ship like the Stark survived two excocets right by her Mk13 launcher with SM-1 and Harpoons. Yet they didn't cook-off. Good design and great damage control training/measures saw to that.

    WireguidedMarine

    ReplyDelete
    Replies
    1. I fear you may be missing my point. I'm not trying to say that a perfect ship can be built with everything on it encased in battleship armor. I'm saying that what reasonably can be protected, should be and I use the example of the Fletcher to illustrate that the main battery (the guns) was protected to the extent they could given weight and stability issues. By comparison, the main anti-ship battery of a Burke (Harpoon) is unprotected. It doesn't have to be. The manufacturer has long offered a VLS capable version which would allow the Harpoons to be stored in the VLS system.

      You cite WWII destroyers as only having 1" armor on the hull and 1/2" for the mounts. Compare that to a Burke which has 3/8" (15.3#) HY-50 on the majority of the hull and 0" on the gun. The WWII hull armor, by the way, was the equivalent of current HY-80, I think, as opposed to the HY-50 we're now using. All that armor for the hull and guns was able to be accomodated on a WWII destroyer of 300 ft length or so compared to the 500+ ft long Burkes. You're reinforcing my point that WWII ships were built much tougher and that we need to return to building "war"ships.

      Delete
  4. The WWII destroyer had very little armor. not just in absolute terms, but in relative terms to the threat. The WWII DD's primary anti-ship weapon, the only one that had a chance aginst a CA or BB, was the torpedo. And the 1/2 inch armor on the 5"gun mounts was barely enough to block out shell fragments exploding on the nearby unarmored structure. The armor was hardly adequate against .50 cal fire; 20mm would defeat the armor. Yet those guns doubled as the prime AAW and secondary ASuW.

    My point is that the older ship's weapons were very unprotected against the threats they faced at the time: direct fire with Japanese DDs and CAs armed with 5" or bigger, and long lances.


    The Burke's prime ASuW weapon when designed was the TASM (Tomahawk Anti-Ship Missile). The Harpoon was considered a secondary ASuw; the Mk45 tertiary ASuW. The fact that the TASM is gone, and Harpoon neglected doesn't change what the Burke was designed for.

    They, along with the Standard, TLAM, and VLASROC, are relatively protected by the hull, kevlar, and halon/H20 suppression systems.

    I'd like to know the source of the hull steel you cite for the Burke. I have seen sources that indicate there are two layers of steel with no thicknesses given in critical areas, along with kevlar. And the steel is HSLA, which is more ductile and less brittle than older, Class A or B steel. And my understanding of hull steel was that a minimum of 25 pounds was need for structure.

    ReplyDelete
    Replies
    1. WW2 destroyers were built in massive amounts meaning..they were expendable.....yes...they were in the reality of todays conflicts corvettes or FAC's. There was simply no way of making them as armored as possible and retaining their speed and agility. They were screens for the larger cruisers, carriers and battleships.

      Todays warships are all hard to replace. All nessesary.

      The US was supposed to have long range strike fighters to launch ASM while the ships could stay out of range.....only problem is those things are gone. The Harpoons, TLASM are either gone or in the case of the harpoon those in US service are way out of date compared to whats sold to others.

      Modern ships are expected to not get hit and if a hit is taken the ship in in for some serious downtime.

      Ships like the DDG-1k are rather worthless as their stealth is worthless and lack the defensive weapons to fight off enemy ships/subs.

      Stealth is useless if every passing ship can see you (these are in the buisiest shipping routes on earth remember) and the enemy can FOLLOW THE 7FT MISSILE ROUNDS YOU KEEP FIRING WITH RADAR......

      Did i mention much of its superstructure is wood?

      Delete
  5. The construction references for the Burke come from a detailed cross sectional structural drawing that spec's out the hull, deck, and support members and was provided by someone in the Navy. The HSLA is spec'ed for use extensively on the main deck due to its ductility. The rest of the hull is HY-50 (HSS) with HY-80 used in the upper strake as shown in the Cole photo I referenced. A few small portions of the hull use 20.4# (the HY-80 strake, for example) or 17.85# but the majority of the hull centered on the waterline is 15.3#.

    The only double layer is an inner bottom void space at the bottom of the hull along the keel. I have heard that the DDG-51 Flt III's may have a double layer construction on the entire hull but I have not been able to confirm that.

    ReplyDelete
  6. Yes and No.
    Systems should be split and redundant to survive some damage.

    The ARA Guerrico lost the use of every weapon after a few dozen Royal Marines shot it with HMGs and Carl Gustavs.

    But was back in action after a spell in dock.

    The ARA Belgrano, formerly the USS Phoenix, a light cruiser from the 30's, armoured against gunfire, torpedo and mine.
    It was hit by two torpedo, the bow was blown off entirely and 20 meter hole was blown mid ship.
    Within twenty minutes, the captain ordered the ship abandoned.

    How do you armour an entire ship against an 800lb torpedo?

    ReplyDelete
    Replies
    1. No one, least of all me, is claiming that a ship can be armored, to the point of immunity, against all possible weapons. What you can do, and this is the point of the post, is to armor against easy kills and provide mitigation of damage effects.

      Delete
    2. Sorry I wasnt exactly clear.

      I was trying to argue two points,
      a ship should be able to shrug off small arms, for lack of a better reference point, up to 20mm.
      Quite why we still maintain the idiocy of sticking the captain and the command staff in a windowed room on top, I do not know.

      Thats the first issue dealt with.
      Snipers cant kill the captain, Carl Gustavs cant blow up the engines and near misses dont require dry dock

      But any serious weapon, a hit is always going to be at best, at mission kill.
      With the obvious caveat that a helicopter fired weapon isnt going to knock out a super carrier.
      But you are not staying in the fight after a heavy fighter or a nuclear sub scores a hit or two.

      Or thats my view anyway

      Delete
  7. Great conversation!! Im in agreeance that it seems our ships are built to deal out lots of damage, yet are awfully fragile. And while damage control training and equipment will probably save a ship, a mission kill is just as good for an enemy, since our ability to repair or replace ships quickly is a thing of the past...

    ReplyDelete

Comments will be moderated for posts older than 7 days in order to reduce spam.