Stardate 20020823.1340

(Captain's log): In response to my comments about the Soviet T-72 tank, Ståle writes from Norway:

You criticize the T-72 for having a smoothbore gun. I assume that this is because you consider smoothbore to be a poor man's substitute for rifling, a logical assumption if you know anything at all about firearms. Well, rifling gives a great accuracy improvement over smoothbores as long as you are firing full-sized HE or AP shells, as they did in WWII. However, the reason that tank guns are almost by definition smoothbore these days is the development of the armour-piercing, discarding sabot, fin-stabilized (APDSFS) round in response to improvements in tank protection. This round is basically a very heavy, very dense small-diameter dart that simply punches straight through tank armour. They do _not_ perform at all well if they are spinning on impact; thus having the round exit the gun barrel spinning would be a Bad Thing. The smoothbore gun in effect buys increased tank-killing capability at the expense of general-purpose shell accuracy. The British Army still uses rifled guns on its tanks (they think that good performance with high-explosive ammo is quite important), their discarding sabot ammunition instead uses a rather complex roller-bearing contraption to keep the spin out of the business end of the round on its way down the barrel - but just about everybody else is using smoothbore guns. I agree that the T-72 has poor fire-control systems but that has nothing to do with the smoothbore gun as such.

We're getting into some pretty esoteric issues involved with tank technology, so for those not fully up to speed I thought I'd provide some historic background before I discuss Ståle's comments. (Besides, it's fun to write about this stuff.)

Tank design has always involved an arms race between the guys designing weapons to kill them and the guys designing the armor and drive train to keep them alive. The first tanks were intended to protect their crews against machine guns, as a way of breaking the stalemate on the Western Front in World War I, but by the 1930's tanks were considered a much more general purpose weapon due, in part, to seminal work by Guderian and Liddell Hart. They were smaller, faster, more maneuverable, and had better armor and weapons, universally including a cannon mounted in a rotating turret.

Originally antitank weapons relied on high explosive (HE) rounds, which is nothing more than a bomb. But the effectiveness of HE doesn't scale well. To penetrate twice as much armor, you needed four times the bomb. That way lies failure.

In response to increasing thickness of armor on tanks, the weapons designers created two entirely new kinds of munitions for their guns, known as HEAT and HESH. High Explosive Anti-Tank (HEAT) is a shaped charge. When HE goes off it creates a spherical effect which expands equally in all directions, which is why it doesn't scale because most of its effect is wasted by going in directions other than directly towards the armor. HEAT fixes that; when it goes off it creates a very high pressure jet of plasma going forward, representing a pretty large percentage of the entire yield of the explosion. If this jet intersects armor, it cuts right through it like a knife through butter (very tough butter). And for the weapons designers, it scaled the right way because HEAT can, as a general rule, penetrate about twice as far as the width of the explosive round, which is to say the caliber of the gun which fired it.

Increasing the size of guns is a lot easier than increasing the thickness of armor. During WWII, the British up-gunned 2100 Sherman tanks with a British weapon that was extremely good yielding something they called the Firefly. (The American gun they replaced was really not as good as you would have liked. It was probably the most important and most correctable weakness in the Sherman which the Americans never did really fix.)

The first M1 tanks had a 105 mm gun, but they were retrofitted with a German-designed 120 mm smoothbore barrel which became the standard NATO tank gun (firing standard NATO 120 mm munitions). Increasing the armor on a tank is much harder because it substantially increases the weight of the tank, which changes the problem for the designers of the drive train and the tracks. If you don't do it right you may end up with something that sinks in the mud or sand instead of traveling over the top, or where the rate of mechanical breakdown becomes intolerable, or produce a tank which is just too slow or difficult to drive.

High Explosive Squash Head (HESH) was, perhaps, an even more imaginative approach. (These are also known as HEP, High Explosive Plastic.) When a HESH round hits a tank, it doesn't go off immediately. Instead, its warhead, which is made of plastic explosive, forms a mound on the surface of the armor, and then it goes off. That does two things.

First, it puts a divot into the armor without penetrating, but that doesn't matter. It also creates a mammothly powerful shock wave, basically a pulse of sound, which travels through the steel of the armor to the inner surface. There it produces a second divot by breaking a lot of the armor loose in pieces which travel away at high speed, a process known as spalling. And the tank crew is killed by shrapnel made from fragments of their tank's own armor. This can also set off the tank's munitions and destroy equipment and generally it's better to read about than to experience.

One of the reasons HESH is interesting is that it can kill without actually fully penetrating the armor. It doesn't leave a hole. More interesting is that it is only slightly affected by increasing thickness of the armor. Steel is an extremely good conductor of sound (far better than air) and increasing the thickness of the armor just means that the pulse has to travel slightly further before causing spalling and killing the crew anyway. That will cause a very small degradation in the shock wave, but the armor would have to be preposterously thick for it to actually matter.

But operationally I believe HESH is less forgiving, because I think you have to get a pretty square hit with it to make a kill. HEAT is easier because you don't have to be as lucky with where you hit, and for thirty years beginning just before World War II, HEAT was the standard for antitank rounds.

In the 1960's, a British research group located in Chobham, England, began to reconsider the entire concept of steel plate for armor. There had been improvements in the quality of the steel and the way it was forged which had improved its resistance some, but that was a matter of diminishing returns, and just piling more metal on was also subject to diminishing returns. So the Chobham group started over. They came up with an entirely new concept, which is now known as Chobham armor, which was shared with their allies including us. All modern NATO tanks, including the M1, use it and that's the reason they all seem to be kind of boxy, with facets and angles, instead of the curves of earlier tanks like the M-60. Chobham armor has to be created in plates and fitted together. The details of Chobham armor are classified but the general approach is known: it is a series of layers of steel, ceramic and air gaps.

When HEAT strikes Chobham armor, as the plasma jet tries to penetrate, it keeps encountering changes in density. In a sense, the jet is a waveform and each time it strikes a change in density there's a tendency for it to reflect and to spread. Which means that Chobham armor converts the effect of HEAT from a thin deep hole which penetrates into a shallow cone which doesn't. The extent to which Chobham armor is resistant to shaped charges is also classified, but as a practical matter modern tanks using a reasonable thickness of it are pretty much immune to HEAT rounds fired by any kind of reasonable gun.

And if Chobham armor is highly resistant to HEAT, it's all but immune to HESH. The concussion that HESH requires to do its dirty work dissipates extremely rapidly as it travels through the layers, and as a result HESH just creates a shallow divot on the surface and scares the hell out of the crew. (And as an added belt-and-suspenders safety measure, the inner surface of the M1 is lined with Kevlar cloth to catch any spalling which does take place, no matter why.)

Well! The weapons designers weren't going to take that lying down, and they again considered what could be done with munitions design. They abandoned explosives entirely and switched to using a solid projectile, which by its nature would be far less affected by all the layers than the plasma jet of HEAT or the concussion of HESH. But a wide projectile is inefficient; what you want is a very narrow one.

The Armor-Piercing Fin-Stabilized Discarding-Sabot (APFSDS) round, as described by Ståle, is the current standard antitank round. They're usually just referred to as sabot rounds (pronounced SAY-bo).

The projectile is a long thin pencil of some extremely hard and dense metal, with little fins on the back and a sharpened point. (We used to use depleted uranium but we're switching to tungsten, which is not quite as good but much less controversial. DU was better because it was denser and because uranium is a pyrophore.) Where older tank rounds relied primarily on the power of the explosives they carried, a sabot round is entirely dependent on the muzzle velocity of the gun which fires it. The faster it's traveling, the harder it hits and the more it can penetrate. And though the details are classified, apparently it's about as good against Chobham armor as against solid steel.

With a sabot round, the size of the pencil is not related to the caliber of the gun barrel. (And there's an optimum size for the penetrator; making it bigger may decrease its effectiveness.) In the shell, the penetrator is surrounded on both sides by two half-cylinders, the sabots, which seal the gun barrel so that the power of the shell's charge can be efficiently converted into kinetic energy. Once the round emerges from the gun barrel, the sabots fall away (which is why they're known as "discarding sabots") leaving the projectile to fly on its way to the target. The trick is making the sabots discard cleanly without deflecting the projectile.

Ståle claims that sabot rounds can't be fired out of a rifled gun. That's not actually true. The original 105 mm gun for the M1 was rifled and could fire sabot rounds. However, I do think that smoothbore guns are better at it, though I'm not totally certain of why. I don't think it's because having the penetrator spinning on impact affects its ability to penetrate. My guess would be that by the nature of a rifled barrel, it presents more resistance to the projectile, so the explosive charge which can be used has to be lower and you get a lower muzzle velocity. (Use too big a charge and the gun will explode.) A smoothbore gun has less resistance and I think you can use a larger charge in the shell, resulting in a higher velocity for the penetrator. With HE or HEAT, muzzle velocity was less important (because it mainly governed the range of the gun) but with sabot rounds you need it to be as high as possible because the kinetic energy of the penetrator is proportional to the square of the velocity.

Regardless of why, it was apparently clear that smoothbore guns were better for sabot rounds, which is why the Soviets put such a gun on the T-72. At that time there were no Western tanks using Chobham armor, but the Soviets knew it was coming and wanted to be ready.

My problem with this is that in many cases half a solution is worse than no solution at all. If you want to haul more cargo through the air, there are a lot of reasons why building a bigger plane is the way to do it. But if you design a bigger plane before engines exist with enough power to move it, you'll end up with something that will be so underpowered as to be dangerous to fly, if it can fly at all. You have to wait until all the pieces are there.

By the late 1960's when the T-72 was in design, the advantages of smoothbore guns were apparent. But the Soviet implementation was incomplete. The advantage of rifling is accuracy; the projectile stays oriented in the direction of flight and doesn't slew as much. A projectile fired from a smoothbore gun doesn't stabilize itself as much and is less accurate.

The other piece which is needed to make this work is, for lack of a better term, "avionics". (We need a term like avionics to apply to tanks.) You have to have a better targeting system. You need sophisticated electronics to aim the gun and compensate for all the things which could cause the projectile to miss, and the M1 has it. The firing system on the M1 is unbelievably sophisticated, with many different sensors and highly complex algorithms. It does things like keep count of the number of rounds fired from the gun, to calculate the effects of wear on the inside of the barrel. It has a direct sensor to measure how much the barrel is sagging. It's doing a lot of other things, too, which they won't tell us about. They can fire it while moving rapidly over broken terrain and score a direct hit on a target which is also moving, at very long range. That's non-trivial.

The M1's gun is uncannily accurate. The T-72's gun isn't. The targeting system in it isn't up to the job. It's not that I think that switching to a smoothbore gun was wrong because rifling is superior, but rather that I think they switched to a smoothbore gun before they had all the pieces to do it right.

Are we agreed that a hit is better than a miss? A hit with HEAT or medium speed sabot is better than a miss with high speed sabot. Maybe HEAT or the medium speed sabot round might get lucky, but you're not going to kill anything if you don't hit it, and at any kind of reasonable range the T-72's accuracy was lousy. (It may be that the Soviets couldn't make a sabot round that worked with a rifled gun. I think it may be a lot harder to make the sabots separate cleanly when the round is spinning.)

But there's more to it than that. Everyone concentrates on tank-versus-tank, but most tanks spend most of their time firing at other things, usually ones which aren't moving, and against those they usually fire HE. When a tank is shooting at a pillbox, or a machine gun nest, or a building, or a concentration of enemy infantry in hard cover, they shoot good old high explosive.

The T-72's smoothbore gun is also inaccurate firing HE; a rifled gun would be much better. By switching to a smoothbore gun before they had a targeting system that could make it work, what they did was to pull most of the teeth on their newest tank.

...but it still looked damned impressive.