The Japanese bomb used at Pearl Harbor (and nowhere else, to my knowledge) was the 775 kg (1709 lb) 41cm (16.1" from the NAGATO and MUTSU guns) Type 99 (Year = 1939) Number 80 (Class Type, not sure of meaning) Mark 5 Bomb (that is, a bomb based on the British Mark 5 APC gun projectile, also used after WWI for Japanese guns, since most of their battleships at the time used British-designed 14" APC shells, and, in the modified form -- new fuze with super-long delay and a break-away windscreen and AP cap tip ("Cap Head") to leave a flat end of half-middle-body-cross-sectional area to prevent ricochet off of the ocean surface and inhibit tumbling underwater -- in the later Mark 6/Type 88 (1928) "diving" APC shell (for below the belt underwater hits), which was replaced in 1931 by the improved Type 91 "diving" AP shell (the latter was capped in the 14" and up sizes, uncapped in the smaller sizes).
Unlike the British shells, which changed over to insensitive Shellite filler after Jutland, the Japanese tried to keep the original British Lyddite/Japanese Shimose trinitrophenol filler using all sorts of tricks to try to keep it inert on punching through heavy battleship side armor (this filler was used at their great victory against the Russians in 1904 at Tsushima and it seems that somebody important utterly refused to allow it to be replaced no matter what problems this caused in naval AP ammo). It took until 1931 for innumerable and exhaustive tests to PROVE that this filler could not be put through heavy side armor in a battleship and not explode in or immediately behind the armor plate, eliminating the highly-desired delay-action fuze function.
The new Type 91 AP projectile had a more streamlined "Type 2" ballistic form with a tapered base and long conical windscreen (the cap tip and windscreen broke away on water impact or high obliquity impact with anything solid just like the Type 88 design), an even longer fuze delay for a long underwater trajectory to hit targets below their armor belt, and, finally, that "important person" was overruled or retired -- :-> -- and a somewhat less sensitive Trinitroanisol (Type 91 Explosive) filler introduced that could, barely, make it unexploded through such heavy armor with thick cushioning in the shell's internal cavity. Only the few cruisers and other ships that kept the older 20cm (7.9") guns still used the Type 88 shell after the Type 91 came out, suddenly creating a lot of obsolete and useless AP shells. This bomb made use of some of these.
The old 41cm APC shell's AP cap, cap head, and windscreen were removed, giving a blunt point (perfect nose matching most Japanese bombs designed for other purposes). The body behind the bourrelet (widest point on the forward end of the shell that touched the gun rifling top surface) was machined down to a tapering cone to reduce the weight from the original APC projectile body weight of about 825 kg (1820 lb) to only 744 kb (1641 lb) -- including filler and fuzes.
A roughly 31 kg (68 lb) sheet metal conical tail and fin assembly was screwed to a modified, rather lightweight base plug, which had **two** side-by-side super-long-delay (circa 0.4 second) base fuzes similar to those used in the 41cm Type 91 AP projectile. This special base plug covered the entire rear end of the heavy bomb body and increased the usable explosive cavity filler weight to 30kg (66 lb) of Type 91 Explosive plus a 0.77 kg (1.7 lb) aluminum cavity tip cushion. The original Japanese enlarged British-design 2240-lb 16.1" Mark 5 APC shell body and cavity, if it had used the same filler and cushion (not Shimose, as it actually did), would have only had room for 54.3 lb of explosive plus that same cushion (that is, up to 2.5% explosive filler by weight if no cushion). Thus, the bomb had a significantly larger filler weight, in the gun projectile Semi-Armor-Piercing range, than the original AP gun projectile could have used, even though the cavity was the same size, other than the thin base plug allowing more room for explosive inside (the lower end of the cavity and projectile base were machined to allow the new base plug to be screwed on, but the rest of the cavity does not seem to have been touched). Note that most foreign AP bombs not also made from old gun projectiles (several US AP bombs introduced prior to WWII were also modified old AP shells, some even keeping their AP caps!) had much larger fillers, since no heavy (equal to the projectile diameter in thickness) armor was being hit and the bomb was always going to hit very close to right angles (30 degrees from right angles maximum).
A small set of holes were drilled in the tapered part to allow lugs to be fitted to hold the bomb to the attachment device of the airplane (one bomb per plane).
The data does not state how the fuzes were armed, since the normal set-back (high acceleration) and spin of a gun projectile in the gun bore were not available. There must have been a pair of wires from the plane to pins in the fuze bottoms that were pulled out when the bomb moved several feet from the plane (clean release) -- I cannot think of any other way to make these hidden base fuzes safely work (no spinning arming propellers could be used here).
The bombs hitting armor cleanly could penetrate the relatively thin deck armor of the older US Navy battleships intact and detonate properly after their long delay, as was shown by the destruction of the ARIZONA by one of these bombs. Other hits gave various results. Most hits on gun barrels and so forth caused proper high-order detonations when the bombs were stopped or slowed down by whatever they hit.
Two bombs hit the single-thickness 4" Class "B" (homogeneous, ductile) main turret roof armor of TENNESSEE and WEST VIRGINIA from the rear direction in each turret. These turrets had overlapping armor plates designed for support from hits from the front (where an enemy ship shooting at you would be expected to be). From the rear, the overlapping design kept the steeply-falling bombs from being deflected by much, since, as the bombs' noses began to rotate away as they skidded along the solid thick plate and bent that plate down under them, the noses immediately hit the squared-off rear edges of the overhanging plates in front of the plates originally hit. In each case, the rear plate being bent down under the bomb's impact force and the forward plate still being flat, this opened up a narrow "mouth-shaped" opening in the overlapping region between the two plates, tearing the rivets and bolts that held them together. The bomb, still moving at a high speed, though slowed down somewhat by the bending of the rear plate and the tearing out of the attachments to the front plate, was deflected along the easiest path into that mouth, bending the rear plate down even more and the rear edge of the forward plate upward, making the mouth larger. As the heavy nose passed into the turret (intact, I think), the upper (forward) plate's bottom surface deflected the bomb's nose downward at a steep angle again (it "ricocheted" into the turret), twisting the bomb so that the base slammed into the sharp bottom corner of the forward plate's rear edge. Since this is where the machining of the bomb had thinned it down the most, it was weakest here and the bomb's lower body snapped apart and was crushed as it went through the between-plates "mouth" after its nose punched through intact. The bombs' fuzes then went off, but since much of the filler was being scattered about by the broken bodies, the explosions were only low order and major fires occurred in the turrets, but rather little blast. There were no gun shells or propellant charges in the turrets to add to the damage (compared to the more recent USS SALEM and USS IOWA major turret explosions, especially the former), so nothing permanent happened and the turrets were repaired along with the other damage from the battle. (US Navy turret/magazine isolation systems seem to work rather well, so these hits had no chance of doing more than knocking out the turrets hit, even if the mounts had been manned and with ammo ready to fire.) I think that the turret roof armor was upgraded in these old battleships during their repair after these examples.
The chance of this kind of hit happening had not be anticipated by the bomb designers (though what could they have done about it?). The fact that this happened TWICE in virtually identical hits is kind of amazing, considering how few bombs were dropped. Note that if the bombs had been slightly to the side or forward or behind the turrets, both of these ships could have ended up like ARIZONA, since these hits were both directly on top of the magazines of these two ships -- only the turrets got in the way, unlike with ARIZONA.
A couple of improved versions of this bomb were developed during WWII, one a direct replacement of very similar design and a second one somewhat larger, when aircraft big enough to carry it were available. They were never used, though, and were only stored in Japan. They might have been used if the US invaded Japan, but that never happened.