Terminal ballistics: ID of WW2 amm holing a Thai bridge


I’ve been looking at a railway bridge here in northwest Thailand which apparently has three shell holes from WW2.

I’ve drafted a webpage describing the situation here: San Khayom Bridge.

I’m looking for guidance regarding what ammunition and guns might have produced the holes, and it was suggested that I query this forum.

I thank you.


Definetly 20mm (20mm Hispano, or “20x110” )

The Guns would have been 20mm Hispano cannon, used on both British and US Heavy Fighters and Fighter-Bombers.

The Neat Holes indicate 20mm solid shot ( a solid steel projectile) with or without Tracer Element.

My experience of viewing these is from similar Railway Bridges in Northern Italy, the targets of P-38 Lightnings (US) and Mosquito (Brit) Planes.

Doc AV, Forensic Firearms Examiner
Brisbane, Australia.

BTW, your analysis of the Bullet strikes is worthy of the best Firearms examiners and Ordnanceengineers around.


You may try a search with a metal detector in the direction from where the attacking planes were coming from. With a bit luck you may find the fired cases from that raid.



Thank you for the compliment.

Some clarifications:

• The “neat” holes would be those at Point A and Point F?
• The source for the hole at Point G would remain undefined?
• The “analysis of the bullet strikes” would refer to the scenarios relating Points A through D?

With regard to the hole at Point G, and on the assumption that it remains undefined: Could a 20mm HE shell have detonated on impacting a 3/8 inch thick structural steel plate (the web of the channel in the built-up member), making the approx 2-3/8 inch hole visible today — and leave no other evidence because the expansive force of the explosion continued in the relatively unconfined space of an open structural member in an open truss? Evidence is defined as that visible at, say, 20 feet and in poor lighting conditions — closer inspection being impossible.

(I have previously discounted the peculiarly distorted lacing members near Point G as having been associated with the damage at Point G itself. That conclusion would still apply in that, if affected, the built-up member would have acted to contain the explosion and would have tended to stretch the lacing members rather than collapse them.)


There are some unusual restrictions here: metal detectors aren’t available on the local market. The assumption is that they’re considered illegal — which seems odd in that “demining” operations are a major effort directly east of Thailand. I’ll test that today (Monday here).

Compounding the problem is the comparatively economically poor area in which this occurred: back then folks always had an eye out for anything of possibly salvageable value. Even in Chiang Mai, and after the war, pieces of an aircraft on display began disappearing to the point that one night what was left of it was appropriated by persons unknown.


Islandee, yes, I am aware of the places “east” of Thailand. I spent time there using mine detectors for the purpose they were made. There too aircraft on (unguarded) display litterally disappear like being eaten by ants - I saw that happening to a US AH-1 Cobra and a UH-1B Iroquis.


I am no expert but I am pretty sure that P-38s were armed with .50 cals. not 20mms.



Four 50’s and a 20 mm cannon according to all spec/reference info I see


This is indeed very impressive work. Thank you for sharing your article with us, it is very interesting.


Islandee: In looking over your list of airplanes that might have attacked this bridge I also found that the Beaufighter seemed to me the most likely candidate. Jack


Is there records of who was operating in the area? That would allow to narrow down certain aircraft types.


Jack’s recemmendation is a good one. Beaufighters are a good candidate. They were active in this Theater of Operations. In fact, Graham Irving, who many of you will remember, was an engine mechanic on them in India and other locations in this area.



The Beaufighter (one of my favourites) was used in many armament configurations - a great machine!


google.de/search?q=beaufigh … 20&bih=933


[color=#008000]Thank you: Beaufighter. Working on that.

Seems 211 Squadron with Beaufighter Xs is the most likely candidate right now. There are also these to go through:

No. 27 (F) Squadron - Beaufighter Vf , then Beaufighter X
No. 89 (F) Squadron - Beaufighter VIf, later Mosquito
No. 176 (F) Squadron - Beaufighter VIf

I’ll brief my results here as they develop and add full commentaries to my website.

Not to forget the P-38s.[/color]


For Point G, with a hole diameter of 60mm, I’m trying to flesh in a hypothesis that an exploding HE shell might have produced a hole larger than the caliber of the shell. It has now been suggested that this explanation might be irrelevant because 20mm HE shells may have had time-delay fuses, that delay being around one millisecond after impact. The delay was intended to allow a shell to penetrate an object before exploding which presumably would then have done more damage. That delay would also have been sufficient for a shell to have exited the built-up structural member entirely to explode harmlessly in an open area of the bridge.

Hence, the question: might WWII-vintage 20mm HE shells, as used in aircraft Hispano cannons, have had time-delay mechanisms?



Hence, the question: might WWII-vintage 20mm HE shells, as used in aircraft Hispano cannons, have had time-delay mechanisms?[/quote]
Yes, they did. The original HE shells used in 1941 did not, and were found to be ineffective because the shells exploded on the skin of the aircraft rather than inside it (the Luftwaffe had the same problem). For a while the practice ball rounds were used instead, as they were big enough to penetrate deeply into the structure and do a lot of damage even without chemical contents. These were eventually replaced by HEI shells with delayed-action fuzes.

You should bear in mind that penetrators could punch bigger holes than their own diameters. I have a piece of armour plate with a hole of approx 6mm diameter punched through it by the (saboted) 4mm tungsten bullet from a 6.5x25 CBJ round.


After some interruptions, I’m back to the shell holes in the San Khayom bridge. I apologize for the length of this; but I don’t know how to shorten it further.

With regard to the hole at Point A, I’m trying to visualize the actions of the aircraft that produced the hole. I’d like to include a sketch on my webpage of the possible / probable path of the aircraft during its attack.

The consensus is that the hole is apparently from a 20mm projectile; and in the Southeast Asia Theater both the P-38 and the Bristol Beaufighter carried HS.404 cannon which used that ammunition. I’ll use the P-38 as a subject here simply because I can find more information about it (the Beaufighter might be the better candidate because, beyond its harmonization range, it would have offered a dispersion pattern with its four cannon: such a pattern might more easily have included Point F, the other 20mm impact point on the bridge. [color=#FF0000]Does anyone know of an on-line presentation, including sketches, perhaps a manual, for harmonizing the guns on a Beaufighter?[/color]).

So, scenario: a P-38 pilot happened upon a “target of opportunity”, a train traveling south out of Lamphun and approaching San Khayom bridge. The pilot reduced altitude so as to fire on the train. Coming into range, he fired for a certain period; then he broke contact and climbed away from the train and the bridge.

Angle of attack: The exit angle of Hole A measures about 12° to the horizontal. In the sketch below, I assume the bullet path was almost unaffected in passing through the comparatively thin bridge plate (projectile diameter was twice the thickness of the target plate). On that basis, I assume the pilot dropped to an altitude matching the 12° allowing an angle of attack of 12°.

A P-38’s single 20mm cannon was centrally mounted in the fuselage and its operation was basically point (the plane)-and-shoot. The range of the cannon was limited primarily by the pilot’s visual acuity and his properly adjusting for bullet drop (excluding weather, aircraft performance, pilot skills, whatever).

I assume an attack speed of 350 mph: a P-38’s cruising speed was 275 mph and max speed was 414 mph. A velocity diagram would look like this for a 12° dive at 350 mph:

The important information here is that, flying at a down-angle of 12° and at a speed of 350mph, the aircraft would have been dropping 107 feet per second.

The idealized flight path would have looked like this:

Of course, the pilot would have disengaged at some point before his aircraft flew into the target.


[color=#FF0000]Do these assumptions and the scenario seem plausible? If not, I’d appreciate being corrected. [/color]

If the scenario is not too far off:

[ul][color=#FF0000]At what range would the pilot likely have started firing?[/color] (I read that one B-25 gunner had his guns harmonized for 1000 yards: he cautioned that this was useful only for stationary targets)
And [color=#FF0000]at what altitude would the pilot have had to disengage (how much altitude would be burned up after the pilot tried to gain altitude; and how much spare altitude should he have allowed)[/color]?
Tying these down would also establish how long the pilot might have fired.

Alternate scenario: With the above presented, there is a piece of data which could contradict this scenario, and I don’t have the background to interpret it: the “floor” of Hole A is at an angle of about 7° to the horizontal. If 7° (not the 12° used above) were the angle at which the projectile entered the plate, and the exit angle was, as actually measured, 12°, then passing through the plate would have bent the projectile’s course by 12° - 7° = 5° (see cross-section).

If the angle of attack were a very lean 7°, then at a range of 1000 yards, the gun would have been only 366 feet off the ground — with bullet drop compensated for by either, an extra 24 more feet for a total of 390 feet, or a slight tilt upward in angle of attack. And at 500 yards 183 feet (compensated for 4 feet of bullet drop). On the other hand, the aircraft’s altitude loss while changing from a down-angle of 7° to actually moving upward would have been less than recovering from a down-angle of 12°. But operating that close to the ground at 350 mph or more seems unlikely (which is why I didn’t include it in my main presentation above); though I am willing to be corrected.

[color=#FF0000]Are there any comments about this alternative?[/color]


As nobody is answering I will try to add a bit to the questions rather than the answers – sorry for this.
I am not really entitled to state on this but from my limited insight and the few years I spent following my interest on aircraft and aircraft armament I can only suggest a few things. Also I think we have some USAF armorers here and even a pilot with combat experience (Mel Carpenter) who may tell us much more (less on WWII though I think) and if so any of the guys in question please correct what is said below.

Regarding engagement distances, duration of attack and break of contact:
There are several factors which do influence the answer on this.

  1. What kind of target is selected?
  2. Moving or still target?
  3. Daytime and day light situ, dusk, dawn, bright day, opposing sun
  4. Weather in regard to heavy rain (snow we can exclude I guess)
  5. Actual situ in air like enemy planes
  6. Actual situ on ground like was the train close to a densely overgrown area (jungle) or a tunnel ahead?
  7. Is the target actively defended/was the attacking plane fired at?
  8. Ordnance load/configuration of the attacking plane and what ordnance was used on the actual strafing run, if rockets or air dropped ordnance were used in conjunction with the guns a safety distance had to be kept as shrapnel would endanger the attacking plane.
  9. Available amount of ammunition at the moment of engagement, means if the full load out was available or less as prior strafing might have depleted the available amount of ammo and only the last remaining rounds were fired during the strafing run investigated here.

So for a better answer to your question or at least for narrowing it down to fewer possibilities it my be helpfull to try to find the combat reports on the area.

  • Who was flying in the region at the time the incident took place?
  • Then finding the Unit and establishing the aircraft type
  • Trying to find the reports of this unit in archives (or even the Pilot’s name)
  • Finding the actual report for the flight in question

As mentioned before it may be well worth to do some metal detecting in order to recover the fired cases which will provide some more info as for the nationality of the aircraft (US or GB) and maybe even telling more about the engagement distance though this will have to include some allowances.

Maybe some of your other members here can provide more insight or correct what I said here.


And my apologies for not catching your response sooner.

The biggest problem with the San Khayom Bridge is so little information is available. This shows up in working with your question guide:

  1. Target: railroad rolling stock (assumed: there are no records available from the railroad).
  2. Moving or still: moving (assumed).
  3. Hour of day / position of sun: unk, as is the event date; no witnesses on the ground survive.
  4. Weather: unk.
  5. Presence of enemy aircraft: unlikely (the only confrontation between Allied and Axis aircraft over northwest Thailand was in Nov 1944, and all five Axis aircraft were downed, with only one Allied lost; the skies were controlled by the Allies except for some anti-aircraft batteries (or equivalent) areas around Chiang Mai and Lampang airports, and around two railroad bridges farther south. One B-25 on a bombing run over one of the bridges, that at Kaeng Luang, was shot down by an anti-tank gun, also in Nov 1944. Those sites were well known to aviators for their defenses (and sometimes notably ineffective: see Attack on Lampang.
  6. Site of attack: railroad passing through open rice fields (assumed: that’s what it is today).
  7. Was plane fired on: apparently so, by a 57mm round which hit the bridge superstructure (Point G), which suggests that the gun was out-of-range.
  8. Plane’s armament: only evidence is two shell holes, apparently from 20mm rounds.
  9. Plane’s available armament: unk.

20mm rounds would narrow the probable Allied aircraft operating over northwest Thailand to two: the Beaufighter and the P-38.

The probable range of dates is established by the presence of these two types of aircraft over Thailand: April 1944-end of war. That covers particularly 211 Squadron flying Beaufighters out of India and 449 Squadron flying P-38s out of China; but that list of two cannot be said to be exclusive. The recordkeeping system, at least via the Internet, is not that precise. Further, the odd Beaufighter or the odd P-38 flying out of its assigned or assumed assigned area cannot be ruled out.

The train should be considered a “target of opportunity”. Structural steel bridges weren’t targeted for destruction with 20mm shells. No specific train could be marked for destruction. Intel wasn’t that fast. Communication wasn’t that good. Even without direct interference from hostile aircraft, trains couldn’t run on any schedule. Locomotives very quickly went through the low-heat content junk wood which was stacked at nearly every station for fuel. Even water towers were being targeted by Allied aircraft. So neither water nor fuel supplies were reliable.

Metal detectors: while they’re not illegal, in this current state of martial law in Thailand, local law enforcers can be expected to err on the side of caution. I’ll wait for a more relaxed military and political atmosphere before making any search attempt. Anyway, I am not optimistic about success since the poverty in the area during and after WWII made metal scavenging a cash generating activity; and the current relative prosperity has allowed the accumulation of quantities of steel refuse from railroad operations along (and beyond) the right-of-way.

To the present: I finally came across a description of a strafing process on the 13th Bomb Squadron website: Combat Tactics: Strafing (Area). The aircraft used was the Douglas A-26/B-26 Invader, apparently for combat in Korea:

“. . . from a low level approach to the target. . . . a desired 320 mph. . . . At approximately 3,000 yards from the target, the pilot pulls up to 300 ft. above the terrain, spots his target, . . . At approximately 1200 yards from the target, the pilot enters a gentle dive, lines up the target in the sight and commences his strafing run. Firing bursts should be short, starting at 1,000 yards, and breaking off at 400 yards. Gently fan the rudder . . . to effectively cover the target area. . . . Airspeed . . . is the best defense against enemy ground fire while on a low-level strafing run. . . .”

I’m incorporating that into the webpage dealing with Point A at Possible sources.

I think with the general lack of information, or more accurately with my lack of access to information, operating as I do here in Thailand, I can’t go much farther on the topic of Point A (and Point F). My effort is necessarily limited to producing (at least) one plausible scenario to explain Holes A and F which I hope I have now done.

I would be interested in comments about the page as it now stands: San Khayom Bridge. I’ve still to update subsequent pages.


Islandee, your work looks well researched, sourced and reasonably explaind and concluded on. Under the given conditions and with the eminent limitations by the many unknown factors and lack of availability of the actual combat reports hardly anything better can be produced.

Maybe you can contact US and UK archives for further research?

Are you cooperating with a govt. museum in Thailand maybe? If so they may grant you an explicit permit for doing some detecting around the bridge?

Did Japanese trains have installed AA guns?

Maybe you can get photos of 20x110 HS cartridge (AP, “ball” and HE types) to show them with the penetrated bridge parts or the diagrams of the cartridges?


EOD: I thank you. First I’ll finish cleaning up / updating my subject webpages with existing information that I have in hand. Then I’ll proceed further, using your suggestions, and check back in as work continues.