Plane crash

All posts tagged Plane crash

Going to start this post with a plug for the event I’ve been helping to plan. Normally I don’t do such things, but I feel as this is, at its core, a blog about history and archaeology, then this is fitting. On March 25 there will be a heritage forum brought to you by Heritage Tomorrow NL (formerly Youth Heritage NL). For anyone between the ages of 18 and 35 and who is interested in heritage, history, archaeology, folklore, museums, etc., this is an event for you. We will be having a heritage skills competition, then really focusing on networking and talking to other like-minded young people, getting to know each other, and hopefully creating networks so that we can help each out. Registration is $10 and includes lunch from Volcano Bakery. Click here for details.

I’ve been jumping around the province a bit, and am heading to Stephenville for this post war incident. No archaeology here as the aircraft crashed on the runway and survived the accident. The main reason for sharing this one is actually the great pictures that were featured in the report at The Rooms. I really wanted to share those with my readers as so many of my pictures from the era are often grainy and copied so many times it’s hard to really see what’s happening.

On 13 August 1948, The Evening Telegram printed a small article about an incident at Harmon Field. The article reads:

Three Escape Death In Aircraft Crash At Harmon Field

Three crew members of a twin-engined Lockheed Hudson narrowly escaped death at Harmon Field this week when their aircraft, while coming in to land, veered from the runway as the pilot lost control, groundlooped and crashed.
The plane was being ferried from Dorval airport in Montreal to South Africa, and was owned by the Hanley Aviation Company of Johannesburg, South Africa. The plane called at Harmon en route.
Parts of the plane will be salvaged by the owners at a later date. The crew of the wrecked aircraft left Harmon Wednesday night for Montreal.

The aircraft in question was a Hudson Mark III, serial number 6386 (also sometimes listed as AC #41-23569), with the South African registration of ZS-DAF and it crashed. The crew consisted of Edward R. MacLeod, pilot from the United States, John P. MacMahon, navigator from the United States, and JohnH. Hluboky, radio officer, also from the United States. None were injured in the crash which took place at 23:53 GMT on 9 August 1948.

The aircraft involved had a long history prior to the accident. It was manufactured by the Lockheed Aircraft Corporation and was delivered to the Royal Air Force Ferry Command under the service number BW-707. After 20 hours flying time with the RAFFC, it was transferred to the Royal Canadian Air Force on 12 February 1942 and retained until 26 August 1946. It then went into storage. At this point, it had a total flying time of 1729 hours and 30 minutes. During its service in the RCAF, the Hudson was used in Canada for flight training and as part of a Search and Rescue Squadron (it was here that it was fitted with an airborne lifeboat). It was removed from storage on 20 May 1948 and flown from Greenwood, Nova Scotia, to Montreal, Quebec. On 16 July 1948, the aircraft was purchased by Simple Aircraft Limited from the Canadian War Assets Corporation, then sold again to Hanlys Aviation (Pty) Limited of Johannesburg, Union of South Africa. Long-range tanks were added to the aircraft, but no documents suggest when that happened. It is assumed it was done in preparation for the flight to South Africa. The pilot said the work was done by Ross Aero at Montreal Airport, but no details could be found for the incident report. The engines were ones put on during the RCAF service, and both were certified as fit for the ferry flight to South Africa.

From McGrath 1948

Prior to the ferry flight, inspections were made determining that the aircraft was fit for the flight, but the ferry permit limited flying to daytime flying. Otherwise, it was certified to make the flight to Johannesburg via Gander. MacLeod, the pilot, was an experienced ferry pilot and claimed to have 2,500 hours, 300 of those at night. In the war, he served with United States Air Transport Command and made 12 ocean crossings. Since the war he operated a flight school and charter service using small aircraft, mostly Lockheed aircraft. Prior to this flight, he did not fly the Hudson, but did inspect it prior to the flight. He stated that it was not until after he was in flight that he noted that the aircraft would not cruise at 170 knots, but the report points this out as a discrepancy. Had he been as familiar with Hudsons as he claimed, then he should know they cruise at 160-180 miles per hour, not 170-190 knots and the pilot believed. MacLeod also did not carry an Aircraft or Route Information Manual, did not know the empty weight or gross weight of the aircraft and carried to information or instrument landing procedures for any of the airports on his route.

The aircraft left Montreal at 1922 GMT 9 August 1948, and, as the accident report states, “even on a very optimistic flight plan could not have reached Gander before sunset” (McGrath 1948). The Hudson arrived over Harmon Field in Stephenville at dusk and requested permission to land at Harmon. Due to the daylight restrictions on the permit, the pilot thought it best to not continue to Gander and have to land in the dark. The request was passed on to the Civil Aviation Division in Gander and permission was granted on the understanding that the aircraft would proceed to Gander the following day and that the need to land at Stephenville would then be investigated. As the aircraft approached, MacLeod noted that the undercarriage indicator light did not come on when he lowered the undercarriage. He asked the Control Tower if the undercarriage was down, and due to the low light, the Tower could not see it and asked that the aircraft fly over at low altitude. With that passing, the Tower confirmed that the undercarriage was indeed down, and so the aircraft came in for a power-off landing.

From McGrath 1948

Witness statements from United States Air Force personnel stated that the aircraft made a very low approach, so low that they were not sure that the aircraft would even make it to the runway and thought it would land along the approach lights. At this point, the engines were heard and the pilot opened the throttles so that the aircraft regained some altitude. The aircraft touched down hard on the tarmac about 15 feet short of the end of the runway, bounced 30 to 50 feet into the air and touched down again about 300 feet from the first point of contact. The aircraft then began to swing to the left and the aircraft ran off the left side of the runway ab out 100 feet from that first point of touchdown.  At this point the left wheel hit and smashed one of the runway lights, but the wheel was not damaged. The undercarriage collapsed ad the aircraft pivoted on the left wheel making a turn of 270° to the left finally coming to a rest about 100 feet from the edge of the runway. There was no fire and the crew were uninjured. The left engine was still idling when the crash crew arrived. The pilot could not explain the swing and stated “that after the bounce the aircraft ‘stayed on’ with the tail down and the rudders became ineffective” (McGrath 1948). The report states that this was normal for Hudson aircraft in a tail down attitude. MacLeod said he tried to correct the swing by using the right brake, but it had no effect and he was afraid to often the left engine in case it cause the aircraft to swing to the right. The report states that there were no marks on the runway to indicate that either brake had locked.

As previously stated, the aircraft had been refitted with long distance fuel tanks. During the investigation, it was determined that neither the empty or gross weight of the aircraft was known. The all-up weight would have been 15,400 lbs. and it was rated as such by inspectors prior to the flight, but at departure from Dorval, with full tanks, the aircraft would have weighed considerable more than that. Due to the lack of data, it was impossible for investigators to say if weight was a factor.

From McGrath 1948

In the accident the aircraft:

sustained major damage on the starboard side; the port side was undamaged. The starboard wing was twisted lengthwise and the starboard engine mount was badly twisted; the starboard undercarriage had collapsed in a forward direction; the starboard wing root was damaged; the starboard flap was badly crumpled; the bottom of the starboard tail fin was bent inwards; the starboard cirscrow was bent and had cut through the starboard side of the nose of the aircraft (McGrath 1948).

It was also noted at this time that the plexiglass nose had been broken at some point prior to the accident and had a piece of fabric stuck over the break to close the hole. There was also an opening in the bottom of the aircraft which was made for the attachment of the airborne life boat for RCAF SAR missions. This opening had not been covered before the aircraft left Montreal. The investigation also showed that the brakes will had holding power, but the drag link on the starboard undercarriage had an old crack in the upper end of the drag link rod near the attachment point to the actuating strut. This link rod appeared to have failed when the aircraft slid backward at the end of the ground loop.

From McGrath 1948

In  the opinion of the investigators, the accident “was a result of failure by the pilot to correct a swing developed by the aircraft following a bad landing” and “tail heaviness of the aircraft may have been a contributary cause”. (McGrath 1948)

There were a number of irregularities found that were brought to the attention of the governments of Canada, Newfoundland, South Africa and United States so they could investigate any breaches in their respective air regulations. Such issues included the fact that the aircraft was granted South African registration before ownership had been acquired by Hanlys Aviation (Pty) Limited, and the South African validation was issued before the Canadian Ferry Permit which it validated. The pilot had no South African documentation authorizing him as a US airman, and the aircraft radio was not licensed. There was no record of work being done to the aircraft after it was removed from storage, even thought long range tanks were installed, and the weight and balance were unknown. There was no copy of the Department of Transport Information Circular T/5/47 on the aircraft as required by law, and the pilot had no Route Information Manual nor information on instrument approach procedures for the airports on his route. The pilot also used fuel from the cabin tanks first, causing the rear tanks to be heavier, especially when coupled with the fact that there was 300 lbs. of freight in the tail. Finally, nationality and registration markings were not painted on the wings. They were painted on the fuselage only (and the previous marking were still visible).

From McGrath 1948

The reports were sent off to the governments and a copy sent to Hanly’s Aviation Office in New York. The report was not completed or sent until October due to the delays in attempting to get the weight information. Finally, the copy that was sent to Wainwright Abbot, the American Consul General had an added note from J.S. Neill, the Commissioner for Public Utilities and Supply, which thanked the US authorities at Harmon Field for their assistance to McGrath in his investigation (Neill 1948).


McGrath, T.M.
1948 Civil Aviation Division Newfoundland Aircraft Accident Investigation Report Number 10. On file PANL Accident to Hudson Aircraft AG/57/14

Neill, J.S.
1948 Letter to the American Consul General, 12th October, 1948. On file PANL Accident to Hudson Aircraft AG/57/14

1948 “Three Escape Death In Aircraft Crash At Harmon Field”, The Evening Telegram. 13 August 1948, p. 5.

Share Button

This project, which took place in the fall of 2008, had a number of firsts for me. It was my first experience as an archaeologist on a recovery project, my first time in a helicopter, and my first time in Labrador! The latter two were wonderful experiences, and the first was a learning experience for me. Some of that learning came years later when I looked back on the project. But why do this work if we can’t learn about history, and about how to preserve the archaeological record in varying circumstances.

Helicopter leaving me (and half of the team) in the wilds of Labrador. Photo by Lisa M. Daly 2008.

The recovery of the Douglas A-20 was a project undertaken by Underwater Admiralty Sciences (UAS) for the recovery of the aircraft. They had gained permission from both the U.S. Department of the Air Force, and the Newfoundland and Labrador Department of Tourism, Culture and Recreation, to remove the aircraft. I was on site as an assistant archaeologist, and then, when my supervisor had to return to St. John’s, the archaeologist on site. 444 Wing Goose Bay provided transportation, as they used our little group as a training, making sure we were prepared to stay out in the field overnight if they were called away. Along with myself, my supervisor and the UAS team, Hollis Yetman was hired as a guide. To be honest, I was a little disappointed that we didn’t get to spend the night in the lean-to he built, but at the same time, didn’t really want to sleep in a small space with a half dozen men!

First seeing the aircraft as we landed nearby. Photo by Lisa M. Daly 2008.

Douglas A-20 (F-3) 39-741 was one of 63 units ordered by the US Army Air Corp in 1939. This one and two others were modified to be prototypes for high-speed photographic reconnaissance aircraft, therefore had the designation F-3. The serial number had been removed from the aircraft by the time of this investigation (not sure if it was done around when it crashed or later), but Mark Allen of UAS said that they found “F-3 #2” painted on the interior sides of the speed rings in the #2 engine. He reported that the F-3 #2 confirms that this aircraft is the second F-3 unit (Deal 2009).

The serial number was cut off of the tail. Photo by Lisa M. Daly 2008.

The aircraft was piloted by Captain Secord on 10 October 1942 when it crash landed in a bog in a remote portion of the Little Macatina River, in southwestern Labrador (Allen 2008; Deal 2009). The crash card cites it as being “85 mi. SW Goosebay, Conn. Capt. Secord forced landing fuel shortage. Rep. 1.” but the site was actually “76 miles on a bearing of 218 degrees true from Goose Bay” (Allen 2008).  The aircraft was on a reconnaissance mission and landed due to low fuel. The crew were rescued three days later, but the aircraft was abandoned, most likely due to the remoteness of the crash site. Because of this, the aircraft was in relatively good shape with little evidence of activity on the site. That said, the propeller from the starboard engine had been removed, the cockpit had be burnt (perhaps done after the crash to destroy any potential remains of equipment), and some wooden pallets were found on site (Deal 2009). Given the wet nature of the area, those were probably more modern as the wood would deteriorate somewhat quickly if left on the surface of the bog.

Location of aircraft. From GoogleEarth 2017.

The goal of this project was for UAS to disassemble the aircraft and ship it to Georgia for restoration and eventually to be a museum display. I could not find an update as to the status of the aircraft, but anyone interested in historic aircraft knows that restorations take a lot of time. As this aircraft was to be removed from the crash site, archaeologists were on site to monitor and record the recovery process, to survey the aircraft and the debris field to create an archaeological map, and to record and recover material culture not attached to the main body of the aircraft (Deal 2009).

“Floating” the aircraft out of the bog so it could be disassembled. Photos by Lisa M. Daly 2008.

The archaeological survey consisted of setting up a site datum and mapping the body of the wreck before UAS could start their recovery. Because we were being air lifted into the site, we had the ability to use large fluorescent pins placed at the end of the wings, the nose at the tail of the aircraft so that a photograph taken from directly above the aircraft could be superimposed on the site map. One of the members of 5 Wing Goose Bay kindly leaned out of the helicopter to take the photograph.

Contour map of the A-20A site in Labrador (Deal 2009).

While UAS worked on taking the aircraft apart (which the archaeologists would often stop to watch) we continued to map the site by laying out a grid system to set up a baseline elevation of the site. As it was a bog, there really wasn’t a significant difference in elevation. In fact, because the site was so wet, we had to set up the pallet that was near the aircraft at the datum to be able to get accurate measurements. The ground was soft enough that standing near the surveyor’s level too long would cause it to sink and throw off the measurements. The pallet helped add some stability, and gave whoever was reading the level a spot to stand where they would not sink as much. Every few measurements one would have to walk away to let the ground rebound. The surface debris was recorded, then subsurface debris was identified. UAS kindly let us use their metal detector (because their was so good, we acquired a similar one to use for recording the Ventura site in Benton, the Hudson in Gander, and other sites) and it allowed us to get a better idea of the debris field. As time (and ability) permitted, the buried material found by the metal detector was uncovered. Some was too deep and were abandoned.

Douglas A-20A FbCj-01. Photo by Lisa M. Daly 2008.

A number of interesting artifacts were found, such as the cockpit escape hatch, part of the cockpit greenhouse (with plexiglass fragments still in the frame), the bomb release frame and lever, and a reconnaissance camera. The camera was of particular interest because it was found in two parts near the wing (where it would have been mounted). The case and cover were separate and there was still a little bit of film in the camera. The layout and ripping across the camera suggests that at the time of the rescue, the camera was probably removed from the aircraft, the cover taken off and tossed aside, and the film ripped out. What was left was packed with sphagnum moss for conservation purposes, then freeze dried when it was brought back to Memorial University. It was the end of the film, so nothing was on it, but it was still a rare find.

Body of the camera with moss packed inside to preserve the film until it could be brought to the conservation lab at MUN. Photo by Lisa M. Daly 2008.

In examining the debris on the site it looks as if the aircraft:

approached the bog at a bearing of 85 degrees to the northwest. It may have hit tail first, tearing away the underbelly. A concentration of artifacts from the cockpit area indicates that the nose of the aircraft hit hard and was torn away, leaving much of it imbedded in the bog. The aircraft then veered to the right and came to a stop at a bearing of 115 degrees to the northeast (Deal 2009).

Looking in the tail section. Photo by Lisa M. Daly 2008.

While the archaeological survey went relatively smoothly, the recovery seemed to have a number of problems, from trying to find and transport equipment to the site, to some of the bolts being too rusted to be able to easily remove, to the engines being too heavy and at first it was thought they might end up left in the bog. As well, when 5 Wing was slinging back parts of the aircraft, the wind would take the cumbersome pieces and cause them to spin. The first time it happened, the slight wound so tightly that it snapped back to unwind, and the crew almost ditched the piece, thinking it was going to cause an accident. Once it was figured out what was happening, the pieces were safely transported. The project started a few days later than expected due to hurricanes and low cloud cover, and, because of the hurricanes, the site was much wetter than anticipated. But, overall, the project was completed, and the aircraft is in Arizona awaiting restoration (Goose Hawk Unlimited 2017). Images can be found at

Taking a moment to pose while examining the debris inside the cockpit. Photo by Robert Mester 2008.

As I said at the start, this was a learning experience, both for myself and for the provincial protection of our aviation material. As more work is done, better legislation is written to help protect this material culture, but at the same time, the interest in not necessarily in just leaving airplane wreckage in the forests and bogs of Newfoundland and Labrador to eventually deteriorate. When projects are serious and will help to preserve material, then there is a chance to be able to remove it in an attempt to better preserve it.

Brining in a section of the aircraft. Photo by Lisa M. Daly 2008.


Allen, M. (Underwater Admiralty Sciences)
2008 A-20A Havoc: U.S. Army Air Corps Serial Number 39-741 Recovery Proposal. On file: Provincial Archaeology Office.

Deal, M.
2009 The A-20 Havoc Recovery Project. Provincial Archaeology Office 2008 Archaeology Review, vol. 7, pp. 30-35.

Goose Hawk Unlimited
2017 Douglas A-20 Havoc. [accessed 22 Feb 2017].

A list of surviving A-20s can be found here.

More pictures of this site and others around Goose Bay can be found here.

A rainbow over the crash site. Photo by Lisa M. Daly 2008.

Share Button