Go Back  PPRuNe Forums > Flight Deck Forums > Accidents and Close Calls
Reload this Page >

4th Aug 2018 Junkers JU52 crashed in Switzerland

Accidents and Close Calls Discussion on accidents, close calls, and other unplanned aviation events, so we can learn from them, and be better pilots ourselves.

4th Aug 2018 Junkers JU52 crashed in Switzerland

Old 9th Aug 2018, 07:25
  #181 (permalink)  
 
Join Date: Dec 2001
Location: what U.S. calls ´old Europe´
Posts: 941
Likes: 0
Received 0 Likes on 0 Posts
Normally you and especial the two experienced pilots would have recovered the bird within 500 to 1000 feet even from a power-on or power-off stall. And it did not crash in a flat spin. There is reason to believe that they had those altitude margin.
Spinning is a quite complicated "maneurvre". It takes an initiation phase before you are in a stable spin and terms like "flat spin" make sense. This phase may already use up 1000 feet in a large aeroplane. During the first phase of a spin it is not unusual to fly inverted or in a steep dive before a stable equilibrum between all erodynamic and dynamic forces is established. Check this picture of the early entry phase into a (fatal) spin for example. In the following second the aircraft will be in an almost vertical dive still acellerating the rotational speed before entering a stable spin after somr 5-10 seconds. If you run out of altitude at that moment, the result looks like in this picture. Which matches the Junkers wreakage quite well.
The recovery altitude given in the handbooks is between starting countermeasures in a stable spin and the full recovery. The altitude between stall and stable spin can not be given, as there are too many different scenarios how to enter it, for example the actual bank angle at the time of the stall.

anyone with even a rudimentary knowledge of aviation could easily come up with several obvious scenarios as to how the aircraft came to the fate it did
For me there are basically two realistic scenarios, stall and spin or elevator control (elevator trim) failure. Although people talk about "high speed" I think the low fragmentation of the wreakage does not confirm that, Impact speed must have been well below 200 km/h (compare it with the oskosh mustang, where only small fragments were left and the aircraft lierally "exploded" at impact purely due to kinetic energy). So for me the stall and spin theory, linked to relatively slow inpact speed, makes more sense. Although a stable spin has never been established due to low altitude, so it is stall and spin entry... A former BAZL expert thinks the same.

The sad aspect is, in this case the investigation has nothing to find. The wreakage will tell no story, there will be no "smoking gun". There will be no corrective action. It happened a thousand times before, it will always happen.
Volume is offline  
Old 9th Aug 2018, 09:31
  #182 (permalink)  
 
Join Date: Sep 2004
Location: Royal Berkshire
Posts: 1,725
Received 76 Likes on 38 Posts
Originally Posted by Volume
Spinning is a quite complicated "maneurvre". It takes an initiation phase before you are in a stable spin and terms like "flat spin" make sense. This phase may already use up 1000 feet in a large aeroplane. During the first phase of a spin it is not unusual to fly inverted or in a steep dive before a stable equilibrum between all erodynamic and dynamic forces is established. Check this picture of the early entry phase into a (fatal) spin for example. In the following second the aircraft will be in an almost vertical dive still acellerating the rotational speed before entering a stable spin after somr 5-10 seconds. If you run out of altitude at that moment, the result looks like in this picture. Which matches the Junkers wreakage quite well.
The video of the seaplane crash into the Swan River last year is worth looking at if not seen before........


.
GeeRam is offline  
Old 9th Aug 2018, 10:14
  #183 (permalink)  
 
Join Date: Oct 2017
Location: Tent
Posts: 916
Received 19 Likes on 12 Posts
Originally Posted by GeeRam
The video of the seaplane crash into the Swan River last year is worth looking at if not seen before........

http://www.youtube.com/watch?v=e5c2U...eYrf3jV9pG5VSc

.
Trying to find the final report on that crash to see what went wrong.

Seems to have a bit of evidence around unlike this crash.
Bend alot is offline  
Old 9th Aug 2018, 10:52
  #184 (permalink)  
 
Join Date: Aug 2018
Location: West
Posts: 4
Likes: 0
Received 0 Likes on 0 Posts
Following FollowTheSupper #168 and PaxBritannica #175:

The spotted aircraft in #168, picture in #175, is in the middle line of view to Piz Aul (left), 28km from panorama camera and Piz Mendel (right), 37km. Both mountains are 16km apart, but at the distance of Piz Aul the view lines have spread of 12km, in the picture 450 pixel

The aircraft in the middle has between 4-6 pixel for a 30m wingspan (assuming it is flying towards the camera/Piz Segnas). As the dark pixels are much wider than high, as can be expected for a airplane flying stable shot from the front, I would assume it is not an camera/optics/resolution artefact giving that size/shape of the dark spot.

Numbers above calculate to a distance between 5.3 to 8.1km to the panorama camera, which is about double distance to the cable on top of the ridge in between. The shortest distance to the crash site is then between 8.5 and 11km, measured from map. Calculating with a speed of like 180km/h this would be just between 1.5 to 2.6 minutes.

If the camera operator can further specify the exact time when the "16:50" picture was really taken, and if the exact time of picture in #157 is known or at least time of crash (16:57 ?), this would allow a sharper calculation of average speed in the approach, but accuracy not better than 20% (from distance and assuming straight-line flight) and still subject to time difference accuracy.

Of course the pixels don't need to be the JU52, but specifically the direction and distance are very reasonable.
MadMax9 is offline  
Old 9th Aug 2018, 21:07
  #185 (permalink)  
 
Join Date: Aug 2018
Location: West
Posts: 4
Likes: 0
Received 0 Likes on 0 Posts
Analyzing picture in #151 with photogrammetric calculations (just in excel):

From 8 different topographic marks on the photo and using lat, lon, height values of those marks from WGS84 system in the map referenced in #157 the position/direction of the camera can be derived by mean-squared-error method:
Camera 46.902233 N, 9.224865255 E, 2639m Az 133.51° Elev 6.7°
This is close to the Segnespass mountain lodge, but 12 m higher. Camera direction therefore is SE with 6.7° slightly upwards, 5° rotated clockwise.

The location of the plane shadow is where the line of sight crosses the terrain.
Shadow: 46.8987249855829 N, 9.23385 E, 2541m, distance from camera 792m

Direction of Sun at position of shadow Sat Aug 4, 16:50 local time can be derived rom several sites like sunearthtools. Of course this is the direction to the plane as well to create the shadow.
Sun: Az 252.88°, Elev 38.3°

Last element is to find the point were sun-shadow line crosses with line-of-sight to the plane derived from the photo:
Plane: 46.8981223562367 N, 9.2309869436985 E, 2721m, distance from camera 657m, 82m above camera
Actually the two lines are not crossing exactly, but have 15 pixel distance, which is half the planes size, likely due to sum of all small inaccuracies, which we have to be aware of.

The terrain height at the planes position is measured again from the map as 2473m. Thus the Ju52 has is in the moment of photo taken an calculated AGL of 248m.

BTW, the height of 2721m is about 100m above the lowest parts of the Segnespass ...

Last edited by MadMax9; 10th Aug 2018 at 09:46. Reason: improved accuract
MadMax9 is offline  
Old 9th Aug 2018, 22:31
  #186 (permalink)  
 
Join Date: Aug 2018
Location: West
Posts: 4
Likes: 0
Received 0 Likes on 0 Posts
Actually, the calculated position would be just 200m SE from the crash site
MadMax9 is offline  
Old 9th Aug 2018, 22:45
  #187 (permalink)  
 
Join Date: Aug 2018
Location: With hardship in an old Astra
Posts: 45
Likes: 0
Received 0 Likes on 0 Posts
MadMax9,



Many thanks for taking the trouble to post a range / time analysis. However, I suspect that the peak which is 450 pixels to the right of Piz Aul in the panoramic image... is actually Piz Terri, and not Piz Mendel. The latter is much further away, and by my reckoning from the map should appear close to a cable-car pylon on the ridge... not the previously-mentioned one (post #168), but the one which stands to the right of the wooden post… even allowing for ±100m uncertainty in the actual camera position. (Also, if you do an internet search for Piz Terri, there is an excellent summit photo taken from a slightly more southerly position, which can be matched well to the summit which appears in the panoramic image.)



As the bearing from the panoramic camera to Piz Terri is closer to the bearing to Piz Aul (than is the bearing you have proposed to Piz Mendel), then that would place the aircraft further away.



In any case, a safer analysis (avoiding any debate about summits!) might be to note that the full 360° panoramic image width is covered by 15707 pixels, making 0.0229° per pixel. Taking the aircraft width as 4 pixels, and assuming that the aircraft is approaching the camera (I agree with you that the lack of height or asymmetry in the dark pixel distribution supports this orientation), then a 30m wingspan would subtend 4 pixels at a range of around 18.7km (or 15km if the wingspan is taken as being spread over 5 pixels). This would then place the aircraft over (or slightly beyond) the Piz Mundaun feature, i.e. about 5km SW of the town of Ilanz.

Also, I suspect that 180km/hr as a groundspeed might be a bit optimistic for these aircraft – particularly on a leisurely sight-seeing flight, the appeal of which would surely be more about the flight than a timely arrival. (Indeed, when I see them flying past my house, their progress is usually – shall we say – somewhat unhurried!) So, if the dark pixels are indeed the JU52 around Piz Mundaun, then it could perhaps be 7 to 8 minutes after the image was captured, before it reached the Segnespass area.
FollowTheSupper is offline  
Old 10th Aug 2018, 12:24
  #188 (permalink)  
 
Join Date: Jun 2001
Location: at the edge of the alps
Posts: 445
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by Volume
The sad aspect is, in this case the investigation has nothing to find. The wreakage will tell no story, there will be no "smoking gun". There will be no corrective action. It happened a thousand times before, it will always happen.
Inspection of the engines and fuselage will show whether there were any engine/propeller/structural issues and analysis of surviving video recordings might help to understand the flight path and at least some of the flight dynamics. There must have been at least half a dozen cameras running in that airplane. There was one in the Embraer in Durango a couple of days ago and that was a normal airline flight without much to see outside, so there must have been more on a scenic flight with an Oldtimer passing spectacular landscape.
Alpine Flyer is offline  
Old 10th Aug 2018, 12:38
  #189 (permalink)  
Pegase Driver
 
Join Date: May 1997
Location: Europe
Age: 73
Posts: 3,657
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by Volume
The sad aspect is, in this case the investigation has nothing to find.
If the aircraft was FLARM equipped ( it should be ) and if the device was not severely damaged by the impact , the flight can be downloaded and the final minutes analysis could give some clues
ATC Watcher is offline  
Old 10th Aug 2018, 15:17
  #190 (permalink)  
 
Join Date: Jul 2007
Location: Switzerland
Age: 78
Posts: 109
Received 7 Likes on 2 Posts
None of the JU-AIR aircraft were FLARM equipped according to the Bundesamt fuer Zivilluftfahrt - the Swiss Civil aviation authority. This is difficult to understand by the way, particularly because they often operated in a region with a lot of glider activity. And the ATC radar of Zurich does not cover the crash region at this altitude. There's a chance that the military primary radar might have picked up some signals, but certainly not with a frequency of 1 Hz like the Flarm.
However, they might have carried a GPS device that recorded the flight.

Last edited by clearedtocross; 10th Aug 2018 at 17:16.
clearedtocross is offline  
Old 10th Aug 2018, 16:31
  #191 (permalink)  
 
Join Date: Aug 2018
Location: West
Posts: 4
Likes: 0
Received 0 Likes on 0 Posts
FollowTheSupper,

agree to your corrections regarding larger distance from Mutta Rodunda.

Regarding the calculated position of the JU52 from the picture #157 (my post #186) it would be only 12 to 16 seconds time from there to fly to the Segnespass (657m distance to the photographer close to the lodge), subject to speed. But not much time to gain significant height with climb rate of maybe 2m/s when fully loaded at higher altitude/high temperature. Maybe somebody can calculate that figure. A northerly head wind coming over the ridge and going down the valley with conservative 30km/h = 8m/s would further reduce the effective climb rate a lot.

Actually the crash site is located less than 50m north of that straight line, which would fit to the left bank attitude visible in the last photo. And with only 200m distance from the last photo position to crash site this gives just 4-5 seconds remaining bevor the tragedy occurs. This does not tell us if the intended flight path was to still go over the pass with a low margin, or into a 180° turn. However, with the JU52 position at that time not far from the middle of the valley, the turn radius would have to be less than around 300m.

You can find a map with positions and view lines if you copy this link to swiss map: s.geo.admin.ch/7c4ffa5367
MadMax9 is offline  
Old 10th Aug 2018, 16:56
  #192 (permalink)  
 
Join Date: Jun 2002
Location: Geneva, Switzerland
Age: 58
Posts: 1,903
Received 3 Likes on 3 Posts
Originally Posted by clearedtocross
N
This is difficult to understand by the way, particularly because they often operated in a region with a lot of glider activity.
Indeed surprising.
One less source if information
atakacs is offline  
Old 10th Aug 2018, 19:25
  #193 (permalink)  
 
Join Date: Jun 2002
Location: east ESSEX
Posts: 4,624
Received 65 Likes on 40 Posts
At 180kph/97kts TAS,30* BANK,,radius-1500ft; rate 6.5*/sec.
At 40*bank,,radius 1100ft;rate 9.5*.sec.
At 220kph/120ktsTAS,30*bank,radius 2000ft, rate 5.5*/sec..
At 40*bank,radius 1500 ft., rate 8*/sec...
close approximation..
sycamore is online now  
Old 10th Aug 2018, 20:58
  #194 (permalink)  
 
Join Date: Oct 2007
Location: Here
Posts: 961
Received 3 Likes on 2 Posts
MadMax9

Thanks for your analysis.

Just to let you know that post numbers can change with time and your posts may be more difficult to follow some time in the future. I assume if posts are deleted then the numbers will get re-calculated. I also imagine that if a user deletes their account that all their posts will disappear. I have observed incorrect post references many times on pprune so I think that posts get deleted quite frequently.

I would guess that the "(permalink)" link beside the post number is intended to be used to circumvent this issue.

Example permalink to #192:-
https://www.pprune.org/rumours-news/...l#post10220187

"https://www.pprune.org/rumours-news/611836-4th-aug-2018-junkers-ju52-crashed-switzerland-10.html#post10220187"
jimjim1 is offline  
Old 11th Aug 2018, 08:28
  #195 (permalink)  
 
Join Date: Sep 2010
Location: by the seaside
Age: 74
Posts: 558
Received 17 Likes on 13 Posts
Spin

https://youtu.be/HosF_xKeeGE
A straight spin in a K6 glider from st remy last year. Note the bank angle.
I found myself inverted in a Beech baron in 1971 from a clean power off stall. The aircraft flicked completely inverted but with very little pitch down.
I similarly found myself inverted whilst instructing mountain flying in a grob glider from the cape gliding club. Figures of 8 below 200ft and with a bang caused by the shear from a thermal rising up the quarry face again inverted with nose more or less on the horizon.
Steve Fossett got into the lee and I've had quite a few surprises in the last 20 years flying in the mountains.
MY local paragliding wind reporter had a gust last week exceeding 140kph.
The guys I knew who flew Tante Ju were ex SR captain's and military..another ball game
blind pew is offline  
Old 11th Aug 2018, 09:38
  #196 (permalink)  
 
Join Date: Jun 2002
Location: Geneva, Switzerland
Age: 58
Posts: 1,903
Received 3 Likes on 3 Posts
It could indeed be as simple as a strong unexpected gust in marginal conditions. I would still be surprised that those guys would have put themselves in such a corner.
That's why I'm trying to get real liffe performance info for a JU-52... It is really hard to gauge how close to the limits they might have been.
atakacs is offline  
Old 12th Aug 2018, 11:32
  #197 (permalink)  
 
Join Date: Jul 2007
Location: Switzerland
Age: 78
Posts: 109
Received 7 Likes on 2 Posts
The "NZZ am Sonntag" today printed an article with the headlines "Die JU-52 flog an ihrer Leistungsgrenze" which translates as "the JU was flying at its limits", a statement I made longtime ago in this forum and was bashed for it by people who thought to know better. The author of the article, aviation journalist Mr. Sepp Moser suggested that a vapour lock might have caused a coughing of one of the engines which made the pilots turn away rather abruptly to avoid rising terrain. Considering the fact that the aircraft was parked in Locarno under a blazing sun until 16:00 and that the Junkers has no gravity fuel feed, its a possible scenario, even more so because booster pumps should not be used at altitude because they make the engines run too fat.
clearedtocross is offline  
Old 12th Aug 2018, 11:54
  #198 (permalink)  
 
Join Date: Oct 2010
Location: Germany
Age: 47
Posts: 402
Likes: 0
Received 0 Likes on 0 Posts
of course speculation but i guess there is not much secrets in this tragedy .

probably they realized they do not have the performance to make it over the mountains infront so they tried a sharp 180 deg turn to save the day . during the turn , loosing more and more speed trying to maintain altitude , one wing stalled and they went down like a rock without being able to recover .

investigation surely will focus the reason for the lack of performance .

( just to heavy on this hot day or engine problems )

i guess thats all .
aerobat77 is offline  
Old 12th Aug 2018, 12:58
  #199 (permalink)  
 
Join Date: Aug 2005
Location: EDLB
Posts: 362
Received 4 Likes on 3 Posts
Don’t believe that. With a slow plane with a 18000 feet(ish) ceiling, you can and will circle all day on the mountainside with upwinds, until you have enough margin to make it over the pass. On that CAVOC day you can make that judgement only with visual clues. That they run out of performance only if they had engine troubles. And even then you let not get it into a corner but start a fly or glide to some decent emergency landing area early. With two experiences pilots that should be no task overload. One flies the airplane and the other deals with the technical problem. With such a relative low tech bird with mechanical control surfaces should be not much problem. I am sure there is more to the story and that will come out.
EDLB is offline  
Old 12th Aug 2018, 14:43
  #200 (permalink)  
 
Join Date: Jul 2007
Location: Switzerland
Age: 78
Posts: 109
Received 7 Likes on 2 Posts
EDLB Dream on about your ceiling of 18'000 ft. With the downgraded engines (500 instead of 750 hp) it would be difficult to coax it over 12'000 ft even on a standard day. By the way, we had 3 fatal accidents in Switzerland in the last 3 weeks, all with the same pattern: hot, high, unable to get over the next obstacle, spun into the deck. 28 dead. Really sad. Heat is a killer even if it shows a pretty face compared with other perils for VFR flights like bad visibility or high winds.
clearedtocross is offline  

Thread Tools
Search this Thread

Contact Us - Archive - Advertising - Cookie Policy - Privacy Statement - Terms of Service

Copyright © 2024 MH Sub I, LLC dba Internet Brands. All rights reserved. Use of this site indicates your consent to the Terms of Use.