Go Back  PPRuNe Forums > Flight Deck Forums > Tech Log
Reload this Page >

Mach overspeed 737-800

Tech Log The very best in practical technical discussion on the web

Mach overspeed 737-800

Old 7th Mar 2010, 06:22
  #1 (permalink)  
Thread Starter
 
Join Date: Mar 2010
Location: china
Posts: 4
Likes: 0
Received 0 Likes on 0 Posts
Mach overspeed 737-800

Hi , just want some thoughts on the following situation. This has resulted in the captain being demoted to FO, and myself (the FO) being suspended from flying for one month with no pay, so to us guys its a pretty serious issue and any thoughts or comments as to is the blame ours, some environmental impacts such as wind changes, aircraft not behaving normally etc.


We were cruising at 38100 feet (China) and the aircraft was a 737-800. The FO (myself was pilot flying)
  • Descended to 33100 feet 1000 feet per minute using VNAV descent. Then maintained altitude until we were 2000 feet plus above the descent path (captains memory) 1000plus feet above descent path (my memory) not sure of the exact amount.
  • Told to descend and started descent using VNAV, at this time we had a tail wind component of -25 knots and speed of Mach .78
  • Drag required warning came on the FMC, relayed to the captain who said no need for speed break because so close to descent path. For previous experience I did not see a problem. I then entered some descent information on the FMC. Looked up and saw a rapidly (very rapidly) increasing green speed trend arrow. Which resulted in a Mach over-speed warning.
  • The following is from QAR and our actions.
  • At 40 seconds after beginning decent vertical speed was 5400 feet/Min (we know its fast) with a pitch down of 5.6 degrees, speed Mach .784 . Tailwind component 25 knots. This is probably about the time I noticed the speed trend arrow and called to the captain “speed” or something to that affect.
  • then aircraft started to pitch up reaching .527 degrees nose down
  • At 42 seconds after the decent began the speed break was deployed. At that point the aircraft maintained the same pitch down rate of .527 for 4 seconds and the vertical speed actually increased by 400 feet per minute and the airspeed also increased.
  • Then the aircraft started to pitch up, but due to momentum the air speed still increased, but at a slower rate.
  • At 46 seconds after the decent began VS was engaged by the FO and a low decent rate dialled up.( this does not show on the QAR but both captain and FO agree it happened.
  • AT 47 seconds after the decent began the captain selected level change.
  • At 50 seconds after the decent began , at 29740 feet we had the Mach over speed warning sound at Mach 8.26, with a decent rate of 3400 feet per minute
  • At 51 seconds the speed was Mach .831 before the speed and descent rate rapidly reduced and the flight continued. The warning sounded for 6 seconds, with us being over speed for 2 seconds and equal to the speed for one second.
Wind for the first 40 seconds was steady and gave us an effective tail wind of 25 knots. At 40 seconds this rapidly changed to a tail wind component of 45 knots in less than 20 seconds. We think the wind change contributed quite a lot to the overall situation.


Any thoughts, especially anything useful we can use in our defence. Also if the blame is totally ours then that something we have to accept. Thanks
InChina is offline  
Old 7th Mar 2010, 06:48
  #2 (permalink)  
 
Join Date: Mar 2006
Location: Choroni, sometimes
Posts: 1,974
Likes: 0
Received 0 Likes on 0 Posts
I'm not current on 737, so what is your MMO limit? What does the AOM say if this speed limit has been exceeded?

For example, on A300 there is only a maintenance action required if exceeding VMO+20 kt or MMO+0.02.
hetfield is offline  
Old 7th Mar 2010, 07:01
  #3 (permalink)  
 
Join Date: Mar 2001
Location: I wouldn't know.
Posts: 4,497
Likes: 0
Received 0 Likes on 0 Posts
MMO limit is 0.82. The QRH basicly says to reduce speed if the speed is above VMO(340)/MMO. It would be a mandatory techlog entry and i would write a report afterwards as well.

Anyway, personally i do not like to use VNAV when more than a couple 100 feet above path because exactly this scenario can happen despite my fleetmanager telling me that flying VNAV has more protections than using simple airmanship. Level change with a slightly higher descent speed usually sorts out the tailwind and above path situation, if needed i can take the speedbrake as well. Approaching the path i can switch then back to VNAV. VNAV usually tries to aggressively recapture the path with up to MMO/VMO and with an additional sudden increase in tailwind you can easily bust the limit which is not really a nice thing.

The measures taken seem pretty harsh though.
Denti is offline  
Old 7th Mar 2010, 07:47
  #4 (permalink)  
Per Ardua ad Astraeus
 
Join Date: Mar 2000
Location: UK
Posts: 18,579
Likes: 0
Received 0 Likes on 0 Posts
I would agree with Denti that the company actions as you have described them are excessive for what is a small exceedance.

Not sure if you have things like 'unions' to assist where you are, but I would look anyway at:

1) The Company Ops Manual/perceived company procedure for descending from high level ie VNAV to be used as primary tool?

2) If so, what guidance/warnings/training you are given on its use

As Denti says, 20/20 hindsight now will have you both using LVL CHG in future, and Denti's Fleet Manager is talking horse **** and would appear to have little practical experience of the a/c! 'Protections' in VNAV are virtually non-existent and what 'protections' there are are only effective when the old horse has bolted and not before.
BOAC is offline  
Old 7th Mar 2010, 08:27
  #5 (permalink)  
 
Join Date: May 2001
Location: A few degrees South
Posts: 809
Likes: 0
Received 0 Likes on 0 Posts
I agree with previous posts. It is however proof that aircraft still have to be flown, even in "reduced automatics", like LVL CH or V/S. Calling yourself a pilot on the B737NG if you are only able to set a new ALT in the window, pressing VNAV, then watch the show and not see that trend vector going past the "zipper"? hmmm..
Watever button you push, you should await the result and airplane`s response. And if necessary correct it.
It shows that too many pilots nowadays try to talk to FMC`s, and let them fly the aircraft. I see it on a daily basis, with identical results. FLY first!
latetonite is offline  
Old 7th Mar 2010, 08:35
  #6 (permalink)  
 
Join Date: Oct 2005
Location: Sweden
Posts: 35
Likes: 0
Received 0 Likes on 0 Posts
I wouldn't care about changing mode in that case to V/S or LVL CH. I would just pull the yoke back, let the airplane go to CWP.

It's the easiest and most effective way of getting the airplane to do what you want.

(or just press the magic button (auto pilot disconnect) )
Andrén is offline  
Old 7th Mar 2010, 08:42
  #7 (permalink)  
 
Join Date: Nov 2005
Location: right here
Posts: 342
Likes: 0
Received 0 Likes on 0 Posts
also agree with previous. clacker comes on at .823 (i KNOW that )
once above vnav-descend profile i (also) strongly recommend NOT to use VNAV to recapture as it will dive heads down for the path. while it is true that there are 'protections' the problem is that the auto-flight system installed works very sloppy. in some situations you can even get into over-speed with LVL CHG set to .82/340 as the auto pilot lacks preciseness.
or in other words: even if your entries on the MCP are 'in limits' the inertia of the system will bust those entries/limits. not sure what your employer makes of this.
FCS Explorer is offline  
Old 7th Mar 2010, 09:40
  #8 (permalink)  
Moderator
 
Join Date: Apr 2001
Location: various places .....
Posts: 7,177
Received 92 Likes on 61 Posts
I can only comment on the thread posts.

However, the event appears to be trivial so far as an Mmo exceedance is concerned.

Suggest you check your AFM for any oddball restrictions but, in general, Vmo/Mmo

(a) ought not to be INTENTIONALLY exceeded by the pilot, other than for

(b) authorised test OR

(c) authorised training.

A minor exceedance is a non-event and quite common during descent unless you are maintaining a moderate margin below the barberpole.

From the thead, I would suggest that one probably ought to have done something to control the VNAV excursion earlier and more positively but that's about it.

As a point of note, I think I only ever did one descent in VNAV (-300), wasn't impressed, and always flew the descent thereafter either in level change or, more commonly, handflew. Sounds like Boeing hasn't improved the toy at all ...

Refer to FAR 25.1505 for information.
john_tullamarine is offline  
Old 7th Mar 2010, 12:52
  #9 (permalink)  
Thread Starter
 
Join Date: Mar 2010
Location: china
Posts: 4
Likes: 0
Received 0 Likes on 0 Posts
Hi thanks for the replies.
At this company we don't have any SOP so as a first officers we find ourself having to fly the way a particular captain wants, that means changing our flying style to suit the captain, landing with lots of power or no power depending on the captain, using Vnav/level change or VS depending on the captain. With this particular captain he likes Vnav so that's why it was selected as opposed to either level change or VS which I prefer....And before people say anything about CRM or pilot flying, if as first officers we don't do the flying as a particular captain wants, that captain will change the MCP mode so we have to fly the way he wants.
Anyway this is off the point, so please don't say how crazy that is, its just the way it is at this company.


What we are especially wondering is did we do anything against anything found in the Boeing manuals. Yes we know we made a bad judgement, but was the course of action actually wrong? Its my understanding that VNav should offer protection against busting Mmo. I know some people have said that that the Vnav protection is rubbish (and I now 100% agree with you) but as pilots have we actually done anything wrong by expecting the Vnav to offer protection?
As I stated we did spot the airspeed trend arrow increasing and we tried to correct accordingly. Again, our actions might not have been the best or as fast as needed, but were they actually wrong? The company are saying it's totally pilot error, nothing to do with the wind change and also that Vnav offers no protections at all.........
It should be noted that the time from noticing the airspeed trend arrow to busting the Mmo was 10 seconds, so in those 10 seconds when we were trying to slow the aircraft it was still accelerating.

Also we think the tail wind component increasing by 20 knots in such a short space of time was also a big factor and this is why the aircraft could not offer the Mmo protection.

You thoughts and comments please.... and I appreciate your constructive comments, +ive or -ive Thanks.
InChina is offline  
Old 7th Mar 2010, 13:32
  #10 (permalink)  
 
Join Date: Jan 2005
Location: France
Posts: 2,315
Likes: 0
Received 0 Likes on 0 Posts
Brief question/note, and probably not relevant....

There's a mention of an abrupt change in tailwind as a contributing factor.
Could that have been accompanied by an abrupt decrease in OAT?
That would cause a drop in the speed of sound, hence an increase in Mach number for the same airspeed.

Just asking because we ran into the same thing with Concorde.

The problem wasn't recognised until the route proving, because the phenomenon doesn't really occur at higher latitudes.
The result of the sudden increase in Mach number was a quite unpleasant pitch excursion as the A/P tried to stay within the M=2.04 limit.
The problem was solved in the end by a major modification of the Max Cruise mode, with both A/P and A/T being made to "do their share" to keep the aircraft inside the limits.

CJ
ChristiaanJ is offline  
Old 7th Mar 2010, 15:23
  #11 (permalink)  
KAG
 
Join Date: Oct 2007
Location: France
Posts: 749
Likes: 0
Received 0 Likes on 0 Posts
Drag required warning came on the FMC, relayed to the captain who said no need for speed break because so close to descent path. For previous experience I did not see a problem. I then entered some descent information on the FMC. Looked up and saw a rapidly (very rapidly) increasing green speed trend arrow. Which resulted in a Mach over-speed warning.
Were you in VNAV speed, or VNAV path on your FMA?


Using VNAV is objectively tricky to use.
Many pilots never use it during the descent, because even with a good understanding of the VNAV mode, many incident happens every year due to the use of VNAV during the descent (and approach...). VNAV has to deal with the flight path, speed, pitch, rate of descent, thrust all at the same time and autonomously as the situation evolves. Wrong wind input or altitude constraint will automatically result in the pilot adjusting the thrust (speed decreasing) or the speed break (speed increasing). As a result choosing VNAV for the descent should imply to take very seriouly a FMC warning "drag required", and a very close speed monitoring.

There's a mention of an abrupt change in tailwind as a contributing factor.
Could that have been accompanied by an abrupt decrease in OAT?
That would cause a drop in the speed of sound, hence an increase in Mach number for the same airspeed.
Very interesting indeed...

Cheerio InChina good luck.
KAG is offline  
Old 7th Mar 2010, 16:08
  #12 (permalink)  
 
Join Date: Dec 2000
Location: on the golf course (Covid permitting)
Posts: 2,130
Likes: 0
Received 0 Likes on 0 Posts
I then entered some descent information on the FMC
Don't be surprised then when the FMC is given 'better' descent info (w/v) that it may pitch down to regain the Vertical profile! Why were you programming the descent at that point?

At the end of the day, if you see a situation develop( like an overspeed), your first reaction should be to fly the aircraft (via a/p modes - select v/s or flch, or handfly) but not to reprogramme the FMC!

I'm sorry, but you have to remember who is in control, you or the computer. I have little sympathy for you (and even less for the captain).

Having said that, your operator sounds like a nightmare, but I'm sure you'll learn your lesson!
TopBunk is offline  
Old 7th Mar 2010, 17:24
  #13 (permalink)  
Per Ardua ad Astraeus
 
Join Date: Mar 2000
Location: UK
Posts: 18,579
Likes: 0
Received 0 Likes on 0 Posts
I have for a long time (20 years) been convinced there is an underlying software 'bug' in the Classic/NG pitch A/P computations. To avoid diverting this tread I am starting a new one on this topic at B737 A/P pitch software - PPRuNe Forums
BOAC is offline  
Old 7th Mar 2010, 17:39
  #14 (permalink)  
 
Join Date: Apr 2004
Location: Planet Earth
Posts: 2,083
Likes: 0
Received 8 Likes on 7 Posts
Agree with previous posters that you need to be more assertive about controlling the Aircraft ,not just wait for the automatics to fly it and then react too late.


However, demoting the Ca to Fo and docking you a months pay for a .06 Mach exceedance



That's quite unbelievable. I think I would be looking for employment elsewhere..
stilton is offline  
Old 7th Mar 2010, 18:14
  #15 (permalink)  
 
Join Date: Dec 2006
Location: Still looking
Posts: 126
Likes: 0
Received 0 Likes on 0 Posts
To use or not to use

VNAV is good at some things and franlky pretty poor at others. Now I know the saying input = output but I'd agree that its not always the smartest box of tricks. I never have understood why it'll happily dive into the clackers. For me as a rule, initially I never use VNAV when past TOD or above profile unless at a low enough altitude to have a good margin from the clackers. Even then one tends to get a dirty dive followed by a correction when at the path that results in an uncomfortable ride for all. If the situation is right use vnav but use speed intervent to limit the speed and hence the dirty dive. As mentioned in the tread before, use CWS if needs be. The slower descent may be a problem if a level restriction applies. If to be level by xyz can't be met without compromising the a/c handling say so and they'll have to negotiate another plan. ATC are not always going to realise the limitations you have to work with.

Think the company's response is less than positive. I would have hoped they would take the lessons learned and use it in the next round of training. If my memory is correct an engineering inspection is required if in the clackers for twenty seconds or more by X amount of knots. There is actually a good margin there before any damage to the wing skins is done.

Skyloone

Last edited by skyloone; 7th Mar 2010 at 18:30. Reason: spelling
skyloone is offline  
Old 7th Mar 2010, 18:52
  #16 (permalink)  
 
Join Date: Jan 2007
Location: EU
Posts: 1,231
Received 0 Likes on 0 Posts
Sorry to hear of your plight InChina.

I would agree with the comment's about VNAV's tendency to dive for the path regardless of speed and the dangers of hitting VNAV and assuming all will be well. I've seen the 'clacker dive' in the sim and on the line. We're taught to use LVL CHG and a chosen speed until back on path to avoid problems.

I should add that our SOPs require the input of best available info on winds during the descent, QNH and temp at destination and encourage regular reselection of the next WPT to update the winds in order to keep the path accurate. As always, one has to be very careful to check VNAV's actions after making any updates.

Our ops manual also mentions that you will see a CDU message warning of 'drag required' if an unexpected tailwind results in a significant increase in airspeed to maintain the path.

Unfortunately, it sounds like the scenario that caught you out. Company's reaction certainly seems OTT given their lack of guidance or SOPs.
Mikehotel152 is offline  
Old 7th Mar 2010, 19:10
  #17 (permalink)  

Only half a speed-brake
 
Join Date: Apr 2003
Location: Commuting not home
Age: 46
Posts: 4,299
Received 3 Likes on 3 Posts
Adding tailwind component would initialy bring reduction in IAS/Mach, no? Only when the aircraft is "established" in the new layer of air it will deviate above (forward of) calculated descent trajectory. The increase in speed is a result of AP guidance attempting to regain VPTH if that is the active mode. Of course it should stay clear of the barber pole. My bet is on temperature change. The existence of air layer with different properties is exhibited by the sudden and significant wind change. Again, positive tailwind shear makes your airspeed drop, not the other way around.

FD (the un-real)
FlightDetent is offline  
Old 7th Mar 2010, 19:49
  #18 (permalink)  
 
Join Date: Dec 2002
Location: UK
Posts: 2,451
Likes: 0
Received 8 Likes on 4 Posts
In addition to BOAC's input, it is possible that the changes, particularly the combination and/or timing, were beyond the capability of the autoflight system – a design limit, which requires pilot intervention.
Autopilots have limited capability when adapting to rapidly evolving situations. Thus, when demanding a descent, the pitch channel will lower the nose and there is a tending to increase speed. Auto trim may subsequently run nose-down for the speed increase.
If the autothrottle is not well matched to the pitch command, then the thrust reduction (due increasing speed) is slightly delayed and when it occurs, may add to the nose-down pitching moment (force), which together with the new trimmed condition (also nose down force) may exceed the opposing auto pitch control (nose-up force), due to lack of autopilot ‘muscle’.
Thus, speed will increase until the nose-up control and delayed auto trim input (always delayed to prevent oscillatory motion) takes effect. During the intervening time, it is possible to exceed Mmo, particularly when cruising at high speed.

The effect may be more noticeable in VNAV if that mode uses a ‘soft ride’ feature – reduced autopilot ‘muscle’ and/or rate of control application, and also if the cg is nose-heavy and/or the airbrake causes an adverse pitching moimnet - often a dominant feature (cf Avro RJ operators).

The situation has to be understood as a limit in the capability of automation, which pilots are expect to know, detect, and correct, but which they might only learn from by encountering the situation – experience!
Thus the incident may have been a limit of expertise.

Blame should not be allocated; the situation is an opportunity to improve understanding. Thus, both the crew and the management should have learnt something – the crew about aircraft / autoflight limits; management about the need for training, and the need to gain experience – often via small errors.

We all suffer error, many occurrences during each flight. Generally, they are inconsequential because the source of the error is detected (understood) or the resultant, adjusted in time to avoid a consequence.

We should only worry about the big errors; most of those originate from management.
In this instance, it appears that the management error is the lost opportunity to learn – about investigating incidents, about aircraft systems, about human factors, and the concept of a no-blame culture.

P.S. If you ever consider the need to ‘override’ the autopilot, always disengage the system and fly manually. Manual flight provides a quicker and more precise feel for the situation; particularly for aircraft trim.
Some aircraft may have nasty surprises when attempting to mix auto/manual flight – the trim may run against you, and this can make the situation very much worse. The aircraft / trim muscle may equal your muscle!

Last edited by safetypee; 7th Mar 2010 at 20:05.
safetypee is offline  
Old 8th Mar 2010, 01:24
  #19 (permalink)  
 
Join Date: Jul 2001
Location: East of West and North of South
Posts: 549
Likes: 0
Received 0 Likes on 0 Posts
InChina:
What we are especially wondering is did we do anything against anything found in the Boeing manuals.
To stictly answer your question - Yes as a crew you did:


OM B vol 2 chapter 11:

ADVISORY MESSAGE:
DRAG REQUIRED

CAUSE:
Airspeed is 10 kts or more above FMC target speed or within 5 kts of Vmo/Vmmo.

CORRECTIVE ACTION:
Use speedbrakes, trim or reduced thrust, as required, to bring the airplane within 5 kts of FMC target speed.
Since thrust was already at idle your only corrective action available according to the manual would be to use the speedbrakes. You did not do so accordingly.

InChina:
I then entered some descent information on the FMC. Looked up and saw a rapidly (very rapidly) increasing green speed trend arrow. Which resulted in a Mach over-speed warning.
This may account for the discrepancy between the recalled path deviation from you and the captain. If you changed e.g. an altitude constraint or descend winds, deleted a waypoint etc, the FMC calculates a new path. One of you may remember the old path and the other the new. If the new path suddenly predicts you much higher, the autopilot will increase vertical speed, maybe contributing to the pending overspeed situation.

As stated by many other posters I too would recommend to abandon VNAV at the moment DRAG REQUIRED is annunciated. This is another good reason to switch to LVL CHG when above the path and messing with the FMC, since you don't know witch immediate effect it may have on the flight path.

InChina:
As I stated we did spot the airspeed trend arrow increasing and we tried to correct accordingly. Again, our actions might not have been the best or as fast as needed, but were they actually wrong?
Before the overspeed condition, I would have pulled the wheel into CWS as some other posters also mentioned. But in the overspeed condition Boeing recommends to leave to AP engaged.

FCTM chapter 8, Non-Normal operation: Overspeed:
When encountering an inadvertent overspeed condition, crews should leave the autopilot engaged unless it is apparent that the autopilot is not correcting the overspeed. However, if manual inputs are required, disengage the autopilot. Be aware that disengaging the autopilot to avoid or reduce the severity of an inadvertent overspeed may result in an abrupt pitch change.
During climb or descent, if VNAV or LVL CHG pitch control is not correcting the overspeed satisfactorily, switching to the V/S mode temporarily may be helpful in controlling speed. In the V/S mode, the selected vertical speed can be adjusted slightly to increase the pitch attitude to help correct the overspeed. As soon as the speed is below VMO/MMO, VNAV or LVL CHG may be re-selected.
You did the right thing in switching to V/S. As can be seen from the above the captain incorrectly switched to LVL CHG. Maybe you can go free on this one, especially since the captain also declined your request for using the speedbrakes as appropriate according to the manual in the first place. But I guess it depends on your morale and if you feel that the captain asked you to operate the aircraft differently than you yourself intended and if this was the cause of this situation.

Good luck - especially in finding a better company. You seem like you have a good attitude, do not shy from your responsibility and have willingness to learn from this experience. All good qualities for an aviator in my opinion.
cosmo kramer is offline  
Old 8th Mar 2010, 03:18
  #20 (permalink)  
 
Join Date: Aug 2008
Location: North America
Age: 64
Posts: 364
Likes: 0
Received 0 Likes on 0 Posts
Automation is great but it is not, nor ever will, be perfect.

There are several things about the NG that I don’t like. One of them happens to be the tendency to over speed. I think Boeing is improving the software some. When we first got these things (NGs) we had many cases of crews over speeding coming out of altitude.

Sorry about the time off from work. It seems a little harsh to me; but then you are in China, it’s their airline, their rules, and their culture.

Hindsight is 20/20 so I will dispense with “you should have done…..” and “the manual says…..” and barking at you to “fly the airplane”. At this point it’s counterproductive, and the event is self-critiquing. There are known issues with VNAV in the NG. The NG will over speed in VNAV, the system is not perfect. Whose responsibility is it to keep this from happening; yours and mine (the folks who get paid to fly it). When the NG gets off VNAV PTH (does it frequently) and tucks into VNAV SPD it will over speed if given the opportunity, as you have unfortunately found out.

What did you have on the descent page, what cost index are you using? One suggestion would be to put in something like .76/280 in the descent page. Doing so should give you a larger margin. Another technique to consider is to start down early in VS (or “descend now”) and then transition to VNAV PTH.

I like the wings, engines, performance, auto throttles and the HUD on the –800. The flight deck ergonomics, 2-axis autopilot, elements of the FMS and the noise level leave a lot to be desired.
Northbeach 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.