B737 Airspeed Unreliable?

[Tomaski]

Originally Posted by galdian

If I have a suspected Airspeed Unreliable why would you do anything except maintain standard takeoff pitch - 15 degrees - and maybe tweak power back to MCT, sit fat dumb and happy until MSA THEN start screwing around with things?

You can do anything you want - as long as you can justify it.
To me achieving MSA then doing the Airspeed Unreliable checklist increases the safety of the aircraft, that's what I'd explain to those lined up against me at the enquiry - just remember the warm and fuzzy feelings Sully and the F/O received! - for them to chew on.

Ultimately the passengers are safe, the aircraft's secure on the ground and I've written/am writing the report(s)....what's your problem??

Cheers.

A significant challenge for creating this NNC is to come up with "memory item" pitch and power settings that work for all aircraft and all flight regimes. Presumably, MCT and 15 degrees does not work for all aircraft and all conditions. The tab data is limited, but the go-around chart suggests that for certain aircraft at heavier weights and high density altitudes (e.g. Mexico City), 15 degrees pitch would be too high. I suspect it might also be unsuitable for a heavy-weight, Flaps 25 takeoff. At the other end of the spectrum, a light-weight low density altitude scenario may result in a flap overspeed.

Keep in mind that once the initial steps have been completed, the crew should then move on to the expanded NNC. Step 7 of this checklist directs the crew to the tab data to obtain the appropriated pitch and power settings for their aircraft and conditions. If executed correctly and promptly, then it shouldn't take long before the aircraft is configured for a safe climb to MSA.

I would caution against intentionally disregarding the steps of an established NNC procedure. It's one thing to decide that a certain NNC does not fit a certain situation. It is entirely another to make up your own procedure. Don't expect your flight department or certificate authority to take such a casual attitude when they review your actions.

That said, I have three concerns with the current Airspeed Unreliable NNC procedures as they apply to takeoff (particularly when accompanied by a false stall warning). First, there is no clear guidance on when to retract the flaps. Second, there is only limited information in the NNC regarding erroneous system effects at either the low-speed (stick shaker, EFSM, STS stall ID, autoslats) or high-speed (overspeed clacker, flap load relief) regimes. Third, flight crews are not being exposed to this malfunction on a regular basis in the sim. It can be challenging situation to handle, but really not much more than an takeoff engine failure. The difference is that we practice one all the time, the other not so much.

[Dropp the Pilot]

The settings are not the same for all aircraft.

The 777 is 10/85 and 4/70.

[gearlever]

Originally Posted by Tomaski

A significant challenge for creating this NNC is to come up with "memory item" pitch and power settings that work for all aircraft and all flight regimes. Presumably, MCT and 15 degrees does not work for all aircraft and all conditions. The tab data is limited, but the go-around chart suggests that for certain aircraft at heavier weights and high density altitudes (e.g. Mexico City), 15 degrees pitch would be too high. I suspect it might also be unsuitable for a heavy-weight, Flaps 25 takeoff. At the other end of the spectrum, a light-weight low density altitude scenario may result in a flap overspeed.

Keep in mind that once the initial steps have been completed, the crew should then move on to the expanded NNC. Step 7 of this checklist directs the crew to the tab data to obtain the appropriated pitch and power settings for their aircraft and conditions. If executed correctly and promptly, then it shouldn't take long before the aircraft is configured for a safe climb to MSA.

I would caution against intentionally disregarding the steps of an established NNC procedure. It's one thing to decide that a certain NNC does not fit a certain situation. It is entirely another to make up your own procedure. Don't expect your flight department or certificate authority to take such a casual attitude when they review your actions.

That said, I have three concerns with the current Airspeed Unreliable NNC procedures as they apply to takeoff (particularly when accompanied by a false stall warning). First, there is no clear guidance on when to retract the flaps. Second, there is only limited information in the NNC regarding erroneous system effects at either the low-speed (stick shaker, EFSM, STS stall ID, autoslats) or high-speed (overspeed clacker, flap load relief) regimes. Third, flight crews are not being exposed to this malfunction on a regular basis in the sim. It can be challenging situation to handle, but really not much more than an takeoff engine failure. The difference is that we practice one all the time, the other not so much.

Will the stick shaker on 737-800 be activated if CPT or FO pitot blocked?

[Tomaski]

Originally Posted by gearlever

Will the stick shaker on 737-800 be activated if CPT or FO pitot blocked?

Yes. A faulty airspeed or AOA input will cause the bad side Stall Management Yaw Damper (SMYD) computer to generate a stall signal. This stall signal will then initiate four system responses:

• Bad side stick shaker
• Speed Trim Stall ID function (trims stab nose down, but not as aggressively as MCAS)
• Elevator Feel Shift Mechanism - increases elevator control forces
• Autoslats - slats will go to full extend with any amount of flaps extended.

All of these functions are useful in a true stall, but can create unnecessary distractions when the stall condition is erroneous. There is no compare function between sides, so one faulty input can create this cascade of system effects in addition to the other challenges of an Airspeed Unreliable scenario.

[gearlever]

Originally Posted by Tomaski

Yes. A faulty airspeed or AOA input will cause the bad side Stall Management Yaw Damper (SMYD) computer to generate a stall signal. This stall signal will then initiate four system responses:

• Bad side stick shaker
• Speed Trim Stall ID function (trims stab nose down, but not as aggressively as MCAS)
• Elevator Feel Shift Mechanism - increases elevator control forces
• Autoslats - slats will go to full extend with any amount of flaps extended.

All of these functions are useful in a true stall, but can create unnecessary distractions when the stall condition is erroneous. There is no compare function between sides, so one faulty input can create this cascade of system effects in addition to the other challenges of an Airspeed Unreliable scenario.

Thx Tomaski.

I always thougt the airbus is a complex beast.....
Hopefully all 737-800 jocks will know about the stall warning design of their bird.

Days are obviouly gone when a stick shaker was a stall warning just by an AoA vane (B727, A300), no computers which may get it wrong.

[Tomaski]

Originally Posted by gearlever

Thx Tomaski.

I always thougt the airbus is a complex beast.....
Hopefully all 737-800 jocks will know about the stall warning design of their bird.

Days are obviouly gone when a stick shaker was a stall warning just by an AoA vane (B727, A300), no computers which may get it wrong.

On the 737NG and later, pitot-static and AOA inputs are processed through independent (left & right) Air Data Inertial Reference Units (ADIRU's) to generate a computed airspeed (CAS) which is displayed on the Primary Flight Display (PFD). The ADIRU's also sends data to the SMYD's (also left & right) which then manages the stall and yaw damper functions. Again, there is no comparator function between left and right sides, so one bad input on one side is all it takes to get the ball rolling. Even if the crew determines that the stall signal is erroneous, there is no published procedure to cancel the various system responses.

[galdian]

Tomaski
You reaised some interesting points in your original response, suppose I was thinking a departure in a "normal" environment where why change something that's still working OK at 15 degrees/MCT until MSA.
Obviously hot and high altitude/mountainous departures have their own considerations.

We very occasionally do flap 25 takeoffs and have 800' acceleration height - I wonder when I'd start doing the Airspeed Unreliable NN checklist, whether you'd retract flaps - and which is the more reliable airspeed indication(s) to do so? - or are the flap speed limitations factored in so you'd just leave the flaps at 25?

Points to consider, suppose that's why we're paid the (not so) big bucks!!

Cheers.

[Uplinker]

Originally Posted by galdian

..................If I have a suspected Airspeed Unreliable why would you do anything except maintain standard takeoff pitch - 15 degrees - and maybe tweak power back to MCT, sit fat dumb and happy until MSA THEN start screwing around with things?.............
To me achieving MSA then doing the Airspeed Unreliable checklist increases the safety of the aircraft...............

Just to be devil’s advocate:

Boeing designed their NNC to work at all weights, all engine variants and all altitudes. Does your procedure work for all such variations?

Will your procedure work at, for example, Las Vegas during the summer? (very high MSAs and very high ISA deviation). Or Geneva? (LSGG).

Best climb is with a clean aircraft, but your proposal is to climb dirty?

[galdian]

Uplinker

You little devil you!

OK - based on which Airspeed Unreliable airspeed indicator(s) are you going to retract the flaps?

And in fairness I DID previously acknowledge hot and high/mountainous etc does throw more considerations into the mix.

Cheers.

[gravityf1ghter]

If you were in need of retracting the flaps, you would make a note of your power and pitch setting for the desired stage of flight you wish to be at with lesser or no flaps from your ‘Performance Inflight’ source (company), and then, as per FCTM, set said power and pitch, and change configuration.

[galdian]

Yes, fair enough.

The question I am asking - at what height will you be taking the time to consult manuals etc to do the same.

800' AGL - my companies acceleration height?
1500'AGL - another standard used by many?
XXXX AGL?

I have no qualms doing the NNC, just a matter of when, in many cases suggest it can be delayed whislt everything's going OK with 15 degrees/MCT to MSA as an example.
Retracting flaps (if required/desired) - that's the next stage for consideration.

Cheers.

[Tomaski]

Originally Posted by galdian

Yes, fair enough.

The question I am asking - at what height will you be taking the time to consult manuals etc to do the same.

800' AGL - my companies acceleration height?
1500'AGL - another standard used by many?
XXXX AGL?

I have no qualms doing the NNC, just a matter of when, in many cases suggest it can be delayed whislt everything's going OK with 15 degrees/MCT to MSA as an example.
Retracting flaps (if required/desired) - that's the next stage for consideration.

Cheers.

IMHO, two of the primary reasons that 737 pilots have so many questions/concerns about the Airspeed Unreliable NNC in the takeoff phase is 1) lack of detailed guidance from Boeing and 2) lack of specific training in the sim. I believe that this malfunction has been seriously neglected because it doesn't happen that often and there is an underlying presumption that the "bad" airspeed will be quickly identified without going through the whole checklist. However, there is no guarantee that any of the airspeed indicators will be accurate. One of the possible scenarios is a bird strike that severely damages the radome (and possibly pitot/AOA probes) which then changes the airflow around the nose. In this case, every airspeed sensor is potentially effected.

I'll say again, this can be a very challenging procedure, and the crew does not have the luxury of figuring these things out on the fly. We really need to treat this as seriously as an takeoff engine failure and know (and have hopefully practiced) what we are going to do and when we are going to do it long before it happens.

Based on the scanty guidance, this is my recommendation on how to execute the existing Airspeed Unreliable NNC during takeoff.

• Identify and confirm. The "IAS Disagree" annunciation is the most responsive indication (difference greater than 5 knots for greater than 5 seconds). A single stick shaker is the most obvious (and most distracting). There is absolutely no need to do anything but establish and confirm a reasonable takeoff pitch attitude and power setting until 400'.
• At 400', if airspeed disagree is confirmed, announce "Airspeed Unreliable" and execute memory items. Declare an emergency. Prepare to execute engine out routing if applicable for the airfield. If you think about it, have PM check Probe Heat and grab QRH. Don't do anything else until your flap retraction altitude.
• Flap retraction altitude. Decide if you need to retract some flaps. If flaps are greater than 5 degrees, retract them to 5. If at 5 or 1 degree, leave them. Personally, I would only fully retract the flaps if I needed it for climb performance (some guidance from Boeing would be helpful here). Otherwise, I would leave them extended until at a safe altitude. If you need to retract the flaps, the PM looks up the appropriate climb pitch and power settings from the tables in the NNC. Set those parameters, retract the flaps, climb to a safe altitude.
• Once at a safe altitude, work through the rest of the checklist. If there are no reliable airspeed indicators by completion of the NNC, then plan to land at nearest suitable airfield with good VMC conditions with emphasis on trying to maintaining VMC enroute.

If the unreliable airspeed is generating a stall indication, be aware of the related system effects - Stick Shaker, Speed Trim Stall ID, Elevator Feel Shift, and Autoslats. Each of these will create an abnormal indication that can be mistaken for another malfunction. I would recommend knowing the location the stick shaker CB's (P18-2, P6-1) if the Captain wishes to exercise his emergency authority to silence this noise maker. As a side note, it is my current understanding that the 737NG simulator does not accurate replicate all of these ancillary system effects. I hope to confirm this my next training cycle.

If the unreliable airspeed is generating an overspeed, you will have an overspeed clacker. There is no way to silence this alert short of shutting off the bad side ADIRU, and you probably don't want to do that. If the left side is faulty, that may trigger the flap load relief function (flaps will not extend fully). This function varies by aircraft, so check your FCOM for detail. Load relief only retracts flaps to the next higher setting, so you can still get to Flaps 30 by selecting Flaps 40. Depending on how far off the airspeed is, you may still be able to find a usable landing flap setting (Flaps 15) that does not cause the overspeed clacker to come on.

As I said, a lot of moving parts to this one. It is unfortunate it has not been given the attention that it deserves.

[sekmeth]

Question regarding the procedure as I found myself in a discussion with our instructor the other day:
we experienced airspeed unreliable just short before minimums on a cat1 ILS, and had to make a go around.
after ap/at off and FDs off, and askint the PM to set the power to 80% Question regarding the procedure as I found myself in a discussion with our instructor the other day:
we experienced airspeed unreliable just short before minimums on a cat1 ILS, and had to make a go around.
after ap/at off and FDs off, and asking the PM to finetune the power to 80%, I asked for gear up.
the instructor asked me during the debrief, where I found the gear in the memory items. According to him it isn’t written anywhere. I tend to disagree, as the values given specifically say “gear up pitch attitude and thrust “.
The instructor said that if Boeing found the gear that important they would have made it a separate memory item, which is fair enough. But when I asked/said that the values given in the QRH are only valid for gear up, and without asking for gear up I was also not flying the procedure, his response was that I was being a smartass and I should listen to him.

I agree this being an academic discussion but I was wondering how you are looking at this. Would you raise the gear or not?
fyi: we were training in Innsbruck, so terrain is always an issue

[FlyingStone]

You won't get any decent rate of climb with F40, 80% and 10deg pitch and gear down. Terrain or no terrain, the gear has to come up for the memory item pitch/thrust values to be valid.

[Dark Stanley]

Quote

The instructor said that if Boeing found the gear that important they would have made it a separate memory item, which is fair enough. But when I asked/said that the values given in the QRH are only valid for gear up, and without asking for gear up I was also not flying the procedure, his response was that I was being a smartass and I should listen to him.

The memory items for a V1 cut with fire or severe damage don’t include an item for gear up either. But it’s covered in the emergency brief much the same as gear coming up is covered in the go around. Sounds like a very egotistical examiner, still around unfortunately and even more surprising they seem to be the younger ones.

Agree that it’s not a good idea to make up you’re own procedures, but it’s still a good idea to have some contingencies ready. The scenario of birds on approach taking out a radome and pitot could potentially also take out an engine. So you now have one engine and no reliable airspeed.Now Try going around with gear down and 10/80 set....

[FullWings]

...his response was that I was being a smartass and I should listen to him.

It’s a shame that there are so-called “instructors” around like this, even if they are a tiny minority. Being unable to explain *why* a particular course of action is desirable and then criticising the student for asking is not a good trait for a teacher.

As I suggested in another thread, the response to these types of scenarios that appears to work best is to nod, pretend to make a note of it somewhere, then move on. Afterwards you can ask someone else that you respect to give you an answer/explanation.

I think the logic used to raise the gear is difficult to disagree with in this situation. Why would you want to leave it down and invalidate the figures in the QRH by doing so? Especially when terrain constrained. It doesn’t say to put the speedbrakes away either, should you be using them, but it makes a lot of sense...

[john_tullamarine]

Sometimes I have to shake my head a little.

Back when, we had to commit to memory the data relevant to the current Type; ie pitch, thrust for altitude and configuration to achieve a speed which was associated with a climb or descent rate. This didn't require memorising books full of numbers but one did need to commit a few useful combinations to memory.

For newbies (eg cadet pilots at around the 200 hour mark converting onto the 737), and knowing that they probably would never have the exposure opportunity again, I used to get them to fly a single pilot circuit towards the end of the endorsement program - raw data, failed pitot statics, low cloud and vis for a recovery via an ILS. I can't recall that anyone ever had a major problem - OK many were not pretty to watch but they got the aircraft back onto the runway in a reasonably satisfactory manner - some did superbly proficient jobs of the task.

This was just a skills exercise which probably would never arise on the line. However, the student confidence building value and confirmation to me as the instructor that they had a reasonably sound grip of the I/F basics at that stage were sufficient justification to me for the time spent. I always thought it unfortunate that, other than for very low time pilots coming onto their first airline jet, the endorsement program hours/sessions were too tightly constrained to permit such sideline exercises.

SOP procedures aside, is it too difficult to commit a few number combinations to memory to save your tail in extremis ?

[Derfred]

Refer to the Flight Crew Training Manual for Airspeed Unreliable on approach.

Normal go-around using go-around thrust and 15 deg pitch until reaching a safe altitude, then do the Airspeed Unreliable checklist.

[turbidus]

great thread!

One has to remember that there were what, 7 revisions to the 737 NG manual, and in the 3rd revision to the 737 NG/MAX manual, and after the joint technical review board brought it up, they will finally correct the spelling and terminology issues!

[FullWings]

SOP procedures aside, is it too difficult to commit a few number combinations to memory to save your tail in extremis ?

In some ways I agree, but the procedures are written on the basis that you may have had an unreliable airspeed condition for some time before you diagnose it, so could be starting a long way away from your desired speed. Setting normal pitch/power for, say, F5/180kts when you are in reality 40kts+ slower could make things worse, not better.

In terms of crew coordination, it works better IMO. As soon as someone says “Airspeed Unreliable” you know what to do and also what should be done by the other pilot, rather than seeing someone else set a performance attitude that is not quite what you expected.