Cabin Altitude (once more)

[simonijs]

Hi Hardy,

We have discussed CAB ALT in other threads, in April 2017 & May 2017. Both resulted in updates (10.1.3 & 10.1.5). Cabin Altitude, however, still behaves strange at times and I couldn't exactly put my finger on when and where that would happen, with the sudden rush of air clearly audible and cabin altitude rates - sometimes - in excess of 8000 or 9000 ft/min.
To me, it now seems related to the altimeter setting (QNH) being (well) above Standard atmospheric pressure at low elevation airports.

For example: departure from SAEZ (to SBKP) and a QNH of 1046 hPa produced a sudden cabin altitude rate of a momentary 9500 ft/min, climbing through ± 17000 ft of flight altitude. Same situation but a QNH of 1033 hPa: cabin altitude rates in excess of +7000 ft/min at a flight altitude of ± 12000 ft. Therefore, the spread between a high QNH and STD seems to determine the flight altitude at which the effect occurs, and how extreme. Selection of either CLB, CLB 1 or CLB 2 influences both the effect and flight altitude. Similar effects are visible and audible during departures from EHAM.
Until the effect occurs, cabin altitude rates are very low (100-150 ft/min), then - with the effect occurring - the system seems to be catching up to a more normal cabin climb schedule by using momentary high rates. There is no indication of the outflow valves, changing position. When the QNH gets closer to - but still above - 1013 hPa, the effect becomes smaller occurring at lower altitudes, or is not noticeable anymore.

With QNH values below 1013, this effect never happened. And cabin altitude rates are more normal (± 350 ft/min early in the climb).

I tried similar scenarios from higher elevation airports (SBKP, SEQM) but here I could not reproduce this problem with similar high altimeter settings.

I also noticed, that the CAB ALT figure on primary EICAS changes with a change from QNH to STD. "Delta P", however, does not change accordingly when switching to STD nor does the cabin altitude rate. This seems strange if one changes from QNH1046 to STD, and the cabin altitude changes by ± 900 ft.
I don't thing that the cabin altitude should change into another value when selecting STD from QNH or vice versa. I believe it is a calculation performed by the FMC to arrive at a cabin altitude derived from a cabin pressure altitude.

I will (try to) include two situ.files: one for a light-weighted departure from SAEZ, a second one for a heavy departure from EHAM. Also...: one screenshot of excessive cabin altitude rates.
If you find some time, I would like to hear from you whether or not you can reproduce these effects as well.

Kind regards,
Simon

http://www.hoppie.nl/forum/var/01_Departure_012_-_SAEZ-SBKP.situ
http://www.hoppie.nl/forum/var/01_Departure_010_-_EHAM-VHHH.situ (both situ's with QNH1033)

http://www.hoppie.nl/forum/var/Primary_EICAS.jpg[/img]]

[Hardy Heinlin]

Hi Simon.

I don't thing that the cabin altitude should change into another value when selecting STD from QNH or vice versa.

The real 744 cabin altitude does change when the pilot changes the baro setting. This has been observed on the real deck on the ground, and it makes sense. How should the cabin altitude controller know what pressure is to be expected at the programmed landing altitude? You should only pressurize the cabin to 1013 hPa when the outside pressure is 1013 hPa. The controller won't get the data from ATIS ...


Regards,

|-|ardy

[John H Watson]

For example: departure from SAEZ (to SBKP) and a QNH of 1046 hPa produced a sudden cabin altitude rate of a momentary 9500 ft/min, climbing through ± 17000 ft of flight altitude.

Here are some typical expected cabin altitude change rates:

During the takeoff roll, the aircraft is pressurized at the nominal rate of 500fpm to 50 feet below takeoff altitude.
In flight, the cabin rate is limited to 1000fpm unless the differential pressure limit is exceeded. In which case, the positive pressure relief valves will open (in the side of the fuselage) or the negative pressure relief valves in the cargo doors will open. When landing, the outflow valves open to allow a 500fpm cabin altitude rate, but will switch to 2000fpm after a certain time period if the cabin pressure hasn't equalised. To get such rapid pressure changes in your described scenario, it seems like you are exceeding the limits and that the pressure relief valves are being activated. I'll have to do more reading on the flight deck effects as a result of this activation.

I don't think that the cabin altitude should change into another value when selecting STD from QNH or vice versa. I believe it is a calculation performed by the FMC to arrive at a cabin altitude derived from a cabin pressure altitude.

Makes sense in some respects. With FMC control, a cabin altitude climb schedule is generated based on field elevation and cruise altitude. Similarly during descent. However, I'm having trouble understanding a few things.

1) Prior to takeoff, how does the system determine the planned rate of cabin climb? When the pilot dials in the baro setting, he's not changing the atmospheric pressure. He's just determining his altitude above sea level. You would think the system would need to know the expected pressure change between the ground and cruise altitude and the expected time to reach cruise altitude (based on the LNAV/VNAV profile).
There is most certainly a corrected baro altitude sent to the CPCs (cabin pressurisation controllers). However, there is also cabin pressure and air/ground inputs. Cabin pressure and outside air pressure should be the same on the ground.

2) I don't know how the system responds to sudden change in airplane altitude (e.g. during the switch from STD to QNH). Could this be treated as a hold or climb?

Hold During Descent
During descent, if the airplane altitude varies by less than +/- 250
feet for two minutes (altitude hold or slow descent), the Cabin
Pressure continues to increase until it reaches either the preselected
landing field altitude or the differential pressure at the start of the
descent. Following this condition, an airplane altitude change of
more than +/- 500 feet the controller reverts to ascent or descent
mode of operation.

The real 744 cabin altitude does change when the pilot changes the baro setting. This has been observed on the real deck on the ground, and it makes sense.

I'm not sure of the full details, but until the aircraft enters takeoff mode (>65kts), the outflow valves should be fully open, so no pressure changes. Who would be playing with baro settings during the takeoff roll?

[simonijs]

Hi,

Several documents, that I read before posting, define the Cabin Altitude as a Pressure Altitude *. A schematic drawing in a 747 AOM mentions an "FMS Barometrically corrected altitude" but that may be read also as inputs are from: FMS; Barometrically corrected altitude; Cabin Pressure; Air/Ground signals. See schematics below. As usual, information in the AOM/FCOM is not very detailed. And none of the documents that I read discuss the switching from QNH to STD. Finally, some aircraft pressurization systems do not use LDG ALT but LDG ELEV (which I thought would come from the FMS). My original posting was about a departure, with the LDG ALT set to the departure airport.

"Delta P", however, does not change accordingly when switching to STD nor does the cabin altitude rate.

Now here, I have been "tunneling" a bit: focussing on only one aspect. The other aspect is, that the flight altitude changes as well when selecting STD from QNH. As a result, "delta P" will remain the same (which I just verified using Excel). So, if on the real aircraft CAB ALT also drops when selecting STD from a high QNH, then PSX is modelled correctly. But I still find it hard to believe that this happens.

When the doors are being closed before push back, the pressure inside the cabin is essentially the same as the outside pressure. This pressure inside the cabin will not change with the altimeter setting nor will the pressure outside the cabin. Once established in the climb, the cabin pressure will slowly become lower and the cabin (pressure) altitude will slowly begin to increase. If the indicated cabin altitude indeed is "the altitude on a standard day, that has the same pressure as that in the cabin" (see the second link below) then only a sudden, momentary increase in cabin pressure would cause a sudden decrease in cabin altitude when selecting STD from a high QNH. That would be rather uncomfortable for the ones on board.

With a high QNH - say 1050 hPa - and a climb schedule based on 1013 hPa, the cabin would have to climb at slightly higher rates than on a standard day (or on a day with a low QNH). In PSX, climb rates are really low early in the climb from an airport with a high QNH. This seems to trigger these very high rates from my first post. And that was my main concern  .

Regards, Simon

PS - Just noticed, that the standby altimeter is indicating NEG 9000 when the altimeter is set to STD on a sea level airport where the QNH is 1050 hPa. The ISFD indicates -1000 ft, as it should...


* https://www.nap.edu/read/10238/chapter/4#34
   https://www.sweethaven02.com/Aviation/MaintHandbook/ama_Ch16.pdf (see page 23 of 62)

   http://www.hoppie.nl/forum/var/Schermafdruk_2019-09-18_10.47.33.png[/img]]

[Hardy Heinlin]

PS - Just noticed, that the standby altimeter is indicating NEG 9000 when the altimeter is set to STD on a sea level airport where the QNH is 1050 hPa. The ISFD indicates -1000 ft, as it should ...

"NEG 9000" also appears on the real standby altimeter. PSX models this correctly. Unlike the LCD instruments, the mechanical "steam gauge" is no endless tape but a finite roll: When you wind it above any full thousand position, you'll get the next "1". And vice versa you'll get the next "9". The "NEG" flag is a separate mechancial feature that indicates that the roll is now below sea level. But the sequence of digits remains decreasing with rising pressure: The roll won't turn backwards nor does it contain negative numbers.


Regards,

|-|ardy

[John H Watson]

When the doors are being closed before push back, the pressure inside the cabin is essentially the same as the outside pressure. This pressure inside the cabin will not change with the altimeter setting nor will the pressure outside the cabin.

I think there is a slight positive pressure. Opening the MEC hatch (pushes inwards) with the packs runnning is a little harder with all the doors closed than it is when the doors are open, even though the outflow valves are fully open. Whether this slight positive pressure is shown on the instruments, I don't know.

The other aspect is, that the flight altitude changes as well when selecting STD from QNH. As a result, "delta P" will remain the same (which I just verified using Excel). So, if on the real aircraft CAB ALT also drops when selecting STD from a high QNH, then PSX is modelled correctly.

I'm not sure I follow your logic. After takeoff, isn't the cabin altitude being ramped up according to an internal schedule using values determined before flight, ignoring outside influences unless some limit is exceeded or the aircraft stops climbing or starts descending? Is your Excel table based on cruise?

[simonijs]

Hi John,

I'm not sure I follow your logic.

I was referring to what I wrote in my first post: "I also noticed, that the CAB ALT figure on primary EICAS changes with a change from QNH to STD. "Delta P", however, does not change accordingly when switching to STD nor does the cabin altitude rate."

Selecting STD from QNH will change the indicated altitude on the altimeter, in PSX it also changes the CAB ALT. So...: after writing my first post, I then realized that if there is a relation between the indication on the altimeter and (on EICAS of) cabin altitude, then Delta P will not change and no Cabin Rate will be shown.


Hi Hardy,

"NEG 9000" also appears on the real standby altimeter. PSX models this correctly.

I didn't know that, no mention of it in the AOM/FCOM. Your explanation - of course - sounds very logical. I wonder how many pilots are aware of this.

Regards,
Simon

[Hardy Heinlin]

Hi Simon.

I was referring to what I wrote in my first post: "I also noticed, that the CAB ALT figure on primary EICAS changes with a change from QNH to STD. "Delta P", however, does not change accordingly ...

This is because the LANDING ALT must agree with the barometric setting. When the QNH at sea level is 999 hPa, and the aircraft has landed at sea level, the instruments should indicate "0" feet. You can achieve this by setting 999 on the panel. If you set 1013, the instruments won't indicate "0", and that would be an error. Shortly before touchdown the LANDING ALT setting and the sensed CABIN ALT should agree. Your baro setting influences the system. If you set 1050 instead of 999, the QNH for the LANDING ALT setting will be wrong, and the CAB ALT will drift to the wrong target too.

The occassional momentary pressurization changes may come from momentary outflow valve movements. When you hear a wind noise, that momentary valve movement is already over.


Regards,

|-|ardy


The cabin pressure sensors on the real 744 are very sensitive; you can see significant momentary altitude rate changes on the ground when someone just closed or opened a door somewhere.

[Britjet]

I have to say that personally I find it difficult to believe that the cabin altitude is affected by a subscale change.
There are many instance in the cruise where one can touch the Altimeter Setting pushbutton by mistake (finger trouble!) and the altimeter zooms off to the (perhaps very) different preset QNH - in such instances I have never been aware of a change to cabin altitude - but I admit I know nothing of this technically..
Peter.

[Hardy Heinlin]

Fact is, in descent the cabin altitude controller cannot know the current QNH at the landing elevation if it doesn't get any data inputs. It must know the data so that it can calculate the difference between cruise pressure and landing pressure. If it were to use ISA pressure all the time, the pressure before landing would be wrong on most flights. It wouldn't know where the flight ends, and it can't simply fully open the outflow valves shortly before landing, as it doesn't know the barometric distance to the landing field if the QNH is wrong. When you land at 9000 ft, the valves won't fully open on short final; they will keep the pressure pax friendly.


|-|ardy

[simonijs]

Hi,

The scenario's or situation files - that I included and hopefully are readable - are all about departures and not (yet) about landing somewhere. When I discovered that the high rates only occur at a high QNH, I ran the same situations over and over again with different settings for QNH, closely watching the ECS page. No movement of the valves whatsoever...

For approaches...: the pre-calculated vertical path does not know about QNH either. Once a high QNH is selected, the aircraft all of a sudden is (well) above this path, which also has to be corrected in some way.
For landings at high elevation airports, the FCOM contains supplemental procedures to set the Landing Altitude Selector. Landing at SKBO (Bogotá at ± 8350 ft elevation, close enough to 9000 ft), one would have to set the landing Altitude selector to 8000 ft at Top of Climb, then to Automatic at Top of Descent (source: KLM FCOM).

For me, aircraft pressurization has always been one big mystery, especially since I am one B I G technical nitwit... Because of these tiny little things.

Regards,
Simon

[Hardy Heinlin]

Hi,

I think we're having a little misunderstanding:

Yes, when you turn the captain's EFIS baro knob in PSX, the CAB ALT indication changes accordingly. But this is just the indication; it's not a physical cabin pressure change. It's just for the reference of the EICAS indication, like on the flight instruments.

In fact, the EICAS cabin altitude can be used as a flight instrument: As an altimeter in the ultra-rare case of a total static port failure. We tried that in the sims at Lufthansa. It works on any aircraft type that uses an outflow valve system. When you're below 10000 ft, fully open the outflow valves. The cabin altitude is now equal to the aircraft altitude. And therefore the cabin altitude indication can now be used to determine the aircraft altitude. The QNH setting must be correct, of course. The indication has a great lag, though, and a resolution of just 100 feet.

There's a sample situation file in PSX, called ...

03 Approach 010 - Auckland - All altimeters inoperative.situ


Regards,

|-|ardy

[simonijs]

Moin,

Just checked the QRH (FCOM) and the Reproduction Emergency Checklist (AOM) for procedures, that deal with situation 03 Approach 10. They are not given. With a total static port failure and flying in IMC, I would not be able to tell when I am below 10.000 ft. V/S will also be unavailable, so one would have to make estimates of vertical speed based on attitude and power. IAS is unreliable as well, making things even more complicated.

Actually, my remark on the change in CAB ALT when selecting STD from QNH was just a side comment to a problem that I found far more serious: cabin altitude rates in excess of several thousand feet per minute (see the screen shot in my first post). We haven't been discussing this problem so far, nor if it can be reproduced and/or traced back to the difference between a high QNH and STD.

Being well aware of your "to-do-list", let's give this a very low priority and perhaps discuss these high rates in due time.

Regards,
Simon

[Hardy Heinlin]

Just checked the QRH (FCOM) and the Reproduction Emergency Checklist (AOM) for procedures, that deal with situation 03 Approach 10. They are not given.

Of course, they are not included :-) It's rather a Kirk-Scotty-Beam-me-up scenario. Chuck Norris might have played a role as well.

But it works! It's an example of how creativity can save the day in a desparate, "hopeless" situation, -- when you can't find a related list in the QRH. Remember all the "absurde" technical improvisations in Apollo 13 :-)

If you can't safely determine whether you are approximately below 10000 ft, you may open your outflow valves slowly. When the CAB ALT rises above 9000, move the valves slowly back towards "closed".

V/S is available by the IRS. The barometric component in the V/S indication is just for long term stability.


Regards,

|-|ardy

[Will]

Any comments about the original issue in this post? I'm seeing this behavior too, with the cabin climbing after takeoff at ~150 fpm until ~10,000 feet and then a momentary WHOOSHHH of air and cabin climb rate of ~9000 fpm lasting for a few seconds, followed by a return to a slower cabin climb.

[Hardy Heinlin]

Guess somebody was on the toilet ...

[G-CIVA]

Any comments about the original issue in this post? I'm seeing this behavior too, with the cabin climbing after takeoff at ~150 fpm until ~10,000 feet and then a momentary WHOOSHHH of air and cabin climb rate of ~9000 fpm lasting for a few seconds, followed by a return to a slower cabin climb.

Yes, I see it often too ... randomly ... most often if I fly several continuous sectors without closing PSX down completely.

[Hardy Heinlin]

Maybe all vacuum toilets were flushed at the same time?

[Jeroen Hoppenbrouwers]

See what happens if you don't release the pax at 10k.

[Will]

Need to pull the lav circuit breakers.