Wednesday, March 04, 2009

Q.E.D.

A commenter recently asked about the altimeter setting and Q-codes. I have probably blogged on this before, but Blogspot doesn't show it in my archives so it's just as easy to explain again. It is to do with the way the altimeter works.

The instrument in the panel that indicates my altitude is a barometric altimeter. It operates purely by air pressure, mechanically comparing the pressure in its sealed reference capsules to the air pressure sensed immediately outside the airplane. For this purpose there are little holes on the outside of my fuselage called static ports, and tubes conduct the pressure from the static ports up to the altimeters.

As the airplane climbs, the outside air pressure drops, but the pressure inside the reference capsules stays the same and this difference moves the hands around the face of the clocklike altimeter to indicate my altitude. Electronic altimeters work the same way, except that the pressure difference is converted to a digital signal. All the altimeter knows is the difference in pressure. If the pressure goes up, the needles show a lower altitude. If the pressure goes down, the needles show a higher altitude. This also happens without the airplane ever leaving the ground, just because the air pressure changes as weather systems move through.

In order to always start at a known place, there's a knob on the altimeter that moves what the altimeter considers to be the zero altitude up or down. So you could be on the ground with the altimeter reading zero, and turn the knob so that the altimeter read a thousand feet, and then crank it back the other way so the altimeter read two hundred feet below sea level. While you turn the knob, and the altimeter hands are moving, a small dial with numbers is also moving, visible through a cutout in the face of the altimeter. On an North American altimeter the numbers usually range from 28.10 to 31.00. On a European altimeter they range from about 950 to 1050. (International altimeters have two windows and two dials driven by the same knob). Turn the knob so a bigger number appears in the window and the hands go up to indicate a higher altitude. Select a lower number in the Kollsman window and the hands point to a lower altitude. Whee! The pilot will have to set it correctly before takeoff.

But what is correctly? There are actually two systems as to what the altimeter should read when you're parked. One is that it should read zero when you're on the ground, no matter where that ground is. Under that system, a circuit pattern flown at a thousand feet above ground level would always read 1000' on the altimeter, regardless of the elevation of the airport. If the altimeter reads two hundred feet, you're two hundred feet above ground. The number that you would set in the Kollsman window to make the altimeter read zero on the ground at a particular airport is called the QFE for that airport. The QFE system is not very popular. I believe the British military uses it, as does Russia, so probably much of the former Eastern bloc. I highly doubt that a Canadian or American controller would be able to give you a QFE if you asked for one, and many probably would not understand the question. Even if you could get the QFE, there are many airports in the US where you couldn't set it on an ordinary altimeter, because the numbers in the Kollsman window don't go that high.

The far more popular way to set an altimeter is using the QNH, but North American pilots don't learn that term, either. We just call it "the altimeter setting." The altimeter setting is the number you set in the Kollsman window so that the altimeter shows airport elevation while you are on the ground. If the airport is 1200' above sea level, then the altimeter shows 1200' when you're parked, and 2200' when you're in a thousand foot circuit pattern.

Neither the QFE nor the QNH for a particular airport is fixed. It varies from hour to hour with the weather patterns. That's why the controllers and the ATIS broadcast the altimeter setting.

There's a third way to set your altimeter, which everyone does one they climb into the flight levels. That is to set 29.92 and leave it, so that you don't have to keep changing it as you go through different pressure systems en route. If there's a Q-code describing this setting I don't know it. We call it "twenty-nine nine two."

Update: The terse and anonymous first commenter below has indicated that twenty-nine nine two is called QNE. And in verifying that, I'm reminded that with 29.92 set on my altimeter, it is showing the pressure altitude but most people don't call it "setting the pressure altitude" unless they are doing it momentarily for the purpose of determining the pressure altitude for a calculation.

Anonymous said...

QNE

Anonymous said...

Your 2nd paragraph reminds me about what happens when static ports are plugged. These ports are important for the altimeter, airspeed indicators, and vertical speed indicators. See aluwing's pt1 movie link from 2:18 onward for the consequences of no such basic instrumentation at night ... in IMC.
The clip begins with calling a Duchess an Cirrus SR20, but it gets better @ 2:18.

Aluwings said...

also:
http://en.wikipedia.org/wiki/Birgenair_Flight_301

http://en.wikipedia.org/wiki/AeroPeru_Flight_603

http://aviation-safety.net/database/record.php?id=19741201-1&lang=en

So many losses from such a "simple" failure...

Julien said...

Hi Aviatrix,

When you say

Even if you could get the QFE, there are many airports in the US where you couldn't set it on an ordinary altimeter, because the numbers in the Kollsman window don't go that high.

did you mean to say "don't go that low"?

Apart from weird locations below sea level such as Furnace Creek airport in the US (-210 ft) or Schiphol airport in the Netherlands (-11 ft), the QFE at a given place is always lower than the QNH. As far as I remember that's actually the only law in meteorology that does not have an exception: pressure goes down as altitude goes up.

One important point about flying "in the flight levels", i.e. with the altimeter set to 29.92, is that the purpose of maintaining a given altitude is primarily to ensure separation between airplanes, and not so much separation between airplanes and the ground.

Julien.

Anonymous said...

So many losses from such a "simple" failure...

Which begs the question: why hasn't more effort gone into designing avionics systems to test the reliability of air data sensors at the start of a flight?

I mean, it's not hugely difficult to implement an altitude crosscheck to take data from IRS, GPS, and radalt sources and throw up a warning flag if the plane has climbed 2000' from the runway but the measured static pressure hasn't changed.

Ditto testing pitots by checking that measured airspeed increases in proportion to measured wheel speed during the takeoff roll.

Johnson said...

Howdy!

I just recently started following your writings, and I have to say I really appreciate your thoughts! I'm way further back just working on my instrument so its really amazing to read about the adventures of a pilot who is 'out there', so to speak.

Well, happy flying and thanks again for sharing.

Unknown said...

The way to keep all this straight is to think (capitalization for emphasis):

qFE: Field Elevation
qnH: Home
qnE: Enroute

dpierce said...

I was taught to remember ...

qNE = Neighbors (are) Equivalent
qNH = Nautical Height
qFE = Freaking Exotic :)

And in the QNH world you have altitudes, QFE you have heights, and QNE you have, of course, flight levels.

Anonymous said...

@anonymous:

If you're climbning into cloud on a dark night, discovering the problem at 2,000 feet is probably already too late.

Pitot measured speeds don't read accurately below a certain threshold. But it is one reason why the pilots do a verbal crosscheck at a defined speed during takeoff - typically at 80 knots.

That being said, there is a monitoring system on modern aircraft to detect discrepancies between the three sets of indicators. But figuring out which one to believe is often "the issue."

In the Aero Peru accident I was wondering why no one was looking at the radar altimeter as it should be completely independent of the air-sensing system. But it's always easier to figure this out from the comfort of an office chair.

And the only way to confirm that the pitot-static system is operating before flight would require moving air through it, which might be feasible? But if the system isn't taped over, or iced over or blocked by mud-wasp nests, it's incredibly reliable.

AndyC said...

And a bit if history is that the Q-codes were invented as a shorthand for use in Morse Code communications, originally in telegraphy and later in radio.

About the only surviving use today outside of limited use in aviation is in amateur radio, where QSL, QSB, QRN and QRM are all commonly used along with many others.

Louise said...

We used:
Quantity From Earth
Quantity Near H2O

We never needed QNE.

Anonymous said...

The QFE system is not very popular. I believe the British military uses it,...

And British civilians, at least in general aviation. QFE in the circuit (pattern), QNH when cross country; the tower gives you the pressures when you leave and rejoin. It comes of being a low level country with few, if any, airfields where you can't set QFE.

Anonymous said...

A very timely subject, as the news today is reporting that the Turkish B737 crash in Amsterdam was most likely the result of a stall when the a/p responded to a faulty radar altimeter input. While that may have been the root cause of the crash (along with inattention to airspeed), it probably also made it survivable because the a/c hit nose high at very slow speed.

Anonymous said...

As an ex-RAF pilot it always puzzled me why anyone would prefer to fly a circuit, or an approach to a runway, on anything but QFE. Knowing the real height above the airfield always struck me as being pretty important, particularly in bad weather. And Julien is right of course, the altimeter can always be set to the local QFE, no matter how high the altitude.

Given that a lot of places don't use QFE, there must be a good reason for it - ATC considerations at busy airports, I imagine. Even so, having the altimeter indicate height above sea level rather than the hard stuff you're about to land on is definitely an affront to logic.

Anonymous said...

About Schipol - I know nothing about 737 automation or CAT-III approaches - but it does sound like the crew let automation and faulty radar altimeter pull the throttles to flight idle a bit early.

The investigation will continue, but early reports from Dutch authorities and the Boeing notice is here

Aviatrix said...

Julien is right of course, the altimeter can always be set to the local QFE, no matter how high the altitude.

The part that Julien is right about is that it would have been clearer if I'd said "the altimeter can't be adjusted for fields that high" or "the Kollsman window numbers don't go that low." As Ed-Davies says, in the UK all the airfields are close enough to sea level that QFEs (I'll be able to remember "freaking exotic") are settable. But if I want to land at Leadville, in Colorado, I'd have to wind TEN INCHES off my altimeter to have it read zero on the ground. And that's not happening with the altimeter in any airplane I fly. I couldn't even set the QFE for Salt Lake City. QFE also assumes that there's someone on the ground to give you an altimeter setting. The QNH (hey, "home" it works) for a given airport is going to be fairly similar to that of an airfield ten or twenty miles away, and the adjustment when using a remote altimeter setting is printed right on the approach plate. But while the air pressure won't change too much over twenty miles, the terrain can. Does the pilot manually calculate a QFE at an uncontrolled airfield? If so, that calculation can't be any more difficult than the "airfield is at 2400' so circuit at 3400', turn final at 2900'" that North American pilots do.

Splendor said...

I'm sorry to break the consensus but QNE isn't actually the same as 1013/29.92 or standard pressure setting. It is the reading on your altimeter when you touch down with 1013 set, if that makes any sense. It is used when the QNH is outside an aircraft's altimeter sub-scale range.

Extreme example: QNH 1053, elevation 100ft. QNE would be -1020ft. (-40 x 28 = -1120 + 100)

A previous comment asks why QNH is used rather than QFE. I would venture the guess that terrain clearance is the answer - if QFE was used all hills/obstacle height figures would have to be in reference to the airfield in question - which would be rather unpractical and in my opinion confusing. QNH allows a single constant datum, and I never found it hard to remember field elevation.

S.

Splendor said...

I should have added that QNE is commonly referred to as the Q code for standard pressure, but that's not it's accurate definition.

S.

Anonymous said...

As others have said, QFE is used in the UK quite a lot for military and GA traffic. However, I believe QNH is used more in busy transport airspace because you can have a common intecept altitude, a common SID climb altitude, etc. for separation between aircraft of different airports. Also, imagine the situational awareness of a pilot on a frequency with aircraft going to Luton, Heathrow, City each referring to different datums.

As far as Q codes go (in the UK):
-QDM is used quite a lot, and an aircraft may call up a tower with a VDF and request QDM, and he will get the bearing magnetic to the tower called!

-QSY is unofficial, but is often used as it is can be a 'neater' transmission.

http://en.wikipedia.org/wiki/Q_code#Aviation