Altimeter needs calibration?

[ScAvenger001]

The barometric altimeter has some drift in non-standard weather.

 

I took off from Batumi (field elev 33ft) in the conditions of the Summer Clouds & Wind preset. By the time my altimeter read 20,000 feet (according to the HUD, didn't check the gauge), the F2 view showed me at 20,920 feet. Do real altimeters behave this way? I don't think so....

 

Conditions were:

temp: 26*c

Clouds: base 1969ft; thickness 2526ft; density 5; no precip

QNH: 29.72

Wind:

26k ft: 14kts @ 071

6600ft: 16kts @ 243

1600ft: 21kts @ 160

33ft: 10kts @ 160

[CrashO]

Did you set the correct barometric pressure (in mbar) before takeoff?

[TheSnark]

Usual procedure IRL is to set the QFE for where you takeoff and land, and once up high enough go back to 1013 so that all the aircraft are on the same altimeter settings. What did you set and when?

[ScAvenger001]

I set it correctly on takeoff. Batumi is 33ft, the hud read 40 on takeoff. The drift crept in gradually as I climbed.

[ScAvenger001]

I tried it some more in various weather conditions.

 

First I did another try in 29.72 pressure and got the same result, so it is reproducible.

 

Default weather (29.92), took off with the baro altimeter reading 40 in the hud, when it showed 20050 the f2 showed 20526.

 

Winter, clear sky (31.10), took off with the altimeter reading 40, when it showed 19990 the f2 showed 18848. Note that this time the instrument is reading too high at altitude while the rest of the time it was too low.

 

Also, it's worth noting that gradually increasing drift was also reflected in the radar altimeter as I flew over the sea. The radar altimeter showed the correct altitude over sea level as I flew over the sea while the baro altimeter lagged behind.

[Frederf]

Real barometric altimeters are simple devices which have a certain needle position for every pressure. If an altimeter shows 1,000' this is not an altitude but instead is a pressure. How this should be read is that "the pressure outside is the same as it would be at 1,000' in a standard model of the atmosphere." How well that pressure indication (in feet) happens to match your actual position entirely depends on how well the weather matches the design of the device.

 

The upper part of this graph can give you a picture that there is an assumed relationship between pressure and altitude built into the design of the altimeter. This is a curve: http://patentimages.storage.googleapis.com/pages/US4106343-2.png

 

We can also imagine that there is a different graph curve for the real world. So there are two curves.

 

There exists the altimeter setting knob which allows us to change the curve for the instrument to better match the real life curve. When we set QNH we are making sure that the curves cross at the altitude of the airport elevation. When we set QFE we are making sure that the instrument 0' point matches the airport elevation point. The knob shifts the instrument curve along a linear transformation until they intersect at a point of our choosing.

 

Unfortunately the curves may be different shapes and it may be (in practice always) impossible to get the curves to completely match at all points. We can force the matching at one point but they will not match at all points.

 

This is the reality of barometric altimeters with only one knob for calibration. It reads pressure altitude shifted by a calibration amount to ensure that the reading matches a desired value at one particular pressure.

 

Normally in real life the weather is well behaved so that the error between the design of the device and the environment is minimal. The curves are close to the same shape on most days. But in a simulation it is not guaranteed that the environment is reasonable or that the curve built into the design is correct.

 

Usually temperature is the cause of the disagreement. In an extreme case at -50C and at 5000' indicated you are at 3500' even though it has the correct altimeter setting! That is an error of 30%!

[Robin_Hood]

Yes, deviation from standard temperature accounts for most of the observed deviation. It has been that way for some time (about when the Su-27 PFM was introduced I think), although I don't know how realistic it is regarding actual values.

 

You can test it with any aircraft, A-10C, F-15, Su-27... Here is an example with an A-10C, with 26°C on the ground (vs 15°C standard), when the altimeter reads 20 000, the aicraft is actually at 20 775 ft. This is also why contrails do not always appear at the same 8 000 m anymore (actually they do, but most of the time your altimeter reads something else) (EDIT: actually, now the contrail altitude is really dépendent on temperature).

 

Now there is still a small discrepancy with the Mirage 2000C altimeter specifically, in which the error is a bit bigger than other aircraft (in the standard atmosphere, it reads 20 000 ft at 20 125 or something).

 

Also attached is a chart that I did around the time I noticed that behaviour, where I checked the readings at different températures (note: altitudes are in meters).

Edited November 28, 2016 by Robin_Hood