Tuesday, September 29, 2009

Everything at Once

Not only are we out of fuel at this base, but as we fly out the last hours before scheduled maintenance, the airplane throws a fit and we have a burst of minor defects we have to defer in consultation with maintenance: the aforementioned CHT gauge, a couple of instrument lights, and a bit of an oil leak that maintenance was right about when they said it would be messy but would not worsen or cause a mechanical problem. (That wasn't done lightly: two pilots, two engineers and the PRM all had to agree on it). It`s pretty embarrassing how much can go wrong at once. You probably don't want to know how many things are wrong with any airplane you ride on as a passenger.

During my last revenue flight before scheduled maintenance, during flight the compass started sitting at an odd angle in the housing. I rarely consult the compass after start up, because I have other instruments that use a different compass to do the same job more effectively, but it's still a required instrument. I'm used to the compass bowl being tilted as an effect of northern latitudes -- I'll do a post on that sometime, as it's rather interesting -- but as I changed heading I saw that the behaviour of the compass wasn't matching the angle of the magnetic field lines at our latitude. The compass read accurately on all headings and led and lagged as expected on turns to north and south, but it just wasn't sitting the way it usually did inside the bowl.

There is much in the way of preventative maintenance that is done on a compass. It's just there. It isn't connected to anything, except it has wires running to its light. It's calibrated once a year, and the couple of degree differences between its indicated and bearings and what it should read are recorded on a card for my use, but it doesn't get oiled or tuned or remagnetized. It's just one more thing for this stupid weekend. I imagine the rectification for this snag will be "Compass told to smarten up and fly straight," or possibly "Pilot headrest adjusted so she can see straight."

My next flight will be a ferry, to take this bag of aluminum bones to a shop where it will have its complaints attended to. They are also going to change out the tachometers for electronic ones, as we've been changing too many tach cables lately. The right tach needle is already oscillating slightly, in the way that says "order a new tach cable now before I break."


Cirrocumulus said...

"You probably don't want to know how many things are wrong with any airplane you ride on as a passenger" Or how many things can be wrong and they still fly it.

Yours is bored.

Anonymous said...

Whiskey compass you don't use it!? The one istrument hat rarely lies.

maintenance is usually the replacing the bearing and adding/changing the whisky. Were there bubbles visible?

Aviatrix said...

There weren't any bubbles visible, but yes, when maintenance examed it, it turned out to have leaked.

I beg to differ regarding your "never lies" description. The magnetic compass lies during acceleration, deceleration, turns, changes in pitch, turbulence, electrical system configurations different from the one used for the calibration card, when the front seat passenger is playing with an electronic appliance, in regions of the country (not all of which are marked on the chart) with magnetic ore deposits, and pretty much everywhere in the far north. Even under ideal conditions of straight and level it lies on almost all bearings. It's the only instrument that is required to come equipped with a chart to tell you how MUCH it lies. It's also inconveniently located for my scan.

I grant you the fact that it is not affected by an electrical or vacuum failure (except that it goes dark in the former), is not dependent on probes that can ice over, and the facility it uses for directional information is never out of service due to maintenance or hostile action, but the only thing I use it for is to check that the more precise instruments are giving reasonable values.

If you can read a compass to two or three degrees accuracy, you're doing fine. But if I fly 0.2 degrees off the required heading for ten seconds, I have made an error.

I do have to look into this boredom thing. What should I give it to play with to keep it more alert?

jk said...

I've got to echo Aviatrix's comments about the issues with a magnetic compass. Its saving grace is that it can give you an idea of where you're going without any external power source. "An idea" is the operative word... reading one in turbulence is an exercise in interpolation at best.

In my airplane, and I suspect the one Aviatrix flies from her post, I have a remote 'flux gate' compass that my gyro heading indicator is 'slaved' to. This requires electrical power to operate... but its amazing. Its spot on (well, as best as I can tell) through any heading and with any of the airplane's equiptment on/off. This is because it's mounted in a location somewhere off in the wing away from electronics and ferrous metals.

By contrast, my magnetic compass will swing 10 degrees or so when I do something like turn on the landing light. Its deviation card has to be read with some knowledge of what switches should be on/off for the deviation card to be accurate.

Aviatrix said...

Likewise, jk. I like to think of mine as a flux capacitor, although it doesn't draw 1.21 gigawatts from the avionics bus.

Plus I have inertial nav and three GNSS units: two GPS and one GLONASS. When I'm lost, I'm lost really precisely.

nec Timide said...

That is quite a line up. Every time I learn a new piece of equipment at your disposal or a new operational constraint (0.2 degrees over 10 seconds, and you eat while doing that, very humbling) I am intrigued anew.

Unknown said...

It's been a few years since I've regularly used the magnetic compass. I'm re-acquainting myself with the mysteries of swinging it to produce the "how much it lies" card, and trying to remember if I refer to that correction card while setting my gyro to the compass, or afterwards for determining my target heading...

Oh well. If my gps goes out I'm lost anyway.

Aviatrix said...

Grant, set the gyro to the compass exactly, and then use the correction card when setting your heading with respect to the gyro. If you're going in a straight line for the whole trip, or resetting your gyro every time you change heading (which is a very good idea), it doesn't really make any difference, but that's the way they teach it in flying school these days.

In practice I have to admit to looking at compass correction cards only when bored, or when picking on students who were good at flying in a straight line.

chephy said...

"There is much in the way of preventative maintenance that is done on a compass. It's just there."

Perhaps you meant there isn't?

A Squared said...

Grant, set the gyro to the compass exactly, and then use the correction card when setting your heading with respect to the gyro.

Maybe I'm misunderstanding you, but it seems that would be a suboptimal procedure. Let's say that you have a compass correction card that says "for 045, steer 050, and for 315 steer 310" (+ 5 correction on NE headings and -5 correction on NW headings)

Say you're maintaining a magnetic heading of 045. (compass reading of 050) You set your gyro to your compass exactly so now your gyro reads 050 also, but your actual magnetic heading is 045. No big deal as long as you're on generally NE headings, and you remember to steer desired heading + 5 degrees.

Now, for whatever reason, (ATC vector, approach segment, etc)you need to turn to a 315 heading. So using the card correction for NW headings, you turn from gyro heading 050 to gyro heading 310,(card says steer 310 for heading 315) which is a 100 degree turn, so you've just turned to a actual heading of 305, not the desired 315.

Seems to me it makes more sense to apply the correction between the compass and the gyro, so that the gyro will always read *actual* magnetic heading (plus or minus precession or other DG errors) rather rather than compass reading (until you turn off that heading, then it reads something that is neither of the two) That way you only have to apply the correction when you are setting your gyro, every time you look at the gyro.

Running through the same scenario, but applying the correction to set the gyro would go like this: Compass reads 050, so set your gyro to 045. Actual mag heading and gyro agree. ATC requests turn to 315, you turn to heading 315 on your gyro. no mental math needed and it doesn't introduce 10 degrees of error. After you've been wings level long enough for your compass to settle down, you check that the compass reads 310, if not adjust gyro to compass heading. Depending one how much your gyro drifts, it should be close.

Aviatrix said...

The correction is rarely as much as five degrees in any direction, and really only makes a difference for a long flight on a constant heading, so in the scenario described, I'd just turn to the indicated heading assigned by ATC.

I don't have a compass card handy to play with, but I don't believe that the add/subtract is symmetric around the circle, such that one gyro setting will cover all the errors. I know for sure that the error is zero on some headings and non-zero on others in the plane I was flying last, which pretty much proves that point without further elaboration.

Sarah said...

I use Asquared's method, setting the DG to the actual magnetic heading including the correction. That way you don't have to apply the correction except when setting the DG to the compass.

Since the error isn't symmetric around the circle, setting the DG directly to the compass builds in the compass error as if it were.

Really, either method works. In practice, I think I'd have trouble reading the mag. compass to within the +/- 2 degrees on my correction card unless it was exactly on a degree line mark.

A Squared said...

The correction is rarely as much as five degrees in any direction,

I've seen them with as much as 10 -15 degrees.

I don't have a compass card handy to play with, but I don't believe that the add/subtract is symmetric around the circle,

They aren't, of course.

“.............such that one gyro setting will cover all the errors.

You need to step back for a moment and think about what you're saying. If the DG is set to the actual magnetic heading (rather than the compass reading) it will show the actual magnetic heading on any heading (plus or minus gyro inaccuracies, which will exist in any case)

Assume non-negligible compass errors

Assume a perfectly functioning Direction Gyro.

You set the DG to actual magnetic heading, using the compass and the correction card. From that point on, the DG indicates actual magnetic heading, without applying a correction. How is that not preferable to setting it to a compass reading, which is in error, and then having to apply a compass correction to the DG reading each time the instrument consulted?