Directional Gyro Indicator (DGI) — Page 162, Lesson 192

Let’s pick up with the Directional Gyro Indicator, or DGI. We’ve already covered how it works in principle, so now I want to walk you through the errors that affect it. These are the things that make the DGI’s reading drift or shift, and you need to know them cold for the ATPL. First, gimballing errors. These are errors in the DGI’s indications that occur when you apply bank. If you plot the errors during a full 360° turn, you get an approximate double sine curve. That means the error goes positive, then negative, then positive, then negative — two peaks of each. And there are four headings where the error is zero, spaced 90° apart, sitting between those alternating peaks. The curve gets more complex if you change pitch during the turn. And here’s a key point: the actual headings at which the maximum errors occur depend on the make and mark of the instrument. So you can’t memorize one set of headings and expect it to hold for every aircraft. Now, why does this happen? It’s all about the geometry of the gimbal system. The gyro rotor axis wants to stay fixed in space. But unless the instrument case — and the aircraft bolted to it — can rotate about one of the gyro’s axes, the outer gimbal itself has to move. And that movement shows up as an error. The good news is that these errors are small, as long as your deviations from level attitude are only moderate. And they disappear as soon as you resume level flight. So gimballing error is a transient thing, tied to bank and pitch. Next, random wander. The gyro rotor axis can change its direction in space — that’s real wander — or it can appear to change direction — that’s apparent wander — or it can suffer from both. Real wander is mainly the random wander caused by manufacturing imperfections. For the DGI, gyroscopic rigidity is high, so random wander rates are low. Let me give you the numbers. An air-driven DGI with the rotor spinning at 10,000 rpm has a drift rate of about 1.6° per hour. A later design with 20,000 rpm has a quoted drift rate of 1.2° per hour. Electrically driven indicators can do even better — only a few degrees per hour. And for comparison, gyroscopes used in inertial navigation systems may have random wander rates of less than 0.01° per hour. That’s a huge difference, and it tells you why inertial nav systems are so precise. Now, apparent wander due to the rotation of the Earth. This is the big one. Let’s set up an azimuth gyro — that’s a gyro with the rotor axis horizontal — at the North Pole. The rotor axis stays rigid in space, assuming zero real wander. But the Earth rotates under it, through 360° in one day. That’s 360 divided by 24, which is 15° per hour. An observer standing still on the Earth moves with it, once around the gyro in 24 hours. So if this gyro is the DGI, its reading will be decreasing at the North Pole at a rate of 15° per hour. At the South Pole, the reading would increase at the same rate. That 15° per hour is the maximum rate of apparent wander due to the Earth’s rotation. Now let’s move to the equator. Set up the gyro with the axis horizontal and aligned north/south. In 24 hours, the observer and the gyro move with the Earth once around the Earth’s axis of rotation. But here’s the thing: there’s no change in the direction of the rotor axis relative to the meridian. So there’s zero apparent drift. That means the apparent drift rate due to the Earth’s rotation is a function of latitude — maximum at the poles, zero at the equator. So to tie it together: gimballing error is about bank and pitch, random wander is about manufacturing imperfections, and apparent wander is about the Earth rotating under a gyro that wants to stay fixed in space. The apparent wander is the one you can predict and correct for, because it depends on where you are on the globe.