Directional Gyro Indicator (DGI) — Page 162, Lesson 196

I want to walk you through the Directional Gyro Indicator, and specifically the part that deals with compensating for the earth's rotation — the latitude nut correction. First, let's set the scene. A gyro that's free in space holds its spin axis fixed. But we're on a rotating earth, so relative to the earth's surface, the gyro appears to drift. That's called apparent wander, or apparent drift. The DGI has to cancel that out, and it does so with a mechanical device called the latitude rider nut. Here's the construction. There's a threaded stud fixed horizontally to the inner gimbal. On that stud sits an adjustable nut — the latitude rider nut. When the nut is in its central position, its effect is cancelled by a counter-balance weight on the opposite side of the gimbal. So in the middle, the nut does nothing. Now, what happens when you screw the nut out a few turns? It applies a downward moment on the gimbal. Apply the usual precession rule to that downward force, and you get — viewed from above — an anticlockwise precession of the gyro, including its scale, in azimuth. And that anticlockwise precession causes the reading in the window to increase. Conversely, if you wind the nut in, you get clockwise precession, and the readings decrease. So the whole idea is this: the earth's rotation causes apparent drift that varies with latitude. The latitude nut produces a real drift — a genuine mechanical precession — that is equal and opposite to the apparent drift, for a given latitude. That cancels it out. And because you can screw the nut in or out, you can compensate for increasing readings — that's the southern hemisphere case — or decreasing readings — the northern hemisphere case. Now, here's a critical limitation. The setting can only be changed under workshop conditions, not in the aircraft. So the compensation is fixed for one chosen latitude. The good news is that errors from latitude changes are usually small compared with the random wander errors of the DI itself. But if the aircraft is moved to a new operating area with a latitude change on the order of 60 degrees, a DGI with the appropriate latitude correction would probably be substituted — swapped out for one set for the new latitude. Let me give you a concrete worked example, because the numbers matter. At 60°N, the apparent drift is minus 13 degrees per hour. That minus sign means the reading is decreasing. Where does that 13 come from? It's 15 times the sine of 60 degrees — 15 sin 60°. The 15 is the earth's rotation rate in degrees per hour, and the sine of latitude scales it. So the latitude nut introduces a real drift of plus 13 degrees per hour. Plus 13 cancels minus 13, and the resultant drift — assuming no random error — is zero at 60°N. But here's the catch. That plus 13 degrees per hour compensation is now present at all latitudes. The graph in Figure 12.13 shows this by renumbering the drift scale. At 60°N, drift is negligible after compensation. But if the aircraft moves south of the equator, the drift values become greater than if no compensation had been made at all. The compensation that was perfect for 60°N actually makes things worse far away from it. Let me summarize the rules, because they're the practical takeaway. Flying north from the corrected latitude gives a decreasing reading — that's a minus drift rate. Flying south from the corrected latitude gives an increasing reading — a plus drift rate. Flying away from the corrected latitude makes the drift rate increase. And flying towards the corrected latitude makes the drift rate decrease. So the latitude nut is a workshop-set mechanical correction that cancels apparent wander for one chosen latitude, using a real precession equal and opposite to the earth-induced drift — and its accuracy degrades as you move away from that latitude, which is why a large move of about 60 degrees calls for swapping the whole instrument.