How exactly are satellites drifted to slots in different orbital planes?

The Iridium constellation is being replaced by new satellites. There are six planes with 11 satellites + one spare each. Since the Falcon 9 launcher can only carry ten satellites per launch, filling the planes is not an easy task. Out of the first launch, eight satellites were inserted into orbit to take the slots of the old satellites. Two vehicles have started drifting to the next plane (westwards). It will take them app. 10 months to get there.

The next launch will insert only five satellites into service orbit, another five will be drifted west- as well as eastwards.

I would like to know how the drifting is actually handled. I somewhere read that as far as I understood, westward drifts are helped by precession. I figured out that the drifting satellites move app. 0.11 degrees more to the west than their non-drifting companions' RAAN changes on a daily basis. The drifters are in a lower 629/611 vs. 783/781) and more excentric orbit (0.0013000 vs 0.0002100 acc TLEs) and have a slightly lesser inclination (85.88 vs 86.4). Do I correctly assume that these factors aid the drift? And apart from that, do the vehicles do side-burns on a continuous basis? And if so, does this require less delta v than - hypothetically speaking - putting them in an equatorial orbit and spitting them out at the appropriate RAAN to make a (nearly) right-angle turn (I know it could not work like this, but just to have a simplified model for my brain)?

I guess it also needs to be kept in mind that the drifting satellites have to end up in very precise positions, since they have to perfectly fit into the slots of the constellation. They have to cross-link with the satellites before and behind them as well as to the ones in the planes east and west.

Anything else that comes into play that I haven't even touched upon?

The precession depends on the height of the orbit. The storage orbit for spare satellites is a bit lower than the regular one (620 versus 780 km) and precession is a different one (not going to do the maths here). This means, this satellite drifts slowly with respect to the operational satellites in the higher orbit. The whole process doesn't need any fuel, despite for changing the orbital height when reaching the right plane. This amounts to about 85 m/s using a Hohmann transfer and the two heights you give in your question.

And apart from that, do the vehicles do side-burns on a continuous basis?

No, that is not necessary. You don't have to care about precession, as it is the same for all constellation satellites. If they all precess (move westwards) at the same rate, the constellation stays essentially the same for all practical purposes.

And if so, does this require less delta v than - hypothetically speaking - putting them in an equatorial orbit and spitting them out at the appropriate RAAN to make a (nearly) right-angle turn (I know it could not work like this, but just to have a simplified model for my brain)?

You can do a simple back-of-the-envelop calculation here: Changing from an equatorial orbit to an almost polar one requires to cancel all "westward" speed and built up "northward" orbital speed. This is a $\Delta v = \sqrt 2 v_{\rm orbit}$, or about 11 km/s. These satellites carry fuel for at most one tenth of this amount. You could do the plane change in a very high and slow orbit (above GEO) and save fuel, but getting there is prohibitively expensive.

I guess it also needs to be kept in mind that the drifting satellites have to end up in very precise positions, since they have to perfectly fit into the slots of the constellation.

They only important point is to get the satellite into an orbit with the same RAAN as the others of the destination plane. The precise position along this plane can then easily be reached by varying the orbital height slightly - E.g. if you change to an orbit which is 1 minute (out of 100 minutes orbit) shorter or longer, you can reach any point on the plane within about 50 orbits. For comparison: The lower of the two orbits you mention is 4 minutes shorter. Naturally, you can time raising to the right altitude appropriately to end up in the right spot right away.

• Thanks a lot, really helpful answer! I still haven't got a few points, though. Let me clarify: SVs 105 and 108 are being drifted from Plane 6 to Plane 5, which has a RAAN of app. 30 degrees less than Plane 6. These two satellites drift westwards faster than the sats in the constellation (by app. 0.11 degrees p.d.). But how do they do it? Is this only due to the small difference in inclination? And does the first part of your answer refer to an eastward drift? Because in that case it would make sense to me that the drifting sats precess more slowly than the rest of the constellation. May 29, 2017 at 19:58
• It's not due to the inclination (mainly), but due to the orbital height. Precession is faster the lower you are. That is, because the equatorial bulge of the Earth has more influence the closer you are. All satellites in prograde orbit drift to the West. May 30, 2017 at 8:32