# How is a satellite constellation launched?

If want 10 satellites chasing each other around the earth (evenly spaced) all one the same plane, these satellites would be evenly distributed around the globe after launch by slightly raise or lower their orbits. As long as there is a few weeks available to do this, the adjustments are small so the deltaV required is minimal.

At first I asked "Does each satellite slightly raise or lower it's orbit to start spreading out compared to one reference satellite?" but comments suggest this has already been discussed in answers to Deploying multiple satellites from one second stage

However, if I want to do full global coverage and have this setup over multiple different orbital planes, would they each need a different launch? I assume the deltaV for a (satellite to make its own) plane change is too high to make it worth it?

Does each satellite slightly raise or lower it's orbit to start spreading out compared to one reference satellite?

Yes.

So off the bat there's a deltaV penalty for distributing the constellation?

The ∆v required to phase the orbit can be arbitrarily small if you're not in a hurry to reach the final configuration, as described in this answer.

I want to do full global coverage and have this setup over multiple different orbital planes, would they each need a different launch? I assume the deltaV for a plane change is too high to make it worth it?

Unless the planes are very similar, that's correct -- separate launches per plane is the way to go.

• I made an edit to the question that tries to maintain it (since you've quoted from it) and simultaneously address the close voters' concerns.
– uhoh
Commented Nov 9, 2018 at 4:01

About the second part: changing the declination of an orbit is always so costly, that no one does it. Consider a 180 $${}^\circ$$ change, i.e. if we would reverse the orbit direction, it would require the $$\Delta v$$ of the double of the second cosmical speed, $$\approx 15.5 \frac{km}{s}$$. The typical $$\Delta v$$ reserve of the deployed satellites is measured more in some hundreds $$\frac{m}{s}$$.

Deployment to the same declination requires only minor changes in the orbit, and then we can simply wait until the satellites get to the expected phase.

• There may be ways to do a 180° using the Moon from LEO that use less. I think there is a Q&A about that here somewhere. I'm not advocating that as a better solution, but only as orbital-mechanical-trivia. Related to that is the challenge Would lunar flyby be less costly in delta-v than direct change from ISS to Hubble orbits
– uhoh
Commented Nov 9, 2018 at 4:03
• If you have a chance, take a look: What might these silent close-voters reasons have been? How might I address them? Thanks!
– uhoh
Commented Nov 17, 2018 at 2:21
• @uhoh Sorry for the late react. My first, largest problem with the question was that it is harder to understand, than writing three good answers. Although this alone is not a close reason, but was an argument for me to strengthen the for-closure decision. Understanding the question requires googling for most users, thus my personal opinion is not a decisive "opinion-based" argument, but a mix somewhere between "unclear" and "opinion-based". I could vote only one, I can't remember which reason I clicked. The question has already 4 reopen votes, I hope it will be made also more clear. Commented Nov 17, 2018 at 20:37
• @uhoh Typically I am inclusionist, far more inclusionist as the community in all SE sites, but not over all limits. I think maybe I should have voted for "leave open", it was a very border case. Somehow you've found a case where I was on the side of the close-voters and I feel a strong cognitive dissonance now, I am nearly always on the opposite side in such debates. I gave the last reopen vote to the question. Commented Nov 17, 2018 at 20:41
• @uhoh Btw, I think the image on the bottom is unneeded, it only confuses the comprehension, I think some link, citation or tabular data would be better. Target to make the post more easily comprehensible. Just to see some Canadian government member, one of them in a quite uncommon clothing, has no net information value. Commented Nov 17, 2018 at 20:44