4/11/2015 8:35pm
Om said:
4/8/2015 3:56pm
...
We understand that Dawn will be adjusting its orbit, constantly.
We are interested in the timing of the big red dot point:
Hi Om,
Remember, this is an artist's rendition. The graphic is based on a design reference trajectory that we are not flying for many reasons: gravity errors, OD errors, execution errors, updates to the DSN schedule which affect thrust on/off times, and periods of forced thrust to reduce hydrazine consumption during periods set aside for statistical thrusting. So you are trying to extract precision information from a representation that doesn't support it. Even the width of the white line is large and hence it conceals details of the plane. I don't have time to dig up the details of the design reference trajectory, so to try to answer your question, I took a quick look at the time derivative of the plane angle of the actual profiles we uplink. I think the break you are seeing is a result of the interruption in thrust for OpNav 6 and the associated DSN pass. That would make sense too, because our thrust vectors (even in the reference trajectory) generally change more at the boundaries between segments than within segments. There are several reasons for this, but a big one is that we represent the inertial thrust vectors we transmit to the spacecraft as Chebyshev polynomials, and we (prefer to) use a single polynomial for a complete thrust segment. So putting the bigger changes between segments reduces the size of each polynomial and hence increases the accuracy for a limited number of coefficients. I hope that makes sense.Since I'm not sure I really answered your question, maybe I can make it up to you with a brief comment on your other questions. I don't have time for thorough answers, but perhaps this will be of some interest.
Om said:
1. Apodemeter
I don't mean to sound like Sgt. Friday, but why was Dawn communicating with the DSN between the hours of approximately:
Mar 17, 2015, 06:25 UTC: Start
Mar 18, 2015, 11: 00 UTC: Stop
28.5 hours!
Parallax?
Doppler shift in carrier signal?
1) We were communicating from Mar 17 at 14:50 UTC to Mar 18 at 10:55 UTC. It was for routine uplink and downlink plus routine radiometrics (Doppler and ranging). This was to refine the orbit knowledge to get a good delivery to RC3. We do this all the time, and there was nothing special about what we did at that time. What was special, however, was that our Monte Carlo analyses had shown that we were a bit more sensitive to the OD accuracy then than usual. This was not a significant effect, but it was noticeable. The DSN does a terrific job, but it is a very, very complex system of systems, and performance is not guaranteed. Glitches occur. High winds, strong rain, a mechanical or electrical problem, another spacecraft declaring an emergency all could have caused us to lose a pass. It happens, and all projects have to learn to live with it. In this case, because it affected our deliver accuracy to RC3, I had us schedule multiple sequential stations in case one station or even one complex had a problem. That made the communications session last longer. As it turned out, everything went perfectly, so we got more radiometrics than the minimum we needed. I hope Sgt. Friday buys that alibi, because it's true!
Om said:
2. HAMO altitude changed from 1470 to 1450 km in this blog.
Is this a change of plans, or just non-nerd number smoothing?
2) The short answer is that it is what your fearless leader would call non-nerd number smoothing, but there is a little more to it than that. I did indeed decide for the purposes of the table that 910 miles and 1470 km were too many significant figures, so I rounded to 900 and 1450. However, my orbit altitudes are always rounded. We don't design on the basis of a single number for altitude, nor would that really make sense. I explained this in one of my Dawn Journals at Vesta, but it's actually pretty obvious. Vesta, Ceres, Earth, and virtually all other solar system bodies are not spherical, so a single altitude is not a good description, especially for a polar orbit. Inside the Dawn project, we talk not about altitude but rather about orbit radius. So every time I present an altitude, I make a choice about how to present it, and usually I just subtract the mean geometrical radius of the body. Then I round it to something reasonable, because specifying an altitude at the implied precision of 1 mile or 1 kilometer makes no sense when the actual altitude varies by so much more than that with our targeted circular orbits. There is another effect as well. I have mentioned in several blog comments that I will give an updated Ceres mass in my May Dawn Journal. I'll leave you in (mild) suspense, but my choice for how to specify the HAMO altitude in the table was influenced by an expectation for what is going to be the new design altitude when we account for what the actual mass is.
I hope some of this information is helpful.
Regards,
Marc