Venus Express Options

[babakm]

Much appreciated update Don.

[cndwrld]

New VEX Science Information Released

On the ESA Science page at: http://www.esa.int/SPECIALS/Venus_Express/
there are two new releases that you might find interesting.

The first shows some great 'movie' images taken by the VIRTIS imaging spectrometer of Venus' south pole. Combining the IR images with a UV image of the dayside, you can see the full rotation of the south polar vortex. Over a period of five orbits, VIRTIS took images centered around apocenter. The page strings them all together, so that you see the five or so shots from one orbit, and the sudden jump is when the data jumps to the images from the next orbit. You can really see the change in the cloud structure over time in the vortex.

Just released is an article about ground-based observations that will be taking place in coordination with VEX. This will extend up to the Messenger Fly-By in early June, when we will be (sort of) simultaneously taking data from Messenger, VEX and the ground.

Over the next few months, Venus will be around its closest distance to Earth (inferior conjunction). Planning began this week for late August through late September, slightly past conjunction, and during that period we'll also be coordinating VEX observations with some ground observations:
- Infrared observations will be made by the Infrared Telescope Facility (IRTF) facility in Hawaii, USA.
- Sub-millimetre observations will be made for two days with the James Clerk Maxwell Telescope (JCMT), located in Hawaii, USA.

[cndwrld]

Venus Express Status

At the end of the last Cebreros pass in DOY 132, 18:00z, Venus Express
was orbiting Venus at 139 million km from the Earth. The one-way
signal travel time was 463 seconds. We are approaching inferior
conjunction, when Venus will be between Earth and the Sun.

The Venus Express spacecraft has been operating nominally, with only
relatively minor exceptions. One of these was a problem related to an
anomaly where a thruster failed to close. The thinking on that is
its "probably due to a missed pulse in the commanding
circuitry". I'd feel better if I knew what that meant, but even then
it would still mean that we don't know. We haven't seen a repeat, and
the best spacecraft problems are the ones that fix themselves.

With the installation of a flight software patch and the update of the
on-board ephemeris, Venus Express was configured for the quadrature
phase. This phase is defined as the period during which the
Sun-Spacecraft-Earth angle is between 75° and 95°.

During the quadrature phase, revised operating constraints on the
VMC camera lead to the necessity for changing the spacecraft attitude to
prevent unacceptable illumination of the VMC camera. To this end, fake
ephemerides on the positions of the spacecraft and the Earth were
uploaded to Venus Express.

On 11 May 2007, for the first time, Earth pointing was achieved
using fake ephemerides with a Sun illumination of ~10 degrees on
the +Y spacecraft face. The first pass with a tilted attitude was
closely monitored by the Flight Control Team (FCT). No anomalies
related to quadrature operations were detected and the performances
of the system was nominal. The flight control team and flight
dynamics teams at ESOC, in Darmstadt Germany, spent a large amount
of time and effort in planning this 'quadrature' exclusion period,
but it all looks good now. We're in Quadrature, in the
exclusion period, and it is all working perfectly. The ESOC flight
control and flight dynamics teams are very good.

Later in the quadrature phase, a swap to the smaller High Gain Antenna 2
is required for Earth communications, as the spacecraft attitude for
continued use of HGA 1 would result in illumination of spacecraft
faces not designed to cope with such exposure. This is scheduled for
01 June, and will immediately drop our downlink data rate from 228 Kbps
to 28 Kbps. Its unfortunate that during inferiour conjunction, our data
rate has to drop so much. Re-using the Mars Express design allowed
the VEX mission to get funded, but one consequence is that the thermal
constraints force things like this.

It looks like we've had our first non-recoverable failure. It appears that
the S-Band downlink path has a problem which significantly reduces our
downlink power on S-Band. That's the bad news. The current indications
suggest that the problem lies in the path between the entry to the RF
switch immediately prior to the HGA 1 antenna, and the antenna itself.
The good news is that radio science, which used to rely on having both
the X-band and S-band downlink, can now use only the X-band downlink
along with models of the Earth ionosphere (which replaces the need for
the S-band signal). There is still a lot of work being done, to try and
figure out exactly what happened, and more importantly, why.

Some delays have been encountered in the data deliveries for ingestion
to the archive, but a first release of (a subset of) Venus Express
data in the Planetary Science Archive (PSA) is still expected for
this summer.

The preparations for a special section on Venus Express results
(some 9 papers), to be published in Nature, are ongoing.

Planning for our sixteenth monthly medium-term plan of operations (MTP016)
are completed. We are just now finishing MTP017, and MTP018 planning started
this week.

[Guest_AlexBlackwell_*]

Ground - based observatories join forces with Venus Express
European Space Agency
23 May 2007

[The Messenger]

It looks like we've had our first non-recoverable failure. It appears that the S-Band downlink path has a problem which significantly reduces our downlink power on S-Band. That's the bad news.

???
Wasn't one of the instruments non-functional from the get-go? PFS? Fourier IR or something?

[cndwrld]

You're right. The PFS instrument
(http://sci.esa.int/science-e/www/object/index.cfm?fobjectid=33964&fbodylongid=1444)
never worked. The mechanical scanner is stuck.

I guess I don't count that because technically, they are still trying to free it. The flight control team is working with PFS to do a test shortly where they try to move it while we are doing a burn. Which is trying to kick it and turn it at the same time.

We don't hold out much hope. But one never knows.

[cndwrld]

There was some discussion about getting data down from spacecraft, and frustration with the amount of time it takes. I think I finally figured out how to post images, so I'll put up a couple examples from VEX to show how this stuff gets downloaded.

When we do our planning, our ESA proprietary simulation tool will create a prediction of how much data the instruments accumulate each orbit, as well as how much we are able to downlink during each comm pass with the Cebreros Station in Spain.

The resulting up-and-down graph looks like this, for our 18th monthly plan which I just finished:



This is a bad, bad MTP, because the data does not go to zero at the end of it. We have some restrictions on our downlink towards the end, and one can see that at the end of the 28 day simulation period, we have accumulated data which has not been downlinked. This causes problems with the planning, not on the spacecraft.

This image is the same simulation output, but for MTP015. The Messenger fly-by was in orbit 411, as I recall. We had a few passes where we could not downlink data, while we were accumulating data in support of the fly-by. Once the Messenger pass was over, we started our downlinks, but it took a while to get the data down.


Some instruments, who don't take large data volumes, had their data down within a couple passes. The instrument with the highest data volume, the VIRTIS imaging spectrometer, took days to get it all down.

I don't know how the Messenger data came down. But I'd assume that they got a certain amount of time on one antenna each day, so the data would get downlinked in pieces over a period of days. Their downlink would look different, but probably similar: A big increase in on-board data, followed by one or more dumps of a few hours each to get the data down.

[brellis]

hi cndwrld

thanks for providing a window into the dataflow workings

"There was some discussion about getting data down from spacecraft, and frustration with the amount of time it takes. I think I finally figured out how to post images, so I'll put up a couple examples from VEX to show how this stuff gets downloaded.

When we do our planning, our ESA proprietary simulation tool will create a prediction of how much data the instruments accumulate each orbit, as well as how much we are able to downlink during each comm pass with the Cebreros Station in Spain."

Is the storage capacity on V.E. heavily-burdened?

As to transmission back to earth, I wonder if there could be a software on the spacecraft similar to the program that updates a web page, looking for numbers that change. For example, once a normal temperature range in a given area over a given time period has been established, the craft prioritizes data that appears outside the norm for sending back to earth.

Similarly, I've wondered how "smart" the Mars orbiters can get as they "pushbroom" their way around the planet. Perhaps they can automatically notice small changes like new gullies.

[cndwrld]

>>Is the storage capacity on V.E. heavily-burdened?
Well, yes and no. The instruments (particularly the imaging spectrometer VIRTIS and the VMC camera) could take much more data than they do. We are usually limited by our data rate, and our ground contact periods, not the SSMM volume.

>>I wonder if there could be a software on the spacecraft similar to the program that updates a web page, looking for numbers that change.

I'd see two problems with this. First, it only would work with housekeeping telemetry, not with science data. And the housekeeping is much less in volume than the science data. We dump a fixed 216 Mbits per orbit for housekeeping data; we dump about 4200 Mbits per orbit of science data. It might be a useful idea in some circumstances, though. Say, on spacecraft on long duration missions in cruise phase, where all you have housekeeping telemetry and no science, or commercial spacecraft.

Two, you'd only get data when it changed. But in that case, if you are looking at telemetry to find a problem, you would have to trust the numbers you don't have. You'd have to believe that if no data came down, that everything was working and the number you got last week is still the number you have now. I would have real trouble doing that. In my experience, data that doesn't exist has an unknown value, and the item generating the data doesn't work unless it proves it works by putting out data. Things 'fail silent' all the time.
Again, though, in certain circumstances, it might be useful.

[cndwrld]

VEX Status, 22 June 2007

At the end of the last Cebreros ground station pass on DOY 167, 18:00z,
Venus Express was orbiting Venus at 97 million km from the Earth. The
one-way signal travel time was 323 sec. We are executing our 15th monthly
plan, MTP015, and beginning the first week of MTP016 next week. We are still
operating on the small high-gain antenna, HGA2, due to Sun angle on the
spacecraft during Earth Pointing.

We are currently in the Quadrature Period, meaning that the angle
Sun-Venus-Earth is less than 90 degrees. Only two faces of the spacecraft
can take long-term solar exposure, +X and +Z. During parts of this
Quadrature Phase, these two faces can only face the Sun if we point with
the smaller antenna. So our data rate, when we are closest to Earth, is
pretty low. Kind of counter-intuitive, one could say.

During short periods of Quadrature, near the beginning and end, when in
normal Earth pointing the +Z face of the spacecraft (the one with all the
instrument fields-of-view) was too close to the Sun. We want to prevent light
falling in the VMC camera field-of-view, as they don't have a shutter and found
out after launch that direct sunlight on their camera CCD causes degradation
to the plastic lenses. So for the past few weeks, we were operating with a
10 degree roll whenever we were in Earth pointing, to keep the Sun out of
the VMC field-of-view. For a spacecraft with tight pointing requirements and
precise pointing systems, it was not easy to find a way to make it continually
position itself 10 degrees off what it knew to be correct. But the team in ESOC
in Darmstadt figured out how to do it. Last week, we finished our need for this
offpointing, so we are back to normal pointing modes now. Still on the darn
small antenna, though.

No new problems have developed lately. Flight control systems and power are
fine; still have plenty of gas. Some instrument glitches occur now and then,
but nothing too worrisome. VIRTIS did a big upload to their instrument to
update their operating system and take care of some small errors they were
getting. The PFS instrument, stuck since launch, is going to be operated
during some momentum wheel off-loadings, when the thrusters are fired, in
the hope that moving it and kicking it at the same time might jar it loose.
The other earlier problems have not gone away: S-band downlink is way too
weak due to expected thermal damage on large antenna or cabling; thruster
abnormal firing which occurred but has not re-appeared; SPICAV shutter
gives odd readings now and then. But nothing new lately.

The radio science experiment team was able to do a Bi-Static Radar test last
week. This is where we turn on a special, ultra-stable transmitter, point the
big antenna at a feature on the surface, and blast away at full power at such
an angle that the reflected signal goes towards Earth, where it is usually
captured at Europe's antenna at New Norcia, Australia. The geometry between
Venus Express, a surface feature of interest and the ground station have to
be just right in order to do it, so we've only done five or six of these so
far in the mission. With low staffing, data analysis has not been what you
would call timely. Upcoming BSR observations have been cancelled because
they have discovered that the latest BSR data are significantly different
from what was expected and won't give them the information they need. It
will take a bit of time for them to figure out what to do next.

All the data from the Messenger fly-by support is on the ground, and being
looked at. We hope to see some press releases soon. The ground-based
observing campaign will be ending soon, as Venus moves further away from
Earth. We had a bit of a panic last week with the Moon's occultation of
Venus. One of our team realized that our downlink files didn't take the
occultation into account, and there was much rushing about. Turns out that
the occulation occurred in The Netherlands and Germany, but the Cebreros
station in Madrid was just outside of the occulation area. Which is why the
files didn't take it into account.

Planning began this week for two monthly plans in parallel, meaning double
the work for us and the instrument teams for the next four weeks. But we
are doing this because most of the teams take a month off for the summer
holidays. We'll work harder for a month so they can take a month off. I
think most of our team will not take a lot of time off during the summer.
Since none of us have kids, we can take our holidays outside of the rush
period in the summer.

In this planning period, we are introducing a new type of pointing that our
software will now support, called 'track pointing'. We should be able to
give a start time, and end time, and the lat/lon/elevation of a point, and
the spacecraft will track the point. Or we can give it
the name of a astronomical object in our database, and it will track the
object. I say that we should be able to do these things. The software has
a few 'issues' that were discovered recently, so we are going to just
allow a couple of them in each of the MTPs we are starting to plan now.
We'll see how it goes. If it works, though, it will probably get heavily
used.

In summary, things have not gotten calm and boring yet, we're doing a lot
of science that people can whine about not seeing but which should be more
published soon, and the spacecraft is doing well. Our next Science Working
Team meeting is near Rome in July, and I'm going to stay over an extra
day to finally go see the Pantheon, and the remains of the Venus temple
in the Forum. I'll pay my respects to Venus while I'm there.

[dvandorn]

>>Is the storage capacity on V.E. heavily-burdened?
Well, yes and no. The instruments (particularly the imaging spectrometer VIRTIS and the VMC camera) could take much more data than they do. We are usually limited by our data rate, and our ground contact periods, not the SSMM volume.

So, here's a question -- is the download rate limited more by intrinsic limits on data transfer rates (i.e., the spacecraft's transmitter power and the size of its dish), or by the relatively short periods of time you have the Cerberus facility available to receive downlink? (I know you have some attitude issues which result from VEX being designed to orbit Mars and not Venus, but I'm looking for average data transfer rates, here, not those that are constrained by other parameters.)

What I'm getting at, here, is whether or not a vastly expanded Deep Space Network would aid appreciably in getting more data off of the vehicle, thereby letting you take as much data as the instruments can actually collect.

I hate to see a valuable resource being under-utilized because of bottlenecks in our ability to get data back from it. And I'm wondering if it just isn't time for an expanded DSN to be started. If it could be funded by an international consortium that gets its seed money from something other than charging rapacious hourly rates, that would be even better... but I keep hearing more and more issues with the ability to get data back from our spacecraft that are out there exploring the unknown, and it just seems bass-ackwards to spend billions of dollars on a well-designed probe and then hamstring its science return because we have an antiquated DSN that can't keep up with demand and yet still charges thousands of dollars (or Euros, or what-have-you) an hour for the privilege of using it.

Maybe we ought to have everyone convert their unused Dish Network dishes into a vast distributed amateur Deep Space Network and offer its services to all comers, first-come-first-serve...?

-the other Doug

[Gsnorgathon]

...
we're doing a lot of science that people can whine about not seeing
...

Well, I should hope so. Imagine the whining if you weren't doing a lot of science!

[cndwrld]

So, here's a question -- is the download rate limited more by intrinsic limits on data transfer rates (i.e., the spacecraft's transmitter power and the size of its dish), or by the relatively short periods of time you have the Cebreros facility available to receive downlink? (I know you have some attitude issues which result from VEX being designed to orbit Mars and not Venus, but I'm looking for average data transfer rates, here, not those that are constrained by other parameters.)

What I'm getting at, here, is whether or not a vastly expanded Deep Space Network would aid appreciably in getting more data off of the vehicle, thereby letting you take as much data as the instruments can actually collect.

Well, let me mumble a bit while I try to think of something to say. Hmmph, hmm.

The limits on the downlink data volume are basically data rate and time of contact.

How to increase time of contact for an orbiting planetary spacecraft? We could have a steerable antenna, so that the only time we didn't have the ability to point at Earth was when we were occulted. We could transmit data and point the instrument platform to take new data at the same time.

We'd have more contact time, and more flexibility, if we could get as much time on the Earth networks as we wanted. VEX is limited to using the Cebreros station. But if Europe had its full network up (i.e. the third antenna, in addition to New Norcia, Australia and Cebreros, Spain), and each station had multiple antennas large enough for planetary probe reception (> 30 meter, say), and we could jump from station to station as the stations rotated out of view, then we could continue our downlinks as long as we wanted. It is worth keeping in mind that most stations are quite heavily booked these days. So it isn't just about upgrading networks, but expanding support would mean building additional dishes.

To increase the data rate, the spacecraft (steerable) antenna should be larger (while also pointing very accurately). And the ground station antennas should be larger, while also pointing very accurately. We use Cebreros, which is 35 m. DSN can give us 70 meters, which we used once and it let us pump a lot more data through. Transmitter power isn't really a problem for VEX, because power is not a problem. On some probes, it might be an issue. But its a design issue; it is easily scaled up for most missions, probably with minimal weight increases. The data rate also changes with distance from Earth, but there's not much we can do about that. I'm gonna assume for the foreseeable future that antennas will be fixed in size.

The problem with all of this, of course, is cost. No surprise there. A 35 meter antenna, with sufficient accuracy and senstivity, is incredibly expensive. A 70 meter is, well, more. Mulitple sets of antennas in the same location multiplies the price. Bigger antennas on spacecraft are expensive, but more (I am guessing) for launch cost than construction cost. You've got to get it in the launch fairing. The extra weight is expensive. And if you are too heavy and get pushed from one class of vehicle to another (say, Delta to Titan, or whatever), the difference in cost can be enormous. The ability to steer larger antennas accurately gets more expensive: motors, sensors, etc. Also more expensive in operations, although it can be largely autonomous now, I think (i.e., if you're on star trackers, you should be able to just say, 'point to Earth'.)

Can we simulate a bigger antenna on a spacecraft if we use two or three small antennas, separated from each other? Someone will have to tell me the highs and lows of that idea. But one promising thing is somehow getting a bigger antenna for smaller launch cost. Some commercial telecom spacecraft (like the Thuraya cell phone system) and military death stars use collapsible (or expandable) antennas, but I don't know much about them and I'm not a comm guy. Since they don't seem to have taken off, I'm assuming there are reasons for that. Cost again, probably.

I would spend all the money that people are willing to pony up, of course. I'm an engineer, and more expensive toys are always more fun. I happen to feel that we are already spending an awful lot of money, and we are getting the science down. So if we hit some constraints along the way, it doesn't bother me too much. When I worked on Magellan mapping Venus (or in my case, looking at radar telemetry while smart people mapped Venus), it was great fun. I hear that Magellan mapped about 73% of the surface. If we had spent an extra $20 million and gotten 80%, would it have changed very much? I guess the answer to that is up to the individual, but I'm pretty happy with 73%. With VEX, we should outlive our design life, and meet all our science goals, for an amount of money that Europe was willing to pay. And they are paying for Rosetta, and Bepi-Columbo, and ExoMars, one of which I hope to have an opportunity to work on. I guess I'd rather have mulitple missions with reasonable contraints than one or two more perfect missions. I'm biased, of course; I need a job.

I hope that answered some of your questions, Doug.

[dvandorn]

Yes, that answers a lot of my questions -- thanks! I guess one thing I was wondering was whether a dozen, or a hundred, little (like half- to one-meter) dishes on the ground could be used in a sort of an "antenna farm" mode to simulate a 70m or larger receiving antenna.

You see, there are literally hundreds of thousands of people here in the U.S. who have at one time or another used a satellite cable system (like Dish Network) but who no longer use that service. That's a lot of parabolic dish receiving antennas that are just going to waste.

If all of those unused dishes were mounted with decent views of the ecliptic (where most of the unmanned probes spend their lives) and tracking mounts, and set to communicate to a central data collection center, we could have a "distributed" DSN network made up of thousands of dishes. And heck -- some of them could be tracking Venus, some Mars, some Mercury, and some Saturn. And some tracking points in between.

I know it's probably a foolish pipe dream, but it would seem that you could vastly expand the DSN (within North America, at least) at relatively little overall cost (certainly at relatively little cost per dish) by making use of these unused dishes.

Your response tells me that such a network of dishes could be useful for getting additional science out of the assets we have in place around the Solar System. That's the information I was looking for. Now all we have to do is find the seed money for outfitting these unused little dishes, and we can get started...

-the other Doug

[djellison]

An upgrade for the DSN based on arrays of modest ( <30m but >1m ) sized dishes is under serious study - with a test dish being used at JPL - (I think it's 12m).

Doug