Help Shape the Future of Space Exploration

Join The Planetary Society Now  arrow.png

Join our eNewsletter for updates & action alerts

    Please leave this field empty

Headshot of Emily Lakdawalla

Talking to Pluto is hard! Why it takes so long to get data back from New Horizons

Posted by Emily Lakdawalla

30-01-2015 9:53 CST

Topics: New Horizons, mission status, explaining technology

As I write this post, New Horizons is nearing the end of a weeklong optical navigation campaign. By taking photos of the Pluto system at regular intervals, New Horizons' navigators can precisely measure the observed positions of Pluto and its moons with respect to background stars, and determine the spacecraft's position. The last optical navigation images in the weeklong series will be taken tomorrow, but it will likely take two weeks or more for all the data to get to Earth. Two weeks! Why does it take so long? It's not like it's all that much data: 10 full-resolution LORRI images per day.

The short answer to that question is: Pluto is far away -- very far away, more than 30 times Earth's distance from the Sun -- so New Horizons' radio signal is weak. Weak signal means low data rates: at the moment, New Horizons can transmit at most 1 kilobit per second. (Note that spacecraft communications are typically measured in bits, not bytes; 1 kilobit is only 125 bytes.) Even at these low data rates, only the Deep Space Network's very largest, 70-meter dishes can detect New Horizons' faint signal.

DSS-43, the 70-meter dish at Canberra, Australia

Glen Nagle

DSS-43, the 70-meter dish at Canberra, Australia
Taken on November 10, 2011 while the Sun was surrounded by a "sunbow."

How much data is in a single LORRI image? (LORRI, which stands for Long Range Reconnaissance Imager, is New Horizons' highest-resolution camera.) LORRI's detector is 1024 pixels square. Like many modern space cameras, when the camera reads out its detector, it digitizes each pixel as a 12-bit number. Twelve million is an awful lot of bits, but fortunately LORRI's images are amenable to lossless compression, especially now when they contain mostly black space; they can be zipped up to about 2.5 Megabits without any loss of detail. They can be made even smaller with lossy JPEG compression, but for optical navigation, precision counts; the pictures have to be returned losslessly.

So, do the math. 2.5 Megabits, at 1 kilobit per second: it takes 42 minutes to return one LORRI photo to Earth. Most communications sessions last about eight hours. That's eleven images per communications session. And that assumes that New Horizons is transmitting only LORRI data, which it's not; there are other science instruments and spacecraft housekeeping data, too. The Deep Space Network has only three 70-meter dishes, and there is a lot of competition for time on them; New Horizons is lucky to get one communications session per day. And while New Horizons is pointing its dish at Earth, it can't point at anything else, including Pluto. It has to choose between communicating and taking data.

What all of this means is that whenever New Horizons is actively taking science data, it's building up a data backlog, which it fails to transmit completely in its next communications session. The New Horizons team wants to go into the close encounter phase with data recorders as empty as possible, clearing the decks for all that juicy data from the flyby. What to do?

They have a neat trick that can nearly double New Horizons' data transmission rate, but it comes at a cost of doing simultaneous science. New Horizons' radio system includes two Traveling Wave Tube Amplifiers or TWTAs (pronounced "twittas," like a Bostonian would say "twitters"). The TWTAs amplify the radio signals before they get broadcast from New Horizons' 2.1-meter dish. There are two TWTAs for redundancy: if one fails, the mission will still be able to return data to Earth. But the two TWTAs are not quite identical. One of them transmits radio signals with left-hand circular polarization, and one of them transmits with right-hand circular polarization.

Because they transmit with different polarization, both TWTAs can simultaneously transmit the same data through the dish antenna. On Earth, special hardware at the Deep Space Network can separately receive the two differently-polarized signals, and then combine them to make the signal stronger. Stronger signal means New Horizons can transmit at a higher data rate, about 1.9 times the rate with a single TWTA.

This two-TWTA mode wasn't developed until after launch; they deployed it early in the mission, and it worked well. But radio transmitters are power-hungry. New Horizons' nuclear power source has decayed since it launched nearly a decade ago, and there is no longer enough power to run both TWTAs at the same time as all the other spacecraft subsystems. If they want to nearly double their data rate and reduce their backlog, they need to shut something else down.

Amazingly, they can shut down their guidance and control system and use the saved power to run the second transmitter. But how can you point stably at Earth with your guidance system shut down? The answer is to turn New Horizons from a spacecraft whose orientation stays fixed in space to one that spins. Spinning spacecraft have incredibly stable pointing. It's costly in terms of precious hydrazine fuel to spin up and spin down the spacecraft, so they don't want to make the transition from three-axis to spin-stabilized very often. And you can't take photos from a spinning spacecraft (not with the cameras New Horizons has, anyway). But it's worth it to spend a little hydrazine and quit taking pictures a couple of times in order to get all the approach data down to Earth before the near-encounter phase starts.

So that's why New Horizons is going to spend two long periods of its Pluto approach taking no image data, with the spacecraft spinning and its high-gain antenna pointed at Earth. The two spin periods are scheduled right after trajectory correction maneuvers, rocket-firings that will fine-tune New Horizons' path past Pluto. The first spin period will last from March 10 to April 4; the second lasts from May 15 to May 27. For image fans like me, it'll be a little frustrating to know that Pluto will be getting bigger and bigger, yet New Horizons is not looking. But the consolation is that New Horizons will completely empty its memory on both occasions, sending everything it's got back to Earth, and making room for more and better data. And while the spacecraft is spinning, its particles instruments SWAP, PEPSSI, and SDC can all still take data.

New Horizons should be able to use this two-TWTA communication mode until well after the flyby, assuming, of course, that both transmitters remain healthy. It'll still take more than a year to get all the science data from the encounter back to Earth; it's a lot of data, and a skinny pipe. And it won't be able to use the two-TWTA communication mode forever. Eventually, the output of the nuclear power supply will decay to the point that even while spinning, New Horizons won't have enough power to run both transmitters simultaneously, and we'll be back to using just one at a time. The Kuiper belt object flyby will almost certainly require data return through only one transmitter. New Horizons is a lesson in patience!


Here's some further reading on New Horizons' radio downlink. Thanks to New Horizons deputy project scientist Kim Ennico for her help with this entry.

See other posts from January 2015


Or read more blog entries about: New Horizons, mission status, explaining technology


Richard: 01/30/2015 03:15 CST

Juggling power between system activities reminds me of the old mainframe Star Trek game from the 70s. Players decided how to power to allocated between shields, scanners, etc. Bits nomenclature isn't limited to spacecraft communications. We talk about megabits, gigabits, and terabits per second in terrestial datacom, too.

Tom: 01/30/2015 03:57 CST

Fun comm fact for Voyager 2 @ Neptune. For the Neptune encounter, they combined the 70m antenna with two 34m antennas, and twenty-seven (!) 25m antennas (those usually used for radio astronomy). I think the idea was to get as much down between Neptune and Triton encounters as possible. Its onboard storage was a tape recorder. 1970s technology after all. Source:, section 6.3.3

Tom Burk: 01/31/2015 01:04 CST

Have they, or are they planning to, test the "spin without G&C" mode in-flight before the real thing? Wouldn't a safing event in that mode be harder to recover from with the G&C off at the time of the event?

Fred: 02/01/2015 12:15 CST

Thank you for that very understandable description! That's jargon even an ancient mariner like myself can understand!

Pioneer: 02/01/2015 04:05 CST

I remember when the Galileo spacecraft was at Jupiter sending data at a rate of 160 bits/sec at most. We had to be even more patient in those days, but the results were worth the wait.

Boyd Waters: 02/04/2015 06:21 CST

The VLA was used with Jet Propulsion Laboratory’s Deep Space Network in Goldstone CA to track NASA’s Voyager 2 encounter with Neptune in 1989. NRAO began planning and preparations in 1982, and test phased array observing on Voyager and planets began several years before the encounter as more VLA antennas became outfitted for X-band. NRAO devoted 340 hours of VLA observing time to system tests and reception of Voyager's telemetry, and there were 40 scheduled telemetry passes from 26 April through 28 September 1989. This November 1985 sketch by Socorro's Paul Harden shows Bill Brundage (NRAO) and Ron Logan (JPL) with the test instrument setup at the VLA

Boyd Waters: 02/04/2015 06:25 CST

My previous comment is a quote from the NRAO newsletter, written by Ellen Boutin. I was a software developer for the NRAO for 10 years. They still proudly display the posters commemorating Voyager at Neptune.

s3nsfan: 07/15/2015 01:26 CDT

Can someone explain to me how we can transmit data from 3 Billion miles away in 4 hours, but it takes 9 1/2 years to travel there. Please don't be rude, this is an honest question I am trying to understand

s3snfan: 07/15/2015 01:29 CDT

Further to my question, if i understand correctly the speed of light is approximately 298,000,000 m/s. If we take 3,000,000,000 billion miles transferring data at 4.5 hours that is a speed of 666,666,666 MPH, converted to m/s is approximately 299,000,000 m/s which is faster than the speed of light. Is that accurate? how can we transmit data at the speed of light? If I am correct in my calculations.

Lance: 07/15/2015 03:32 CDT

s3nsfan, Let me try to help: 1)The reason it take 9 1/2 years to travel there is because the spacecraft travels at about 31000 miles per hour not 671 million miles per hour like light. 2)I think you mean 3 billion miles NOT 3,000,000,000 billion miles. 3)The reason we can transmit data at the speed of light is because we use radio frequency (RF) electromagnetic radiation to carry the data. And electromagnetic radiation travels at the speed of light through space.

Sbob: 09/12/2015 08:44 CDT

Hi Emily - lovely post. But I fear your "short answer" may be one of the sources that is used to create the many variations on this paragraph I keep reading in the press: "Because New Horizons is so far away, more than five billion kilometres, data trickles back hundreds of thousands of times more slowly than over a fibre optic broadband on Earth." I find this explanation annoying because it is misleading. Should we not say it takes so long because it is too expensive to put a more powerful transmitter on the probe, rather than because it is so far away? Surely the distance creates the wait for the start of a signal, but the budget dictates how much information the signal can transmit. Would I be correct to suggest that a more accurate "short answer" might be to imply that it's a function of power and distance (rather than just distance)? i.e. the shortest sensible answer (to my mind at least) would be: New Horizons has limited broadcast power and is transmitting from a great distance. As distance increases the signal spreads out and becomes too weak to carry significant data. A vastly more powerful transmitter could send the data at an arbitrarily higher rate, but at significantly higher economic cost. For instance if the probe had been designed to crash into Pluto rather than perform a fly-by, then obviously data rates would have been the key concern because we wouldn't have leisurely months to send the data back. In the extreme, if NH had been designed with a broadcast power, say, 150 orders of magnitude higher, then surely it could send back all its mission data in a very short time. Science tells us that we can't avoid the wait for the start of the transmission to reach us. But is the data per second delivered not completely arbitrary and based on budget?

Suzanna: 09/22/2015 12:00 CDT

My students and I were wondering: If Pluto is four light years away, how can data get back to us in two weeks? We were confused. Thanks!

Suzanna: 09/23/2015 03:42 CDT

Hello, I just wanted to follow up in case anyone encounters my last comment. We realized that Pluto is ~4 light-HOURS away, not light years, and this we now know how it is possible to get data back from Pluto in a short period of time! (Lesson for the day: Units are important.)

Leave a Comment:

You must be logged in to submit a comment. Log in now.
Facebook Twitter Email RSS AddThis

Blog Search

Planetary Defense

An asteroid or comet headed for Earth is the only large-scale natural disaster we can prevent. Working together to fund our Shoemaker NEO Grants for astronomers, we can help save the world.


Featured Images

LightSail 2 and Prox-1
Bill Nye at LightSail 2 pre-ship review
LightSail 2 pre-ship review team photo
Swirling maelstrom
More Images

Featured Video

Class 9: Saturn, Uranus, and Neptune

Watch Now

Space in Images

Pretty pictures and
awe-inspiring science.

See More

Join The Planetary Society

Let’s explore the cosmos together!

Become a Member

Connect With Us

Facebook, Twitter, YouTube and more…
Continue the conversation with our online community!