13
$\begingroup$

When using a camera and and not being stable in respect to the target, picture smear occurs. Pluto is 2370 km wide and flyby occurred at speeds ~14 km/s and from distance of ~10,000 km.

To put in perspective, let's scale that down to Earth terms: 14 km/s ~= 50,000 km/h => 60 km/h is a scale of ~840. Scaling everything down linearly:

  • Speed: 14 km/s => 60 km/h
  • Pluto size: 2370 km => 2.8 km
  • Distance: 10,000 km => 12 km

I could not find the zoom factor applied, but given that some images are zoomed to ~240 km, I would assume it is at least 10x.

So New Horizon's Pluto flyby is like going 60 km/h in a car and taking a picture of something of the size of ~2.8 km from the distance of ~12 km with a 10x zoom.

How is it that the pictures are so clear without any visible smear? Looks to me I have some concepts, data or calcs wrong - some clarification would be great.

$\endgroup$
  • 2
    $\begingroup$ The exposure time is essential to calculations related to motion blur .. $\endgroup$ – Andrew Thompson Jul 17 '15 at 14:53
  • 1
    $\begingroup$ It would take quite a bit of movement at 60 km/h to get a motion blur of a 12 km distant object 2.8 km in size, don't you think? And to remove some of it for longer exposures, you can still adjust your camera angle (pan, tilt,..) relative to some focal point. That said, I think you answered your own question in it. $\endgroup$ – TildalWave Jul 17 '15 at 15:13
  • $\begingroup$ @TildalWave Combined with high zoom, long exposure time and no smear at the pixel-level, I was not really sure if that's really true :) $\endgroup$ – steady rain Jul 17 '15 at 16:40
  • $\begingroup$ @steadyrain Exposures weren't that long, but it's pretty standard to maintain a nadir fixed attitude with respect to the target of observations. It would be mighty odd if New Horizons didn't do that. $\endgroup$ – TildalWave Jul 17 '15 at 16:43
23
$\begingroup$

New Horizons pans during shots to eliminate motion smear. See this related question. The Voyager mission proved that this was possible:

Between Voyager 2’s encounter with Saturn in 1981 and Uranus in 1986, controllers developed a technique called “image motion compensation.” This involved moving the scan platform at slow rates, which was found to be possible despite the damage, in conjunction with thruster firings to rotate the entire spacecraft at a rate that would allow a target to be tracked long enough for imaging.

Because New Horizons doesn't have a scan platform, they rotate the whole spacecraft to achieve this.

New Horizons uses these thrusters:

  • 12 Aerojet MR-103H 0.8 N thrusters (those were used on Voyager too)
  • 4 Aerojet MR-111C 4.4 N thrusters

These are all hydrazine monopropellant thrusters.

The MR103-H can be run in very short pulses, down to 4 ms. This gives a minimum impulse bit of 0.0026 N-s at 100 psia.

The requirement:

In particular, the RALPH instrument requires that the observatory be capable of setting up and maintaining a scan rate of +/- 34 μrad/s about any given axis.

Paper on NH's thruster system

$\endgroup$
  • $\begingroup$ New Horizons doesn't have a scan platform. Did they rotate the whole spacecraft? $\endgroup$ – Vedant Chandra Jul 19 '15 at 19:44
  • 3
    $\begingroup$ Yes. And they don't have reaction wheels either, so they have to use thrusters to do this, which is quite impressive. $\endgroup$ – Hobbes Jul 20 '15 at 6:54
  • $\begingroup$ Wow, what kind of thrusters can achieve such a low impulse and perfectly controlled burn? $\endgroup$ – Magic Octopus Urn Sep 9 '18 at 1:23
  • 1
    $\begingroup$ Added that to my answer. $\endgroup$ – Hobbes Sep 9 '18 at 8:13
4
$\begingroup$

Keep scaling linearly if you like:

Speed: 14 km/s => 60 km/h  => 60m/h
Pluto size: 2370 km => 2.8 km => 2.8m
Distance: 10,000 km => 12 km => 12m

That's like a car sized object 4 car lengths away. With that scale in mind 60 meters per hour doesn't seem very fast at all.

Edit:

Here's some additional information I have gathered from an informative article on the LORRI imager I found here.

Combined with the fact that the imager is combined with a scope of focal length 2630mm, there is some significant chance of blur.

To summarize those specs:

  • FOV: 0.29° × 0.29°
  • Nominal exposure time: 50-200ms
  • Focal length: 2630mm

The fov of LORRI is 0.29° which is around 5060μrads. At 1024 pixels that's about 4.94μrads per pixel. That number is confirmed in the linked PDF on page 5. The nominal exposure times are 50-200ms. Using the numbers from above (14km/s at 10000km) I get 280μrads of radial distance moved at the longest exposure. That's 56 pixels or about 5% of the FOV. That IS large enough to cause blur except for the fact that the vehicle does maneuver to keep the target in place.

The article explains how they deal with another phenomenon referred to as "readout smear" which is caused by the way the electronic data is flushed from the CCD.

$\endgroup$
  • $\begingroup$ Checking the math, the radial distance moved in 0.25s is 350 urads, well over 1 FOV for LORRI. Smear is a very real possibility. $\endgroup$ – PearsonArtPhoto Jul 18 '15 at 17:26
  • $\begingroup$ A valid point. I added more info to my answer to address your comment. $\endgroup$ – Octopus Jul 19 '15 at 5:03

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.