Tag: 2014A

Tonight was an interesting night. We got on sky pretty quickly and proceeded to taking science images with ~1″ seeing. Everything was working well until the seeing blew up to >3″ for about an hour, essentially killing our first target.

The AO fought admirably throughout the horrible seeing bursts but eventually we just had to wait it out. The seeing eventually died down to a steady 1-1.5,” allowing us to make some great science observations for a few hours.

But the night wasn’t done messing with us. Our next obstacle: clouds. They came in and pretty much obliterated all flux hitting the wavefront sensor for about about 2 hours in the second half of the night. After some good old-fashioned waiting and looking outside with our own eyes to inspect the clouds, we were convinced there were enough non-cloudy holes to move to the next science target.

But then we encountered our third obstacle: a telescope malfunction. This effectively killed the rest of the night for us.

However–while we were on sky we did make some great observations of some very interesting targets. All in all, despite the various problems, I would call tonight a successful night. Everything is semi-awesome.

Pictures:

Quotes:

Various people at various times: “Everything is not awesome when the seeing is bad”; “Everything is awesome when the clouds disappear!”

Laird: “I can’t work in these conditions. I can’t do this.” (referring to operating the AO in the horrible seeing)

Laird: “It’s not me, it’s you” (claiming it’s not the AO’s fault)

Laird (redacted)

Jared (in response to redaction): “You know, you can just make up quotes for the blog.”

Goodnight/morning everyone.

I’m very happy about how we switch between our target of interest and its point spread function (PSF) star, so I’m guest blogging about it. We want the same AO correction on both objects, even if they are slightly different magnitudes (usually we choose a slightly brighter PSF). We also want to be very efficient, of course, and change between them quickly.

Keeping these in mind, I look for a PSF that is 1) nearby on the sky, i.e. within a few degrees of the target, and 2) has a similar temperature to the target (so the wavefront sensing in visible and the IR camera still see similar brightnesses).

When we are ready to go to the PSF, it takes everyone working together to make the change efficient. We make sure the two science cameras are done integrating, and then we tell the AO operator (AOO) to “pause” the AO. This opens the loop, but saves the gains (the strength at which the AO modes are driven). Once the AOO confirms that the AO is paused, the Telescope Operator (TO) moves the Magellan guider from its park position (GPARK) to its center position (GCENTER). Once the guider is in, it occults the wavefront sensor, but we can see in the guider field of view exactly where the star was placed when it was centered on the pyramid. Since the AO wavefront sensor is off axis, the location of the star as seen by the centered guider changes with time and position on the sky, but since our PSF and target are close together, the position of the PSF should be very very close to where the target was. So, the TO marks the position of the target in the guider.

Now, the TO slews to the PSF star and brings it to the marked position on the guider. He then moves the guider out of the way, back to its park position, and voila, the PSF appears on the pyramid. Then the AOO can unpause the AO system, and we’re locked on the PSF with the same gains (level of correction) as we had on the target.

This is super efficient. From the time of our last integration on the target to our first integration on the PSF is less than 3 minutes! That is a testament to the efficiency of the Magellan TO and telescope as well as the efficiency of our teamwork on the AO pausing/restarting.

In other words, “Everything is cool when we’re part of a team.”

We’re not just singing “Everything is Awesome” on our PSF switching. We also sing it when we’re able to stay closed loop on a star when it passes only 46.5 min from zenith. Magellan is the only Alt-Az telescope I know that tracks so well at zenith and without vibration. And Laird says MagAO is the only system that can keep the loop closed when the rotation is changing so fast. How awesome is that?

By virtue of doing my afternoon conference call from the Vizzytor Support Building, I got a good look at our sleepy friend Vizzy. An energetic relative of his (hers?) was hopping around the rocks just west of Clay as I left the dome yesterday morning.

I wish I could sleep as late into the afternoon as Vizzy.

Tonight we said <cough> goodbye </cough> to Alfio, who is headed back to Italy tomorrow (despite the efforts of the Lan Chile domestic pilots strike). Kate chose this farewell song in his honor. We trying not to let the AO system know he’s gone, since it’ll be 24 hr before we can call him with any desperate pleas for help.

And finally, for some science. Here is a cool quick data reduction complements of Kate Follette that shows the re-detection of the accreting (H-alpha bright) point source from last year’s commissioning run (HD 142527B in Close et al. 2014, ApJ ). The new H-alpha and continuum filters are working well (see yesterday’s blog post showing the differential imaging).

 

 

One of MagAO’s specialties is high contrast imaging in the visible using simultaneous differential imaging (SDI). For this run we bought a new H-alpha SDI filter set, and tonight we really put them to use. This image compares our on and off H-alpha PSFs. These are formed from 1 hour of 45 second exposures. No strehl selection or any other shady tricks.

We close tonight with deep thoughts by T. J. Rodigas: “What I’m worried about is human knowledge after the Earth is destroyed . . . we should beam it up to the space station.”

Our first non-Arizona visiting astronomer was taking data with Clio and VisAO tonight. His name is Brett Addison of the University of New South Wales, Sydney.

Also Alycia arrived safely today.

Today the seeing was bad to very bad for almost the whole night.

Alfio made a new mirror diagnostic last year that shows the mirror commands sampled at a very high speed. See the video here: It’s a really cool one!

Then this run he added the plot to the right that shows the maximum mirror command at each time step. I liked this plot so much because it shows you both how stable the loop is as you are adjusting the gains, and also gives a diagnostic for what happened if the loop breaks open. I told Alfio it was my favorite plot, so he made a quick update to the code:

In honor of our knocking out targets numbers 1-10 tonight (although we had to skip target 3 due to seeing and faintness, so it was 9 targets total…) here is Tina Dico singing her song Count to Ten:

We had another somewhat so-so night. Some high winds made it interesting, but there were few to no clouds and seeing was at least mediocre. But right at 4 am, when we switched targets, seeing blew up to more than 1.5 arcseconds. Sorry DZ — we aren’t doing this on purpose.

There have been several calls for the return of the quotes of the day. Here’s a sample from the last few days:

Jordan: “The problem with this AO system is that it stays closed for so long that it’s hard to find a bathroom break.”

Alfio: “I went there once to take their cheese.” (explaining that he has been to the Baade telescope).

Laird: “You fooled me by taking the data incompetently.” (to T. J.)

Finally, my song for the night is absolutely full of astronomical references. Free MagAO sticker to the first person who emails me a list (no googling, on your honor). In any case, you should listen to this with the volume at 11. Trust me, it’s better that way.