Hi all! So at the end all the measurements are done…
First we verified the settling time of the ASM for a reduced set of modes. Here there are some examples of the step responses (the slowest mode and the most overshooting one) and a resume of the general behavior in terms of settling time:
After that we checked the ASM ability of tracking turbulence. A turbulent-like sequence of commands is sent and in the meanwhile we record the shell position using internal metrology. Here is the tracking error as a function of number of mirror modes.
Icing on the cake, we tested the FastLink connection between the NAS and the ASM (WFS Frames Counter running…):
Tip for users of pyramid wavefront sensors: you are not very sensitive to focus with a large modulation amplitude. For MagAO in particular, you have to get amplitude down to ~.1 milliradians before you really have any z-stage resolution.
Tip for users of the VisAO camera: there is a huge factor (it appears to be about eight) between z-stage motion and VisAO focus stage motion. So failure to heed the above PWFS tip will cause you to tear your camera apart several times in one day trying to find why your focus position shifted. You will feel very dumb later.
So that’s the story of Laird and Jared’s day. Because we forgot these two things, we were completely and totally baffled by the results we were getting when running focus tests on our camera. Since we replaced our gimbal mirror (the motorized mirror that lets us steer stars around our detector) we naturally assumed that was the problem. We even made up a bunch of reasons why it could be and proceeded to apply made up solutions. But in the end, we just hadn’t been careful enough with focusing on the pyramid, and so our numbers were meaningless and our time wasted. The good news, I guess, is that we earned a bunch of intuition for how our stuff works. When life gives you lemons . . .
We pulled the Gimbal back out of the system (I won't admit how many times) to try to adjust its position.Tightening down the bolts on the Gimbal mount.
This story started last night, when we figured out how to tighten the beam splitter wheel – making it much more repeatable. This changed the angle of our beam slightly, causing us to adjust our gimbal angle a little.
It's a bit terrifying, given the tight space and delicate optics nearby.
See Katie’s post here for a demo on why we’re so obsessed with our gimbal and beamsplitter. Those two devices need to be reliable and repeatable for us to efficiently acquire the stars we use to measure turbulence. They also need to be as free of vibrations as possible for VisAO science operations.
All this visible light focusing meant working with the lights off in the lab. That didn’t stop the ASM testing.
Armando and Marco are verifying that various mirror calibrations are still good.I came across this spider on my way to work this morning. Nobody panic, I gingerly poked it with a long stick and it didn't move.
An argument broke out at lunch today over how many days we’ve actually been here. Luckily we have a blog, so we just looked it up.
As the Guanaco noticed, Mario and Frederico have left us after more than pulling their weight. They took the early bus to La Serena.
Frederico, Laird, and Mario just before the boys from Microgate left for their long journey back to Bolzano. Thanks guys.
Now that the TSS is fixed, we moved on to mounting our windscreen. This is just a metal ring that encircles the bottom of the ASM to protect the shell from high wind. We also removed a set of shims from the mounting system.
The windscreen about to be lifted into place. This requires some fine alignment since our calibration system hangs from it.If you're going to have someone climb on top of your ASM, it should be Armando.The ASM with the windscreen.
In the afternoon Laird spent some time perfecting the routing of cables to the W-Unit. This is a hard problem – the whole thing translates in 3 dimensions, and has to be able to turn upside down, and cables tugging on optics generally ruins nights.
We had to make a few adjustments to the cable housing to get enough flexibility.The final product.This is for Enrico. After dinner we cleaned some dust off our WFS side fold mirror, and recorded these white light images. Pupils look great, and the SDSS i' is in focus. Click to enlarge.
Today’s quote: “You know Laird, you need to seriously think about hiring an exorcist.” (Armando Riccardi, commenting on Jared’s project long bad luck with electronics).
The remaining 4. Click for a rotated image - we leave it as an exercise for the reader to determine the correct orientation of this image.A nice shot by Laird of newly snow-capped peaks. The big storm improved the view a bit, once we could see it.
The clouds went in and out with the TSS problems. AO team discovered a problem with the Thin shell Safety System (TSS) that helps the mirror against the wind effects. One crate was not applying any TSS current.
Both Microgate and Arcetri teams at this point turn the system upside down and started bravely to act on the rear of the crates…
The currents setting of the TSS where slightly decreased in order to avoid a DC DC converter to go crazy. After that a cleaning is a must.
The system were restored in the working position.
… and finally verified all good….
At this point we can say good by at our Microgate friends that are going home.
The storm is finally gone, and the beautiful clear sky is back. The TSS is fixed too, and after we installed our gimbal mirror motors the W-unit is complete.
It's nice to see the sun again, and to not feel like you're going to blow away.The white light calibration source, in focus in the SDSS i' filter. Note the 3.2 pixel FWHM.
This (kind of lame) video shows our little W-Unit robot responding to a Board Setup command. This tells it to go to a set of pre-determined positions which makes it ready to operate.
The Burros came back after the storm. We hadn't seen them for a few days. A Guanaco walked by at lunch, so they're around somewhere.Saturday was GTL night.
At this point most of our urgent critical path stuff is done, and we are starting to relax a little. We have a few more days of ASM testing to do, and some mundane odds and ends in the W-unit to take care of, like cable management. We also plan to do some testing of our software interfaces with the telescope control system.
