Unhappy Faro :(

The Hex River Valley's autumn colours are long gone & it now wears a spectacular wintery look...

The Faro Arm's the latest member of the team to go down with a nasty bug... Extensive testing of this essential piece of equipment's shown that the measurements aren't quite as accurate as they should be. The arm's being sent back to the agents in Joburg for more tests & possibly repair or replacement. Since we can't really proceed without it at this stage, we're returning to Cape Town until further notice :(

We leave Francois alone in the clean room to puzzle out how to set up & measure the position of a crosshair on the CGH.

Paddling around between M2 & M5

Having vanquished the virus that was doing the rounds, the team dived back into the M5 alignment process!

The mirror we'd been using on the back of M2 (to get the alignment telescope set up in tip & tilt) wasn't flat enough so we now have a high quality mirror for the job. Since we need to be able to slide it in between M2 & M5 & move it in & out of position over the hole in M2, the mirror was set within a flat paddle.

The mirror protrudes 0.5 mm below the paddle so that the alignment telescope could be set with the mirror resting perfectly level on the back of M2. But the paddle needs to hold the mirror above the hole in M2 & still keep it parallel to the back of M2 so jacking screws with incredibly fine pitch threads (250 microns per turn!) were used to lift the paddle & get the mirror perpendicular to the alignment telescope's optical axis again.

A temporary mount for the alignment telescope was arranged on the aluminium ring of the NRS simulator so that it could peer down just inside the birdcage to see the mirror resting flat on the back of M2.

Once the jacked up mirror had been levelled again, it was time to haul out the nail varnish (Charlie: Ruby Red #018 - thanks Sam!) & lock the 3 screws in position.

Although Our Captain appeared to quite enjoy painting the screws, he's clearly more at home back behind a telescope...

True to astronomical form though, the human at the eyepiece was soon replaced with a small CCD camera, allowing the determination of centroids & offsets to be done more accurately by Francois' software.

A familiar scene from a new perspective - the god's eye view!

While all the fuss gets made about the IQ side of things, don't forget about Peter & the gang still doing battle with that scary looking spectrograph next door!

The M2 null lens gets its cell!

All the bits are ready so it's time for Ockert to assemble his cell & integrate the M2 null lens...

It's a perfect fit!

Now toss it up in the air & see if it's heads or tails. Well, not really - but its fun to imagine ;)

On goes the flexured retaining ring so that it can't get away...

Then along comes the beefy adjustment system that takes care of tip, tilt, decentre & despace.

All set - with safety clips & all... Beautiful!!

Aligning M5 with M2

The first step in this procedure is to set the alignment telescope up so that it's aligned with the optical axis of M2. Reflections from a mirror resting on the back of M2 & from the alignment target in the hole in M2 need to land on the graticule of the alignment telescope. This is achieved by carefully (& tediously!) adjusting the cones on the alignment telescope's mount & the tip-tilt stage on the fold mirror (that allows the telescope to look downwards).

One then has to view both the pinhole & the image of the pinhole as those two points give you M5's optical axis. That was made easier by etching a crosshair onto a piece of glass & attaching it to the top of the pinhole carrier.

The crosshair marks the location of the pinhole, which can then be viewed with the alignment telescope, as can the image of the pinhole (aka "the spot"). The pinhole carrier position is tweaked to get the best image quality for the spot & these two points then land roughly on the alignment telescope's graticule.

At this stage, the only degrees of freedom left belong to M5 & so the mirror itself needs to be moved to get all these points precisely centred on the graticule. M5 movements are effected by placing shims under the spherical bearings where the tie rods attach to the mirror.

That's the story in principle & all the fixturing's in place - now we just have to DO all this to the required tolerances... A task made tougher by the fact that Simon's led IT to quite the comeback on the pool table this past week!

The Anti-Gravity Quest

The Texans are back - this time to copy the anti-gravity drive on our tracker & to see what other useful ideas we have down here...

They seemed impressed with SALT's CO2-snow mirror cleaning system, but Adelaide's curry vetkoek took the cultural exchange to a new level!

Frosty Friday

Woke up to heavy frost all over this morning!

Eben brought his new baby in to show us how it's coming along - see the 12 July entry for how it all began. The handle's made of buffalo horn.

The remaining polishing will be done by hand...

It's been a good week up here - the M5 test's complete & we're confident that we'll be able to align the null lens for the M2 test when the time comes. So now, with the temperature rocketing up to almost 3 degrees already, we're heading home for the weekend :)

Playing with the M2 null lens

With the tests of M5's surface figure completed, the next mirror on our list is M2. We have the null lens that the manufacturers used to check M2 & a plan as to how to go about the test has been formulated. So now we need to establish whether we can meet the alignment tolerances in order to make a meaningful measurement...

The back of the lens is flat & the front surface is concave so the plan is to use the interferometer to get fringes off the back to align it in tip & tilt. Decentre looked like posing more of a challenge, but the shape of the lens comes to our rescue: you get circular fringes from the front surface thrown in for free!

Getting the lens squared up with the interferometer in tip & tilt kills the straight fringes (off the flat back of the lens) & then one can centre the circular fringes (from the curved front surface) with respect to the reference beam by translating the lens in X & Y - too easy :)

The remaining requirement is to get the axial spacing from the interferometer to the null lens, for which you need to know exactly where the point source focus of the interferometer is. A cunning trick for doing that is to place "a ball bearing on a stick" in the laser beam (close to where it emerges from the diverger) & adjust the focus until the circular fringes disappear. The centre of the ball is then at the point source focus & its position can be measured with the Faro Arm.

Midst all this fun stuff, there's still time for some funky photos :)