Sci IPT 19 March 2008

Emerson, Nikolic, Hills, Testi, Wotten, Reid, Hales, Mangum, Indebetouw, Chris, Liu, Lucas, Laing, Myers, Remijan, Peck, Hibbard

Transporters are now moving around, as of yesterday. Brackets are on first Melco and Vertex antennas. Early April for trying to move first antenna. Antenna tests moving along.

Board meeting. Search for Director, Chair of Committee Bob Williams visited. NAASC Search ongoing. CSV: Mauersberger and Lucas have accepted.

ATF: Lucas has been working on telcal and working on asdm information. RL: Working on telcal. Successfully tested phase cal reduce delay measurements, offline tests for BP not ready, amplitude, not atmospheric. Wanted to advance on pointing and total power, not working yet. Perhaps before end of week. RH: ASDM files were missing pointing table, new version 5.0.3 is not released yet should offer improvements. Most of what we need for telcal is there. RL are having some stability problems with ATF. Phase glitches and drifts also are seen. RL: Integrate for 30 mins, you see mainly LLC resets, RR: ASDMs with PSI scripts running to correct LLC glitches. After running scripts, jumps did not show up so much. RH: Darrel, what is situation fitting for a single baseline. DTE: My info is out of date. Successive baseline fits to fairly long runs gives 1mm scatter in fitted antenna position. I'm still tracking down the source of this drift. RR: Did do another baseline run with sources around the sky; to look at this data one must translate it to understand it. RH: Will reading ASDMs be the method of choice in the future? At the moment our biggest interest is checking software and instrumental instabilities in phase. I think both these are high priorities. RR: This is a step in the direction we need. RH: Line stretcher item. Current design has servo to keep fiber length constant. Is this the best approach, given the effects we see at the ATF? Perhaps we should reconsider this approach. RL: What is the magnitude of this correction in phase at 100 GHz. RH: The range of the LLC is about one millimeter. Now one waits until this reaches the end of range and resets, then LLC produces a number of amount of stretch which is used to provide a phase correction, about a range of pi in phase. Normally most of the fiber is underground. The part in the antennas is most exposed. Over hours the fiber changes by few mm in hours. RL: If most of the variation is thermal we need to know that. RH: I think we should have a focussed set of measurements to do this, despite the ATF setup being not final. RL: The setup may be different but the setup is the same. RR: Did try to separate temp from other variables. When slewing the LLC doesn't seem to register changes. This prevented research into comparative time scales. RH: Readout is every 4 seconds? RR: Fastest time between glitches is 20 minutes. 4 sec should be good enough. This is time between resets. RH: The servo has wide bandwidth 500 Hz or so. This is to correct very small variations as it is closing the loop on an optical fringe. To some extent this is not a question for science, but it is not something other telescopes are trying to do. All others measure then apply corrections. RL: Not true for IRAM. We correct every second based upon the previous second. RH: Should we correct in software or adjust LO phase.

Polarization. B7 was not meeting spec, which was changed over my protests. Further investigation has been made. Some improvement can be made by changing mfg technique on IR filters. Also can put twists in waveguide. We will then have receivers which have different polzns, a complication in operation. We need to decide to ask IRAM to change design or not. Two Questions: How important is it to improve the polzn. This polzn is due to the design so should be constant. But this will mix with terms which do change. How important to go from around 18-19 dB in power terms to 25-26 dB. RR: If big crosspol then it affects I. RH: Yes, CB and George noted this. We think the breakpoint is in this region. We think if we get better than 20 dB total power would be OK. RH: Assume we endorse the change. For first few years we would have the hybrid situation. Can we live with this complication? RR: I have not worked this out for linear feeds but diversity in feeds could be a good thing. RH: This is true. All the interesting numbers come from the cross polzns. Does this outweigh. RL: I would argue against this based on VLBI. SM: You would have to get the bookkeeping correct. RH: Should all be changed? We would have to retrofit ones already built. SM: Significant off-axis structure? RH: Rx is only 100mm offaxis. If perfect, this structure is -45dB for an ideal telescope. Still in this range with gravity considered. Offaxis polzn will be dominated by the receiver. A direct picture with real receivers we do not have. We need to FT the beam pattern to see what the pattern is on the sky. SM: Will these be on an antenna soon? RH: Not before we have to make a decision. SM: Also an item on QW plates. RH: This is related but only weakly. The effect is to transform the polzn from linear to circular. This is good, stability of channels relatively speaking doesn't affect. We can use the mechanism to hold halfwave plates to improve with a hit on system temperature. SM: First do you want to do the conversion at all or live with a simpler straight out system. I would not worry about the interim. RH: A vote. How many. CV votes yes. Seems to carry the day.

FE pointing change request. May take some time before FE responds.

B10 review. Laing--groupd did an amazingly good job starting in Oct 05 as a small group doing a very difficult job. Unsurprisingly a few major technical areas. SIS junction fabrication. There are good results even considering state of the art CHAMP+ or HIFI. Concluded that none would meet ALMA requirements. They have some mixers which can meet spec but only over a small fraction of the band. They should adopt a two-pronged approach. Try higher current density approach. That has potential and will go ahead. We need to consider the necessity for tradeoffs of Trx and bandwidth. We recommend looking at this in one year's time. At that point Sci IPT and ASAC need to have their view together on noise BW tradeoff. We can get 8 GHz continuum bandwidth but we need to consider how important it is for the whole 787-950 GHz band, Technical on concerns--waveguide coupler may need to be replaced with quasioptical coupling.thought the group may be too small.

Will we end up with similar results on all rx or have various. RL: Hifi has flat response, by tuning each rx. One could align all resonances instead of filling in wholes.

MH has a good new design.

BN: Main change is to introduce n pads. RH: Lood at pages 2 and 3 of note. Ground does constraints. This is near the pipeline but it is believed this will be viable. MH will finish this work. We will get a smaller intermediate configuration, then we need to test the imaging properties of the various arrays. LT: Idea is to replace some pads and add additional ones. RH: Numbers are not in but we prefer to move pads. We might need a slightly higher number of pads than at present. RL: Looks like substantial progress.

Also pay for towers, installation.

RL: Simulations workshop. Probably be 8-9 Sept in Grenoble.

-- AlWootten - 16 Apr 2008