Attending: Brian, Julian, Shaul, Karl, Qi, Roger, Kris, Jacques

Agenda:

• optical system + noise model + focal plane options (Young, Wen)
  • 1.4 m Open Dragone Options
  • 1.2 m Crossed Dragone
• TM rates for large/small aperture systems (??)
• Scan angles
• Scan angles + full sky scan simulations + integration with systematics (Gorski, Delabrouille, + all)

Notes:

Open Dragone: optical system + noise model + focal plane options (Young)

• Size of pixel defined on lowest or middle frequency band.
  • middle band means smaller pixel, higher sensitivity
  • 10 dB edge taper + middle band size, overall the best in terms of mapping speed
  • lowest band has 4.8 dB edge taper
  • edge taper is at 4K stop
  • Using 10 dB at middle of band moving forward.

• For Open Dragone, the DLFOV is small. The outer area of the focal plane is not usable for high frequency, thus it’s hard to trade low -frequency detectors for more high-frequency detectors.
  • We don’t know what we need for low frequency, it depends on the synchrotron structures.
  • Plan is to optimize mirror shapes to increase ~150 GHZ DLFOV so trading low sensitivity for high sensitivity is an option.

• Open Dragone Optics
  • V3.D: focal plane further from stop, linearly 10% more focal plane compared to baseline
  • looks good, has room for mechanical structure of focal plane
  • Karl is working on coma correction to this telescope, and hopefully we will have larger DLFOV (AI)
  • V4: 10% less DLFOV compared to baseline. Overall smaller telescope, allows roughly 2-3 deg. increase in alpha.
  • sensitivity calculation uses edge taper defined as lowest band, can be done using middle band

Large-aperture Cross Dragone (Wen)

• Follow-up to matrix from previous week. Now we have a baffled, large cross dragone.
• Reminder: Why 50cm? De-scoped to save cost, if imager + spectrometer is emplemented.
  • Still questions as to whether just 140cm – 50 cm saves cost.
  • Idea here is to provide options to cost and be decided on by the EC.
• Cross Dragone has large DLFOV if not limited by vignetting (blockage, mirror sizes etc)
  • current case is limited by vignetting and blockage, not image quality.
  • unlike Open Dragone case, we can trade detectors at low frequency for high-frequency detectors more easily
  • ~ 3K detectors (Open) VS ~ 5K detectors (Cross)
  • F number: 1.5 (Open) VS 2.5 (Cross)
  • more detectors + larger F#, means Cross Dragone has larger focal plane

• We have filled the matrix of telescope categories
  • optics design is wrapping up. Polishing will be done to improve sensitivities.
  • the large/small, open/cross options provide input to trade-off decisions.

Scan (Kris)

• Simulation
• note WMAP Q/U map in V band; Probe with alpha = 22 deg, beta = 73 deg, is close to WMAP
• fast spin + different precession (fast, slow, very slow)
  • for T spin 20s (3 rpm) is fast, Kris do 1 rpm (AI)
  • slowest precession, more non-smooth features, higher resolution needed
• LiteBird & Core use same convention of alpha and beta, Kris will change his definition
• The simulations are the average over 1 year
• simulating over other different periods will be very useful, give more opportunities to look into the scanning.
  • Jacques: maps with no holes over week timescales are useful.
  • Kris will do two-week simulation (AI)

Scanning strategy

• Julian: 6 + 1 parameters; precession angle alpha, spin angle beta, 3 rotation rates (spin, precession, HWP), radius of L2 orbit, + data (or sample) rate.
• Jacques will lead to make table of scan drivers, wants, and evaluation metrics in wiki (AI)
  • Shaul to make table of hardware or engineering limits on 6 + 1 parameters (AI)
• Brian: there may not be many options of the L2 orbits
  • there could be, we need more information input from project
  • Jacques: Planck had radius L2 = 300,000 km, small. Needed 380 kg fuel for insertion.
  • Brian and Amy to check details of L2 orbit (AI)
• Amy told Shaul that we will probably have a steerable antenna for tele-communication; also fly wheels for pointing
• Kris: demonstration of full scan + systematics + map making is also important
  • Julian: Agree, the machinery exists to do these sims. They need to be in time domain. They're computationally cheap. Need limits on the input parameters and evaluation metrics.
• Shaul’s comments on this scan parameters study: 1) optimization of parameters; 2) systematics given parameters
• When mapping + noise correlation? Julian: As soon as we have the tables from Jacques, we will have a range of parameters.