Differences

This shows you the differences between two versions of the page.

--- imagerteleconnotes20171031 [2017/10/31 13:43] – kyoung
+++ imagerteleconnotes20171031 [2017/11/07 13:46] (current) – kyoung
@@ Line 1: / Line 1: @@
 ====== Telecon 20171031 ======
-Attending:
+Attending:  Brian, Shaul, Karl, Brendon, Qi, Joy, Al, Jamie
 __Agenda:__
-   * {{::optimizing_baseline_20171031.pdf|Higher throughput, lower noise optics and focal plane}} (Young)
+   * {{::optimizing_baseline_20171031.pdf|Higher throughput optics, lower noise focal plane}} (Young)
    * {{::probe_scanning_20171031.pdf|Scan optimization + simulations}}
+   * Orbit radius issues: {{::screen_shot_2017-10-31_at_1.50.35_pm.png?50}}
    * Noise + systematics simulations
 === Notes ===
+Higher throughput, lower noise optics and focal plane (Young)
+  * correcting optics for coma
+  * get to 7350 bolos or 15030 bolos (depends on pixel size and edge taper)
+  * pixel size set by middle band or lowest band.
+    * middle band gives smaller pixels, worse spillover efficiency.
+    * Brendon: far sidelobes a systematic concern.  Can we estimate what these might be?
+  * alternative pixel band structure (slide 5)
+    * reduces spillover variance within a pixel.  Not clear this is a major problem.
+    * NET penalty, narrow bands to avoid overlap.
+      * Jamie: not clear that overlap is a problem.  As long as band shapes are known.
+    * Jamie: what does single band pixels look like?  Fewer broad bands.
+      * **A/I** Karl to check.
+  * Technical or science issues with alternative band structure?
+    * nothing from people currently.
+Scan optimization + simulations (Kris presentation by Shaul)
+  * single detector, alpha = 22, beta = 75
+  * was at 3 rpm spin, now 1 rpm spin, varying precessions
+  * suggests precession < 10 hrs
+  * alpha, beta dictate large scale features
+  * Zoom on N_obs panel, shows striping at < 12 hours.
+  * Jamie: pixelization effects? are there gaps in scan on small scales?  Shaul: plan to do detailed scans next, w/Julian.
+  * rings on sky in 1 day, full maps per detector in 6 months.
+  * Jamie: thinks no resource problems with fast precession.
+    * Still need check with Amy.
+  * Aside: Brendon, adding reaction wheels to systematics. Can have vibrations.
+    * Jamie: Can have 'momentum wheel' to balance total spacecraft momentum and take load off reaction wheels.
+Systematics, simulations, noise:
+  * How to combine systematics and imager work?
+  * Brendon: Systematics WG has list of systematics and risk level for each.  Goal to prioritize systematics.
+    * table on wiki. SRF rating of 5 is worst.
+    * Most worried:  Far sidelobes, gain stability, pol angle calibration.
+      * combination of we know least about it, drives instrument design, or limited past experiments
+    * others also, but more confident they can be dealt with.
+    * WG has found simulation capability, e.g. TOAST.
+      * Shaul: TOAST was plan to simulate noise + scan.
+      * systematics can be added here as well.  Work done for LITEBird, CORE.
+      * example: adding far sidelobes is straightforward
+      * T-->P leakage sim by CORE.
+        * QuickPol looks at main beam mismatch.
+        * bandpass mismatch (Ranajoy)
+        * in CORE systematics paper.  They are able and interested.
+    * Joy: many simulations give you large values for systematics, but they will be partially removed by analysis.  This needs to be considered.  Brendon: Yes, TOAST has this type of analysis tool built in.  Needs to be kept in mind.
+  * Jamie: Good approach. Don't worry about everything, priortize based on Planck, CORE, LITEBird work.
+  * Shaul: Goal, get inputs systematics group needs.
+    * far sidelobes, working on input via Brad and Amy
+    * polarization calibration inaccuracies:  Initial calibration.  Errors on calibration. Dependence on spectrum used to calibrate, or dependence on source observed
+      * no solid calibration plan yet, so key errors not yet known or quantified. (not ready for simulation)
+    * gain stability: ability to continuously measure gain.  4 yr calibration on dipole is great. But can't see dipole constantly (scan strategy input). CORE folks have worked on this, Tomassi volunteered but not until January.
+      * Joy: need instrument inputs on gain stabiliy (bath, temperature variation, etc.)
+        * can simulate observation, calibration on dipole with scan, quantify how well this can be done.
+      * Jamie: can this simulate deprojection? It is very useful.  Brendon: Planck had dipole S/N > degree scale S/N, so not useful.  Could be better this time around.
+         * not clear this capability exists.
+  * Brendon, Shaul, Joy to meet this week and coordinate on systematics plan.

User Tools

Differences

Page Tools