Apr 9, deadline for finalizing the author list and material
TeamX meeting
2nd Instrument meeting; Mission meeting
Overall cost is the the most important product
no significant revelation
Focal plane
converged to one configuration; most compact, proper strehl ratio for high frequency
the other two, we have to change the optical design, to accommodate with high frequency
this focal plane has largest number of detectors; the difference compared to other two is very small though.
Noise budget
V3.2: very little change from V3.1; noise 0.63 uK arcmin before, now 0.61 uK arcmin
compare values with CORE and LiteBird calculation, single detector
we are using two independent codes to calculate. Karl at UMN; Roger at JPL. Agree within 5%, slightly difference. Reasonable confidence.
various assumptions
to what extend our assumptions agree with other experiments
in some bands, noise lower than LIteBird, some bands higher than LiteBird.
everything seems consistent, no major change
Psat, safety factor of 2, we haven chosen a factor of 2, it’s used by JPL folks, also by SPIDER AND KECK. We thought it’s reasonable.
Quantitative calculations will be done by next week: what T of elements (mirros) could bring us to a factor of 2
Efficiency
It varies between the bands
Detector: lenselet + antenna + bolo, 70%
Edge Taper: center 10 db, lower band and higher band have different values; low, center, high: 70%, 90%, 99%.
End to end, 50-70%
Optical efficiency can mess up the estimate on safety factor
Emissivity
We are using measured emissivity from Planck. We already know, for highest bands, we may need to change to be more conservative
Margins
Space Mission, “current best estimate”, e.g. 100% yield; we are not likely to do better than 0.61 uK arkmin
What margin is the judge, successful or not.
Two paths:
1. start with assumptions
each one assumption
“worst case”
2. We can propagate backs from sigma_r 2*10^-5.
3. maybe a 3rd one, some factors on NET.
Kris: instrument is one thing, how to get data to science is another
Raphael used mapping speed, assuming delensing on our own resolution. 1*10^-4 is the limit that Shaul said in AAS, which has margins from calculated value 2*10^-5.
It's not a trivial calculation
LiteBird
does have margins
“large sections”, different categories, based on what's typically achieved in lab and by ground experiments
quote r with margin in it
nominal performance and worst performance. Give both to science team. Worst case did not give too much science. Lately, increase number of detectors to improve “worst case”
Shaul is going to check with Planck team
Workshop
suggestions:
Kris: 1) without modulator 2) arguments on margins