Telecon Notes Aug 8, 2018
Agenda
Notes
PICO imager telecon - 8/8/2018
Participants: Kris, Qi, Roger, Tom, Jeff, Jacques, Toki
PICO Focal plane: Finalizing mechanical design (Qi)
On Indium conductivity
slide 6 Indium conductivity value between literature value and Bill's number are different by factor of x500
slide 6 For 1.8e-5 and 1.44 e-7, units are W/cm K
slide 7 Thermal conductivity goes as 1/t^2 at low temp. Extrapolated to 100 mK using data points from Sladek
slide 7 Assuming thermal conductivity from electron only - Phonon contribution will only add
slide 8 Using Sladek's value, 10 mW of loading will change temperature by 6 mK
slide 9 Indium bump bond has lower thermal conductivity. 9.9e-7 W/cm @ 100 mK - This means 20 uW of loading will increase FP temperature by 8mK
slide 9 Jacques asked “Where do you have cooling power requirement?”, Qi answered “requirement not stated in this slide”
slide 9 Toki asked: “what is expected optical load on focal plane?”, Qi answered “I don't know” Karl may know?
slide 9 Tom “~20 uW at focal plane is reasonable”
slide 9 Roger - This (indium bump bond) will work okay. We need to work on how we explain it. Talk about strategy to build proto-type and test
slide 9 Tom - We could also design focal plane with ribbon bond to further mitigate this issue
slide 9 Kris - If the bonding was not enough there is room to put gold wires
slide 9 Roger - Through wafer via is used in PIPER. Are there any thermal problem? PIPER did lab testing. So far no major problem. Paper on this is not out yet. Test was done @ 100 mK
slide 9 Roger - We should point to PIPER as existing proof of working focal plane with similar configuration.
slide 9 Jacques - Are there write up on estimates on mass on cold stage & requirement on cooling power for the focal plane?
slide 9 Toki - There is mass estimate at end of Qi's slide
slide 9 Roger - JPL have thermal engineer working on this
slide 9 Jacques - We should write down Write mass and cooling power requirement of the coldest stage
On back termination
slide 10 Qi - Cavity does not need to be stepped. Idea is to make non-resonant cavity
slide 10 Toki - There is a paper on pyramidal backshort for sinuous antenna.
slide 10 Roger - We put absorber in back before. About 10% of radiation goes in backlobe. This is not the biggest issue.
slide 10 Qi - lambda in the slide is for central band. Cavity is electrically quite small for lower frequency band
slide 10 Toki - Simons Observatory is trying similar reflective backshort idea. Their backshort is ~1 mm away from antenna for 90/150
GHz slide 10 Toki - If we use absorber, we can then point to existing example. But this absorber has to be very thin.
slide 10 Qi - Via's height limit is 400 um~ 500um via conversation with Tom: Stack multiple layers with indium bump bond
slide 11 Qi - More indium bump bond can be added for heat sinking
On conical horn spacing
On focal plane mass
slide 16 Roger: Mass seems low
slide 16 Roger: Scaling from SPIDER seems heavier
slide 16 Jeff: SPIDER weights 15~20 pounds with magnetic shielding
slide 16 Jeff: Cover plate weights a lot, cut back on B-field shielding
slide 16 Roger: We are not sure if we can get rid of B-field
slide 16 Jeff one to one comparison to see where delta is
Roger's update - Verbal update
Technology maturing schedule - How to mature technology up to TRL 5 by Oct 2023
4 sections
LF detectors
Focusing on this section
Pointing to existing experiments - for ex. SPT-3G's three color pixel
Talk about other approaches BICEP/ ACT, giving overview of technology in the field
Systematic effects are setting r at 1e-3.
Hope that over next decade we will push this down to 1e-4
Experiment will need to be supported to push down systematics through actual observation
List of experiments to mature this technology before 2023
HFT
Readout
Interested in having Hannes help Roger on this
Conservative TDM option. Does not mean FDM won't work, but TDM is close to what we need (x64)
PICO need x128, explain how we make this increase
Thermal design
Roger not involved as much - ask Amy on where this is
ADR system is being considered - They work, but some works to iron out for ex. B-field
Continuous dilution fridge version maybe interesting, but doesn't exist
Roughly 3~4 pages point
We can make detectors to work, but demonstrating systematics in actual experiment is necessary to convince reviewers
Schedule seems OK
IDS = Inflation and Dust Surveyer