Stony Brook Update Cross Roads T K Hemmick

Stony Brook Update Cross Roads T. K. Hemmick for himself 1

The Glovebox Water Spiked • Our intuition that the purifier was filling due to the long time with a heavy water load was correct. • The glovebox spiked to about 40 ppm before the windows could be applied over the HBD. • All GEMs put inside HBD along with two chicklets to monitor conditions. • We’ll use the chicklets to assess whether the other GEMs have lost QE. 2

No progress on making a GEMstack. • New production GEMs are completely different from last year: – No charge-up curve. – Requires nearly 3800 V for a gain of 5000. • After two full weeks of never sustaining 5000 gain for more than a few minutes, the bottom GEM finally began a current draw of 5 microamps at 500 V (completely stable at that point) and is now done. 3

Personal Opinion • We are wasting our time trying to build the HBD to the original design. • Even if we get a few stacks operating now they will fail during long term use. • Don’t use the “success” of last year’s voltage holding as evidence of anything. – The “success” was only after most strips had their resistor chain modified. – The loss of single strips compromises the performance of every blob (in arm or partner) that overlaps it. – A repeat of last year’s performance over the whole surface of the HBD will make a detector will poor performance. 4

Dust • I volunteer to wear a tee-shirt to every QM for 5 year stating that: – The original design was perfect. – Stony Brook is dusty. – The only failure ever was Stony Brook dust. – A perfect design that anyone else in the world should use was changed because of SB dust. • I will do this ONLY if we give up in the present design and make changes so that the detector will work. • If we make no changes I have zero confidence in the detector. 5

The proposed changes: • Make a 4 -GEM stack. • Use the body of the HBD as a field cage to get the bias in either forward or reverse. • Independently power each level with CAEN power supplies which internal logic drops the whole stack when any single channel trips. – 55 Fe source illuminates all pads. – Gain photo-electrons by losing grid. – All stacks will work easily at low gain/GEM. – Some loss of Blindness. – Need to make top surface of all stacks equal V. 6

Field Cage Potential (Volts) Field (Volts/cm) • The HBD is basically a cylinder. 7

Field Cage Preliminary Design • Match both end walls to the ln(r/r 0) profile. • Match the face to same profile. • Bingo…bias with uniform field HBD works. 8