Notes on E0 tests, 29.5-27.6 2001, made on 28.1.2003 by T Lodge, D Cockerill

 Test 12, only 2m of HT cable, noise 10,000 - 12,000 electrons.

 Test 13 showed a level of electronics noise of 10,000 - 14,000 electrons
 using 150m of HV cable, and with a series resistor of 1Mohm at PSU end and 1kohm
 in series on the screen, and by improving HV card ground to cover of SC housing
 using braid on chs 5,10,15,20,25.

 This level of noise was reduced significantly (test 14), to 4,000 - 5,000 electrons
 by putting the 150m HT cable inside a screened box, connected to main star point
 (at the preamp chassis).
 

 Shortening the braid between the SC housing and the preamp chassis (the star point) further
 reduced the noise to 3,000 - 3,500 electrons (test 15).

 This work showed the importance of having a good solid earth between the housing and 
 preamp chassis. By screening the umbilical itself the noise was brought down to 
 1,300 - 2,000 electrons (test 16).

 This was further improved to 1,500 - 1,600 electrons (no notes on what changed, test 17).


 NB
 All noise figures measured on a single channel at a time on a white noise meter.
 
 NB
 With protection back to back transisistors/channel noise went from 500-600 up to 1300-1600
 (no umbilical attached). Add the internal signal lines from preamp to chassis D bulkhed,
 noise up from 500-600 to 600-800.

 NB
 50 single preamps cards. Grounding between them, and to the common from P/S - important noise 
 effects reducing from 3400 to 1500 electrons with optimal braiding.

 NB test 11 shows preamp gain dependance vs temp (put box in environmental oven).
 results 780mV at 18.0 degC to 870 at 66 degC. This corresponds to an 11.5% rise for +48 degC.
 Change is +0.24%/degC.