HAPPEX DAQ / ADC Tests
Luis MercadoUMASS4/17/09
Full DAQ System
Crate currently used in electronics lab for testing.
Plan to move crate into the Hall in early June.
Left Arm Detectors.
Will house Luminosity Monitor connections, Cavity Monitors and Right Arm Detectors.
Proper Lumi-ADC setup was tested Jan’08 to achieve minimum pedestal noise.
Needs to be revived. Will have different
configurations with and without Qweak Crate to facilitate standalone tests.
Should we only use Helicity signals in this crate like during HAPPEX run?
Recently underwent a test with 1 kHz MPS sequence.
DAQ starts to breaks down at 3 kHz.
Concluded we can run at 2 kHz, or 200 Hz with ovs of 10, or 240 Hz with ovs of 8.
Maybe higher but needs to be investigated.
Current Plans
Revive synchronization checks. Replace many 16-bit ADCs with
new 18-bit boards. Basic DAQ should be ready for testing
by mid-July.
Future Tests
High rate deadtime. Checks of new helicity sequence
board. Synch checks. Pedestal Studies (helicity-correlated
noise).
Latest 18-bit ADC Tests
New 18-bit ADCs
Received four brand new boards a couple of weeks ago.
Performance matches previous tests with prototypes.
Will be receiving 10 more new boards by June 1st.
Testing of 18-bit boards.
Setting and recording proper pedestals offsets.
Studying pedestal noise with respect to gain setting (as low as 12ppm).
Testing linearity of current mode for use with Luminosity Monitors.
Comparing results with previous study done with 16-bit ADCs.
Integration Time (IT) Study
Have had problems in the past with gross nonlinearities at high end of ADC range.
Integration Time (IT) Study
Redid test using new 18-bit ADCs, with some changes: Sample more IT values by taking smaller
steps up the IT ladder. More data points for higher IT values to
get better error bar. Go up to 230-240K signal to see if it gets
worse at the higher end.
Integration Time (IT) Study
• Residuals areless than 0.1%
• Local slopes vary by ~0.4%
Integration Time (IT) Study
• Residuals areless than 0.1%
• Local slopes vary by ~0.4%
PMT Linearity
Past tests have shown nonlinearities in PMT output dependant on input and gain.
Want to make sure we get similar results with new 18-bit ADCs.
Showed that results matched in good and bad linearity regions.
PMT Linearity (Formalism)
A PMT’s non-linear response can be expressed as follows:
Experimental asymmetry can be approximated:
NNNPMT 1
0exp 1 NANN
NNA true
PMTPMT
PMTPMT
Experimental Test Setup
PMT Linearity (w. 16bit ADC)
Input = 12nA
HV = 370
Output = uA
B*N_0 = -5.25%
PMT Linearity (w. 18-bit ADC)
Input = 12nA
HV = 370
Output = uA
B*N_0 = -5.14%
Good PMT Linearity
Input = 2nA
HV = 727
Output = uA
B*N_0 = 0.7%
Good PMT Linearity
Input = 2nA
HV = 727
Output = uA
B*N_0 = 0.77%
Ongoing Developments
Revive all HAPPEX DAQ crates and test their performance.
Will need to modify 18-bit ADCs to match replaced 16-bit boards.
Want to document good linearity regions for all available LUMI PMTs.
Acknowledgements
Krishna Kumar Kent Pashke Dustin McNulty Robert Michaels HAPPEX/PREX Collaboration