Difference between revisions of "CDC transfer func attempt"

From GlueXWiki
Jump to: navigation, search
m (Text replacement - "http://argus.phys.uregina.ca/cgi-bin/private" to "https://halldweb.jlab.org/doc-private")
 
(39 intermediate revisions by one other user not shown)
Line 1: Line 1:
 
{| border="0" cellpadding="2"  
 
{| border="0" cellpadding="2"  
−
|[[Image:pulser_2mV_input_single.jpg|thumb|x250px|2mV leading edge of sq wave step up (yellow) and trigger from pulser (green) - single ]]
+
|[[Image:cdctransfnsetup1.jpg|thumb|x250px|Setup ]]
−
|[[Image:pulser_2mV_input_many.jpg|thumb|x250px|2mV leading edge of sq wave step up (yellow) and trigger from pulser (green) - multiple]]
+
|[[Image:cdctransfnsetup2.jpg|thumb|x250px|Setup (GS 4001)]]
−
|[[Image:pulser_2mV_input_trigger.jpg|thumb|x250px|2mV leading edge of sq wave step up (yellow) and trigger from pulser converted to ECL (purple) - multiple ]]
+
 
|}
 
|}
 +
 +
<h4>Circuit</h4>
 +
{| border="0" cellpadding="2"
 +
|[[Image:cdc_charge_inj.png|thumb|x100px|Charge injection circuit from Fernando's [https://halldweb.jlab.org/doc-private/DocDB/ShowDocument?docid=1364 GlueX-doc-1364] ]]
 +
|}
 +
 +
Actual component values used were... 51.1 Ohms, 908 Ohms and 2pF.
 +
 +
The aim is to inject a charge impulse into the electronics chain and measure its response.
 +
 +
<h4>Early results</h4>
 +
{| border="0" cellpadding="2"
 +
|- Repeated, minus a few attenuators (used attn x5 to go from 0.5V to 0.1V)
 +
|[[Image:pulser_100mV_input_av.jpg|thumb|x250px|100mV leading edge of sq wave step up (yellow) and trigger from pulser (purple) - average ]]
 +
|}
 +
{| border="0" cellpadding="2"
 +
|[[Image:shaper_output.jpg|thumb|x250px|output of shaper - average ]]
 +
|[[Image:shaper_output_z.jpg|thumb|x250px|output of shaper - expanded time scale ]]
 +
|-
 +
 +
|[[Image:fadc_output_average_new.png|thumb|x250px|output of fadc - average ]]
 +
|[[Image:fadc_output_average_new_z.png|thumb|x250px|output of fadc - expanded time scale ]]
 +
|}
 +
 +
<h4>Final results</h4>
 +
Borrowed Phillips 5771 (no attenuators needed)
 +
{|border="0" cellpadding="2"
 +
|- Input pulse (100mV square wave)
 +
|[[Image:pulse_single.jpg|thumb|x250px|pulser output - single ]]
 +
|}
 +
{|border="0" cellpadding="2"
 +
|- Input pulse (100mV square wave)
 +
|[[Image:pulse_av.jpg|thumb|x250px|pulser output - average ]]
 +
|[[Image:pulse_av2.jpg|thumb|x250px|pulser output - average ]]
 +
|}
 +
 +
 +
{| border="0" cellpadding="2"
 +
|[[Image:pm5771_shaper_single.jpg|thumb|x250px|output of shaper - single ]]
 +
|}
 +
 +
{| border="0" cellpadding="2"
 +
|[[Image:pm5771_shaper_av.jpg|thumb|x250px|output of shaper - average ]]
 +
|[[Image:pm5771_shaper_av2.jpg|thumb|x250px|output of shaper - average]]
 +
|}
 +
 +
{| border="0" cellpadding="2"
 +
|[[Image:run_31491_adc.png|thumb|x250px|fADC data ]]
 +
|[[Image:run_31491_upsampledadc.png|thumb|x250px|upsampled fADC data]]
 +
|}
 +
 +
<h4>Fitted function</h4>
 +
{| border="0" cellpadding="2"
 +
|[[Image:run_31491_meanupsampledadc.png|thumb|x250px|mean of upsampled fADC data]]
 +
|[[Image:run_31491_fit.png|thumb|x250px|fitted function]]
 +
|[[Image:run_31941_fit2.png|thumb|x250px|blue=refit with powers fixed, pink= pars 137, 118, 8, 91.4, 9]]
 +
|}
 +
 +
Beware typos in filenames, this was run #31941!
 +
 +
 +
Input charge: 2 pF x 103 mV = 206 fC
 +
 +
Input current: divide by 12ns peaking time of preamp: 206 fC / 12 ns = 17.2 uA
 +
 +
 +
Peak value of output data is 635 adc_units ~ 635 * 495 mV / 4096 = 76.7 mV
 +
 +
Electronics gain is 76.7 mV / 206 fC = 0.37 mV/fC 
 +
 +
(Gain of preamp alone measured earlier (before modifications) by FJB as 0.57 mV/fC)
 +
 +
 +
Convert to current using 50 Ohm resistance: 76.7 mV / 50 Ohm = 1.53 mA
 +
 +
 +
Current gain is 1.53 mA / 17.2 uA = 89.0
 +
 +
 +
Integrated value of output data is 19.03 adc_units.us ~ 19.03 * 495 mV / 4096 = 2.30 mV.us
 +
 +
Convert to charge using 50 Ohm resistance: 2.3 mV.us / 50 Ohm = 0.046 mA.us = 0.046 nC = 46 pC
 +
 +
 +
Charge gain is 46 pC / 206 fC = 223
 +
 +
 +
To convert histograms of summed ADC data (with 8ns sample time) to charge, multiply by 8 ns * 46 pC / 19.03 us = 19.3 fC
 +
 +
 +
Garfield uses uA for signals, but I want to see the resulting histograms in adc units, for direct comparison with cosmics data.  So... I need to convert my impulse response to input of 1 uA, ie use (1 / 17.2) * fitted function above as the transfer function.??  (Fitted function already includes current gain x 89).
 +
 +
The remaining gain factor needed to match garfield adc data to real life data is the gas gain.

Latest revision as of 16:48, 24 February 2017

Setup
Setup (GS 4001)

Circuit

Charge injection circuit from Fernando's GlueX-doc-1364

Actual component values used were... 51.1 Ohms, 908 Ohms and 2pF.

The aim is to inject a charge impulse into the electronics chain and measure its response.

Early results

100mV leading edge of sq wave step up (yellow) and trigger from pulser (purple) - average
output of shaper - average
output of shaper - expanded time scale
output of fadc - average
output of fadc - expanded time scale

Final results

Borrowed Phillips 5771 (no attenuators needed)

pulser output - single
pulser output - average
pulser output - average


output of shaper - single
output of shaper - average
output of shaper - average
fADC data
upsampled fADC data

Fitted function

mean of upsampled fADC data
fitted function
blue=refit with powers fixed, pink= pars 137, 118, 8, 91.4, 9

Beware typos in filenames, this was run #31941!


Input charge: 2 pF x 103 mV = 206 fC

Input current: divide by 12ns peaking time of preamp: 206 fC / 12 ns = 17.2 uA


Peak value of output data is 635 adc_units ~ 635 * 495 mV / 4096 = 76.7 mV

Electronics gain is 76.7 mV / 206 fC = 0.37 mV/fC

(Gain of preamp alone measured earlier (before modifications) by FJB as 0.57 mV/fC)


Convert to current using 50 Ohm resistance: 76.7 mV / 50 Ohm = 1.53 mA


Current gain is 1.53 mA / 17.2 uA = 89.0


Integrated value of output data is 19.03 adc_units.us ~ 19.03 * 495 mV / 4096 = 2.30 mV.us

Convert to charge using 50 Ohm resistance: 2.3 mV.us / 50 Ohm = 0.046 mA.us = 0.046 nC = 46 pC


Charge gain is 46 pC / 206 fC = 223


To convert histograms of summed ADC data (with 8ns sample time) to charge, multiply by 8 ns * 46 pC / 19.03 us = 19.3 fC


Garfield uses uA for signals, but I want to see the resulting histograms in adc units, for direct comparison with cosmics data. So... I need to convert my impulse response to input of 1 uA, ie use (1 / 17.2) * fitted function above as the transfer function.?? (Fitted function already includes current gain x 89).

The remaining gain factor needed to match garfield adc data to real life data is the gas gain.

Retrieved from "https://halldweb.jlab.org/wiki/index.php?title=CDC_transfer_func_attempt&oldid=80234"