ASIC design optimizations

From GlueXWiki
Jump to: navigation, search

Traces of cosmic rays in coincidence with the scintillators

Here's how the trigger's formed:

Trigger formation.


Here are some infinite persistence traces:

Traces of straight cosmics.
Traces of straight cosmics.


Here are some infinite persistence traces:

Traces of straight cosmics.
Traces of straight cosmics.
Traces of straight cosmics.

Fe-55 measurements

Cross talk

To have a quick look concerning cross talk let's show some scatter plots:

Fe-55 peak amplitude spectrum (1800 V - begin - upper ).
The straw next to it.
The straw underneath it.
The electronic channel next to it.
The other electronic channel next to it.

There is some electronic cross talk.

Raw fADC spectra

Raw fADC spectrum (1800 V - end - upper ).
Raw fADC spectrum (1800 V - end - upper ).
Raw fADC spectrum (1800 V - end - upper ).

ADC-channel versus time(ns)

Upper means the straw that does not need extra support.

"Energy" spectra

"Energy" spectrum (1750 V - begin - upper ).
"Energy" spectrum (1800 V - begin - upper ).
"Energy" spectrum (1850 V - begin - upper ).

High statistics run:

"Energy" spectrum (1800 V - begin - upper ).


x-axis is in ADC channels

This measurements correspond with the three measurements with different HV on the straw that does not need extra support.

Different HV (Straw supported):

Amplitude spectrum measured with the 55-Fe source (1750 V - begin).
Amplitude spectrum measured with the 55-Fe source (1800 V - begin).

Different positions (straw supported,1800V):

Amplitude spectrum measured with the 55-Fe source (1800 V - begin).
Amplitude spectrum measured with the 55-Fe source (1800 V - middle).
Amplitude spectrum measured with the 55-Fe source (1800 V - end).


Different HV (Straw not supported)

Amplitude spectrum measured with the 55-Fe source (1700 V - begin).
Amplitude spectrum measured with the 55-Fe source (1800 V - begin).
Amplitude spectrum measured with the 55-Fe source (1850 V - begin).


Are the measurements reproduce-able?

Amplitude spectrum measured with the 55-Fe source (1800 V - begin - supported).

supported/not supported (middle) + TILTED (55 degrees)

Amplitude spectrum measured with the 55-Fe source (1800 V - middle - supported).
Amplitude spectrum measured with the 55-Fe source (1800 V - middle - not supported).
Amplitude spectrum measured with the 55-Fe source (1800 V - middle - 55 degree tilted).


What about measuring another straw (one that does not need support - and does not have dead spots): Different positions (1800V):

Amplitude spectrum measured with the 55-Fe source (1800 V - begin).
Amplitude spectrum measured with the 55-Fe source (1800 V - middle).
Amplitude spectrum measured with the 55-Fe source (1800 V - end).

Reproduce-able?:

Amplitude spectrum measured with the 55-Fe source (1800 V - begin).

Different-HV:

Amplitude spectrum measured with the 55-Fe source (1750 V - begin).
Amplitude spectrum measured with the 55-Fe source (1800 V - begin).
Amplitude spectrum measured with the 55-Fe source (1850 V - begin).


To finish a high statistics run:

Amplitude spectrum measured with the 55-Fe source (1800 V - begin - high statistics).

Chamber at 0 degrees

  • Measurements with the 55-Fe source
Amplitude spectrum measured with the 55-Fe source (1750 V - low threshold).
Amplitude spectrum measured with the 55-Fe source (1800 V - low threshold).
Amplitude spectrum measured with the 55-Fe source (high threshold - 1800 V).


  • High Voltage plateau === done

Gas: Ar-CO2 87%-13% gas mixture. Setup: the signal after the shaper was inverted and fed to a discriminator. The HV was changed and the signals above threshold where counted (normalized to 100s measure time). Update: results with 3,4,5 sigma pedestal cuts:

HV-curve - 3sigma cut, tracked events, normalized to a 40' measurements.
HV-curve - 4sigma cut, tracked events, normalized to a 40' measurement.
HV-curve - 5sigma cut, tracked events, normalized to a 40' measurement.

The pedestal means (the error bar is the pedestal width):

HV-noise-curve - Pedestal mean as a function of applied HV, the error bars represent the pedestal width



The result:

HV-curve,the count is normalized to 100s measurements.
HV-curve-plateau, tracked events, normalized to a 40' measurement.

So things start to show up (above threshold, the threshold was set just above noise) at 1500 V and start to break down at +2000V. We see a plateau starting at 1750V. All hv-plateaus are ahown in following figure:

All hv-plateaus.


Next the scope traces of the (maximum) noise level after the shaper is shown on the left side. This pickup happens mostly on a different time then a signal. The right side shows a signal at 1800V, also after the shaper. To trigger on the signal I set the threshold to 350 mV.

Trace of maximum noise after shaper.
Trace of a signal at 1800 V after the shaper.

Let's compare signals at different stages in the electronics chain. Left plot is the signal right BEFORE the pre-amplifier, middle plot is the signal AFTER the pre-amplifier (0.366 times attenuated because of the differential probe), right plot is AFTER the shaper.

    • The noise
Noise before preAmp is dominated by the noise pickup of the probe.
Noise after preAmp.
Noise after shaper.


    • 1600V
N/A because of noisy probe.
Trace of signal after preAmp of channel 2.
Gallery of pulses of channel 2.
    • 1700V
N/A because of noisy probe.
Trace of signal after preAmp of channel 2.
Gallery of pulses of channel 2.
    • 1750V
Trace of signal before preAmp of channel 2.
Trace of signal after preAmp of channel 2.
Gallery of pulses of channel 2.
    • 1800V
Trace of signal before preAmp of channel 2.
Trace of signal after preAmp of channel 2.
Gallery of pulses of channel 2.
    • 1900V
Trace of signal before preAmp of channel 2.
Trace of signal after preAmp of channel 2.
Gallery of pulses of channel 2.
    • 1950V
Trace of signal before preAmp of channel 2.
Trace of signal after preAmp of channel 2.
Gallery of pulses of channel 2.
    • 2000V
Trace of signal before preAmp of channel 2.
Trace of signal after preAmp of channel 2.
Gallery of pulses of channel 2.


What exactly does the shaper? Well you can see this in following traces: from left to right: three signals before/after (green/red) the shaper @ 1800 V: (remember that the signal before the shaper is attenuated (0.336x) )

Trace of signal before/after (green/red) shaper.
Trace of signal before/after (green/red) shaper.
Trace of signal before/after (green/red) shaper.

The same measurement was performed but now with two probes measuring the differential input signal and MATH function on the scope. Here are three scope traces:

Trace of signal before/after (purple/black) shaper.
Trace of signal before/after (purple/black) shaper.
Trace of signal before/after (purple/black) shaper.



Below galleries of pulses are shown for different HV (on the left channel 2, in the middle channel 15) -- remember everything above 1V is an overflow on the fADC, also the b/w traces are taken with infinite persistence, the colored ones with 10 s persistence. The amplitude spectra are also shown: notice the growing overflow peak around channel 3000 (they are pedestal subtracted):

    • 1600 V
Gallery of pulses of channel 2.
Gallery of pulses of channel 15.
Amplitude spectrum of ch 15 (counts versus amplitude in ADC channels, with shaper).
    • 1650 V
Gallery of pulses of channel 2.
Gallery of pulses of channel 15.
Amplitude spectrum of ch 15 (counts versus amplitude in ADC channels, with shaper).
    • 1700 V
Gallery of pulses of channel 2.
Gallery of pulses of channel 15.
Amplitude spectrum of ch 15 (counts versus amplitude in ADC channels, with shaper).
    • 1750 V
Gallery of pulses of channel 2.
Gallery of pulses of channel 15.
Amplitude spectrum of ch 15 (counts versus amplitude in ADC channels, with shaper).
    • 1800 V
Gallery of pulses of channel 2.
Gallery of pulses of channel 15.
Amplitude spectrum of ch 15 (counts versus amplitude in ADC channels, with shaper).
    • 1850 V
Gallery of pulses of channel 2.
Gallery of pulses of channel 15.
Amplitude spectrum of ch 15 (counts versus amplitude in ADC channels, with shaper).
    • 1900 V
Gallery of pulses of channel 2.
Gallery of pulses of channel 15.
Amplitude spectrum of ch 15 (counts versus amplitude in ADC channels, with shaper).
    • 1950 V
Gallery of pulses of channel 2.
Gallery of pulses of channel 15.
Amplitude spectrum of ch 15 (counts versus amplitude in ADC channels, with shaper).
    • 2000 V
Gallery of pulses of channel 2.
Gallery of pulses of channel 15.
Amplitude spectrum of ch 15 (counts versus amplitude in ADC channels, with shaper).

Chamber at 60 degrees

Tilted.jpg

Below you can find some pulse galleries and amplitude spectra (low statistics)


    • 1600 V
Gallery of pulses of channel 2.
Gallery of pulses of channel 15.
Amplitude spectrum of ch 15 (counts versus amplitude in ADC channels, with shaper).
    • 1650 V
Gallery of pulses of channel 2.
Gallery of pulses of channel 15.
Amplitude spectrum of ch 15 (counts versus amplitude in ADC channels, with shaper).
    • 1700 V
Gallery of pulses of channel 2.
Gallery of pulses of channel 15.
Amplitude spectrum of ch 15 (counts versus amplitude in ADC channels, with shaper).
    • 1750 V
Gallery of pulses of channel 2.
Gallery of pulses of channel 15.
Amplitude spectrum of ch 15 (counts versus amplitude in ADC channels, with shaper).
    • 1800 V
Gallery of pulses of channel 2.
Gallery of pulses of channel 15.
Amplitude spectrum of ch 15 (counts versus amplitude in ADC channels, with shaper).
    • 1850 V
Gallery of pulses of channel 2.
Gallery of pulses of channel 15.
Amplitude spectrum of ch 15 (counts versus amplitude in ADC channels, with shaper).
    • 1900 V
Gallery of pulses of channel 2.
Gallery of pulses of channel 15.
Amplitude spectrum of ch 15 (counts versus amplitude in ADC channels, with shaper).
    • 1950 V
Gallery of pulses of channel 2.
Gallery of pulses of channel 15.
Amplitude spectrum of ch 15 (counts versus amplitude in ADC channels, with shaper).
    • 2000 V
Gallery of pulses of channel 2.
Gallery of pulses of channel 15.
Amplitude spectrum of ch 15 (counts versus amplitude in ADC channels, with shaper).

To Do

  • (absolute) Gas gain as a function of High Voltage

Source measurement using Sr-90 (e-) or 55-Fe (X-rays) (if we have it) / or using cosmic rays? -- Estimated time: 1 day -- if we have a good source


  • amplitude distributions of tilted chamber

1 week measure with tilted chamber => In progress


  • Collection of gallery of pulses

I found 4 pulse galleries, measured directly after the pre-amplifier made by Gerard when he was here, they are shown below:

Gallery of pulses measured after the pre-amplifier
Gallery of pulses measured after the pre-amplifier
Gallery of pulses measured after the pre-amplifier
Gallery of pulses measured after the pre-amplifier