2023-01-18
Outline of today's measurements
h=1 h=2, test AWG script and if it works take beam current measurement
Repeat yesterdays's measurment with a flat top (2ms) with max enegry rev frequency of 3.91MHz
need new AWG script for this which has a max energy corresponding to 3.91MHz
Sweep through kicker voltages of 10,20,30 etc
Repeat at differnt energy
might need to move kicker?
10:28 accelerator start procedures
Initially using standard KURNS RF programme to verify normal operation.
Normal operation confirmed
Kicker voltage sweep to replicate yesterday's results
The kicker amplitude seems to be unstable for 30kV but shows normal behaviour for other numbers, unclear why. For the meantime we will use 31kV instead of 30kV and 0 , 5, 10, 20, 40, 50 and 55.
Take data with 100 MeV script with 2ms flattop, this AWG script as a top frequency of 3.89 ( ie the same as the measured rev frequency when we saw good side bands using the nominal KURNS AWG script)
11:22 RF offset set to 2149 micro seconds (same as yesterday)
Kicker trigger set to 28.850 micro seconds (aiming for middle of flat top to enable comparison of data before and after kicker firing)
11:32 SME (septum magnet) field could be affecting the beam. Some of the septum magnet field leaks so we turned it off to avoid the problem. (?)
Kicker voltage sweep: 10, 20, 31, 40, 50, 55kV. Results shown below
Data taken with 10kV kicker voltage. Sidebands visible in first and second harmonics (sensitive to windowing). Tune 0.1995309940142849
Data taken from 55kV kicker: tune 0.19815261978114412
13:50 New test series
Reduce kicker radius
Run nominal KURNS RF script and find frequency at kicker position
Develop new flattop RF for new frequency
Sweep kicker voltage
Aim: Maximise amplitude of beam oscillations to increase amplitude-dependent tune shift
14:00 Linac issues.
14:23 Linac issues corrected
determine kicker position by sweeping kicker timing, measuring beam loss, and looking at intersection on frequency analyser
New kicker position frequency 3.762MHz (radius reduced 1cm)
Generating new flattop script
New flat top script in use, rf offset changed to 2.153
Data below taken from 50kV kicker with new script. Tune could possibly indicate a tune increase due to increased amplitude...? However ~10% difference in energy should not produce as large an amplitude change as e.g. factor 2 in kicker voltage.
Beam loss is large for kicker voltage >10kV...
It's difficult to get data from this set of measurements -- unsure how to reduce beam loss in this case.
We think the beam loss is caused by very large kick from kicker magnet -- before we were not in 'sweet spot' of kicker. Now with new position relative to kicker the kick is too large and beam is lost? However, signal is very poor for
15:12
Started harmonic 1 vs harmonic 2 test
I remade the AWG scripts for the test so that the top energy is 84MeV. The septum is at the radius corresponding to 107 MeV. The other AWG scripts accelerate to ~120MeV which is too high.
The harmonic 2 script over loaded the RF amplifier so we reduced the length of the AWG script again such that the top energy was ~40 MeV which corresponds to 10ms of acceleration with phi_s=20.
The RF profile for the latest AWG script is shown below (h=2). There is a variation in the amplitude -- possibly due to the RF amplifier?
The RF profile for h=1 is:
As we were taking data for h=1 vs h=2 with the 10ms (40MeV) AWG script we notices spikes in the FAB monitor data even when the RF amplitude was reduced to zero. We will try to emlimiate the noise before taking data.
15:51 unsure of the source of the noise
Recorded FAB data for h=1, h=2
16:15 New test series: Schottky measurements
Thinking of ideas on how to improve ADTS measurements or harmonic 1, harmonic 2 measurements so proceed with more experiments while we try to come up with something
First test: Abrupt debunch from 4kV
Data taken (file #38)(CH3:RF, CH4:FAB)
Second test: Adiabatic debunch
Third test: 3kV abrupt debunch+Measurement of bunched beam for tomography
Fourth test: 3kV abrupt debunch
Fifth test: 0.9kV abrupt debunch
FAB monitor noise likely RF noise. Record data with beam on, beam off using Schottky script
noise looked much less significant in h=2 data, supports hypothesis of RF noise
Return to h=1, h=2 test
h=1 data is for some reason better (#47)
h=2 beam not captured
Potentially small signal at the start, but later signal is dominated by RF noise
Adjusting RF offset. New value 2.410ms
RF amplifier tripped
Capturing h=2 beam with the new offset but the h=2 does not survive for long
Took some data for the short time where the beam survived
Took data of h=1 and h=2 FAB signal with the beam off to measure the noise on the FAB monitor induced by the RF
Tomorrow plans:
Experiment:
Possibly determine 'sweet spot' of kicker/kicker profile measurement (?)
Try nominal rf programme, sweep kicker timing
Possibly see change in beam loss with changed kicker position, see if beam loss can be determined
Why is there no sidebands ?!?!
Amplitude correction for input pattern h=1, h=2
h=1, h=2 60MeV pattern
Analysis:
Reduce windowing on noisy datasets. try short window immediately after kick before beam 'settles'