Minutes of meeting held on 07 May 2020
==========================

slides are at:
https://beamdocs.fnal.gov/AD-public/DocDB/ShowDocument?docid=8365

(1) Jeff presented his local beta bumps idea
      (i) The idea is to create local beta bumps that impact one
      section of Booster without impacting the whole of Booster.
      (ii) Jeff used a simplified 24 identical period Booster for his
      analytic model.
      (iii) For a single QL error which maximally affects the vertical
      beta without strongly affecting the horizontal beta or the
      dispersion, the effects are:
         (a) A QL11 error affects all the vertical betas.
	 (b) The local bump consists of 5 QLs.
	 (c) Result is very local beta y bump with small effects on
      dispx outside the location of the bump.
      (iv) For a single QS error, effects are on beta x as well as dispx.
         (a) Interestingly, there are already dispersion bumps in
      Booster using 3 QSs.
	 (b) These dispersion bumps can correct the dispersion locally
	 (c) The effect on the horizontal beam size looks to be
      strongly dependent on dispersion and not beta_y!
         (d) Tan found it a surprising result because the beta beating
      will cause 1/2 integer resonances thus blowing up the beam near
      the resonance. Jeff says that the particle far away from the
      resonance should not be affected by the 1/2 integer.
         (e) Yuri says that adding dipoles into the dispersion
      correction may make bumps better.
      (iv) After beam comes back:
      	   (a) See whether losses can be reduced at sensitive
      loations.
           (b) When QL11=0, can it be compensated using adjacent QLs.

(2) Chandra gave a talk about the longitudinal beam dynamics, beam
loss and mitigation
     (i) Prior to 2016, injection was done at Bmin.
     (ii) After 2016, early injection was implemented.
     (iii) ESME simulations were done without external impedances like
     laminations, or RF impedances. Only impedance from space charge
     was included.
     	 (a) Chandra says the simulations reflect what is seen in real
	 life even without those impedances.
    (iv) Simulations showed that early injection shows no beam loss
    and only 10% emittance dilution.
    (v) TCLK improvements reduced jitter from +/-40us to +/- 5 us.
    (vi) RF phase errors between cavities were discovered and
    corrected.
    (vii) Digital paraphase capture will be cirtical to control
    paraphase errors.
    (viii) Present operations show that beam capture is non-adiabatic
    because of the neck.
    	    (a) Why is the neck needed? Yes for operations.
	    (b) Appearance of the neck was after LLRF upgrade after
	    2018 shutdown.
	    (c) Prior to upgrade, no neck was demonstrated to give efficient
	    capture.
   (ix) Suspicion is that the phase loop or the radial loop is yanking
   the beam and blowing it up.
   (x) Additional issues
   	(a) Injection energy jitter.
	(b) Frequency and energy mismatch.
	(c) Balance between A and B stations are still not optimal.
    (xi) Transition crossing. Present ops is due to bucket
    mismatch. Correction will be added to LLRF.
    (xii) Extraction. Snap bunch rotation causes bana shaped beam in
    longitudinal phase space.
    (xiii) Beam loading compensation is simulated in ESME. Results
    agrees with analytic calculations: Present ops can work without
    beam loading compensation. Beam loading compensation is required
    in PIPII.
    (xiv) Dampers are being upgraded for PIPII.
    (xv) Chandra has been studying coupled bunch modes using data that
    he has collected. The growth rate of each mode can be extracted
    from the data.
    (xvi) ESME simulations for 6.7e12 for PIPII shows that there is a
    large emittance dilution after transition
    	  (a) Tan is skeptical about this result because this problem
	  is not seen in Chandra's simulations with 6e12 for present
	  ops.
	  (b) Chandra will continue to investigate.