08 & 09 Sep 2020
============

Eduards talk:

Simulations were originally done with no SC to Booster. It's also hard
to imagine that with the low charge per bunch would give a different
result with SC. However, later simulations showed that SC has an
effect on dp/p and thus bunch length.

Then Francois worked sim with SC. There is a little bit of beating and
line needs to be retuned. dp/p is a problem with SC. 2e-4 (0.02%)
energy spread without SC.  In the linac, SC was always in the linac
sim. Meiqin has never been told to sim with SC in the BTL
lattice. Full 3D SC model.  SC increases dp/p to 0.04% Bunch length
grows. to 40 mm at the end of the BTL. Spread as dp/p.  (RMS values
quoted here.)

Requirements Do not want energy spread at all.  dp/p Errors > 1e-4
causes losses in original injection scheme.  On energy injection has
higher acceptance. dp/p of 4e-4 has no tolerance for errors.

Need sims to simulate injection with SC.

Mitigation:
RF gymnastics in linac -- can be difficult wihout sacrificing quality
and energy gain.

Bunching cavity: use 325 MHz  or 650 MHz cavity. (any harmonic of
162.5 MHz)

Where to put cavity. End of Linac, Bunch is short. Not much
correlation between energy and length. Doesn't do much. Really need
the bunch to drift. Possible locations are before the bend and after
the bend.

Voltage required for the RF gymnastics.  Lower
voltage with 650 MHz and lower further downstream.

For the future, it would be beneficial to have the buncher cavity
further down stream because if we go to 1 GeV, there will not be
enough space for the cavity.

energy spread is minimal after compensation.

There are several cavity choices. SSR2 (325 MHz), PIPII HB (650 MHz),
or copper JPARC ACS buncher (CCL structure). (800 MeV is about beta
=0.85). Cryo may need to be extended if installed in 50 m. R&D NbTin
coated cavities Cryo cooler at downstream locations.

Cryogenics near Booster should not be a problem. Both machines are
connected by a sewer pipe. No direct personnel access.

Residual energy spread is not quite linear because there are tails. A
simple estimation assumping (dp)height*dz = constant. Assume initial
dp/p wth SC is 4e-4 with RF cavity compensation.  RF cavity near the
end, dp/p can be < 0.4e-4.

Buncher is needed but not part of baseline. It has to be seriously
addressed because of losses.

Energy control plan is to use linac cavity or set of linac
cavities. Changing voltage in cavity to change energy.  Control loop
30 to 50 kHz. Energy and power problem for fast corrections. A few
degrees of phase corection is possible. If changes are small, it can
be quite fast.

 TOF as part of Linac energy feedback loop is being discussed.

How does all these energy/pahse corrections interfere with injection
painting?

Right now, it seems that 550 us beam pulse cannot be corrected. Only
20 Hz pulses can be corrected.