User name R. Michaels
Log entry time 09:29:07 on August 9,2005
Entry number 151245
Followups:
keyword=cavities du jour (current scan)Last night there was a test of the beam current ramp, run 4224.
Here are some cavity monitor results. Cavities allegedly work
since Thurs Aug 4 at 17:00, so let's confront this claim.
Fig 1
BCMCAV2 and BCMCAV3 vs BCM1 -- looks pretty good.
Some evidence of a nonlinearity at low currents, but this is expected.
Freyberger says the cavities need automatic gain adjustment. His group
will implement this, but in the meantime the settings are only good
around our operating current (ca 35 uA).
Fig 2. Position correlations during the same run.
Cuts on bcm1>8000 to ensure the beam is high (reason: see above).
If I don't make this cut the results are not as pretty.
A) BPMCAV2X vs BPM4AX looks well correlated, but...
B) BPMCAV2X vx BPM4BX is not well correlated. I'm not as worried
about this now, see comments below. BTW, this implies that BPM4BX
is not well correlated to BPM4AX (stripline to stripline) and
nobody is worried about that, right ?
C) BPMCAV2Y vs BPM4AY looks sort-of correlated.
D) BPMCAV2Y vs BPM4BY is more correlated than plot C).
The above 4 plots were concentrating on CAV2 positions.
Now, running out of room in my plot, I plot the 2 prettiest
correlations for CAV3:
E) BPMCAV3X vs BPM4AX
F) BPMCAV3Y vs BPM4BY
Discussion: Between each monitor the beam is changing a lot
(need for a big phase advance). What's more, the beam is moving
rapidly, and there are electronic phases between devices which are
a significant fraction of the raster period. Finally, each device
has its own integration time. I think that combining these effects
cause the imperfect correlations we see, and it is not a problem.
To test this, we could do a "bulls eye" scan, where the raster is
nominally off and the beam is moved slowly and parked in different
spots. Regarding the need to have raster-off data, my understanding
is that this would still be useful to fine tune the phase zero
offset, but the present value used should be pretty close.
As for the info I had Friday about phases being unstable, this was
explained in
halog 150700.
According to Musson this does not affect our measurements.
It makes me worried but they say we should be ok. We'll see.
FIGURE 1
FIGURE 2
