Wakefield and Beam Centering Measurements of a Damped and Detuned X-Band Accelerator Structure Page: 3 of 4
This report is part of the collection entitled: Office of Scientific & Technical Information Technical Reports and was provided to Digital Library by the UNT Libraries Government Documents Department.
The following text was automatically extracted from the image on this page using optical character recognition software:
each step, the drive beam position was moved parallel
to the structure over a +/- 2 mm range and the result-
ing deflections of the witness beam recorded. The
deflection angle per unit drive bunch offset was then
computed and converted to a wakefield strength.
Finally, the wakefield amplitude was obtained by
fitting the wakefield strengths to a sine function with a
frequency equal to the mean value for the lowest
dipole band (15.1 GHz).
The results from 35 such measurements over a 200 ns
range are shown in Fig. 2. They are plotted versus the
square-root of the relative beam time so values at short
times are more discernible. Although the initial falloff
of the wakefield is large, it is expected to stay below
about 0.5 V/pC/m/mm out to 20 ns, instead of rising
to values near 4 V/pC/m/mm as is observed. This
difference is likely the result of the cell fabrication
errors being larger than the design tolerances. The
effect of such errors are included in the predictions in
Fig. 1, which were obtained from an equivalent circuit
model of the structure . In the top plot, the errors are
based on limited cell QC measurements made prior to
the structure assembly. They are essentially equivalent
to random cell-to-cell, 5 MHz rms, Gaussian frequency
errors, and are larger than the - 2 MHz rms design
tolerance. In the bottom plot, the errors were increased
to best match the wakefield measurements, which re-
quired 12 MHz rms values. Thus, the actual cell errors
may be larger than that inferred from the QC results or
more systematic in nature, which can enhance their
2 4 6 1
[T (ns) ]
Fig. 2: Horizontal (crosses) and vertical (diamonds)
wakefield measurements and predictions (solid lines)
including 5 MHz rms cell frequency errors (top) and
12 MHz rms cell frequency errors (bottom).
effect. The theory itself has worked well previously,
yielding predictions in good agreement with the DDS1
measurements without added frequency errors .
The decrease in the wakefield that occurs at longer
times is due to the mode damping. In contrast, the
wakefield measured in a similar detuned-only structure
slowly increases in time . Another comparison of
note is the similarity of the DDS3 horizontal and
vertical results even though the upstream manifold
ports for the horizontal modes were shorted for this test.
This was done to assess the effect of removing the
upstream ports in the future. However, the lack of a
significant difference is not too surprising since the
dipole mode power is expected to go mainly (99%) to
the downstream ports (this was experimentally verified
as well). Finally, we note that the short-range (< 50 ps)
DDS3 wakefield measurements agree well with a
prediction that includes contributions from higher band
3 BEAM CENTERING
Although the manifolds were nominally added to
damp the dipole modes, they also yield signals that
provide a measure the beam's transverse position in the
structure. Moreover, the beam coupling to the modes is
fairly localized (2 to 10 cells) so filtering the signals
by frequency should yield beam offset information at
different regions along the structure. The NLC goal is
to use this signal information to keep the beams
centered in the structures to about +/- 20 gm.
To evaluate these signals, data were taken in which
the beam was stepped transversely across the structure
while the signal from a 15 MHz slice of the 14-16 GHz
dipole spectrum was processed. Figure 3 shows data
taken at 15.0 GHz. As expected, a 180 degree phase
transition occurs when the signal power goes through a
minimum. The solid lines in the plots are fits to the
data: a parabola to the power and an arctan function to
the phase. In each case, a fit parameter was included
to account for the addition of an out-of-phase signal
component. The size of this component is conveniently
expressed in terms of an equivalent beam offset, that
is, the beam displacement relative to its position at
minimum power that increases the dipole power by an
amount equal to the out-of-phase power. The phase
data yields the best measure of this offset, 5 +/- 2 im.
However, values as large as 100 gm were initially
observed. After some study, it was realized that an X-Z
correlation along the bunch (a- Z = 0.7 mm) was the
likely source. This correlation was reduced by tuning
the dispersive properties of the beamline to decrease
the horizontal bunch width. Thereafter, the measured
offsets were generally smaller than the NLC beam
centering tolerance, which is desirable since it will
simplify signal processing in the NLC.
Another parameter obtained from the fits to the
dipole signal data is the beam position at the minimum
* * **9
2 4 6 8 10 12
Here’s what’s next.
This report can be searched. Note: Results may vary based on the legibility of text within the document.
Tools / Downloads
Get a copy of this page or view the extracted text.
Citing and Sharing
Basic information for referencing this web page. We also provide extended guidance on usage rights, references, copying or embedding.
Reference the current page of this Report.
Adolphsen, Chris. Wakefield and Beam Centering Measurements of a Damped and Detuned X-Band Accelerator Structure, report, September 14, 1999; Menlo Park, California. (https://digital.library.unt.edu/ark:/67531/metadc624169/m1/3/: accessed April 21, 2019), University of North Texas Libraries, Digital Library, https://digital.library.unt.edu; crediting UNT Libraries Government Documents Department.