Alcator C-Mod Weekly Highlights
			March 21, 2005

FY2005 weeks of research operations planned: 17 weeks, completed: 2.5 weeks.

Operations
----------

Plasma operations continued at Alcator C-Mod last week. Four run days were
scheduled and 3.5 were completed, including 2.5 days of research operation and
one day of RF conditioning. A total of 101 plasma discharges were produced,
with a startup reliability of over 90%. The typical H/D ratio continued to
improve, reaching values below 5% on some discharges.

This week will be a scheduled maintenance period.  No plasma runs are planned.

Tuesday's run was dedicated to ICRF Conditioning (MP#355). All three antennas
operated simultaneously to high power, with brief periods above 5MW total
coupled power. Operation at 1MW per antenna (3MW total) was reliably
achieved. Significant interference on diagnostics associated with operation of
J-port was noted, and subsequently identified as being due to RF leakage from
the DC breaks in the external transmission line.

Installation on Wednesday of ~50uF capacitance in parallel with the breaks has
greatly reduced the RF pickup on diagnostics, allowing use of the J-port
antenna for physics runs. The run on Wednesday was scheduled for MP#326, an
evaluation and comparison of the three ICRF antennas. This experiment
contributes to our Level 1 (JOULE) target, "Measure plasma behavior with
high-Z antenna guards and input power greater than 3.5 MW". A problem with one
of the poloidal field supplies delayed the start of this run, and some
additional time was required to implement and validate the modification to the
DC breaks. Only half a run-day was accomplished on Wednesday, and the
remainder of the planned experiment was re-scheduled for Friday. The antenna
heating effectiveness appears about the same for all antennas within 10%.  The
J antenna may be slightly less effective than D and E and D may be slightly
better than E.  From a few discharges, the heating efficiency was measured
from change in slope in stored energy and the range of efficiencies is from
0.5-0.7.

The J antenna neutral pressure limit appears unchanged from last campaign
(~0.25 mTorr) at 0.8 MA and increased to between 0.7-1 mtorr at 1 MA.  This
suggests that post-boronization conditioning should be done at the higher
current.

The run on Thursday was devoted to MP#414, "Divertor conditioning with Li
pellets in near double-null discharges". The concept underlying this
experiment is that scrapeoff layer flows in lower single null discharges
transport Li leaving the core to the inner strike point, and perhaps to the
the outboard limiters. Little of the escaping Li would be expected to be
deposited at the outer (low-field side) strike point, where the dominant heat
flux appears. It is conjectured that this effect could account for the lack of
beneficial effects of lithium conditioning in experiments on C-Mod and on
DIII-D, in contrast to the very positive experience on TFTR, which used
inner-wall limited discharges for which the expected deposition location and
high heat flux points coincide. In Thursday's experiment, we operated a
standard ICRF-heated lower null discharge for the first half of each shot, but
then varied the equilibrium to a nearly balanced double null, which eliminates
the flow from the outboard side to the inner strike point. Lithium pellets
were injected into this phase of the discharge, and subsequent discharges were
monitored for any effect of conditioning on the plasma performance during the
heating phase. A total of 21 pellets were injected; however, in most cases the
plasma disrupted following either the first or second pellet. There were no
obvious effects of the lithium on any of monitored plasma parameters,
including radiated power, h-factor, H/(H+D) and core moly radiation
(HIREX). Chromex measurements clearly saw neutral Li emission from the lower
outer divertor. Mo I emission from this region was weak even before the first
pellet, and did not appear to change much. While the numerous post-pellet
disruptions may have affected the result somewhat, the initial conclusion is
that the pellets had little or no effect in terms of wall conditioning.


Physics
--------

An analysis of data taken during the 1050303 run, "Scoping experiments for the
upper divertor cryopump" (MP323), has been performed. Neutral pressures and
conductances at three locations in the vaccum vessel were compared in a set of
near double-null discharges that are thought to be prototypical for upper
cryopump operation: (1) behind a simulated cryopump baffle in the upper
chamber, (2) at a vertical port at the top of the vessel, and (3) behind a
lower divertor module. Neutral conductances were inferred by puffing known gas
flow rates at these locations via capillary tubes and recording the resultant
local pressure rises. In addition, the active deuterium inventory in the
vessel was monitored. From these data, a characteristic neutral pumping time
was estimated based on ideal cryopumps located at the three
locations. Location (2), which simulates a 'pumping slot' design
option, was found superior to location (1) for an upper cryopump. Gas
throughputs there are projected to be insensitive to upper strike point
location and to exceed those in location (1) by a factor of ~1.7. Location (2)
was also found to yield the highest possible throughput for an ideal cryopump
on C-Mod. Based on the in-situ gas conductance measurements, an array of 25
'slots' in the upper chamber would have a combined conductance to the
cryopanels of 48,000 liters/s while the 10 slots in the lower divertor yield a
value of 18,000 liters/s.

Using the Thomson scattering (TS) system, considerable amounts of dust have
been observed in the C-Mod vessel while the machine is backfilled with
deuterium gas. Such backfills are customary procedures, used for
density calibration of the TS system using Raman scattering from gas
molecules. However, on two occasions during the present campaign, the level of
dust was too high to allow a successful calibration. We suspect the dust
particles are on average quite small, lightweight enough to be stirred and
supported by natural convection of gas in the unevenly heated
machine. Interestingly, the frequency of laser-dust interaction was roughly
four times higher during the second calibration, suggesting that the
concentration of dust is rising with time. We cannot positively identify the
source of the dust, or whether it significantly affects plasma performance. We
are continuing to look for more signatures of dust and try to understand the
source.


Lower Hybrid System
-------------------

A second cart (four klystrons) was brought into operation last week with up to
180 kW of lower hybrid power coupled into both plasma and vacuum from eight of
the total twelve klystrons.  Drive leg calibration of the last four klystrons
is underway. The installation of the rear launcher probes is nearing
completion.

ICRF Systems
------------

All four transmitters are operational and tuned to operate near 80 MHz (78,
80, and 80.5 MHz). The ICRF system was used successfully in all four runs this
week.  Much of the RF interference issues were resolved by adding 50 uF
capacitor in parallel with the J antenna DC break.  We will continue to survey
for RF leakage in an effort to reduce the interference problem further.

Long Diagnostic Neutral Beam
-----------------------------

During the past week all of the DNB systems were energized and tested, at
first individually, and finally all together to generate the first hydrogen
beam pulses.  By the end of the week, the DNB parameters had reached 50 kV and
6.5 amps for pulse lengths of 100 ms.  Longer pulse lengths totaling 600 ms
with 100 ms on/100 ms off modulation were achieved at 45 kV and 4 amps.
Auxiliary functions, such as the automated helium cryostat refills, are
currently being tested.  Most of the Russian team will be departing shortly,
leaving their principal engineer and principal physicist to condition the beam
to full pulse length during the upcoming (and final) week of installation.


Diagnostics
-----------

The upper 16 channels of the FRCECE diagnostic (0.74m < R <0.84m at 6T) are
now operational. Raw data is being acquired using CAMAC channels while a
problem with the PCI data acquisition is being resolved.  The local oscillator
for the lower 16 channels failed a couple of weeks ago and was returned to the
manufacturer for evaluation.  We are awaiting a response.

Background experiments were conducted to determine the limits on the time
resolution for measurement of ion temperature with the CXRS spectrometer.  The
integration time was reduced to 5 ms with signal levels from ambient emission
that appear adequate.  Initial interpretation for ion temperature appear
successful.  During the coming week, more detailed comparison will be made
with existing measurements using longer integration time.

The full spectrum observable with the CXRS spectrometer was measured for a few
shots during Friday's run.  Without the spectrometer's internal interference
filter which is required to produce the full 45 channels, the instrument is
limited to 15 channels, but each channel has a greater spectral range.  Each
channel may have a range of either 4930 A to 5040 A or 4845 A to 4955 A.  The
principal line emission identified within this range and relevant to recent
experiments is from B II, B V, Ar II, He I, and H I.



Travel and Visitors
-------------------

Joshua Stillerman traveled to Greifswald Germany from 3/16-3/18 to serve
on the Wendelstein 7-X data systems design review committee.

Stewart Zweben (PPPL) was at C-Mod last week to take data on edge turbulence.
Normal turbulence images were seen in the GPI view using the ultra-fast PSI-5
camera at 250,000 frames/second.  An intensified Xybion camera was also tried
in order to image the toroidal vs. poloidal structure from the side, but the
signal level was so far too low for clear imaging at 5 microsecond exposures.
However, this camera and view were used at long exposure times (2-10
milliseconds) to image the Helium II ion "plume" exiting from the inner wall
gas puff capillary.  Images such as these may be useful to estimate the
parallel flow velocity in the far SOL near the inner wall.

Earl Marmar participated in a meeting of the Fusion Facilities Coordinating
Committee on Monday of last week in Germantown.

Earl Marmar, Miklos Porkolab, Martin Greenwald and Bruce Lipschultz attended
the Budget Planning Meeting in Gaithersburg on Tuesday and Wednesday. Marmar
made a presentation on C-Mod results, plans, budget and schedule for
FY04-FY07; Porkolab made a presentation on plans for the PSFC.

Gerd Schilling and Randy Wilson were at C-Mod during the week of March 7th,
and participated in the successful lower hybrid startup experiments.
_______________________________________________
Cmod_weekly mailing list
Cmod_weekly@lists.psfc.mit.edu
http://lists.psfc.mit.edu/mailman/listinfo/cmod_weekly