_______________________________________________________
What I have done is to make predictions for a current
scan from 0.5-1.1MA. To do this I have started with the
inputs from the high power phase of shot 1090902024
(t=1.32-1.42s). The values of the 8 EPED1 inputs for
this case are:
Bt=5.364T, R0=.6707m, a=.2123m, kappa=1.484, delta=.471,
betaN_global=0.578, neped=15.2e19 m-3, Ip=0.8093 MA
I've run EPED1 to make predictions for this set of
inputs and then for a current scan with
Ip=(0.5, 0.65, .8093, 1.1 MA). To do this scan
I've kept the Greenwald fraction constant. That
is, I've scanned neped proportional to Ip (the
other 6 inputs are fixed). My understanding is
that fixed Greenwald fraction is a good approximation
of what we expect in the experiment. I've also scanned
the density a bit around the base point, and found
it has only a relatively small effect on the predicted
pedestal pressure and width.
The plots show the EPED1 predicted pedestal
height in terms of pressure (which we usually
compare to the measured value of 2*neped*Teped)
and in terms of betaN,ped. The third plot
shows the predicted width in normalized poloidal
flux units (average of ne width and Te width
in normalized poloidal flux). As you can see, the
model predicts the pedestal pressure should increase
with current, but somewhat less than linearly,
because the pedestal is predicted to be narrower at
high Ip. The predicted betaN,ped value is
in the vicinity of 0.3, and decreases slightly
with current.
I've also done a scan in global betaN
(that is to say, global Shafranov shift),
which finds that as we increase the
global betaN value from (0.578,0.8,1.2,1.6)
that the predicted pedestal height in
terms of betaN,ped also rises somewhat from
(0.285,0.302,0.343,0.389). This scan is
done from the base point (Ip=.8093, neped=15.2e19).
This is the usual expected result because
peeling-ballooning stability generally
improves with increased Shafranov shift.
These predictions are all for the Type I ELM
regime. If we don't achieve Type I ELMs
at some points, we'd generally expect the
pedestal pressure to be below the predicted value here.
Got some ELMy H-modes, 2nd one seems a little over-dense.
RF is on before the final equilibrium is obtained, so we will delay onset of RF
until 0.75s on next shot. Also, Wukitch would like a smaller right gap.
For next shot, decrease right gap.
No changes for next shot. Yijun will try to get an improved tune.
Pedestal density is nearly 2e20, which is on the high side for this kind of
experiment. Looks like a reasoably strong QCM underlying the ELMs on
magnetics/reflectometry/PCI.
Let's call ourselves tuned up and move on the power steps of 3/4/5MW.
I think we have too much radiation early, and we are unable to establish
ELMy H-mode. Fairly strong QCM in evidence, however.
2nd part of discharge has lots of Prad spikes and RF faults.
to get the strike away from the floor of the divertor by going to the
grazing configuration.
Oops, PO moved the X-point out instead of in.
suggests we're not on the EF1 shelf yet. Injection is still there early.
EDA with some ELMs.
I'm not sure the strike is consistently on the corner of the EF1 shelf.
Have Ron shift the x-point 2mm in for next shot.
Injection from J seems to have killed the ELMy H-mode, and reduced the density
control. For the next shot, reduce demand on J-port, lower target density.
Also, delay NINJA puff to 1.4s.
Working on a set of EPED1 runs for parameters close to those in the initial shots.
neped ~ 20e19 m^-3, betan_global ~ 1, Ip ~ 0.9 MA
(other inputs as in base case)
Gives an initial EPED1 prediction of:
p_ped~27 kPa, beta_ped ~0.24, betaN_ped ~ .31, wid_psi~ 0.03, wid_mid ~ 4.5mm
We're not likely to get steady ICRF above about 3MW with the ELMs.
We will lower power demands for next shot. Keep everything else the same.
Amanda reports actual Te drops at Elm onset.
EPED1 results for shots similar to 014 and 015.
Inputs: neped~15.2e19, betaN_global~1, Ip=0.9 MA
(other inputs as in base case)
Output: p_ped~28 kPa, beta_ped~0.25, betan_ped~0.32, wid_psi~0.03, wid_mid~4.5mm
(higher predicted pedestal than in 1090902024 based predictions due to increase in Shafranov shift, betaN_global from ~0.58 to ~1)
Great looking ELMs on H-alpha.
Edge Te is substantial.
First H-mode has good ELMs. 2nd phase is plagued by fast injections.
current, I am asking PO to increase elongation for next shot.
apparently. Essentially a repeat of 18.
Good steady H-mode data.
2 stages of H-mode density.
Also had the PO increase the target density.
Looks like EDA with maybe some small ELMs.
Looks very EDA.
QCM on PCI, ch16 is showing some frequency doppler-shifting behavior
that is directly correlated with changes in the edge Te from GPC.
If this ever went into H-mode, it was extremely brief.
ICRF and the first wall aren't getting along very well. There's a good
chance the gaps aren't exactly what we think they are, due to the lower
Ip. It's interesting to note that the Te pedestals on the last few shots
are falling outside of where EFIT says the LCFS is.
For next shot we will try to go the higher current point.