sPHENIX discussion of the superconducting solenoid

[Kin Yip <kinyip AT bnl.gov>]

Hello,

Today, the first main struggle was that voltage tap VT11 seemed open.   It went back to good but people don't trust it anymore.  In fact, I reminded them of the
attached file "NegMagnetLeadJump_Down_Expanded2.jpg"  that Carl made on Jan. 18 (and the data was on Jan. 16) and you can see the "mating" (on-off) feature
in the plot.  In any case, Carl and Zeynep went to use VT10, a bit closer to the coil.

Pablo took some time to do the compensation for the power supply.   It's not perfect yet but they worked with it in the afternoon.

So, they went up from 100 A, 200 A, 300 A ...(~0.5 A/s)  and then decided to 500 A at 1A/s.   The first time we reached there, after 1 minute or so, we experienced a quench/fast
discharge.  And we haven't analyzed it yet as the current labelling (channel 1, 2, ...) seems to make it difficult to know which is which in the "sddsview".

At that time, Cryo was already going towards 10%.

During the 2nd trial, Carl suddenly remembered that Pablo had an "overcurrent" setting in the power supply which was set too ~low (600 A).   As we have dump resistor
in parallel with the power supply, the power supply saw more current than the magnet current ~500 A, apparently > 800 A (not sure how much it's due to imperfect compensation).
So, Pablo set the "overcurrent" to 900 A.  And this time, it stayed at ~500 A (it's more like 495 A due to imperfect "compensation").

We'd like to stay longer but Cryo was running out of Liquid Helium in the tank.   So, after 3 mintues or so at ~500 A,  we ramped down, using the slow discharge (going through the
free-wheeling diode).   I estimated that it took 1000 seconds to go to 0.

Carl will analyze the fast logger data and show us in this Wed.'s biweekly meeting.   In fact, it's good that the fast logger mechanism was actually working the first time that it needed to work :-)

I attach the plot of magnetic field (which reads >1367 G at the highest).   The Magnet readings were taken at once per minute (I didn't make it more frequent today).  You saw a spike in current
PS_current when there was the quench.

Kin

PS:  Unfortunately, Carl tried to be a bit "fancy" today and tried to take 25 data point with his laptop for his precise meter readings across 3 pairs
of voltage taps.   But it seems that his computer went to the "sleeping" mode and he lost that part of the data (which is outside our standard DAQ
system) :-(

Attachment: NegMagnetLeadJump_Down_Expanded2.jpg
Description: JPEG image

Attachment: 2018-1-29.gif
Description: GIF image