Agenda and powerpoint files can be found :
https://indico.bnl.gov/conferenceDisplay.py?confId=3291
1.
We review together the powerpoint materials given by Jon Hock, Carl Schultheiss and Paul Orfin.
John’s powerpoint (including one slide of 5 Gauss lines from Wuzheng Meng) includes stuff that he’s shown before, mainly deflection and other stress analysis. D. Phillips reminded us that we need to measure any expansion or lateral movement of the wall probably
by some surveying laser during the High Field Test, as J. Tuozzolo has asked us to do.
Carl shows a new block diagram for the overall circuit and reminded us that a quench at full current (~4600 A) would cause an increase of 37 K from the BaBar paper (IEEE TRANSACTIONS
ON APPLIED SUPERCONDUCTIVITY, VOL. 9, NO. 2, JUNE 1999). Dave Phillips has just updated (in PDF) the location of power supply, dump resistor and Cryo racks and I’ve updated on the powerpoint that we’ll show tomorrow. Again, when he mentioned
the quench system, we discussed about when to do slow discharge and when to do fast discharge. Apparently, it seems that in most situation, we’ll do slow discharge (except when his watch dog timer doesn’t work). Carl also informed us the his front-end electronic
boards are due to arrive on July 11, 2017.
Paul mostly used the slides we showed in the ASSRC/ESRC review before the Low-Field Test in Dec. 8, 2015. The Cryo group will need to add about 30’ of pipe due to the new location and height of the Magnet and Valvebox. And because of the new Extension, they
have to make new splice joints to connect with them. They have also bought a current lead heater. After the meeting, Paul told me that because they don’t know what materials the coaxial cable is really made of, they won’t push for high pressure in the High-Field
Test to try to speed up the cooling. When people asked about what happened during quench, Paul reminded us the PCSS review has already considered all the scenarios for the Cryo system including quenches and approve our Cryo system. So, this has been covered
there and he embedded the
2.
Apparently, Jon Hock still needs to work with the surveyors et. al. (and Ray Ceruti’s technicians are not really available), we won’t move the Extension until next week. P. Rosas
is in vacation and he’ll test the power supply with the two PSI’s when he comes back in mid-July. I understand that the Control Group couldn’t provide the software before he left.
3.
I gave a talk about slow and fast discharges from the full current ~4600 A. Verified with C. Schultheiss’ circuit diagram software calculation, I’ve generalized M. Berndt’s slow
discharge equation with freewheeling diode and written a webpage
http://www.c-ad.bnl.gov/kinyip/sPHENIX_Ramp_Down.html to help calculating. Owing to the fact that our cable resistance of 0.25 mOhm is much smaller than that of BaBar’s 1.25 mOhm. In the worst case of our estimate of resistance (cable etc.), it may
take 1.5 hours and 2.5 hours to go to 1300 A and 0 A respectively. Obviously, larger resistance would help and Dave is trying to see whether we can use only 8 or 9 MCM cables instead of 12 and make them as long as possible
J
I then showed that at 4600 A, for slow discharge, the intial di/dt is about -1 A/s, compared to -132.5A/s for the fast discharge. This is one way of explaining why it didn’t quench at slow discharge but did at fast discharge. For fast discharge, even at
1300 A, the di/dt is still -37.5 A/s and you have to go to 100 A for initial di/dt to be less than 3 A/s so that it may avoid quenches.
I have calculated that at 1300 A, energy release (½*LI^2) is about 2 MJ. Assuming conservatively 50% going to dump resistor and another 50% going to the Magnet, Roberto Than has calculated that it would give rise to a temperature rise of 20K. And their refrigerating
capacity of 100 W, this corresponds to 10000 seconds for the temperature to go back to 4.5 K (assuming no disaster such as burst disk).
Including this Cryo recovery consideration, I formulate the total time required algebraically for ramping down and cryo recovery and tried to find the minimum time by looking for derivative = 0. With diode voltage ~ 0.7 V and resistance ~ 0.25 mOhm, I came
up with a minimum time of ~8735 seconds at 261.3 A. This is not too useful as the dominating time is due to the Cryo recovery at 100 W. { But it’s fun to calculate all these based on advice from C. Schultheiss, B. Lambiase and D. Phillips. }
