sPHENIX discussion of the superconducting solenoid

["Phillips, David B" <phillips AT bnl.gov>]

If we go with the lowest resistance option now and decide we want to add more resistance later we could replace sets of 535kmcm jumpers with water cooled conductors as Paul G had suggested for the connection to the magnet (at the time he suggested 2 ea 750kmcm in parallel).

 

For example if in 1008B we replaced 25’ of 12 in parallel 535kmcm jumpers with 25’ of 2 in parallel 750kmcm water cooled jumpers the total resistance would go from 0.73mOhms to 0.87mOhms.

 

So I propose that we proceed with the modification of the water cooled bus in 1008B, positioning of the PS & Dump Resistor and hook-up of the 535kmcm to try to achieve the 0.73mOhm target resistance, any objections?

 

Dave

 

From: Yip, Kin
Sent: Thursday, March 28, 2019 3:24 PM
To: Phillips, David B <phillips AT bnl.gov>; sphenix-magnet-l AT lists.bnl.gov
Subject: Re: Brieif Summary of today's Magnet meeting (Mar. 27, 2019) at 902A conference room

 

Hi,

Thanks for all the works so far.   But I like your choices !

> If we want the option to add a resistor (to reduce ramp down time) we should have a plan and leave space for it now.

This sounds good.  I guess Carl may be happy about this, too.

With the same parameters ( L=2.33 etc. ), let me show the ramp-down time  (from 4600 A) to 1000 A or 1300 A.  

From  \\bnl.gov\c-adnas\babar\BaBar P.S. Manual\Babar P.S. Section1.pdf, p.15 (PDF page), in their ramp-down example, they did fast discharge at 1338 A.
{ A note there said "500 A to 2000 A" is acceptable. }  Fast discharge at 1300 A would save a few to several minutes compared to 1000 A.  ( I imagine we
could do that after some experience of stability. )

Total-Resistance (mOhm)                 ramp-down time to 1000/1300 A                  di/dt at time=0 (A/s)
0.73                                                     3329/2874                                                        -1.74
0.86                                                     2963/2548                                                        -2.00
0.92                                                     2819/2421                                                        -2.12


The delay (once in a while ?) is not comparable to the risk of having a high risk quench.    I like the choice of 0.73 mOhm very much :-) 

By the way, you can try to verify those calculations from my webpage: https://www.c-ad.bnl.gov/kinyip/sPHENIX_Ramp_Down.html
{ I know it's not as appealing as facebook.com :-)   But my website doesn't have any advertisement and it can't dig out your personal information and track
your location etc.   ... At least not as harmful :-)    }

Kin

On 3/28/19 1:39 PM, Phillips, David B wrote:

For item 7, I assume that there will be entry to the IR with the magnet at full field (except for anyone with medical implants affected by magnetic fields). My concern was for time delay waiting for “magnet off” before opening a pole tip (maybe this isn’t a concern).

 

Some details for item 5:

Estimates of resistance of connections between the Power Supply & Magnet (water cooled bus and cables), I tried to be conservative with lengths and temperature of conductor so actual resistance may be slightly lower.

There are 4 pairs of PHENIX water cooled bus (WCB) from 1008B to the IR (2 pairs 2” schedule 80 and 2 pairs 2” schedule 160 copper pipe).

I considered two different paths to reach the magnet, #1 jump from the west wall of the IR to the magnet platform (basically follow the cryo lines), #2 go to the floor (below the rails) and up the side of the platform.

All of these estimates of resistance include 535kmcm (12 in parallel) from the Power Supply to the Dump Resistor, from the Dump Resistor to the WCB, from the building WCB to the Platform WCB and from the platform WCB to the Magnet Leads (total resistance of these 535kmcm jumpers is 0.13mOhms).

For path #1a using 2 pairs of existing schedule 160 bus (modified at the ends of course) from 1008B to the west wall of the IR and then one new pair of schedule 160 bus from the wall to the magnet the total resistance is 0.73mOhms.

For path #1b using 1 pair of schedule 160 bus and 1 pair of schedule 80 bus from 1008b to the west wall of the IR and then one new pair of schedule 160 bus from the wall to the magnet the total resistance is 0.86mOhms.

For path #2 using 2 pairs of existing schedule 160 bus (modified at the ends of course) from 1008B to the base of the platform and then one new pair of schedule 160 bus up the platform the total resistance is 0.92mOhms.

 

I assume we want to go with option #1a?

Kin – can you tell us the ramp down time for 0.73mOhms?

This resistance could be lowered by using a lower resistance bus from the West wall to the Magnet, or by adding more 535kmcm jumpers (in parallel).

The resistance could be raised (within limits) by removing 535kmcm jumpers.

If we want the option to add a resistor (to reduce ramp down time) we should have a plan and leave space for it now.

I would like to start the modifications to the WCB in 1008B soon (during this RHIC Run), so we need to solidify a plan.

 

Attached is the proposed layout of the PS & Dump Resistor on the mezzanine in 1008B.

 

Dave

 

 

From: sPHENIX-magnet-l <sphenix-magnet-l-bounces AT lists.bnl.gov> On Behalf Of Yip, Kin
Sent: Wednesday, March 27, 2019 5:08 PM
To: sphenix-magnet-l AT lists.bnl.gov
Subject: [Sphenix-magnet-l] Brieif Summary of today's Magnet meeting (Mar. 27, 2019) at 902A conference room

 

Powerpoint (Carl Schultheiss) : https://indico.bnl.gov/event/5935/ 

 

 

1. Carl with his presentation got us to discuss about two different of works related to voltage taps.  First, since we have many spare channels left at the MPF

connectors at the bottom of the doghouse/junction-box attached to the Valvebox, he proposed that we re-route some of the voltage tap signals from the
magnet which originally went up to the chimney and came out of the connectors at the top of the Valvebox,  to now be connected directly to the MPF connectors
at the bottom of the junction-box.   There is a small chance that the flaky VT11 is due to connection in the chimney/Valvebox and this re-routing may solve a problem. 
Even if rerouting doesn’t solve a problem, this is still almost like a no-brainer.   There weren’t too many questions about it after people understood what Carl meant.

2. For the second part, the voltage-tap VT11 was one that showed intermediate circuit-opening during the high-field test and we abandoned it at the end.   So far, we
   haven’t found any fault about it but we haven’t actually reached it.  To reach it or to add spare close to it, we have to open more on the magnet junction-box.
   If we add spare for VT11, we may as well add it to VT05.
3. Joe D’ambra pointed out that we should talk to Ray Ceruti.   I then walked to the 902 floor and found Ray (whose hand was still dirty and dark) and
   brought him up to the conference room.  We talked with Ray and after a few back and forth, Ray felt that he could do it without too much risk.   It’s just a matter of
   time.  He’ll try at least to open one side (VT11 side or the VT5 side) and hopefully we could reach the other side.   He estimated that it may be a day or two to open it up and

another 2 days to close it back.  All in all, it may be a week of his time.   He’ll send us an email when he can find a time to do it.  This is OK as it’s not really that urgent.

4. It’s not clear whether there are still unused wires which go through the Extension.  Joe said that he’ll look for them after Ray finishes opening the junction-box.
   If there is no spare or not enough unused wires, we’ll use some sort of snake to put wires through the Extension.

 

5. David Phillips got us to talk about the long cable/bus connecting the power supply to the Magnet current leads.   Carl and I were advocating for lower total resistance
   to prevent the initial di/dt being too large to cause quench.  ( You may see my email sent yesterday : https://lists.bnl.gov/mailman/private/sphenix-magnet-l/2019/000902.html )
   Even with the small resistance of 0.666 mOhm, the ramp-down time was like just almost one hour, compared to 40.5 minutes
   Dave was worried that too low resistance would cause the ramp-down time to be too long.  Carl said that we could always add some resistance though Dave said that it’s
   not as easy as people thought.   But other than David, people generally worried about the quench more than the ramp-down time being too long.

 

6. After the meeting, I discussed about the above addition of spare voltage taps and the works involving Ray Ceruti (with the presence of Jim).   He also agreed that we should do it
   as long as the expert like Ray doesn’t feel there is any significant risk, which was the case.  

 

7. When I discussed with Ed in another occasion about the lower resistance and lower di/dt vs longer ramp-down time, Ed also didn’t worry too much about the one hour vs 4x minutes.
   And in fact, Ed pointed out that even if we don’t ramp down the Magnet, there are only a few spots close to the poles which are at 100 G or so and we may enter the Experimental Hall
   even when the Magnet full current is on.   Whenever I emphasized the risk of quenching, people seemed to be threatened by that more.
   
   
   

 

Kin