Interim Assessment of Solenoid for GlueX

Indiana University Cyclotron Facility
November 30 - December 1, 2004
South Large Conference Room
AGENDA     CHARGE

Assessment Team


Background Information for the Reviewers

This is an appropriate time to assess whether the plans for refurbishment of the coils of the solenoidal magnet are complete; whether more extensive rework should be considered; and take a fresh look the complete list of tests which are scheduled for the magnet before installation in Hall D.

The magnet assessment will be conducted at the Indiana University Cyclotron Facility in Bloomington, IN, on November 30 and December 1. We will begin on Tuesday morning at 9:00 am and be finished by noon on Wednesday.

More general information about the GlueX project can be obtained from from the project web site. Chapter 5 from our design report has background information on the solenoid magnet including plans for refurbishment. A description of the scope of work of the GlueX solenoid contains an evaluation of the status of the magnet before any coil refurbishments, initial list of recommended modifications, detailed assessment of the components, and early estimates of costs.

Pictures of the Magnet

Below is a brief description of the goals for the assessment as well as details of the recent work and status of the superconducting coils.

Travel

For any travel arrangements, please contact Linda Ceraul at 757-269-7621.

Background

Two of the four coils of the MEGA/LASS solenoid have been refurbished at IUCF and have been stored for shipment to JLab early next year. The other two coils will be refurbished this fiscal year (FY05) as soon as contract terms are agreed to, expected by October 1, 2004. Coils three and four are already in the IUCF Solenoid work space and the technical team has been ready to start work since Sept 1, 2004.

This is a good time to conduct an assessment of the ongoing work on the coils as well as have experts comment on the long term plans to upgrade this magnet for GlueX. The timing is motivated by the recent discovery of internal corrosion of LN2 supply stainless piping during the refurbishment of coils one and two. Coil one had LN2 system leaks that were all traced to corrosion sites on the internal supply piping while coil two had no shield leaks at all but had a Helium leak. Coils one and two have been refurbished, leak tested, pressure tested and the shields have been cold tested. Both coils are currently leak free to highest industry standards. Coils three and four have been extensively tested and leak tested to very high sensitivity. Location of the known shield leaks and final refurbishment remains TBD for coils three and four.

A brief description of what we would like to achieve during the assessment follows.

Purpose

Notes:

Status and Detailed Questions on Coil Refurbishment

Report by Paul Brindza

The work to refurbish the GlueX Solenoid is at an interesting point where we have a significant choice to make about how to proceed. The first two coils are completely refurbished, tested LN2 tested and in dry nitrogen storage. They have new shield instrumentation, new strain gauges, new conflat flanges on all internal connections, no leaks and they have passed a pressure test. However coil one was at this stage around Memorial Day and it sprung a leak in the LN2 supply pipe at a pressure of only 30 pounds gauge. This is below the JLAB LN2 system operating pressure. The leak was found and fixed and coil one has since passed all of the above tests. However the leak source was found to be surface "pits" in the stainless steel LN2 supply line. Chemical analysis confirmed the existence of chlorides and the "pits" had visible red inclusions at the bottom. All this is evidence of corrosion, possibly due to brazing fluxes left over from the initial assembly or subsequent repairs.

We are about to perform the same refurbishment on the remaining coils three and four. These two coils are known to have large LN2 shield leaks at as yet unknown sites.

The question is "Should we proceed with the R&R activities as planned and rely on further testing to catch any future leaks due to latent corrosion, or should we "bite the bullet" and remove all the suspect stainless piping that may harbor the cancer of corrosion"?

If we decide to do a more thorough rebuild, essentially a complete remanufacture of the four cryostats, it will cost us significant funds. We estimate that the R&R of coils cost will easily double. The extra cost could be at least 250K$ to 350 K$ over and above what we have and will spend to get all four cryostats completely remanufactured. This is partially offset by not having to conduct the months long search for the location of the leaks in Coils three and four.

If we choose not to do the extra work and rely on periodic retesting to make sure we don't assemble leaky coils in Hall D then we run the risk that we spring a leak perhaps after the GlueX experiment starts. Every experiment that has used this magnet has had to contend with leaks. The corrosion pits that we have seen and the leak that opened during pressure testing ( greater than 10^-4 warm) may exceed our ability to pump.

Are the risks worth the extra costs?

Are there extra risks if we remanufacture the cryostats that may reduce the final value?

If we choose to remanufacture the cryostats, should we then replace the copper shields with all welded stainless inflated panels?

(This is at a small extra cost relative to the labor cost to perform the remanufacture).

Is there anything else that should be considered if we choose to remanufacture?

If we choose not to remanufacture then what further testing could be considered to buy more confidence?

Two of the Coils (coils one and three) have resistive shorts to ground. These shorts are historical and MEGA ran with them. However the value of the short in Coil one has changed while the short in coil three is nearly identical. There is thus a resistive short of 2 ohms across about half the solenoid when in operation. During charging there would be of order 5 volts thru two ohms thus dissipating approximately 2.5 watts, mostly in coil 1 at the site of the short. This assumes that the shorts remain at their room temperature values. If the short(s) are metallic then the resistance would decrease and if the shorts were due to oxides then in general they would become more resistive and if the shorts are due to foreign objects then anything can happen including the disappearance of the shorts.

Given this and considering the other work proposed, is it worth the obvious extra work and risk to open the coil cases and try to remove these shorts?



Elton Smith
Physics Division,
Jefferson Lab
12000 Jefferson Ave., Mail Stop 12H, Newport News, VA 23606
Office: (757)269-7625,   FAX: (757) 269-5800