Magnet problem

Status
Not open for further replies.

Mark R

Diamond Member
Oct 9, 1999
8,513
16
81
Came into work the other day, and found water about 1 inch deep on the floor, and a torrent pouring out from under of the machine room doors.

A cooling water hose had become disconnected from the cooling circuit for one of the magnets. The BMS showed cooling circuit pressure starting to drop at about 3:30 am, and by about 5:00 the pressure was 0 (but the emergency mains top-up system had kept working).

1 equipment rack was completely flooded out (the one with the heat exchanger between building cold water circuit, and equipment coolant circuit), and 2 were a bit damp.

More importantly, the helium compressor was swimming and the PSU for the pulse-tube refrigerator had been for a refreshing shower.

Thankfully, the magnet seems to be reasonably happy, even if most of the monitoring equipment has now been switched off to dry out.

Still, it could have been worse. The following story was relayed to me from another site nearby:

A major electrical fire started in the plant room for one of their MRI scanners. This had the effect of burning though the 400V 3 phase cable, and all the control cables going from the machinery racks to the scanner (as well as causing severe damage to the building structure).

The fire dept and building engineers wanted to get in to inspect the damage - but the magnet had got through pretty much unscathed, and was still energised. So, for safety reasons, the scanner room had to remain out-of-bounds to non-medical personnel, and absolutely no safety equipment/metal inspection equipment could be brought in.

The problem was how to de-energise the magnet, now that all cables to it had been severed by fire. The manufacturer was called, and stated that a controlled shutdown may not be possible as the fire had damaged the magnet PSU connector, and it would not be safe to connect the PSU to discharge the magnet. They suggested an uncontrolled magnet quench may have to be performed. Unfortunately, the emergency quench control panel had been burned-out. An engineer would have to connect a battery direct to the quench heater terminals on the magnet.

But there was another problem. The fire had caused severe structural damage to the building, and the concern was that it had damaged the emergency quench vent pipe, which would carry away the explosive blast of ultra-cold helium ejected from a quench. If the pipe was damaged the magnet might suffer over-pressure and burst it's "burst disc" pressure release valve into the scanner room, resulting in total replacement of the oxygen in the room with helium, and the rapid suffocation of anyone in there.

And, of course, as the magnet was energised, it wasn't possible to bring breathing equipment in either.

In the end, the fire department came up with the idea of building a plastic tunnel (like in those outbreak type movies) from the main building, into the scanner room,right up to the magnet terminals. The engineer could then enter the room through the tunnel, initiate the quench, but be somewhat protected from sudden asphyxiation by an eruption of helium.

Rather him than me!
 

Analog

Lifer
Jan 7, 2002
12,755
3
0
Should of had a cooling water detector at a low point in the room. Even my basement has a cheap one near the sump...
 

Jeff7

Lifer
Jan 4, 2001
41,596
20
81
Scotty could've done a cold restart on that thing in 10 minutes.

Just reverse the polarity and then decouple the plasma flow regulators. That always does it.



In the end, the fire department came up with the idea of building a plastic tunnel (like in those outbreak type movies) from the main building, into the scanner room,right up to the magnet terminals. The engineer could then enter the room through the tunnel, initiate the quench, but be somewhat protected from sudden asphyxiation by an eruption of helium.
Guy's just sitting in his office or lab, putzing around with some new cooling apparatus, and then with one phonecall, his week gets very interesting.
"It did what? Wait, I have to go where now?"


"Will there be oxygen there?" is also not a question I'd usually think to ask when visiting a customer site.
 

Mark R

Diamond Member
Oct 9, 1999
8,513
16
81
Should of had a cooling water detector at a low point in the room. Even my basement has a cheap one near the sump...

Yup. I wasn't involved in the installation, and I was surprised that they didn't install one.

Anyway, the connection that failed was one that was the responsibility of the magnet supplier (it was the connection between the equipment cabinet and the vendor supplied cooling water hose; not the connection between building water supply and the manufacturer's hose), and we have a full-coverage maintenance contract with them, so we're not out of pocket (except for magnet time and inconvenience).

And just for random entertainment, here's a video of a magnet quench (quench heater initiate at 2:45), but moderately entertaining video before this.
Link
 
Status
Not open for further replies.