Below is a summary of the many constructive responses regarding alternatives to shutting down our Gemini 2000, as well as advice for how to safely quench the magnet if we need to decommission it. Thanks for the many generous offers of decommissioned parts; this is truly an amazing group.
We received several offers of decommissioned replacement computers, consoles, and boards, and are grateful for the receipt of a working replacement computer. We also received a response from a more local NMR service company which can offer service calls at lower costs (Open Technologies, Inc.). With these resources, we're in the process of re-attempting console communication for much less of a cost.
Regarding how to safely quench a magnet if we should still need to decommission it, we received the following replies:
How to run down the magnet power slowly:
Contact Agilent NMR (who bought Varian) and see if you can get a power stick for the magnet with instructions for running down the magnet slowly rather than quenching it.
I have, in the past, brought down magnets with a power supply, and with diodes. That is more controlled, takes longer, and spews fewer cryogens. I have done that with users around.
The safest way to decommission a magnet, if you have access to a charging stick is to bleed down the current through crossed-diodes. I've done it a few times though it was a little unnerving the first time.
If you have the connector rod for the magnet, it is possible to safely run down the magnet. You need a set of resistors and a heat sink to connect across the magnet circuit, and you can use a 9V battery to open the switches. I have done this three times to warm up magnets.
I don't know specifically about this magnet. But in most cases you can run down a supercon magnet with a diode box and a 9V battery, if you know where to hook up the 9V. I have done it with more than one magnet in the past, a wide bore 360 and a wide bore 400. I have also managed to quench an old 250 magnet while putting in the charging rod to do the same. I think that magnet was low on helium when we attempted this procedure. It was not mine, I was just trying to be helpful. The quench of these small magnets are quite uneventful. Still, if at all possible, I wouldn't just let it die because of low cryogen level, that could sometimes be dangerous, if someone happens to be nearby. I think some Varian engineer should be able to tell you how to hook up the 9V. Do you still have the charging rod and a diode box? You need those.
To quench by breaking the vacuum:
If you quench it by breaking the vacuum then let in DRY nitrogen gas. If water or helium gets into the vacuum you will not be able to get a good vacumm again, if someone wanted to use the magnet again.
I scrapped 3 magnets in the last year, a 400 and two 250s- all standard bore. (No one wanted them either.) In each case, I opened the vacuum valve. It was all over in less than 30 minutes. I tried to have the cryogen levels pretty low before I did this- but one of them had quite a bit of both nitrogen and helium. Yes, the plumes were pretty impressive. Two were each in a small room alone. I opened the vac valve, and walked out into the hall. I left the door open, and we watched from there.
Or you can always do a big show like we did in our lab (if you don't want the magnet to be reused): After the nitrogen is gone open helium port and open the vaccum valve. There will be some hissing sound and the helium will come out quite quickly. It is not as "safe" I guess. And even if you do a safe quench you will have to keep students out for a while just in case. It will depend on how large your lab is in volume!!
To quench by draining the cryogens, blowing in warm helium, or not refilling:
If you have to bring it down, I'd suggest waiting for He level to go way down and open the Helium port. You can blow helim gas into it to speed it up. When ours went down, I only saw a bit more plume than usual, lasting for about 1-2 mins. It is almost like what you see when helium is filled up to the neck. Nothing spectacular really. Again, that's assuming the helium level is very low already.
If you have access to an e-stick, you can run the magnet down on diodes to save it for rebuilding and re-use. Otherwise, there is not much energy in the magnet to cause damage or injury, so you can let it quench on its own. To control the timing of the quench, I would blow out the LN2, then the LHe late one Friday afternoon. If it doesn't quench immediately, just leave it to quench sometime over the weekend. A Gemini 200 quench is much less than spectacular, and not worth waiting around to watch. Quenching dry, the protection circuitry on top of the magnet will burn out, so it could stink up the room. The magnet is not heavy, especially without LN2, and 2 persons can carry it to another room or out to the parking lot before blowing out the LHe. Carry it sedan style with 2 boards slid under the leg mounts, or roll it on a furniture dolly so it does not quench into someone's face. Caution, do not pour LN2 or LHe onto asphalt, due to the condensation of LOx.
In regards to a safe quench, I did one a couple of years ago on a IonSpec cryostat. We did not have problems. Our lab is large and I stopped students from coming in until I finished this. I allowed the nitrogen to boil off first, you can insert a long stainless tube into one of the nitrogen ports (make sure it does not sit on the bottom-lift it up a bit), at the top of the tube attach a hose pointed somewhere away from students/doors. You wait for the liquid/gas to pour out.
The helium will take time if you don't release the vacuum a bit:
1. Make sure the valve is shut.
2. Take the cap off the valve then replace it and tighten the clamp
3 With the cap in place briefly open and shut the valve.
The small amount of air will help the boil off.
Last summer we had to decommission two older NMR systems that were still working but the space was needed for other uses- an old Varian VXR500 and a GE Omega 400WB. We simply stopped filling with cryogens (skipped a helium fill that was scheduled for the beginning of August and stopped filling with nitrogen) and they slowly quenched over a period of several weeks (one over Labor Day and the other a few weeks later). It would probably be good if you could let your system quench during the summer when there may be fewer students around.
To record the event for posterity:
If you quench, leave a steel rod on the magnet's side (to announce when the field is dead), get a tripod stabilized video of the event, leave the doors open and maybe sell tickets.
Rich Shoemaker filmed a quench and put it on youtube. I don't know the url.
Other useful information on quenching:
You do need to be wary that the cryostat vacuum space may have some accumulated frozen air that will create pressure once the cryogens are gone.
I left the door to the room open, and I didn't let anyone in for a few minutes after it was over. I would not do this with anyone going into or out of the room. But, since you can plan this, pick a time when no one is around. I did one at 8 am and the others on Friday afternoon. I would have at least two people around, just to be safe. Make sure there is nothing you care too much about just above the magnet- the plumes go up pretty high. (I was worried I would freeze the sprinkler head above the magnet, but it was
fine.)
We also received information on some additional alternative NMR instrument companies:
You might consider a couple of other options than the Anasazi -- TecMag has a lapNMR that will solve your computer problems (although not your cryogen problems) for a fairly low price. They are also providing a rebuild on an old Varian 90 MHZ permanent magnet for me. We are actually getting almost a 2 channel system because of the fast switching that is available. Also, if you really just need a teaching machine, you might look at the PicoNMR (out of Colorado).
From: Ruth Nalliah
Sent: Tuesday, June 21, 2011 2:50 PM
To: 'ammrl_at_ammrl.org'
Subject: AMMRL: NMR euthanasia decision
AMMRL colleagues -
We have a Gemini 200 MHz NMR which we successfully maintained 7 years without a service call... until the acquisition status became Inactive despite many reboots and cable repair. What we think is the problem is the Ethernet board in the console, which would cost $3,000 and is non-returnable if that should not fix the problem. Service calls are too expensive. Recently the Sun workstation also died, so we will probably have to get a new hard drive and re-install the operating system and software.
Cryogen costs have more than doubled since we commissioned the instrument in 2003. We assume that cryogen costs are likely to continue increasing; therefore we are at the point where we are reluctant to invest a large amount of funds in instrument repair. If software re-installation, a new cable, and a new small transceiver box (connecting the cable to the board) do not fix the problem, we will need to scrap our instrument and go with a non-cryogenic Anasazi 60 or 90-MHz.
The magnet is still fine and has held up very well over the years. The instrument has also done well except for this recent problem. So far, two used instrument companies are uninterested in decommissioning and taking our instrument due to its vintage (Gemini 2000).
We have two questions:
1) Are there any additional alternative suggestions to letting the instrument die? We want to make sure we've explored all of our options.
2) If we do need to let the magnet warm up ourselves (allow it to quench), what would be the safest way to do it? If we were to let it run out of cryogens during the school year, it's housed in a central instrument laboratory where students are going in and out. Have others brought their own magnets down in a controlled fashion?
If replies are directed to me, I'll summarize for the list.
Thanks -
Ruth
Ruth E. Nalliah, Ph.D.
Professor of Chemistry
Huntington University
Huntington, Indiana 46750
(260)359-4203
rnalliah_at_huntington.edu
Received on Tue Jul 26 2011 - 05:07:22 MST