Hi AMMRL,
Thanks to everyone who responded to my request for magnet lifespans. The following Excel file contains the data I received on 160 magnets:
https://uoguelphca-my.sharepoint.com/:x:/g/personal/malabdul_uoguelph_ca/ET-zTBWdRWlFvqB5w2ifPvQBkMS4D63Lwl_HHIgaAF2TaA?e=2cXiQa
One item that I neglected to ask (but many of you were kind enough to include) was whether the magnet had been restarted after being quenched or decommissioned. In the "Current State" column of the Excel file, a (Q) indicates the magnet had a quench and was restarted afterwards, and a (D) indicates the magnet was decommissioned/moved and restarted afterwards.
I also received a number of insightful comments which I have included below. To summarize the majority of comments: if the magnet is shielded and has a good boiloff rate, keep it!
Best regards,
Sameer
Comments (Anonymized and cleansed of vendor information).
*****
The lifetime is limited by the sealings, mainly the big rubber ring at the bottom. This might be one month, this might be 40 years.
The more interesting questions are
- how large is the stray field of your existing magnet
- what is the field homogeneity of the existing magnet
- what is the helium refill interval
If you are satisfied with these points, keep your magnet. Your vacuum sealing might good for another 20 years and the new one might fail after 1 and a half year. Nobody knows. Of course the opposite might be true as well.
*****
I would suggest if budget is allowed and old magnet is not in good condition, new magnet has many advantages: 1. shielded, save space; 2. save much liquid He and easier to refill/maintain; 3. the o-ring of old magnet will age and the repair cost is also significant; 4. other features, such as showcase, etc.
*****
The tough thing is that if you are changing vendors it may be hard to get the new vendor to guarantee probe lineshape specs unless you can demonstrate lineshape and such to them right before they decommission your current console.
*****
Magnet replacement nowadays is done for mostly two reasons.
1. You have an unshielded magnet and space is valuable. Replacing a unshielded magnet with a shielded one will result in space savings of near 75%. It also makes the NMR safer as the 5G line is now much closer to the magnet and less problems will occur.
2. The new magnet will use significantly less helium than the existing magnet. This is an investment for the future as helium prices will go up the question is just how much.
If a magnet has not changed the boil off rate there is a good chance that it will live for a very long time.
There are systems in the field with magnets that approach 40 or more years.
*****
Magnets are usually fine for a long time. More determining of whether to replace them are factors like:
- helium availability and consumption
- if it is an unshielded one, whether it takes expensive real estate
The O-rings are expected to last 20-30 years after which they get bridle and need to be replaced. This requires warming up the magnet and rebuilding it, which is costly, but by far not as expensive as a new magnet and worth it, if the other two points above are no issue.
*****
I don't know if you want any additional information, but there were a few items from when we rebuilt the 11.79 magnet in 2017 that the engineer was kind enough to walk through for the age of the magnet. The race along the bottom plate is a potential problem area over the years, not just the o-ring it holds. Keeping it from oxidation by condensed water is a good measure for longevity of the magnet. Putting check valves on the nitrogen stacks is a good way to protect from water damage, or just having cut sponges to catch the water is good enough. Drift rate and boil off for helium and nitrogen are a measure of magnet health, or at least something to watch. When the magnet was recharged, I was asked to watch the drift rate for about a week to make sure it got down to about 4 Hz per hour. If a magnet is acting up, the drift rate can vary widely hour to hour (which is normal for a freshly charged magnet, I think I saw up to 20 Hz per hour at first). Boil off should likewise be mostly stable, barring the occasional pressure fluctuations from storms. If the helium boil off is higher than usual, double check the nitrogen boil off as well. If both are off, something might be happening with the magnet or the cryostat.
*****
I find magnets do not need replaced unless you want a shielded Magnet or in most cases the newer magnet may have a smaller footprint. Some of the newer magnets may have a longer hold time but this is probably do to larger LHe reservoirs. If it is a pumped magnet if possible I would replace it. less worries. There are a lot of magnets that are 20-30 or more years old and working fine with new consoles.
One problem is the Vendor replacing the system ,may not agree to use the old magnet or not guaranty spec. One thing look at is some of the new magnets have worse drift specs but newer console can better handle that. In general if the magnet is 10 years old and working good and you keep it filled with cryogens it will be just as good in 10-20 years or more.
In summary magnets don’t age . You want to make your decision based on Drift rate , LHe hold time, size, shielding, and the one that will most likely get you, is will the Vendor agree to it and how much do you save and what happens to the specs.
What type magnet do you have.? If it is a from Vendor X and you go Vendor Y, if they would agree you would still need to buy a new RT shim coil and sample upper barrel assembly. If it is a Vendor X magnet and you go with the same Vendor they may agree but then the other factors I mentioned should be considered.
Not age.
*****
I can give these comments:
- moving a magnet cold or on field (hence without exchanging the O-rings) can be risky if the magnet is quite old. O-rings can be OK for 30-40 years without problems as long as the magnet isn’t moved.
- the magnet should be helium filled using the correct procedure such that the helium dewar isn’t open to air for more than 1-2 seconds at a time, no ice buildup
- the magnet should be properly cleaned (no moisture buildup on lower flange which could lead to contact corrosion and compromising of the main magnet seal)
Upholding these rules I have seen magnets getting really old without any problems (recent examples were a 40-year old 400 and a 30-year old 600 which stayed on field until the last day when they were deenergized - the 40-year magnet got relocated, a set of new O-rings in the process, and is happily working again).
I do not think the age has really to do anything with magnetic field, vendor or shielding - only with the kind of attention it gets…
*****
Over my time I have used many 20+ and even 30+ year old magnets. To be honest those older magnets were often great. The magnet we decommissioned suffered from a slightly larger drift, but for short experiments had amazing homogeneity. Most of the magnets we currently have are under 20 years and will have their o-rings changed when we do our move simply as part of the process of de(re-)energising them, but none show signs of having an issue with the o-rings or vacuum.
*****
END OF COMMENTS.
--
Sameer Al-Abdul-Wahid, Ph.D.
Manager, Nuclear Magnetic Resonance Centre
University of Guelph
50 Stone Road East, Guelph, ON N1G 2W1, Canada
Tel: 519-824-4120 x58914
Web: http://nmr.uoguelph.ca/
Received on Wed Oct 23 2019 - 11:19:38 MST