Hello:
I would like to thank everybody taking your precious time to read and answer my questions.
The answer is clear:
Don’t replace the magnet seal unless you have a reason, such as higher helium boil off rate and shorter helium hold time.
Even if the helium boil off rate is higher than normal, pumping the magnet vacuum with a turbo-pump may also help to bring the helium boil off rate back to normal. This is less expensive than replacing the seal.
I am happy to hear that many magnets have been at field for more than 20 years and their helium boil off rate is still normal. Since I do not see any significant change of helium boil off rate for my magnets, I will not do anything at this time.
I have attached the original responses because many other people are also interested in this topic.
Thank you again for you input.
Weixing Zhang
$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
The seals can last twenty years or more, if nothing has happened to them. As long as Helium consumption does not increase it is not worth doing anything to the magnet. If there is a increase in He consumption then I would first pump the magnet to see if this can bring it back down. If this does not help you can consider replacing the seals.
Seals can be damaged if for example liquid nitrogen is being splashed on to the flange of the magnet for a longer time.
Clemens
$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
Are these Bruker magnets or Oxford or Magnex? If Bruker, can you please tell us the order numbers of the systems so that we can check the type and build of the magnet. Generally speaking, the time to replace the O-rings comes when you see a significant increase in N2 boil-off and/or He boil-off.
Best wishes, Mike !
$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
If there is no significant difference (say 20% increase of the boil off rate),
simply do nothing! And even if you observe a significant increased boil off rate, first try to recover the vacuum. Changing the seal is very expensive and very risky.
> What is the longest time that the seal can last?
There is no number available. We have a magnet with a 25 year old seal.
Greetings
Rainer
$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
I can tell you from experience that we only changed seals on a functioning magnet when there was a reason - like a magnet move. That said you may get some numbers back so that you can plan. I think that seals on modern magnets last much longer and you don't often hear about an increase in the boiloff rate and a magnet quench. Our Inova 600 has been a happy magnet for 1o years. We do monitor the rate at which the level goes down. Magnet engineers tell me that we should all be monitoring the boiloff, which of course depends on the barometric pressure, but over the long term you should see an upward trend if there is a problem with the magnet.
It really costs too much for most people to warm up a magnet, etc, when they don't absolutely have to.
Sara
$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
We have an Oxford 500/54 magnet that was commissioned in 1989. It's He and LN boil off rates have not changed in twenty years.
I suppose that there are environmental conditions that could degrade the seals. Sloppy LN fills where the liquid spills on the flanges is also a concern.
We check our boil off rates weekly. I would begin to become concerned if either of these rates increased by a factor of five.
Maybe this helps.
Walt
$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
Would you please share your responses with me as I'm also curious what the varying opinions are on this subject.
Tanya Young
$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
As long as everything with the magnet is stable, like the He/N hold time, boil off rate, there is no reason to replace the seals in the magnet.
If you relocate the system, or notice some problem, then a seal change is recommended.
If you a have a quench, but do not loose vacuum inside the magnet, then you have to decide if you want to incur the extra time/expense of a seal replacement. I use 10 years as a breaking point. If the seals are over 10 years old I would probably do a seal change.
I recently had a quench on a 400MHz system that was at field for 6 years. Since it lost vacuum, I replaced the bore tube seals, but did not replace the belly band seal (you need a hoist to do that one).
I also have a 600MHz round bottom magnet that (knock wood) has been at field for 17+years. The boil-off is around 16 lph and the refill interval 2 months and it uses about 50 liters. There are no plans to do anything to the magnet.
Sincerely,
Joe
$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
If cryogen boiloff remains low, there is no need to reseal a magnet.
Magnets can stay at field for more than 20 years.
Jeff
$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
We have a magnet that has lasted 15 years with replacement of the seals, but I have heard that some have lasted 23 years.
Ted Burkey
$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
what I heard is that you should not touch it as long as it's working. But would like to hear what other people say.
We have a magnet that has been discharged and charged back, and then its field is falling fast.
$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
If you could, please share your findings with me from this question. I have 4x magnets in my lab, two of which are 20+ years old. I've had seal work done on the other magnets in the lab when their vacuum's were called to question, but I've had no reason to question the vacuum's on the two older magnets to this point.
$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
We've just moved some magnets and had to make the same decision. We replaced the seals on an Oxford 500 that was installed in 1992, so ober 16 years old. The seals on the magnet were ok, just hard, but the TMC leg seals failed after the move and had to be replaced. We also replaced the seals on a Bruker widebore magnet that was installed in 1996. There was significant corrosion and discolouring of the o-rings.
This is because condensate dripped off the N2 outlets and down the sides of the magnet on particularly hot and humid days. The Bruker magnet did not have the heat exchange tubes on the vents. (it was also not under my supervision at the time)
I hope these data points help,
Tim
$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
I have an Oxford(500) magnet that was energized in 1992 that is still stable at field, and have no plans to bring it down to replace the seals at this time. Several years back, I did have an Oxford(270) early model that had a leaky vacuum valve. The boiloff rate increased to at least 3 fold and was increasing rapidly daily. I brought the field down with a diode pack, and had Oxford replace the valve and O-ring seals and it was stable for several years until we retired the system.
Ed Ezell
$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
You don't have to replace the seal if the magnet is still good.
I have one here, it was installed in 1988, so far it is still very good.
swu
$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
We have here a magnet that is 25 years old and has never had the seals changed. Our criteria for changing is cost. We have lost some vacuum due to the seals or for whatever reason. Thus, the He boil-off has increased significantly. Yet, it is still cheaper to fill He on the 400 on a monthly basis rather than spend > $10,000 to de-energize, replace the seals and re-energize a magnet that is already near the end of it's life. It's all about economics and the answer is going to depend on what reduction you expect in your cryogen costs compared to the cost of replacing the seals over your expected life-time for the magnet. In our case, my rough estimate is that it would if our boil-off returned to spec, it would take about five years to recoup the cost of the re-build. I'd certainly like to know the answer to your last question, especially in relation to the expected life time of a magnet in general. In my limited experience, a 25 year old magnet is the oldest 'high-field' instrument that I've seen in operation. It seems that caring for an NMR is similar to caring for an automobile. When one is considering the costs and benefits of repairs and maintenance, one must think about whether they are going to be driving that same car in 5, or 25 years. On the other hand, we do have "insurance" against catastrophic loss of the seals in the form of quench insurance from the manufacturer.
Hope that helps.
Matthew Devany
$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
When you receive your responses could share? We have a Bruker DPX300 and DRX500 both are 11 years old. Peter Ziegler at Bruker told me 15 years is a long time on the seals. He said they would just keep increasing their boil off rate until you couldn't keep He in.
$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
We have 20 year old magnets that are fine, but I have also seen magnets develop leaks. Unless you have a reason to suspect a leak, like increased boil off or condensation (besides the LN2 port), don't touch it.
$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
In my lab, we have an old Oxford/Bruker magnet that is over 20 years old. It had been losing He at faster and faster rate, eventually requiring a He fill every two weeks (we fill at ~50%). We decided to re-seal the magnet (a costly and time consuming process, but cheaper than buying a new magnet). The magnet was de-energized, re-sealed, then brought back up to field.
For a 'normal' magnet, He fills should be carried out every 3-4 months.
Now we are filling the magnet about once a month, which is still more frequent than normal, but better than every two weeks.
Rather than having the magnet resealed, you could try hooking up a pump to the valve on the magnet to drag out the life of the magnet. You could contact the manufacturers to see what they recommend based on your helium loss rate and age of the magnet. Ten years doesn't seem that old to me, but perhaps your magnet is not very good. What is your He-loss rate?
Hope that helps,
Shelley.
$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
I would be interested in a summary of your responses. I'll tell my understanding, but on these kinds of items, others might be better informed, or add to important knowledge.
Traditionally, seals will last anywhere from 12 to 20 yr. 15yr would be considered typical. I think many labs--as ours has done--replaces the seals when upgrading or replacing consoles, which should occur in a similar timeframe. In any event, you should be fine for another couple years minimum.
You should, if not already, start monitoring your LHe boiloff. Measure the actual boiloff rate via gas flow meter, but also start plotting the amount of LHe used for each fill. If either increases, then you should be concerned that the o-ring and vacuum it is protecting are degrading.
Sometimes, occasional vac pumping on the magnet can extend the time to refurbishing, but sometimes an emergency discharge is needed to prevent a full quench (typically here people will observe a sharp rise in the He gas boiloff rate).
I have heard but not had confirmed that new o-rings may exist that last much longer, and may prevent the need for such rebuilds. One of the items that would be nice to hear more about...
Charlie
$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
I have a spectrospin 300 MHz widebore that was installed in January, 1985 and has very similar He, N2 boil off when installed. Some magnets can go 20-30 years. I also had a 500 MHz spectrospin narrow bore magnet that went for 15 years before there was an accident and the magnet quenched. However, that magnet was performing quite well also.
I would think that if you have no problems, then leave them alone.
Best Regards,
Kurt
$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
Don't worry! We have an OXFORD-400 magnet running since 1992. We never changed the O-rings of the dewar since 18 years.
On the other hand, our 600-Magnet was rebuilt (warming up, replacing all seals/O-rings) already after two years - for our luck on guarantee of the manufacturer (with capital "B"). The told us that they had changed the producer of the O-rings and their material was not suitable.
We noticed this by an increase of the helium consumption (i.e. worse vacuum).
In my opinion this is the only criterion. Notice the He-Level once a week, draw it in a sheet vs. time axis - if the slope of the curve does not change : all is OK.
If the slope gets larger in time - decide for your own, when the limit is reached (in dependence of the frequency of refilling He - ie. increasing costs) - Nobody wants live with an refill interval of 2..3 weeks, ain飔?
Yours
findy
$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
Watch and track your boiloff rate. If you start to notice a long term increase, then change the seals. If you move the magnet, change the seals then.
Martha Morton, Ph. D.
$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
Our Varian 400 MHz has not had the seals replaced for almost 20 years. Our boil off rates for nitrogen (180) and helium (2-4) are normal for this Oxford magnet.
Keith
$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
The NMR vendors will tell you that the average lifetime of a set of seals is around
12 years. Of the 6 magnets I have here at Louisiana State University, 3 of them have
been at field longer than that. In fact, one of them has been at field since about
1988 AND I moved it once while it was at field with no ill effects.
All I do is carefully watch the helium and nitrogen consumption. Until you see a significant
increase in the use of cryogens I would leave things well enough alone.
I hope this is of some assistance to you.
Kindest regards.
W. Dale Treleaven, Ph.D.
$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
The seals can last forever, under the right conditions. Do not consider replacing them unless there is a known leak, because replacement requires the magnet to be de-energised, warmed, and disassembled. This is expensive and should not be considered unless there is some other problem which requires magnet de-energisation and warming. Over time, helium gas can dissipate into the magnet vacuum insulation. This causes boiloff to increase slowly over time. This condition can be relieved by repumping the vacuum, and does not require seal replacement.
I would not be concerned until the boiloff increases beyond the magnet spec.
At that point, I would connect a turbopump and helium leak detector to the magnet vacuum port. While renewing the vacuum, check for leaks. If leaks are located, then you need to replace the seals. If leaks are not located, then just continue pumping until the boiloff goes down.
Cheers,
Jerry Hirschinger,
$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
I have one magnet that has been at field 25+ years. No noticeable change in boil-off. So it is possible that they can go a long time. The main thing to watch is be sure that you have good ventilation as any build-up of He in the local environment can diffuse through the seals (even new
seals) and spoil the vacuum.
-Kirk
$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
We have two magnets that are 16 years old and going strong. I think you just wait for the magnet to quench. If you carefully monitor the boiloff rates (N2 and He) you will probably see them begin to rise as air leaks into the vacuum space, but often the process can be very fast and you just come in one day and see that your magnet no longer has any field. Our approach is just to wait and hope for the best. The magnets I'm talking about are 200 and 300 MHz, maybe some other people will have different advice for a 600.
I think 10 years is not a long time and you shouldn't have to worry for a while if you do a careful job of cryogen filling each time. I've heard people say that 13 years is the average lifetime of a magnet.
To replace the seals you will have to de-energize and warm up the magnet anyway, so you might as well wait for it to quench on its own.
Neil
$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
I have had several magnets at field without repair or seal replacement for 20 years . If your He hold time is less than half of the original you might want to consider repumping.
George
$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
Generally, one of three things will indicate a magnet rebuild is
forthcoming:
1. The drift rate changes suddenly.
2. The boil-off increases suddenly.
3. You arrive one morning to find you no longer have a magnet!
Without any of these conditions, it's usually better to think "if it's not broken, don't fix it."
Regards:
Paul Cope
$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
We have a Bruker/Oxford 360 MHz magnet which was installed in 1989. The seals have never been replaced, and the hold time and boil off rate do not seem to have changed.
I would not replace anything until I noticed a change. There are always risks involved in powering magnets up and down.
Cheers,
Chris
$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
We have a 300MHz (Oxford) magnet that was installed in 1988 and still has its original seals - He boil-off rate did increase slightly about 5 years ago and I thought that was the first sign that I would need to get the magnet refurbished. Since then the boil-off has remained constant & in every other aspect the magnet continues to perform well.
In my opinion, only if the He boil-off rate begins to increase or if you have need to de-energise the magnet for any other reason should you consider replacing the seals.
Best Regards
Ian
$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
The short answer is 10-20 years. My guess is that closer to 20 is the answer you'll end up with.
Now the long answer:
In my previous job, we decided to budget to replace the seal every 15 years. If your magnet is boiling off a lot of He, you may want to replace the seal asap. The rise in boiloff rate may be slow, or it may be sudden. In the slow case, just define a threshold for He boiloff rate (eg. 2-5 X the historical baseline rate) and do the replacement when the threshold is exceeded. If the rise is sudden you should plan for near-term replacement.
Hope all is going well,
Bill
$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
Economically, you would replace the seals when the cost of the extra He and N2 over a period of time is greater than the cost of ring replacement. Practically, when an ice bridge thermally connects the he can to the room temperature bore and it starts freezing. Just plot your helium consumption (L/day) over time to get an idea of where your magnet is. We have a 300 that was brought to field in 1987 which still has the same boil off rate.
I *think* its most important to vent the helium boil-off gas out of the NMR room because it doesn't freeze if it gets by the o-rings.
dave scott
________________________________
Email Disclaimer: www.stjude.org/emaildisclaimer
Received on Thu Jun 11 2009 - 10:46:03 MST