It was not clear - Did you check for bad o-rings on the top of the magnet
itself Look at the top of helium towers and check the safety relief valves.
We have had o-rings go bad in both places.
David
Associate Professor
Dept. Of Pharmacology
University of Colorado
School of Medicine
Tel (303)-724-3600
On Apr 2, 2026, at 1:49 PM, Monika Ivancic via groups.io wrote:
Hi there fellow spinlanders and helium recovery gurus!
I've been meaning to post this for a while now, though hoping to find a
solution prior to this, to no avail.
Here at the University of Vermont (UVM) we have a small scale helium recovery
system, manufactured by CryoMech, which is capable of producing 15L of liquid
per day (LHeP15). This system was installed in summer 2021, and aside from
several hiccups along the way, has been functioning well. However, one major
issue that we've been dealing with is that we haven't been recovering steady
state boil-off from the two NMR magnets in quite a while now (at least several
months). We have two NMR magnets connected to the recovery lines (a Bruker
Ascend 500 and an Oxford 500) and each magnet has it's own manifold, with
heat exchangers, etc. The copper piping leading into the recovery room, which
is in the basement, was installed when the building was built in 2017. The
other magnet that is connected to the system is part of an MCD, which is
cooled down and ramped up only when the researchers are collecting data, and
we also have an EPR that is hooked up, though used only very occasionally.
According to the magnet book for the two NMR magnets, we should be boiling
off about 0.6L of helium per day, or 4.2 per week total. Unfortunately
when I plot the totals in our recovery system, they are not increasing, and
the curve remains flat.
So....where is this helium going?
When I do the magnet fills, and there's a lot of pressure in the system, I
recover about 25L from filling both magnets. During the fills, I open up
the large flow meter for the large volume to flow through, while during
steady state the path goes through the smaller 'normal boil-off' meter. We
have a high quality helium gas detector, and I've wanded the manifolds on
numerous occasions. I have also followed the path of the pipes searching
for leaks, especially since there are at least two other labs that have
connections to the piping, although no instruments are hooked up there.
The ultimate test that we did to figure out if the piping is leaking, was
isolating the piping from just beyond the NMR magnet manifolds, all the
way into the recovery room in the basement, and pressurizing it with
balloon grade helium gas to 21psi. The 21psi pressure held for over 24h,
which means there's no leak in the piping leading to the basement. We've
wanded from the shut-off valve in the basement to the manifold where several
lines for the recovery system connect (leading into the bag, coming from the
transfer dewar) and haven't detected a leak there.
Our system does have a recovery bag, which sits in a room next to the recovery
system room, and an insulated hose (formal name?) leads into the bag, while
another one goes from the bag to the gas compressor, where the bag gas gets
compressed into cylinders. We have wanded the bag when inflated, wanded the
connections into and out of the bag, we've observed it fill up until it hits
the laser at the top, and didn't detect helium in the air when the bag is
completely inflated. Some oddities of the bag are that a couple of years
after the system was installed, somehow the bag was reseated and after this
it would never get as big as it did when it was brand new. Because of this
we lowered the location of the laser. Back when the system was new, we'd
let the bag inflate all the way, and once it would trigger the laser at
the top, the compressor would go on and the bag would be emptied. At some
point, for some reason, we thought that the bag was losing helium when
inflated, so a decision was made to 'bypass' (or empty) the bag daily during
the week. At that point we believed that that solved our recovery issues,
but when looking at the plots of the totals in the recovery system, it
still doesn't show that this is the case.
So... I am putting this out there to you, many of whom have a recovery system,
to solicit ideas of what I could try to figure out where this 0.6L per day
are going. Another test we thought of trying is to isolate and then inflate
the bag to see if we detect an obvious leak. Could this be done by just
shutting the compressor off, so that it doesn't start processing the bag
once the bag is full? And of course shutting a valve before the gas enters
the bag, and using balloon grade helium to pressurize the bag.
The steady state boil-off has got to be going somewhere. The 'normal' boil-off
flow meters do give a reading and this reading has been the same since the
system was installed. I'd appreciate any ideas/suggestions that you may have!
Thank you and hope to see you at the Asilomar ENC very soon!
Cheers,
Monika
-=-=-=-=-=-=-=-=-=-=-=-
Groups.io Links: You receive all messages sent to this group.
View/Reply Online (#3333): https://urldefense.com/v3/__https://ammrl.groups=
.io/g/main/message/3333__;!!PvDODwlR4mBZyAb0!U1hbG1pse1jAdFpQenk9NJCLHMVMqO=
91HHXUnJI73sW2cyjL8_Z516a4aRUF6PoT7pBfBbtRjF3pRUZCiNT89OJR4nzY$
Mute This Topic: https://urldefense.com/v3/__https://groups.io/mt/118639575=
/7559972__;!!PvDODwlR4mBZyAb0!U1hbG1pse1jAdFpQenk9NJCLHMVMqO91HHXUnJI73sW2c=
yjL8_Z516a4aRUF6PoT7pBfBbtRjF3pRUZCiNT89N9vTMkS$
Group Owner: main+owner_at_ammrl.groups.io
-=-=-=-=-=-=-=-=-=-=-=-
Received on Fri Apr 03 2026 - 07:37:14 MST