Hi all,
Here's the summary from my Dec. 15 query, "doing Helium recycling
correctly."
There appear to be a couple points of consensus:
1) recovering helium can introduce field stability/homogeneity
problems, and may not be suitable at all sites
2) it can be done well under certain circumstances, and this usually
requires placement of a manostat in between the magnet and helium
recovery system; the manostat uses a precise pressure gauge and a
precise chip-controlled valve to keep pressure in the helium dewar
constant to high fidelity.
Greg Labbe at the University of Florida has some wonderfully explicit
photos, diagrams and plans showing various successful configurations
of one-way valves, pressure gauges, etc.:
http://www.phys.ufl.edu/~labbe/photos/ucla
Aside from the value of conserving one of Earth's nonrenewable
resources, the decision of whether to recover helium rests on cost,
and this is going to vary from site to site. It seems to make most
sense in institutions where the NMR facility is close to physics labs
that use a lot of helium, and when the NMR systems use relatively
insensitive low-field magnets. NMR labs with lots of high-field
magnets may seem to be good candidates for recovery systems, but the
sensitivity of 500+ MHz magnets to fluctuations in helium pressure has
led several labs to stop their recovery programs.
It seems the key to success is to use a manostat, but these are
expensive relative to the valves and gauges one could otherwise use to
hook up the . If you have a helium recovery system available, and you
have an old magnet, it may cost much more to purchase the manostat
than you will save on you helium bill. However, if you are purchasing
a new magnet, adding a manostat should be manageable because it's
coming from a large capital equipment fund/grant rather than an
operational/repair part of the budget.
For my own facility, I will determine whether our 600 needs a manostat
for stability purposes, and if it does, I'll hook up the vent to the
recovery system and dial up the setpoint pressure higher than the
highest expectd recovery system's pressure. Our lower-field magnets,
which are primarily used for short-term organic compound
characterization, I'll leave hooked up the recovery system sans
manostat.
Thanks to everyone who responded. You've been a great help.
- Josh
ORIGINAL POST:
I've recently been noticing some lineshape instabilities in our 600,
and I'm wondering whether it could be caused by pressure fluctuations
associated with the helium recovery system we have in place. I suspect
that we either have a leak somewhere, a faulty valve, or our perhaps
our system design is flawed.
I'd appreciate your insights regarding how to properly configure our
helium recovery system to minimize artifacts. Please note that our
facility shares this building-wide system with a number of individual
labs, some of whom aren't doing NMR or EPR, and some of whom remain
hooked up to the system when they fill helium.
Thanks.
RESPONSES:
Josh, Do you have a manostat on your system? If so, you would probably
want it set at a pressure a bit higher than the highest back pressure
you
get in that helium line. Normally they are set on delivery to a
pressure
a bit higher than the highest expected barometric pressure in the area
where the magnet is installed.
***
I solved a similar problem at XXX by buying one for an old system,
and it worked very well. I surveyed the NOAA website to see what the
high and low pressures were over the last few years so I could set a
good pressure. When we got a new 500, it came with one, and the Varian
people told me they don't sell 500+ magnets without them any more. It
seems that Bruker doesn't have that policy, but I'm not certain about
that.
See:
http://cdo.ncdc.noaa.gov/qclcd/QCLCD?prior=N
***
Maybe it is not caused by the Helium. I had once the lineshape
instabililty which caused by the magnet leg damping system. It is
worth a try.
***
I know we used to have a He recovery system at XXX when I was there
many years ago but it was discontinued initially for the high field
(500, 600) instruments due to pressure fluctuations, and later
decommissioned due to high cost of maintaining the compressor.
I recently looked into helium recovery system for our new lab (we
moved due to building renovations) but after talking to applications
lab and magnet labs at Bruker in Germany and Switzerland I decided
against it ( I was going to give recovered helium for free to physics
who will get new compressor/liquifier).
Reasons for this was Bruker only recover at magnet testing facility,
none of the applictaions or magnets used for testing probes and
consoles were connecvted due to fluctuations and performance issues so
I thought that if they can't get it right, we would have no chance!
Perhaps you could install a pressure manostat on your systems to
prevent back pressure and other fluctuations, and still collect helium?
***
This is all anecdotal, and very outdated... it may not be useful to you.
Many years ago (mid-1990s) we had a recovery system at XXX, which was
located in the Physics building some distance away. The Helium was
delivered via 1.5" PVC pipe that went through the steam tunnel. I
didn't
have a 600 at the time, but sent helium from a 200, 100-Widebore-Solids,
300, 360-Widebore, & 500. The system was based on a mechanically
assisted
gas-bag (rubber bladder, the size of a small room) that used a motor/
pully
to assist in inflation to prevent back-pressure. It was somewhat Rube
Goldberg, but it worked. It would raise the bag, maintaining neutral
pressure until it reached the ceiling ('full' condition), and a switch
would
activate a compressor that would compress the helium into a room-full of
high-pressure cylinders. Once they were fully pressurized, the
liquefier
would re-purify and re-liquify the helium. The cost of running this
system
was about 3x the cost of buying helium at market prices (but we felt the
conservationist benefits were worth it).
I had many problems, but the main ones were:
1) failure of compressor/bag motor to prevent back-pressure (this would
cause back-pressure down the line, and fluctuation in magnet
pressure. Most
noticeable on the 500 and 360WB... especially in sensitive experiments
like
NOE-difference experiments or NOESY experiments.
2) Compressor would fail to turn off after emptying the gas-bag, and it
would start pulling vacuum on my magnets (the stacks would ice-up, and
if
not caught could well have caused a quench). This problem is self
explanatory.
Solution to both (don't laugh too hard)... Between each magnet and the
main
pipe, the PVC pipe was cut with a ~1-inch gap. Over this gap, a
standard,
non-lubricated Trojan condom was cut (making it a very thin latex
tube), and
sealed with duct tape on each side of the gap. If the compressor
pulled a
vacuum, the condom would collapse, and seal-off the magnet (blocking the
vacuum). On the other hand, if there was back-pressure, the thin latex
would inflate at very low pressure, and burst before any problems were
caused. This was a very simple but viable solution (except for
ordering a
case of Trojan's through central purchasing.... that was very
interesting).
Most Oxford 600s that I know of uses manostat-controlled helium
pressure...
they are very sensitive to pressure fluctionations. If you don't have a
manostat, I would suggest you might benefit by having such a system
(maybe
more sophisticated than condoms) "downwind" (so to speak) of the
manostat
output. This would also isolate from any vibrations being transmitted
from
the pipe... though I expect your problems are due to cryostat pressure
fluctuations, rather than mechanical vibrations.
***
In my case I remember that I could even notice the magnet swinging. In
your case, I would repeat the lineshape tests for overnight with the
legs down to see if the phenomenon eliminated or reduced a lot
***
I have seen pressure fluctuations inside the helium can on a NMR magnet
cause problems with shimming. On one system the problems were so severe
that there would be a loss of lock when the helium can pressure was
vented. A typical problem is there are vibrations produced from the
"burping" of the vent valve.
Problems of this nature are easily diagnosed by venting the helium can
while observing the lock level.
Oxford magnets had a manostat control system which was used to control
pressure fluctuations of this type. I do not know if these devices are
still available.
***
I strongly urge you to contact Bruker regarding this, even if your
magnets
are not made by them. They have a very large He recovery system in
Switzerland, and they have gone through many useful experiences with it.
They gave a talk about this at last year's pre-ENC user meeting. I am
sure
they could help you save a lot of time and energy tracking this down.
From
what they said, you have to be quite careful setting the right flow and
pressures in such a system. They even said that if the backpressure is
too
large it can cause a quench or even a dewar explosion.
***
We had pressure fluctuations on the 600 when I started working here.
They
manifested more as seeming vibrations than lineshape problems,
although I
think it did make lineshape somewhat unstable as well. In our case, I'm
fairly certain is was a faulty check valve - since I removed it over 2
years
ago the problem hasn't resurfaced.
***
There is a guy named Greg Labbe at the University of Florida who is
very experienced at He recycling. I visited his setup a few years ago
and
he took me to lots of labs on campus that use his facility, including
the
NMR lab in Chemistry. Here is a link to his web site:
http://www.phys.ufl.edu/~cryogenics/
I think he could be very helpful about details.
***
As long as the recovery system pressure is low enough to allow a
manostat
to hold the pressure constant in the NMR magnet, that is the solution.
A bypass system around the manostat for helium transfers is needed.
Here are some photos for different set-ups for recovery from a cryostat.
http://www.phys.ufl.edu/~labbe/photos/ucla
In the schematics Option #2 or 3 the manostat goes in where the
flowmeter
is shown. (replacing or in addition to)
You can see I use standard NW, QF flanges, orings, clamps, swagelock
fittings, Tygon tubing (not latex tubing) brass or Stainless ball valves
stainless steel flex hose and refrigeration charging hose (Yellow
Jacket)
for the piping at the cryostat and magnehelic pressure gages.
We have several 600MHz cryostats hooked directly to recovery through
Dwyer flowmeters and no one has complained about fluctuations yet. But I
understand that you may have a more sensitive situation. But if you are
not already going through a flowmeter with long lengths of 1/4" Tygon
tubing to recovery, you might try it as a first step.
The only pressure release valves I use are the big quench valves already
on the magnet manifold from the factory.
The check valve is at each lab's gasmeter and, if the lab wants to keep
the factory supplied relief exhaust valve in place on the manifold, I
install a "T" in from of it for a transfer recovery line with a shut-off
valve and just run Tygon tubing from it to recovery on the other side of
the shut-off valve.
***
I would suggest taking the system off of the recycling line for a few
days and run lineshape checks over that time. Our 600 (not on a
recycling system) is less stable near the baseline (non-spinning and
on vibration isolation legs) than I would like to see when comparing
single shot transients over a series of acquisitions but the lineshape
looks fine with the averaging of just a few scans.
I have heard of problems like this that were caused by sticky check
valves. Do you track boil-off?
I am interested in any comments you receive on how to set up a good
recycling system. The university I attended as a graduate student had
one operated by the physics department and as best as I can recall it
was the source of periodic problems in the NMR lab.
***
Possibly. We just replaced the pop off valve on ours. Also replaced
the pop off on the LN2 well. We did this because of 2 spontaneous
quenches. I would suggest disconnecting the the magnet from the
recovery system for few days and see what happens.
Josh Kurutz, Ph.D.
Instructor and Senior Scientist for NMR
IMSERC, Chemistry Department
Northwestern University
2145 Sheridan Rd.
Evanston, IL 60208-3113
847-467-1681
fax: 208-978-2599
Facility: www.chem.northwestern.edu/imserc
NMR Blog: www.imserc-nmr.org
Twitter feed: twitter.com/imserc_nmr
Other: www.joshkurutz.com
Received on Wed Jan 20 2010 - 09:03:03 MST