Hi Ulli,
I have a few thoughts:
Do you have MICS running on that spectrometer? It is possible to save a long-term
log from that and send it in for analysis. I know the Bruker guys have a utility
that will estimate the boil-off of both cryogens (if you have an N2 level sensor
as well) and could potentially see a change in the N2 boil-off as well. Having
said that--If your magnet was installed in 2004, I don’t think N2 sensors
were standard back then but you conceivably could have added one with a console
upgrade. If you don’t have an N2 sensor then the MICS log won’t help.
It would confirm that the He boiloff has increased but that’s it.
Having said that—if you don’t have an N2 sensor, how are you
confirming that N2 consumption has not increased?
Is the SLCB calibration correct? If you plug in either the black calibration
plug or the other one and run a helium level measurement, do you get the
sensible 0% and 100% readings when you measure the helium level? Do you have
a flow gauge on the Helium exhaust? If so, is that showing an increase
corresponding to the change in your data? I would think that a tripling
in the boiloff would be easy to see on a flow gauge.
Icing on the helium sensor I think is unlikely. That usually manifests itself
as a period of time that appears to be zero boiloff followed by a jump in the
helium level. Changing the slope of the boiloff curve is not something I would
expect if it were ice on the sensor.
Given the plot of helium consumption, my first guess would be that the vacuum is
deteriorating, and since your magnet is around 20 years old, that’s not
impossible. It is possible to pump on the magnet while it’s energized so
it might be worth contacting the manufacturer and seeing if they would recommend that.
If the o-rings need replacing, though, that would only be a temporary solution
without rebuilding the magnet. However, if that did improve things temporarily,
that would be a good indicator of what the problem is.
Cheers,
Mike
> From: main_at_ammrl.groups.io <main_at_ammrl.groups.io> On Behalf Of Ulrike Werner-Zwanziger
> Sent: Wednesday, December 6, 2023 12:46 PM
> To: main_at_ammrl.groups.io
> Cc: Josef Zwanziger <Josef.Zwanziger_at_dal.ca>; Ulrike Werner-Zwanziger <ulli.zwanziger_at_dal.ca>
> Subject: [AMMRL] magnet helium evaporation increasing steadily
Dear Colleagues,
May we please draw on your expertise for help with our (700 MHz) Bruker magnet
problem? Here is the issue: Our daily evaporation rates have continually
increased starting about March 2023. I show a graph to that effect at the
end of this e-mail. What I plot there is the average slope between magnet
fills. From when the magnet was set up in 2004 until about March 2023 the
slope of the helium content has been on the order of -0.33 percent per day.
We are now at about -0.8 % per day tending towards 1% per day..
We are stumped on what is going on and how we can get back to a lower evaporations.
Here are some tests, observations and data we have:
1) Vacuum not getting worse: The nitrogen consumption is unchanged. Unless
the helium coil is in its own vacuum chamber, we think it is unlikely that
the vacuum is deteriorating. Do you agree?
2) Cryoshims are not getting bad: Without any RT shims, the liquid lineshape
of a static liquid sample is about 93 Hz wide, FID lasts 0.5s, which is
plenty good for our solid state NMR. Using our "latest" shim file (you will
laugh, from March 2022) the lineshape easily resolves 6 Hz. The shims are
good for ssNMR. We checked every single time and had no reason to change the
RT shims. Also, if a cryoshim would have gotten bad, I would have expected a
short burst and then no change again.
3) Shim coil temperature? Without any RT shims, the coil temperature reads
25 degree C. With the "latest" shim file the temperature reads 26 degree C.
4) EAPD : We took it out of the system for a while (one week or so). The helium
evaporation does not change.
5) Helium recovery: The helium liquifier system was installed in Feb. 2022,
more than a year before the evaporation rate increased. In our rather noisy
data analysis and limited data, our recovery fraction has not changed, meaning
we are recovering a lot of the increased evaporation. (Thanks! We would be in
even more trouble than we are in already).
6) Overpressure: Because of the helium recovery system the 700 MHz magnet was
always at slightly elevated pressure, compared to atmosphere, due to the liquifier
header pressure. I don't see how internal frost could have built up when the
system was bleeding helium out rather than getting air in.
7) SLCB sensor board, Yes, our sensor board in the BSMS was flaky, for years
already. We always had to take a big average (5 readings) to get a sensible
state of the helium level. However, even completely unhooking the sensor line
from the magnet (terminating the cable with the black plug) did not make any
difference to the evaporation rate (testing for about one week). We recently
exchanged the board and I am impressed how reliable the measurements are now.
But also, the evaporation rate has not improved.
8) O-rings: A Bruker engineer was here to install a new spectrometer for an
other lab. He came to our lab in preparation for a new installation, which
we will get soon. He looked at the O-rings on the fill stack and found them
to be in good shape. We did not look at O-rings on other stacks.
9) Unfortunately, he also mentioned that our ceiling is not high enough to
take out and defrost the sensor. Maybe we can move it up a bit and try to defrost
it then, in the hope that that does not introduce more frost than can be removed.
It seems to me a risky maneuver overall, not just because of the electrical impact.
10) Leak testing: We have tested for helium leaks as much as we could and found
that the Bruker one way valve leaked out some helium. We have now added teflon
around the joints. Since the system is on helium over-pressure to the outside
air pressure that should not have caused frost to get in?! A handheld leak
tester was purchased recently, but had to be returned . We hope to use it
for detailed tracing when we get it back.
11) What happened in March that could have brought about such a change? Frankly,
we don't know. Around that time, we had our one year helium liquifier service.
BTW, our other magnet behaves just like before, so we don't think
that the liquifier system has introduced a problem. Also, we did Variable
Temperature experiments, which for years we have not done, but the temperatures
were not dramatic. Since then, the VT unit is not connected to the probe heads.
Do you have any suggestions of what we can do to get back to the much better
evaporation rates? It baffles us that the rise in evaporation rate is so slow
over more than half a year by now. Do you have any suggestions on what could
go on here and how to remediate it?
Has anybody seen anything like this before and what did you do about it?
Thanks for your insights.
Happy resonating.
Ulli
--
Dr. Ulrike Werner-Zwanziger
Adjunct Professor
Department of Chemistry, NMR-3,
Dalhousie University
6274 Coburg Road
PO BOX 15000
Halifax, NS B3H 4R2
e-mail: Ulli.Zwanziger_at_Dal.ca<mailto:Ulli.Zwanziger_at_Dal.ca>
Tel. 1 902 494 8085
FAX: 1 902 494 1310
-=-=-=-=-=-=-=-=-=-=-=-
Groups.io Links: You receive all messages sent to this group.
View/Reply Online (#843): https://urldefense.com/v3/__https://ammrl.groups.=
io/g/main/message/843__;!!PvDODwlR4mBZyAb0!XYC5WnJl9MvEWPoUK49zlYGoztbZgK4Y=
dMG3n3Hcl0RAx_wlQwJt4a_crSV0IdHsKsvUOu4PoALtDpdqgg$
Mute This Topic: https://urldefense.com/v3/__https://groups.io/mt/103020605=
/7559972__;!!PvDODwlR4mBZyAb0!XYC5WnJl9MvEWPoUK49zlYGoztbZgK4YdMG3n3Hcl0RAx=
_wlQwJt4a_crSV0IdHsKsvUOu4PoALzb039mg$
Group Owner: main+owner_at_ammrl.groups.io
-=-=-=-=-=-=-=-=-=-=-=-
Received on Wed Dec 06 2023 - 12:19:27 MST