Hi Margaret,
I know we’ve discussed some of this before, but I didn’t know (or forgot)
that you were using such a low transfer pressure. Why? Do you keep it that
low for the whole transfer?
We depressurize our 150L dewar from 6psi down to 3psi for transfers. The fuller
the production dewar is, the longer this takes. When it is maybe 60% full or
less it will take a couple of hours, so I start the depressurizing remotely
when I get up, and by the time I am into the lab the system is ready to go.
When it is fuller than that we need to activate the “fast depressurize option”.
Initially I was scared by the “May cause helium loss” flag, but once we
discovered it isn’t really lost, just boils off and is collected into the MPS,
we routinely select that option if we want to speed up the depressurizing process.
Maybe 5L or so boils out. That can still take a good hour- and it always creates
some ice around the top of the production dewar (which I do keep meaning to ask
about… it melts when the fast depressurize is over).
Once the transfer starts our system is supposed to push the dewar pressure
back up to 3.5 psi but depending on a wide variety of factors (and particularly
if the purifier needs to be regenerated) it will either quickly ramp up to that,
or slide down to 2psi and build back up a little as time goes on. We did play
around with the starting transfer pressure a fair bit when we first got the
system, and initially something wasn’t right in the way the system adjusted
pressures during a transfer, but QT made some corrections to the software and
now all is good.
I would suspect part of the reason for a lot of gas generated during the transfer
is the slow rate? At one point, long before we got the recovery system, I started
getting worried that there was a problem with the 800 helium transfer line as
for each dewar of liquid helium I ordered, less was ending up in the magnet.
Turned out that over time I’d been dropping the pressure that I was pushing
the liquid out at. When I went back up to ~ 2.6psi vs 2.0 psi that I had drifted
down to there was much less loss.
Another thing we have observed is that if we start with a pretty much empty, but
cold dewar there is way less gas generated. We used to be really worried about
letting the liquid level in the dewar get too low. But now we just use it all
up (well, pulling up the magnet transfer line an inch or so, just in case) and
if we don’t get around to transferring in again until the next week, no problem.
Tara
*********************************************
Dr. Tara Sprules
0.6 FTE, Working days: Monday, Wednesday and Thursday
QANUC
Quebec/Eastern Canada High Field NMR Facility
http://www.nmrlab.mcgill.ca
phone: (514) 398-1721
fax: (514) 398-8254
3420 University St., Rm 023
McGill University
Montreal, QC, H3A 2A7
SHIPPING ADDRESS:
Department of Chemistry, McGill University
Rm. 41 801 Sherbrooke St. W
Montreal QC
H3A 0B8
*********************************************
> From: main_at_ammrl.groups.io on behalf of Eastman, Margaret via groups.io
> Date: Monday, October 21, 2024 at 10:53 AM
> To: main_at_ammrl.groups.io <main_at_ammrl.groups.io>
> Subject: [AMMRL] Quantum Technology all-metal helium recovery system questions: transport dewar and transfers
AMMRL:
We have a Quantum Technology all-metal (bagless) helium recovery system and
started running it earlier this year. Although we are technically out of the
start-up phase and have experienced a good bit of the learning curve, there
are still questions and concerns. To anyone else who has such a system that
I have not bothered already:
1. Our mother dewar size is 250 L. I wonder if we are the only ones with
this size mother dewar, or do other systems have that size? I’m particularly
wondering if some of our troubles with transfers are due to this mother dewar
size, since the standard size is 150 L. For transfers to the transport dewar,
to get down to 2 PSI or less pressure in the mother dewar takes about 5-7 hours.
We have so far most often used a target of 1.8 PSI, and this has taken 5.5 –
6.5 hours. Is this what you see also, or not? I’m curious about whether this
relates to mother dewar size, so either way, what is the mother dewar size in
your system? If your wait time for a transfer was long, but you found a way to
reduce it, how did you do that?
2. Our transport dewar has two necks on top (meaning those pieces that extend
up above the top surface of the dewar, where connections are made). One central
neck is about 3.5” diameter and a second off-center neck is larger than that.
Measurements vary but putting a round number on it we get about 7 L/day loss from
this transport dewar, compared to a specification of about 2 L/day. The high
boil-off brings a lot of troubles, such as more helium must be transferred to keep
the dewar cold, maybe for just one week, while that amount would keep it cold much
longer with lower boil-off, making for extra transfer work. I suspect it also has
a detrimental effect on filling the transport dewar efficiently. I’d like to
hear from all of you who have this type of dewar, to know what sort of boil-off
you get and see if this design sometimes works or in general does not meet the
spec. Necks are not so well insulated as the main part of the dewar and the usual
approach seems to be to have only one and make it as small in diameter as practical.
If you have a single-necked dewar, do you have a liquid level gauge? I’m including
a picture of the dewar top (our liquid level gauge is on the bottom right in the
picture):
1. Continuing about transfers, we are still struggling to get them to go
well, and quickly. If anyone has tips on setting the parameters, or how to
adjust them during a transfer to fill at a good pace, or other procedural
choices, we would appreciate advice. Not only do the transfers take too long,
one hour or more, but far too much helium boils off during a transfer. QT
suggested it may take 50 L of boil-off to fill a warm dewar, but our last
transfer, which increased the volume in the (cold) transport dewar from 21L
to 138 L and took 75 minutes, boiled off about 63 L.
So, I’m asking a lot, but maybe I can offer one little bit of info that might (?)
have value to someone: We started out with a recovery inlet pressure of about
0.27 – 0.30 PSI, which meant that we would often have a pressure inconveniently
close to the standard start pressure for the recovery compressor (0.30 PSI).
(Our system has a pump for moving helium into the liquefier during regular daily
boil-off, and the recovery compressor that can handle magnet fill flash boil-off.)
We also had too much venting due to over pressure during transfers, and too strong
a cold blast coming out of the magnet inlet ports at the end of fills. We have a
pressure control valve in the recovery compressor, for us PCV302, and the first
advice was to back its adjustment out to lower the pressure, but we backed it
out all the way and still had high pressure. Recently QT offered a new solution:
a new spring for the valve that is much softer. It was one of the most rewarding
moments in my recent life when I popped that new spring in and found that with it,
we have ~0.20- 0.23 PSI inlet pressure, often right at the expected book value of
0.22 PSI! Ahhh- I think my blood pressure went down too. But now it is going up
again over transfers and the transport dewar.
Margaret
-=-=-=-=-=-=-=-=-=-=-=-
Groups.io Links: You receive all messages sent to this group.
View/Reply Online (#1776): https://urldefense.com/v3/__https://ammrl.groups=
.io/g/main/message/1776__;!!PvDODwlR4mBZyAb0!R_-w2_bRgISRb986DdaAy1RjMPKdot=
5NlVrhqJp4aD9lbsFiSZV2S9SWWA_MXqhWeXlv6IPs1CNa3ZcDVsecUkU$
Mute This Topic: https://urldefense.com/v3/__https://groups.io/mt/109133895=
/7559972__;!!PvDODwlR4mBZyAb0!R_-w2_bRgISRb986DdaAy1RjMPKdot5NlVrhqJp4aD9lb=
sFiSZV2S9SWWA_MXqhWeXlv6IPs1CNa3ZcDlqNydMc$
Group Owner: main+owner_at_ammrl.groups.io
-=-=-=-=-=-=-=-=-=-=-=-
Received on Mon Oct 21 2024 - 14:56:42 MST