Hi,
FWIW:
1. I would definitely invest in a ball-float flowmeter, and put it at some convenient place in the VT line. It's very helpful if you want to know what's going on.
2. From what I've measured, the needle valve restrictors in the PNK board deliver the rated software flow rates into free space. Attaching hoses, probes, etc. introduces additional downstream restriction that further restricts flow, so the measured rate can be quite smaller than the set rate. At high flow rate settings, I've found that the needle valve restriction can be comparable to the downstream restriction, so the physical flow is about half of the software setting at higher settings. The flowmeter will tell you.
3. Because cryoprobe sample cooling is caused by radiation, it is best mitigated by using flow rates as high as possible, for maximal heat transfer. You can run it just below where the sample just begins to lift, as indicated by the appearance off lock level instability. Using heavier ceramic or kel-f spinners can help here. If you have a lot of samples, an inexpensive alternative is to weight the normal blue spinners to the ceramic or kel-f weight by placing Tung-Fu (a tungsten-based weighting clay used by fisherman) inside the spinner rim. But of course, higher flow rates will ice up the BCU-X that much sooner.
4. The dew point specs for the BCU-X are very stringent, requiring moisture below the ppm level, below the range of most air dryers.
5. It's possible to run "dry" air until ice forms as you describe. If you can monitor the flow you can periodically compensate, but a mass flow controller will do this automatically, until you have to stop and purge the ice from the BCU-X.
6. An alternative is to use boiloff from "gas pack" LN2 dewars, which provide very dry N2 gas by evaporation. I don't know whether these will last for days for cryoprobe experiments, but I do know you can run gas through a BCU-X at MAS flow rates for at least several hours. I suppose you could calculate how long you'd expect a given tank to supply the N2 by comparing measured flow with the LN2 volume multiplied by the 700x gas-liquid expansion factor. Or, if you have 2 of these or a spare N2 gas tank, you can swap and refill as needed. If you use a gas-pack, be prepared to deal with the heavy water condensation from the air that will drip off the sides of the dewar.
Best of luck - Gerry
-----Original Message-----
From: Torbjorn Astlind [mailto:torbjorn_at_dbb.su.se]
Sent: Wednesday, November 09, 2011 1:11 AM
To: ammrl_at_ammrl.org
Subject: AMMRL: Gas flow for variable temperature?
Hello,
A little odd subject:
I would like to have better control over the gas flow for the
temperature, especially in our cryoprobes.
We have several different Bruker NMR systems, two of them with BCU-X
(-80 C), but common to them is that the flow passing the sample is
known only with low precision.
The flow is set by a combination of needle valves, and calibrated in the
factory years ago.
Since the flow resistance can vary, taking the BCU, and the probe/sample
into consideration, I am afraid that the difference between the sensor,
and the sample temperature is not stable enough for some experiments
over several days. Mainly I think that flow resistance in the BCU-X is
increasing with time from deposit of small amounts of frozen water from
the gas supply. ( it is dried with a high quality drier, and the N2
separator)
The temperature difference between the sensor, and the sample, including
calibration errors, in one probe is 6.5 C and thus even a small change
in flow might influence the sample temperature.
Any views on the temperature gradient across the sample, and its
variation with gas flow?
To measure the flow with god precision I am thinking of adding a
massflow meter just before the BCU.
Any views, or recommendations on this?
Torbjörn
Received on Wed Nov 09 2011 - 14:14:27 MST