Liquid level sensor help

Discussion in 'The Projects Forum' started by Slicko, Oct 23, 2009.

  1. Slicko

    Thread Starter New Member

    Nov 11, 2008
    7
    0
    I have a project to be completed within around 6-7 months, for the final year of my degree. Basically, the project is designing and constructing a Liquid Nitrogen automatic refill system. I'm only just starting it so i'm currently only on the research phase, though time is going to get pretty tight if I start falling behind.

    I just want to know peoples thoughts on the best way for measuring the level of liquid nitrogen in the tank.[SIZE=+0] [/SIZE]Using a pressure transducer in some kind of bubbler system seems to be the best strategy at the moment, I just don't know how i'm going to implement it into the tank and use it effectively (and making sure it can withstand the -196 C temperatures of the liquid nitrogen). Even simple liquid level systems that I've looked at so far seem too difficult/time consuming for me to make myself.

    I suppose i'm asking if it's feasible to design and build a liquid level system myself or whether I should see about buying one (which I would have to discuss with my project supervisor), and if so, what would be the best one to buy (I have a budget of £100 for the project).

    6-7 months sounds like a reasonable amount of time in itself, though I have 4 other modules to concentrate on, as well as exams, progress reports, literature reviews and writing the thesis for the project.

    Any help or advice on what my best options are would be greatly appreciated :rolleyes:
     
  2. jpanhalt

    AAC Fanatic!

    Jan 18, 2008
    5,684
    900
    I don't have any good answers for you, but do have a question and comment:

    1) Do you need you need to measure the level, or just determine when the level reaches the refill point?

    2) The bubbler seems to be a good approach. We used to use a glass tube with a tulip-shaped top. A rubber diaphram over the top allowed us to probe the LN2 level and you could tell when it was reached by a change in the diaphram with your finger. If you actively pass bubbles, be sure to use nitrogen or helium, not air. Oxygen in air will condense in LN2, and over time, can lead to problems.

    John
     
  3. Slicko

    Thread Starter New Member

    Nov 11, 2008
    7
    0
    thanks for the quick reply, any input is appreciated.

    To fulfill the basic specification of my project a sensor that just determines when the level reaches the refill point would work fine, and I'd be happy to fall back on that if I can't find a more accurate way of doing it. I think probably I was being a bit ambitious, but i've included in my plan as a possible extra if I have time, to have a display attached to the system that outputs the current level of liquid nitrogen inside the tank.
    Obviously it's better to do something less complicated but get it finished, rather than half complete something more complicated.
    I'm doing more research on it at the moment, there's actually quite a lot of information about it on the internet, so i'll post again if I find something better.
     
  4. KMoffett

    AAC Fanatic!

    Dec 19, 2007
    2,574
    230
    For a literature search Google: LN2 level detector indicator

    Ken
     
  5. hgmjr

    Moderator

    Jan 28, 2005
    9,030
    214
    If your container is non-ferrous, you may be able to use the technique outlined in this application note on hall effect liquid level sensing.

    hgmjr
     
  6. Blackbull

    Well-Known Member

    Jul 26, 2008
    70
    6
    I saw a device that is placed on the outside of butane gas bottles to indicate the contents level - probably no use to you, but what do I know.
     
  7. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    You might be able to determine the level using the liquid nitrogen as a dielectric for a capacitor that is used to control an oscillator or timer circuit.

    Liquid nitrogen has a dielectric coefficient of around 1.538+or-0.025 from about 0.5GHz to 10GHz and around 1.4 near DC.
    Air's dielectric coefficient at 1 atmosphere is 1.00059; 100 atm. is 1.0548.

    So, if two parallel plates had a capacitance of 100pF when air was the dielectric, when immersed in liquid nitrogen the capacitance should be roughly 154pF.

    The trouble would be finding materials that wouldn't simply fall apart or shatter at that low temperature.
     
    Last edited: Oct 23, 2009
  8. THE_RB

    AAC Fanatic!

    Feb 11, 2008
    5,435
    1,305
    Measure the weight of the tank. Ie put the tank in a "frame" then zero the weight (or use a constant value for a known empty tank weight) then weigh the tank as it is being filled.

    It's a good system used a lot in factory filling tasks, one of the main benefits is that the tank (or bottle) does not need to have any sensors etc attached to it.
     
  9. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    Weight would be a very good option; you wouldn't need to penetrate the tank.

    Something that might need to be considered is the relative humidity and temp range of where the tank will be stored. If the humidity is fairly high like it is here in Florida, it is probable that you will wind up having large accumulations of ice on the outside of the tank surface. Thick insulation may help to prevent that situation.

    At a former employer, they had a large liquid nitrogen tank outside the building. It had so much ice on the outside of it that it looked more like a huge igloo than a storage tank.
     
  10. KMoffett

    AAC Fanatic!

    Dec 19, 2007
    2,574
    230
    I agree. One of my cryogenics physicists has a large LN2 tank in his lab sitting on a low-profile floor scale. Noy cheap, but it works perfectly.

    Ken
     
  11. bertus

    Administrator

    Apr 5, 2008
    15,647
    2,346
    Hello,

    Our company also uses liquid nitrogen to cool the temperature regulation gas.
    It consists of the nitrogen container with a heater that boils off the liquid nitrogen.
    The amount of power given to the heater (a couple of highpower resistors),
    determens the flow of the gas coming out.
    For protection there are two diodes mounted on the heater assembly.
    One close to the heater for the "empty" status, this also shuts off the hrater.
    A second one is a couple of centimeters higher mounted, this will give a warning to refill.
    When the diodes are in the liquid nitrogen the voltage accros them will be nill.
    When the diode comes out of the liquid nitrogen, the diode will have a voltage drop.

    Greetings,
    Bertus
     
  12. someonesdad

    Senior Member

    Jul 7, 2009
    1,585
    141
    Your first task is to go to the library and do some research -- find out what's already known. Then the smart engineer hops in the car and goes to the local welding supply place (or whatever) and kindly asks for some help/knowledge to learn about the current industrial practices. A great strategy is to call the owner/manager, as they will often make time for you when the underlings won't (or he/she will delegate it to someone who will make the time). This will also exercise your people skills.

    You don't say what size of tank(s) you want to deal with. A common one in industry/labs is the LS-160, which are steel tanks of 160 liters. We used to get pretty good estimating how much LN2 was in them by tipping them -- i.e., by mass. Measuring the mass would be my choice, as it's foolproof, cheap, and non-invasive. When you start getting to the truck-sized tanks and larger, though, it's not really an option.

    A nifty "toy" project would be to make an ultrasonic detector that you'd move down the side of a tank and detect where the liquid level is. The stud sensors for walls use this technique -- they detect a change in the acoustical impedance. You could have fun repackaging an existing stud sensor to fit e.g. an LS-160 bottle and you'd make friends with lab staff who deal with LN2 bottles, as now they can quickly and easily measure tank levels. Since you wouldn't spend much engineering effort on designing the guts, your time would be spent on qualifying how and where to use it, how to check it, when not to trust it, etc. You might wind up with a design that you could market to companies like VWR (lab supply houses), but the market is tiny. Better, make your project's output a documentation package and show lab rats how to make their own. Be thorough and figure out how to make it work with different types of tanks. Talk to the guys who deliver truckloads of LN2 and ask them how they know how much they have in their tanks (they probably use a volume metering method on the truck). They might know of additional application areas.

    You could also make it work with propane tanks; that's a larger market. But it would be difficult to make and sell cheaply, as the consumer market demands very low cost --some Chinese manufacturer would make them and Harbor Freight would sell them below what you could get them made for. And there would be no patent protection, because the idea is in the public domain as soon as I press the "Submit Reply" button. :) Designing and making the thing is easy; selling them at a profit is hard.

    An obvious extension is to use an ultrasonic sensor inside the tank to sonically range to the liquid level. But the engineer in me doesn't like that, as it would probably be prone to failure and hard to service.
     
  13. thatoneguy

    AAC Fanatic!

    Feb 19, 2009
    6,357
    718
    I'd suggest going with the weight. A strain gauge between two platforms, also known as a Scale. Could be run by an 8 pin uC, strain gauge, serial LCD and LED for refill time.

    The Tape that is used on butane/propane bottles works because there is only one wall. Liquid Nitrogen is stored in a Multi-Walled vessel, like a Thermos, so there is a vacuum between the tape and what it is trying to "see" for a change in temperature.

    I'd try to stay as non-invasive as possible for both contamination prevention and reliability.
     
  14. Slicko

    Thread Starter New Member

    Nov 11, 2008
    7
    0
    Cheers for everyones intput, very helpful :) Using the weight of the tank does seem like the most reliable and cheapest idea so far, though i'm still going through what everyone has said so far. I'll post back once i've fully decided what'll work best for me.
     
  15. RimfireJim

    Member

    Apr 7, 2008
    22
    1
  16. bertus

    Administrator

    Apr 5, 2008
    15,647
    2,346
    Hello,

    I do not know how the sensor will react on the low temperature ( - 196 ° C ).
    Also mounting can be a problem.

    Greetings,
    Bertus
     
  17. Slicko

    Thread Starter New Member

    Nov 11, 2008
    7
    0
    the weighing option does sound the best so far, although there are a few things I think will cause problems.
    To initiate the transfer mechanism once the LN2 requires refilling, the scales will need a digital output which will be the input for the controller which will inititiate the transfer mechanism. I only have a budget of £100 for this project and as far as I can see, digital scales like this are pretty expensive. any ideas of how to do this on the cheap?
     
  18. thatoneguy

    AAC Fanatic!

    Feb 19, 2009
    6,357
    718
    http://www.freescale.com/files/sensors/doc/data_sheet/MPXV5004G.pdf

    About $9 each in single quantities from Digikey, select the pressure range you need (size of tank).

    The one above is for water column, unless tank is over 100lbs, sensors are relatively inexpensive. Since it is refill only, options for having the "scale" fill a bellows or smaller vessel for measurement will also work.

    Add $50 for PIC, Programmer, and board (PicKit2, 44pin Demo Board).
    $10 for LCD
    $15 for enclosure and misc parts.

    That's around half your budget after Euro conversion.
     
    Last edited: Oct 28, 2009
Loading...