1. Limited time only! Sign up for a free 30min personal tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Low pressure barometer

  1. Dec 17, 2016 #1
    I am working on a barometer that will operate t-ed in to an experimental vacuum chamber. The fluid column is composed of vacuum oil so it doesn't boil. I have calculated that given the oil's density of 0.92g/cm3, each cm of oil height indicates 90 pascals of pressure. The oil has a vapor pressure of 10^-6 mbar. Given a height of 12cm, it should operate between ~0 and ~1kPa.

    The problem is that there is gas evolution happening. When I vacuum it down, the fluid level drops corresponding to pressure, but continues to drop slowly even as I hold the pressure constant. This continues until gas bubbles out from the bottom of the column.

    I have already de-gassed the parts to the point where no more bubbles form. I'm not sure where this gas is coming from or how it's pressure could be greater than that of the barometer's environment? I.E. forcefully displacing fluid in the course of pushing the fluid level in the column below that of the basin it sits in.
  2. jcsd
  3. Dec 17, 2016 #2
    I can only think of two possible options either the vacuum oil or apparatus is contaminated with water or other liquid/gas , or the oil itself is fake.

    I'm assuming you made sure to get all the air out from the closed end of the barometer.
  4. Dec 17, 2016 #3
    The oli is a well-known brand, so I think there must be some contamination, water perhaps. I have gently heated the oil for a while to drive off any foreign substances and it as behaving much better actually. The column height is more stable but still slowly creeps down however.

    I have also noticed that I need to allow a very small bubble to remain in the top of the column. If the column is completely filled with fluid, then when I re-orientate the barometer upright, the fluid does not drop. I know that since the void in the top of a barometer is filled with the fluid's vapor at its vapor pressure, I am thinking that with no gas pocket to begin with, the fluid becomes sort of 'superheated' and remains in the liquid phase despite the hard vacuum it is subject to. Leaving a small bubble gives sort of a nucleation site for vaporization to begin. Does this sound reasonable?
  5. Dec 17, 2016 #4
    "I have also noticed that I need to allow a very small bubble to remain in the top of the column. If the column is completely filled with fluid, then when I re-orientate the barometer upright, the fluid does not drop.".... this must mean you're not getting a low vacuum , leaving a bubble in will lead to false readings definitely not a good idea... ...

    surely the VP of this oil is so low you can ignore it

    If you you leave all the oil in vacuum for a few hours this must pull out all the volatiles and fix things ...good luck , keep in contact if you need more advice..oz
  6. Dec 18, 2016 #5
    I'm not sure what is going on.

    I have tried it t-ed in to the chamber with boiling water approaching 0*c and the temp dropping as it boils, so I know I am getting into the sub 1kPa area. The column does track the the vapor pressure of the water at the measured temperature, but reads about 25% low. It's not possible that the oil just isn't vaporizing at the top of the column?

    Doesn't there have to be a nucleation site in the top of the column to allow the fluid to vaporize and drop?

    Using water sparingly to prevent contamination.
    Last edited: Dec 18, 2016
  7. Dec 18, 2016 #6
    The pressure range you are working at , 1 Kpa is 10 mBar ..... vapour pressure of your oil is 10^-6.... 10 to the minus 6 ... mBar !! You can forget all about the vapour pressure of the oil (assuming it's pure). effectively it's not emitting any vapour .

    I'm not clear what your doing with water ... If you can explain your whole set up ....apparatus in detail ,pump, how low it can go, what exactly you're doing , then I'm sure I can offer more ideas.
  8. Dec 18, 2016 #7
    The pump is rated down to 5Pa, but I will only get close because of imperfect sealing. I put a container of cold water into the system to test whether it would boil, which indicates approximately the pressures I am reaching, being sub 1kPa.

    The only role the vapor pressure of the oil plays is in the fact that when the column drops, assuming there is absolutely no air in there, then the void will be filled with the oil's vapor at its vapor pressure. After all, there is no such thing as a perfect vacuum.

    The barometer is constructed such that I can invert it to fill the column with oil and re-orientate it upright while under vacuum. My theory is that there must be a small pocket of gas inside the column to provide a liquid-to-gas boundary to serve as a nucleation point necessary to allow the oil to vaporize, thus allowing the fluid to drop. I find that the size of the bubble I leave inside the column has almost no effect on the height of the fluid when positioned upright again.

    I have also found the source of the gas that was evolving. The column is a clear plastic pipe, that I hermetically sealed at the top. I placed another piece of plastic that had been melted this way in the oil basin, and it produces bubbles under vacuum as well. So the gas is actually being sucked out of the porosity of the plastic where it had been melted.

    Here is a video where I purge the column leaving a small bubble, and the fluid drops. Keep in mind this is just a home project so it's kind of rough.

    Last edited: Dec 18, 2016
  9. Dec 18, 2016 #8


    User Avatar
    Science Advisor

    Those gas bubbles must be the air that was in the barometer column when you started pulling the vacuum. The column will not function as a barometer if there is any air in the column. Condensation nuclei are not needed to evaporate a fluid. Cavities will form where there is lowest hydrostatic pressure, near the top of the fluid in the column. The cavities will merge, then fluid will flow down around the cavity. There may be interesting effects due to surface tension that might delay initial cavity formation.
  10. Dec 18, 2016 #9
    That makes sense, but I am removing all air pockets when it is already under vacuum. I will try leaving no bubbles in the column and see if the fluid will drop given enough time.

    That's what I see happening when I allow a small bubble to enter the bottom of the column. It rises and as soon as it reaches the point where the fluid level drops to, it expands to fill the top of the column.

    If I do leave an air pocket, it is at the barometer's operating pressure, and the bubble's size has little effect on the fluid height after dropping.

    I am sure now that the gas that was bubbling out of the bottom of the column is escaping from the porosity in the plastic where I melted it to seal the top. This explains why it continues to appear despite repeated purging and a constant vacuum. It has been under vacuum long enough now that this gas evolution has almost completely stopped.
    Last edited: Dec 19, 2016
  11. Dec 19, 2016 #10


    User Avatar
    Science Advisor

    As an air bubble rises in the column, it reaches the top of the column and the top of the column drops by the air bubble volume. But if you then halve the pressure in the chamber, the bubble will double in size while the column will halve in height. I can see no way to simply separate those two effects. The bubble will begin to escape from the zero height column long before the chamber pressure reaches zero.
  12. Dec 19, 2016 #11
    I see that but I am removing/introducing air bubbles and standing the barometer upright while under vacuum the whole time. The pressure which the barometer is experiencing does not change before or after introducing the bubble. I just tip it sideways so the bottom edge of the tube reaches the surface of the oil in the basin.

    I do exactly this in the video above.
  13. Jan 4, 2017 #12
    Has anyone ever done this before?

    For additional testing I have placed a radiometer with the glass envelope punctured into the chamber that the barometer is t-ed into.

    The Crookes Radiometer will not start to spin until the pressure is below about 300Pa, which is what my barometer reads when it starts to spin as the pressure drops. This is inarguable proof that the barometer works and is fairly accurate, and I am in fact reaching the operating pressure range.

    I still don't understand why the oil will not drop, though, even if left for hours. Anyone have any ideas, like the surface tension of the oil or something?

    I have also verified that leaving a small bubble in the column while already under vacuum does not meaningfully affect the pressure reading.

    I will post up a video documenting all of this on youtube soon.
  14. Jan 5, 2017 #13


    User Avatar
    Science Advisor

    Sorry, I don't have video bandwidth here.

    I suspect there is a minimum pressure at which the surface tension of the oil prevents further movement of the oil. This effect may be quite complex near the open end of the barometer tube.

    The small bubble you introduce is a bubble of partial vacuum, not of air. Does that bubble change size as you continue to pump down the chamber? If it does not change size then the pressure in the oil is not changing which implies that either the pressure is not falling further or the oil surface tension is blocking the flow of oil.

    How accurately can you read the oil barometer height? How wide is the tube? What if you increased the diameter of the barometer tube? Is the meniscus of the oil reasonably flat in the centre of the tube or does it have a parabolic curved meniscus surface?
  15. Jan 5, 2017 #14
    The bubble's size does respond to changes in pressure. The tube is 7.9mm id(5/16 in), stands 12cm tall, and I have marked 1cm increments in height from the bottom. I can probably discern 2mm changes in level. The oil surface shows a noticeable concave meniscus.

    With the size of the bubble having little effect on the reading, that is comparing two different sized bubbles at the same pressure. The level seems to change only by the difference in size between the two bubbles.

    Increasing the diameter sounds really interesting. I will have to set that up hopefully with tubing of 1 inch or more.
  16. Jan 6, 2017 #15
    Did a couple more tests. My original testing was with a clear plastic tub.

    First I used a glass column that had a small hole near the top(from a previous experiment), which I patched with resin. Using this, the oil actually dropped on its own, instantly!

    But having the patched hole there was not ideal, so I made a new column with intact 1" glass tubing that I fused closed at the top. With this the oil does NOT drop.

    So I think it comes down to the interaction between the oil and the surface of the tubing. The oil probably dropped in the column with the hole because a jagged glass edge was exposed which provided a nucleation point. Whereas the smooth glass of the second column has no such surface.

    I know you said the oil does not need a nucleation point to vaporize, but it looks like that's what it takes from what I'm seeing. Do you think this is reasonable?
  17. Jan 6, 2017 #16


    User Avatar
    Science Advisor

    You are calling it a nucleation point as in condensation, but I believe it is a different effect. I think the energy needed to break the surface tension to form the cavity will be important in explaining what is happening. I would expect the pressure difference needed to initially "pull" open the cavity of vapour in a liquid may be some function of the surface profile. Consider a needle point, a step, a knife edge or a 'V' notch. Also consider including an oil-phobic material, maybe Teflon, at the top of the tube as a way to initiate a cavity with minimum energy. To understand the profile effect, if it is significant, will take some research into the surface contact energy and cavitation.

    The argument for using a wider tube is that surface tension of oil against the barometer tube wall will be most significant near the wall while the pressure will be indicated by the level of a reasonably flat central area. If there is no flat area then there will be a minimum pressure detectable that is determined by the wall to oil attraction and the oil surface tension. I think you must study the oil meniscus with different tube materials to identify the best tube material and tube diameter to get a reasonably flat central area. You can do those tests at atmospheric pressure with oil drawn by air into a short open tube, like is done with a pipette.

    How tall is your oil barometer column? There is an argument for using a taller column as it would form a meniscus with a full tube diameter before the pressure in the chamber fell to the point where a small cavity could no longer be initiated. It would not surprise me if the cavity was to “pop” open as the pressure fell below a critical point determined by the oil to tube attraction.
  18. Jan 13, 2017 #17
    One thing I have not seen addressed. How are you filling your column? My assumption would be that you are pouring in the oil. If so it is possible that there is some agitation and that microbubbles are getting entrained. These would then reveal as the pressure dropped.
    Is it possible to fill it either with a syringe or a pipet from the bottom up? If so use process where the tip stays submerged and keep your oil as quiescent as possible. This should prevent the effect if that is the cause.
  19. Jan 13, 2017 #18


    User Avatar

    While this is an interesting discussion regarding fluids, it appears a much simpler solution to the original problem would be to simply use a mechanical vacuum gauge.
  20. Jan 16, 2017 #19
    I have found the source of the out-gassing I originally mentioned to be from porosity in the plastic where I had sealed the top of the tube. it's amazing how much gas can appear to come from nowhere when it is expanding ten thousand times or more.

    I decided in this case that a barometer would be a better solution, because it is free to make and shows excellent resolution(+/- 10Pa) in the pressure range I am interested in, 10Pa - 1kPa.
  21. Jan 18, 2017 #20

    jim hardy

    User Avatar
    Science Advisor
    Gold Member

  22. Jan 23, 2017 #21
    Never heard of that device. Looks perfect for this but I wouldn't want to mess with the mercury and it's unfortunately too expensive for my little hobby.

    Here is that video. Doesn't address the issues I have had with the barometer but shows it in use.

Share this great discussion with others via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted