Water levels - Precision! - Page 2
Close
Login to Your Account
Page 2 of 2 FirstFirst 12
Results 21 to 23 of 23
  1. #21
    Join Date
    May 2017
    Country
    FINLAND
    Posts
    1,595
    Post Thanks / Like
    Likes (Given)
    456
    Likes (Received)
    734

    Default

    Quote Originally Posted by sfriedberg View Post
    That's astonishing. I never would have imagined the effect could be anywhere that large. Really tiny tubes, I presume.
    Its not about tube size but "drop height" with density errors like temperature differentials or partly salty water.

    IF you can run the tubing really close to measured height the density error from temperature (or salty water) is lot less. IE if you are leveling machine feets on concrete slab you might be able to set the measurement points only 2" above the concrete slab and keep the lowest point of tubing only 2" below reference level. Any density error would be only 1% compared to unfortunate case where the tubing drops 16 ft down from reference level.

    Temperature makes quite a difference if you are setting 2nd floor roof line and the tubing runs all the way down to ground and back up to 2nd floor roof.

  2. Likes sfriedberg liked this post
  3. #22
    Join Date
    May 2017
    Country
    FINLAND
    Posts
    1,595
    Post Thanks / Like
    Likes (Given)
    456
    Likes (Received)
    734

    Default

    Water has volumetric expansion coefficient of 0.0002 per degree C at around room temperature so 0.2mm per meter of height per degree C.

    If you arrange the hosing to have something like 50mm (2") drop from reference surface and the hose sits on stable temperature within +-1 degree C its possible in theory to keep the error under +-0.01mm.

    On the other hand if you run the hose around the building and one side is 20 degrees C hotter than other in the sunshine and hose drops 5 meters down to ground you'll get 20mm error.

  4. #23
    Join Date
    Sep 2004
    Location
    south SF Bay area, California
    Posts
    1,993
    Post Thanks / Like
    Likes (Given)
    150
    Likes (Received)
    639

    Default

    I'm swamped this week, so can only provide cursory responses right now.

    chevy43 --

    Working fluid in a communicating-vessel level can very wildly, depending on need and availability. The early earth-tide researchers often used mercury as the working fluid, and deionized water or "200 proof" alcohol may be required in a cleanroom. In a workshop environment, tap water dosed with a surfactant (and sometimes dye) is fairly common, if mildly corrosive if not thoroughly cleaned up after a spill. A mild petroleum liquid such as mineral spirits (aka paint thinner, Varsol, and liquid paraffin) works well, and will evaporate cleanly, so if preferred by some users.

    I've seen three different approaches to probing the liquid surface: 1) Using a "hook gage" (J-shaped) probe that is submerged in the liquid, and then raised until the pointed tip of the J deflects the liquid surface without breaking through (which has been described in old-time literature as "raising a pimple" on the liquid surface. 2) Lowering a pointed probe from above to dimple the liquid surface without breaking its surface tension. 3) Using a pointed spindle from above to pierce the surface and then retracting the spindle until the liquid breaks free from the surface. I have used methods 2 and 3, and STRONGLY prefer 3 when the ambiet lighting is less than ideal.

    ballen --

    I've only had time to glance at the SLAC document I posted, but based on that glance I'm thinking that the electrolytic tilt sensors are used to set and monitor the attitude of the individual vessels, exactly analogous to the tubular vials of the H&W vessel and the bullseye vial of the Wyler vessel.

    The change of height of the fluid within the individual vessel is determined as a change in electrical capacitance between the fluid surface and a surface fixed to the body of the vessel . . . analogous to the micrometer of the H&W and Wyler vessels.

    With these types of instruments, a tilt of the individual vessel creates a cosine-error perturbation of the fluid height change, which darn-near-trivially degrades the vessel-to-vessel fluid height difference that is the basis of the pseudo-geodedetic tilt of the structure supporting the vessels.

    For a simple numerical illustration, let's assume that there are two vessels, slightly more than 206 inches apart. At their installation, the heights of the fluid within the vessels is baselined and verified. A week later, fluid height measurements of the two vessels show that one of the two vessels has a 0.005 inch greater fluid height (which, at first glance, suggests that the vessel itself might be lower), while the other has a 0.015 inch greater fluid height.

    Both vessels having a greater fluid height sounds wrong, but can be caused by a number of things, including temperature change, addition of fluid to the system, or a pinching of the fluid line connecting the vessels.

    On the other hand, the 0.010 inch difference between the individual-vessel changes may be caused by a tilting of the structure supporting the vessels. That tilt -- assuming it's "real" -- can be quantified as the ArcTangent of 0.010 inch / 206 inch, which is very nearly 10 arcseconds.

    John

  5. Likes sfriedberg liked this post

Tags for this Thread

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •