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 Reynolds Number Flow Visualization Techniques

  1. Jun 6, 2013 #1
    Just as the thread says, I am looking for some good flow visualization techniques at low Reynolds numbers. I am working on the design of a small recirculating water tunnel that I think can be built for ~$100. I cannot get a specific RE at the moment as I am still doing a preliminary feasibility survey. So far things are looking up. I'm going to look up some technical papers later, but for now if anyone knows of some methods off hand that would be great.
     
  2. jcsd
  3. Jun 6, 2013 #2

    boneh3ad

    User Avatar
    Science Advisor
    Gold Member

    That's not terribly descriptive. Are you just trying to visualize the flow over an object you put in the test section? If so, is there any reason you aren't just going to use a dye injector?
     
  4. Jun 6, 2013 #3
    For now my goal is to do just simple visualization around various bodies. Later on I may upgrade the tunnel to make it worthy of taking measurements. There is no reason I couldn't use dye injection, however I've never had to do an experiment using it so I do not know its experimental limits. Also, if anyone knows where I can get a decent 20-50 watt water pump that would be great. All my search results yield these stupid solar pumps that cost over $200.
     
  5. Jun 6, 2013 #4
    Update, a feasible model would have a reynolds number of about 2500...Very low
     
  6. Jun 7, 2013 #5
    I'd go with the dye injection method as boneh3ad said. It's easy and cheap. Something else that I'm thinking of is to equip the model with tufts to visualise the surface flow but I don't know if they work in water tunnels to be honest.
     
  7. Jun 7, 2013 #6

    boneh3ad

    User Avatar
    Science Advisor
    Gold Member

    Tufts would work, but they have the disadvantage of being more invasive. Dye of specific gravity 1 is almost certainly the best bet since it will be neutrally buoyant and follow streamlines. The only downside to dye is that after you run for a little while, you have to change the water since your water will slowly change colors due to the dye.
     
  8. Jun 7, 2013 #7
    The configuration of my tunnel would make it very hard to use tufts. I have calculated that, using 3" PVC pipe with water filled to about 60% of the maximum height (imagine a top down view of PVC pipe arranged in a racetrack fashion, with water filled to 60% depth and an opening along on of the long stretches.) would require a pump with about a 600GPH capacity to obtain a flow velocity of .25 m/s or .55 mph or .8 ft/s. The tufts would have to be made out of something that is very flexible that does not dissolve in water. Maybe something hair-like?

    Also, another question:
    1) What is the best type of dye to use?
    2) What is the best injection method. For example, would several small syringes work nicely to display some streamlines or would one thick line of dye show flow visualization through mixing?

    I was also thinking about doing hydrogen bubbles too, as that only requires a wire with very high voltage, but that has safety issues.
     
  9. Jun 12, 2013 #8
    I seem to recall that a fun competition for a flow tracer was announced by a marine biologist out California way. The winning thing was some kind of neutrally buoyant little spheroids, that might have been sea urchin eggs or something biological like that. I have seen folks check ultra-slow airflows by making soap bubbles filled with helium, when the size is right these apparently are neutrally buoyant in air.

    Wonder if there is any neutrally buoyant immiscible oil, and if you were injecting at a point to visualize a streamline and ultimately clouding your water, maybe it could be filtered out or cause to rise by heating.
     
  10. Jun 12, 2013 #9
    Come to think of it, the way my tunnel will be configured I could use soap bubbles if the model only partially penetrates the surface. Dye could even be injected on the surface too!
     
  11. Jun 12, 2013 #10

    boneh3ad

    User Avatar
    Science Advisor
    Gold Member

    Of course your flow field won't necessarily be correct that you are seeing because you are looking at it under the influence of the free surface in air. It wouldn't be a huge effect qualitatively, but quantitatively it would be substantial.
     
  12. Jun 13, 2013 #11
    If that is the case then why are soap tunnels (not sure if that is the correct term) valid tools in fluid mechanics analysis? The only major difference I see would be the presence of gravity waves when used as a film on a water tunnel.
     
  13. Jun 14, 2013 #12

    boneh3ad

    User Avatar
    Science Advisor
    Gold Member

    Now, I am not going to claim to read every piece of literature on wind and water tunnel research, but I don't know of any modern research being done in a water tunnel where the model is only partially submerged with the exception of research on the flow around ships or for flow visualization. For ships it would obviously be quantitatively accurate since that is how a ship is designed to operate.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook