Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Simple flowmeter design

  1. Jan 24, 2009 #1
    Hello :)

    For a undergrad design course I'm asked to design and manufacture a device for measuring the peak urine flow rate to indicate possible signs of prostrate cancer. Basically, the allocated budget is too low to use any electronics (transducers and data acquisition), which was my initial idea. Hence we're limited to mechanical flow rate measurement.

    Now, the device has to allow a male adult to self-measure his flow rate, so he should be able to read off a scale or something. I'm currently looking into rotameters, or spring-based rotameters, as a possible solution. I can't really use differential pressure measurement because that uses two manometers, and you have to subtract the heights of fluid, etc which is too complicated for the patient to use. One professor suggested we just "measure the height of a column of urine"... how does that work? Would that be something like measuring a given volume and using a stopwatch, then dividing the two? That would give the average (not peak) flow rate.

    Otherwise, what other simple mechanical flow-measuring techniques could we use? I'm quite confused at the moment! Any help would be greatly appreciated, many thanks in advance :)
  2. jcsd
  3. Jan 24, 2009 #2


    User Avatar
    Science Advisor
    Gold Member

    Fill a fixed volume and measure the time to fill it?
  4. Jan 24, 2009 #3
    indeed that's the simplest solution, but 1) that gives the average flow rate 2) it requires calculation after the measurement, which is in disagreement with the requirements, which are for the patient to self-measure their flow rate directly via some calibrated scale or similar... but yes, at the moment i can't think of any better solution than that of fixed volume.... any ideas?
  5. Jan 25, 2009 #4


    User Avatar
    Science Advisor
    Gold Member

    Read the time and correlate it on a chart. No calculation. That's how I'd do it.
  6. Jan 25, 2009 #5


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    Could you detect a small current on a loop though which the stream of fluid passes? You can with clean water, may work well with urine.
  7. Jan 25, 2009 #6


    User Avatar
    Science Advisor
    Homework Helper
    Gold Member

    Assuming the outlet pressure on an orifice (or any given restriction) is atmospheric pressure, the flow across that orifice is strictly a function of pressure. The higher the inlet pressure, the higher the flow rate.

    Imagine pissing into a funnel that led into a long, verticle tube. At the bottom of the tube was an orifice. The instantaneous flow rate would simply be a function of the inlet pressure on this orifice. And the inlet pressure on the orifice is strictly a function of the liquid height. So if you were to urinate into this tube with an orifice at the bottom, the instantaneous flow rate would be a function of the height of the column of urine in the tube. Note that here, I'm assuming this primary tube has a small enough diameter such that the urine column height rises and falls fast enough such that it accurately corresponds to flow rate.

    Now you need to figure out a good way to measure the column height and/or record it some how. You want to do avoid using any electronics?

    Let's say you have a second verticle column, parallel to the first, but with only a single tie connection to the main column at the bottom near the orifice. The top of this second column is also open to atmosphere. The column height in this second column would then rise from the bottom as urine came in and then decline as it stopped and the primary column emptied. It would do this in unison with the primary column. In other words, this parallel column would rise and fall exactly like the primary column.

    This second column could have a float inside that was made to register the highest point somehow. The float would come up and not go down.

    Note that I'm suggesting the use of a second, parallel column, because the urine stream will need to be 'poured' into this primary column from the top somehow, so it may be that this urine entering the primary column might interfere with a float mechanism that registered maximum height.

    Instead of a float, you might also consider something like a 'dip tube' that was wetted and the highest liquid level could then be read. Maybe a strip of chemically treated paper for example, that changed color when exposed to urine, thus giving a solid color bar up to the maximum column height.

    In any case, having a second, parallel column in which urine only comes up from the bottom and never enters from the top, might be a good way of eliminating any kind of erroneous readings as urine sprays down from the top of the primary column.
  8. Jan 29, 2009 #7
    hi everyone, thanks for the ideas, they're all practical solutions. Q_Goest, i had thought of something similar to what you suggested, although it's more clear now that you've suggested it. The only complicated issue is the geometry of the float itself, and how exactly it could be made to rise but not go down. The other issue is that the device would be relatively difficult to calibrate, and be read/understood by the user.

    Well apparently the lecturer who wrote the design requirements for us, has just realised that the device is more complicated than he'd at first thought, especially since the device has to measure to an accuracy of at least 1ml/sec if not better. So now, transducers have become an option. my idea is to connect either a differential pressure transducer, whose output goes into any signal conditioning (if necessary) and then into MATLAB to be graphed. This is evidently the best solution in terms of accuracy.

    The other solution, which I'm not yet sure about, would be to have a cylindrical beaker and measure the flow rate using an op-amp differentiator, which would find the rate of change of fluid height in the beaker.
    ie Q=area * height rate of change = (pi*r^2) dh/dt

    We'd need something like a weight/pressure transducer, or even a float connected to a potentiometer, to detect fluid height, then a differentiator to get dh/dt... connect to MATLAB, multiply by the area, and graph it. I'm not sure which transducer to use here...

    And finally, my most confusing issue: I have no idea how you actually "connect" a transducer to a liquid flow... surely a differential pressure transducer won't be directly in contact with the urine will it? How exactly is that done?

    Many many thanks for all your help :)
  9. Jan 29, 2009 #8


    User Avatar
    Science Advisor

    Could you put the container on an electronic scale of some sort (e.g. a strain gauge) and then just log mass readings? While the density probably changes from batch to batch, you could have the user enter the volume after he (I presume) is done. Then you could figure out the flow at any point based on the mass readings you took.
  10. Jan 29, 2009 #9
    sounds like a solution! but that would require user input, and somekind of feedback, as well as the container to be volume labeled, etc... whereas just plugging pressure or height change readings into MATLAB would be a little simpler, but I don't know how to connect the transducer to the flow
  11. Jan 29, 2009 #10


    User Avatar
    Science Advisor
    Gold Member

    A crude way of measuring a fluid lever is to have a small tube at the bottom of the container/cylinder and bubble air through it. The pressure in the tube is equal to the hydrostatid head of the liquid.

    A less crude way would be to use a Ultrasonic Level Sensor.

    Here is an ipaper on the subject of level sensing:
    http://www.scribd.com/doc/2836046/Level-measurement-the-basics" [Broken]
    Last edited by a moderator: May 3, 2017
  12. Feb 1, 2009 #11
    thanks for that digoff, useful paper. i'll try to renegotiate our budget and see what i can do. although im still confused about how sensors are integrated into flow, as in, you dont just put a pressure sensor immersed in urine... how is it connected?

    thanks to all for the help :)
  13. Feb 2, 2009 #12
    My cousins and I conducted this experiment occasionally out by the back fence when we were kids. I remember it more as a comparative measurement than an absolute one however.

    Would Bernoulli's principle work here? I'm thinking of measuring the height the urine rises in a vertical tube connected to the tube through which the urine flows.
  14. Feb 2, 2009 #13
    A problem with any direct methods of measurement is that these may severely disrupt the flow rate being measured.Another problem is that depending on what the device is the patient may be put off from effectively performing the task in hand.I wonder if it would be possible to make weighing scales sensitive enough to measure the patients reduction in weight as he takes a pee.If so it would be easy to build in some electronics which indicates and records the flow rate and how it changes.
  15. Feb 2, 2009 #14
    Dadface: indeed, we had thought of weight reduction measurement, but you need high accuracy transducers, since the weight reduction is minimal. but in fact the prototype we're building will be tested with water coming out of a gardening can or similar, so it has to be direct measurement... thanks for the suggestion, it's still one of the best ideas out there :)
  16. Feb 4, 2009 #15
    I had this idea, based on dadface's suggestion.... could we have the patient urinate into a bucket of known dimensions and then use a weight transducer to measure the (changing) mass of that bucket, then an op-amp differentiator to differentiate it into mass flow rate?

    Then we could find Q = (1/rho) * (dm/dt) and graph this using a MATLAB script or similar?

    Are there any issues with this design suggestion?
  17. Feb 11, 2009 #16
    First time on the site. Using standard, purly mechanical equipment: pressure regulator, eductor, rotameter. Set up a "zero" flow of , say 2 GPM tap water through the regulator, eductor, rotameter, discharging into the drain. Use the reg to maintain a steady pressure/flow. Urinate into the suction side of the eductor. There should be a corresponding rise in flow through the rotameter. Somewhat sanitary, self flushing, everything goes down the drain and the flow measurement is instantaneous.
  18. Feb 16, 2009 #17
    thanks williej, good mechanical solution. Why do you need the "zero" flow of water though? Is the urine flow too weak to make the rotameter float rise or something like that?
  19. Feb 16, 2009 #18

    The urine pressure is low and the water pressure is used to provide the energy to flow the urine with no resistance on the person's bladder. The urine must exit the person at atmoshpheric pressure, as done when urinating normally. There is a pressure drop through the measuring system - i.e. piping, rotameter - that must be overcome. The flowing water provides the energy to cause a suction at the venturi in the eductor, so the person can urinate at atmoshpheric pressure into an open funnel, just like a urinal. Hope this helps.
  20. Feb 16, 2009 #19
    that explains it, thanks again! :)
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook