• Welcome to engineeringclicks.com
  • Dedecting cavitation by using step response answer.

    Discussion in 'Calculations' started by Bert, Jun 23, 2013.

    1. Bert

      Bert Member

      Joined:
      Apr 2013
      Posts:
      10
      Likes Received:
      0
      is there a way to detected cavitation or a full filter by measuring the pressure and calculating first and second derivative of the pressure?
      Not by measuring the pressure drop over the element that cause the cavitation but doing some fluid dynamics calculations and taking care over a changing step response answer.
      Can some point me in the right direction?
       
    2.  
    3. vidgolob

      vidgolob Active Member

      Joined:
      Jan 2012
      Posts:
      35
      Likes Received:
      0
      Why wouldnt you use the pressure drop measurement? Its accurate and not that expensive...
       
    4. Bert

      Bert Member

      Joined:
      Apr 2013
      Posts:
      10
      Likes Received:
      0
      Because then we need two sensors, I know this is crazy but the costumer only want to pay for one.
       
    5. vidgolob

      vidgolob Active Member

      Joined:
      Jan 2012
      Posts:
      35
      Likes Received:
      0
      One sensor... First thing that comes to mind is a pressure drop measuring unit. This unit has two tubes, one for the referential pressure and one for the actual pressure inside a vessel. Maybe you could try placing the referential tube on one side of the filter and and the actual pressure tube on the other side... It is still only one unit. I have worked with Kimo devices that use this kind of pressure drop measurement.
      Another thing you could do is measure the pressure drop inside the vessel for a longer period while using clean air ( same for other kind of fluids, just make sure you are not clogging the filter ). After this period ( lets say two, three weeks ), you take the average of that measurement and you basically have your pressure drop measurement using only one pressure sensor. This method is not as accurate or reliable and the rise of pressure may not mean that the filter is cloged, but it definitely means that something inside your system has changed.
      You could instal a ultrasound measuring device, but that doesnt allways work and is hard to set up. Expensice as well. If the customer wont pay for two pressure sensors, forget about it.
      Another way would be to measure the flow which is directly linked to the pressure drop in your system. But again, it would be much better to use two sensors.
       
    6. bradely

      bradely Member

      Joined:
      Jul 2013
      Posts:
      8
      Likes Received:
      0
      One sensor... To begin with thing that strikes a chord is a force drop measuring unit. This unit has two tubes, one for the referential force and one for the real force inside a vessel. Possibly you could attempt putting the referential tube on one side of the channel and the true force tube on the other side... It is still one and only unit. I have worked with Kimo apparatuses that utilize this sort of force drop estimation.
       
    7. maniacal_engineer

      maniacal_engineer Well-Known Member

      Joined:
      Jul 2009
      Posts:
      137
      Likes Received:
      0
      Shhhhh

      if you are actually interested in cavitation, couldn't you just listen for it?
       
    8. vidgolob

      vidgolob Active Member

      Joined:
      Jan 2012
      Posts:
      35
      Likes Received:
      0
      Nice one. But remember - only one person can be listening!!!
       
    9. maniacal_engineer

      maniacal_engineer Well-Known Member

      Joined:
      Jul 2009
      Posts:
      137
      Likes Received:
      0
      Doesn't cavitation make noise? I thought that was why submarines had to be careful of cavitation, and also why it destroys valves.
      If your pressure sensor has rapid enough (audio frequency) response then why could you not just listen, with the sensor, using a high speed (audio frequency) DAC, and hear cavitation?
      A microphone is just an audio frequency pressure variation sensor, so if your pressure sensor is a semiconductor strain bridge, or a piezo sensor, then you could use it also as a microphone and listen for turbulence or cavitation.
       
    10. vidgolob

      vidgolob Active Member

      Joined:
      Jan 2012
      Posts:
      35
      Likes Received:
      0
      As i mentioned before this is an option, but if the customer wants the cheapest solution possible I believe this is not an acceptable answer. Another thing - this kind of a system is not that easy to install ( so that it actually works ) and as far as I am concerned you need to know exactly what you are doing to do something like this.
      The way I see this problem - someone with little or no understanding of this field got this idea stuck in his head and is now convinced that every other solution is just an attempt to get him to pay more money than he should. Two pressure sensors is in my opinion by far the simplest and the cheapest way to go here. I could be wrong but in most cases like this one that is the main problem. And in the end it always turns out that the customer had even bigger expenses than he would have if he listened to people who actually know what they are doing.
      The only explanation to suggest otherwise would be if this is a REALLY specific case with some limitations I just cant see right now.
       
    11. maniacal_engineer

      maniacal_engineer Well-Known Member

      Joined:
      Jul 2009
      Posts:
      137
      Likes Received:
      0
      I completely agree. Adding a cheap pressure sensor, or a relative pressure sensor instead of an absolute pressure sensor, would be the simplest and easiest and most robust way to do it. So present the options to the customer - here is what it will cost with two cheap sensors, here is what it will cost with one sensor and a complicated DSP to listen in - and let them see that they should just let the engineers do their job.
       

    Share This Page