Data logging in Ugandan toilets - tricky one!

Discussion in 'The Projects Forum' started by NicNakNoe, Nov 24, 2013.

  1. NicNakNoe

    Thread Starter New Member

    Nov 24, 2013
    4
    0
    A tricky one for you: I’m working on a project and am quite stuck.

    It is for an NGO in Uganda who want to know to get a better idea of what goes into their basic pit latrines (essentially a hole in the ground where waste is flushed into via flushing with a bucket of water).

    The setup is pretty much this: http://www.unep.or.jp/ietc/publications/techpublications/techpub-15/2-4/4-1-3.asp

    The waste enters via a 110 mm pipe (angle is variable but can be fixed for a trial). The NGO would prefer if the device is something which fits on/around the pipe
    I want to measure:
    1. At the very least – an approximation of what total combined volume of liquids plus solids is entering the pit (plus timings would be great)
    2. In the ideal world – the proportions of water vs urine vs solids entering the pit (and at what time)

    I guess the time is useful because maybe through some manipulation of the data I can start to link up a ‘use’ with a volume e.g. if the ‘switch’ which say something passes is activated 3 times in one minute – this could be attributed to one user (the distinction between ‘uses’ and ‘volume’ is difficult in my mind.
    This would ideally be transmitted via phone or have collectable data. No mains electricity available.

    Any help greatly appreciated – this data will hopefully enable the NGO to work more effectively for provision of sanitation. Keep in mind the basic and rough conditions in a Ugandan toilet pit!

    Nicola

    p.s. I am not being payed for this work - just trying to help them out.
    p.p.s I have some ideas but this post was too long so if someone joins me for a chat about this I'll explain my initial thoughts.
    p.p.s please be patient - i have no electronics background - civil engineer
     
  2. sirch2

    Well-Known Member

    Jan 21, 2013
    1,008
    351
    What is the pipe material (plastic, ceramic)? Is it acceptable to drill the pipe to fit sensors? Would weighing or a tipping bucket arrangement be possible?

    I am currently debugging a low power data logger that I have designed that would record events against time so it would do what you want if s suitable senor could be developed.
     
  3. NicNakNoe

    Thread Starter New Member

    Nov 24, 2013
    4
    0
    Plastic pipe - definitely drillable.

    I had drawn up some sketches thinking about weight and tipping bucket but I think both are too risky due to toilet roll/sanitary pads/unknown objects passing down the pipe and sticking in/on both.

    I had considered a flap at the end of the pipe. In a very basic version - having a movement of the flap counted as a use...in a more advanced, the distance the flap opens being related to a volume (also risky if it blocks/stuff sticks on it - and whole system is very slope dependent)

    I also thoughtabout some kind of system where some sensor is set up (maybe IR/ultrasound) where a circuit is made as waste leaves the pipe...breaking a beam of light or something, and the amount of time that the beam is broken could be counted...the most advanced somehow telling the difference between water and solids...

    I went to Maplins today and left confused, and with a relay and a timer!
     
  4. MrChips

    Moderator

    Oct 2, 2009
    12,449
    3,365
    You might be in luck. I have done work in this area which I will elaborate later.
     
  5. MaxHeadRoom

    Expert

    Jul 18, 2013
    10,565
    2,379
    Hope you aren't up to your neck in work at the moment!:eek:
    Max.
     
  6. MrChips

    Moderator

    Oct 2, 2009
    12,449
    3,365
    I am. How did you guess?
     
  7. NicNakNoe

    Thread Starter New Member

    Nov 24, 2013
    4
    0
    Great - if you get time at all I'd love to hear about it -timeline for this is tight. I'm under a $50 budget for the simpler device...and up to ~$200 for the more in depth...thats probably worthy of note.
     
  8. KJ6EAD

    Senior Member

    Apr 30, 2011
    1,425
    363
    I posted this to your duplicate thread on ElectronicsPoint:

    An array of capacitive proximity sensors, a strain guage and an infrared thermal sensor attached to a fixed length of pipe may provide the data you need but it won't be cheap.
     
  9. bance

    Member

    Aug 11, 2012
    315
    34
    What is this data for? There may be a simpler way to aquire it!

    Steve.
     
  10. MrChips

    Moderator

    Oct 2, 2009
    12,449
    3,365
    Sorry, can't do on both counts.
    I can't be done for under $200 and would take more than a month of effort.

    I did two research projects that are directly applicable.

    The first was to determine the level of solid material in slurry at a waste treatment plant. This was done using a laser beam by measuring the depolarization of the back scattering.

    The second is using EIT (Electrical Impedance Tomography). A series of electrodes are embedded around a cylindrical pipe. By measuring the electrical conductivity across pairs of electrodes one can capture 3D images of what is in the pipe, very similar to brain scans using MRI techniques. This is very doable for sewage in a pipe.

    There is a civil engineering team at the University of Johannesburg that is working on the latter technique. I can put you in contact with that group.
     
    kubeek likes this.
  11. GopherT

    AAC Fanatic!

    Nov 23, 2012
    6,067
    3,837
    A microphone attached to the outside of te pan-seal drain will be very easy. If you are logging data one peak detect over one second period, log peak, reset peak detector and wait one second for next peak. Can be done all in software (peak detect, record and wait until next datapoint).

    Review data and check if voice and other backround noise is interfering - then set resold higher. Microphone directly on the pipe can easily discriminate from all other local noise, it will be loud for in the microphone.

    The best part of this method is, no biological interference of sensors. It is simple and everyone can relate. Components are widely available. An iPAD app could be written if you need to monitor for a short time.

    For urine content, you can assume 1 liter to 1.5 liters per person per day. Your flush volumes will be higher in most cases (250 ml per event or less).

    For BM, flush volume will be higher than urine flush. calculating solids content can be tricky. Again, you may want to take some averages - even making some sample measurements. Mass will be very dependent on local diet (fiber content) and average body mass index.
     
    NicNakNoe likes this.
  12. wayneh

    Expert

    Sep 9, 2010
    12,153
    3,059
    Twelve posts in before someone makes the obvious "logging in the toilet" joke ? :D
    It's like it was all teed up for me.

    I'm assuming there is a substantial vertical drop as the waste flows? I think that energy could be used in a static centrifugal separator to separate - for measurement only - the floaters and sinkers. Having separated, more defined streams might simplify the measurement techniques. Just a thought.
     
    GopherT likes this.
  13. kubeek

    AAC Fanatic!

    Sep 20, 2005
    4,670
    804
    Would you happen to have some books or links for any of those techniques? I just can´t get my head around any of them, maybe the depolarization a little bit.
     
  14. vk6zgo

    Active Member

    Jul 21, 2012
    677
    85
    I chuckled at the thought of a "strain" gauge!:D
     
    wayneh likes this.
  15. MrChips

    Moderator

    Oct 2, 2009
    12,449
    3,365
    Ha ha. Good one.
     
  16. MrChips

    Moderator

    Oct 2, 2009
    12,449
    3,365
  17. sirch2

    Well-Known Member

    Jan 21, 2013
    1,008
    351
    One of the things that I had in mind when I asked about drilling the pipe was to put electrodes in the pipe walls as Mr Chips suggests. I guess this would be OK for fluids and solids largely submerged in fluids but could be get false readings from slow moving solids. It may still work though without going for full-on tomography - does anyone want to measure the conductivity of their pee?
     
  18. NicNakNoe

    Thread Starter New Member

    Nov 24, 2013
    4
    0
    Ha - I'm surprised how far we got without toilet jokes too!

    Hmm I never considered the 'slow moving solids' aspect - in a country where diarrhoeal diseases are common this is likely to be quite an issue.

    If I go back to basics - the idea is to get a general idea of what's coming in to the toilet. The NGO do not know how many people visit the toilet - and how much waste/water/volume each 'visitor' contributes. Some pits are filling much quicker than others...and they can't figure out why...the probability is that in some areas a toilet cubicle is the only private place around so lots of people may be showering in there. (the easier option might seem to just ask but this is data is necessary for an overall water balance they are working on)

    So - number of users plus volume contributed per user would be great (and is far more important than breaking down solids/water etc).

    This EIT idea makes sense to me - but the slow moving s*** aspect rather messes up the simplified version of this I had in my head.

    Could I put a laser over the outflow of the pipe and just measure the length of time its broken?

    Time scale is tight - but still have ~2 months to figure this out.

    Pipe will only be about 1m long (after the flushing pan) at approx 40 deg to the horizontal...so a a drop of about 0.6m.

    For an example toilet see: http://sanitation.tap.waterforpeople.org/photo-gallery/pour-flush-latrine-ntinda-kampala-uganda-1

    (Thanks for your help already - I really appreciate you taking the time to contribute to this discussion)
     
  19. sirch2

    Well-Known Member

    Jan 21, 2013
    1,008
    351
    What about a strain gauge in the toilet seat - sorry, couldn't resist the pun.

    You could separate visit counting from volume measurement - put a flap/broken beam on the end of the pipe to record visits and put an ultrasonic distance gauge under the lid of the pit to measure how full it is.

    All that said I have worked on several sewage works and they are fairly harsh environments - you are going to need to protect anything close to the effluent pretty well. I doubt you are going to get anything very reliable much under $200.
     
  20. wayneh

    Expert

    Sep 9, 2010
    12,153
    3,059
    Rather than measuring the "output", how about offering the water in aliquots (say, 1L) and then just count how many each person uses. (Aliquots, because of the potential for abuse of a steady stream.)

    The output of an average human is a known thing. I think using the standard values would actually give you more precision than almost anything you might construct in the field. You might think of ways to detect outliers from the expected results.
     
Loading...