Shop OBEX P1 Docs P2 Docs Learn Events
Check the clarity of your water — Parallax Forums

Check the clarity of your water

cessnapilotcessnapilot Posts: 182
edited 2010-07-05 05:00 in Propeller 1
Hi,

Light scattered by the particles allows particles to be detected in solution, just as sunlight passing through a window is scattered by dust particles in the air, allowing them to be seen. Pick up a glass of water and hold it to the light. Can you see any finely divided, insoluble particles suspended in the water? Or does the water seem hazy? If so, the water is turbid. The amount of light scattered and absorbed by the suspended matter in the water is reported in units called Nephelometric Turbidity Units or NTUs (from a Greek word nephelo meaning "cloudy"). The more particles that are in the water, the higher the NTU number.

The measurement of turbidity is a key test of water quality. You can measure the water’s turbidity with an electronic turbidimeter. The basic turbidimeter has a light source and a photoelectric sensor that measures the light scattered by suspended particles in a water sample. The scattering of light results in a change in the direction of the light passing through the liquid. Light is scattered backward, sideways and forward. If the turbidity is low, much of the light will continue in the original direction. Because of the physics of light scattering and the optical properties of a glass of water as a lens, we can construct a simple, low-cost, but sensitive turbidity detector with a pointing laser, with a variety of optical sensors and with the Propeller microcontroller. Except the cheap pointing laser and it's on/off control circuit, all other main electronic components of this DIY turbidimeter are sold by Parallax inc.

One should not mix color and turbidity and should not associate clarity with healthiness. Dark brown water can be clear and extremely healthy, like a good orange pekoe tea, and crystal clear water, or a deliciously looking glass of wine can be deadly poisonous. This second illusion changed the course of human history many times.

The turbidity associated with the concentration of 1mg of SiO2 or Kaolin clay per litre is equivalent with 1 NTU unit of turbidity, but the silica or Kaolin clay used had to meet certain specifications as to particle size and weight. In a household finely grained coffee powder can be used as first substitute. Approximately 1 g of very fine coffee in 1 liter of tap water well stirred up, gives a solution with about 100-150 NTU of turbidity. Nescafe or any kind of instant coffee does not work (at least) here. If you have only instant coffee at home, finely granulated garden soil applied in the same amount as the coffee
powder, will do.

Some possible applications of the DIY turbidimeter:

At home:
=======

Tap water:
Cloudy tap water can be caused by tiny air bubbles in the water similar to gas bubbles in carbonated drinks. Wait a few minutes to see if the bubbles clear up. After that the water should be clear and will be clear. However, low, but unhealthy levels of turbidity are generally invisible to the naked eye. Most well owners will not notice high turbidity at levels at or below 5 NTU in their drinking water. However, health standards are much lower, usually lower than 1 NTU for potable water. Higher levels of turbidity can interfere with the disinfection process and provide a medium for microbial growth. Even some brands of bottled water have been found to contain high levels of turbid contaminants (besides plastics chemical leaching from the bottle).

Aquarium:
The water in aquarium must be kept clean and contain adequate amounts of dissolved oxygen. Any aquarium tends to build alkali level and get cloudy when water is not changed at the right time. Clarity is more important in sea water tanks than in fresh water ones. Tropical environments in particular contain very few dissolved solids and they are the most transparent waters on the planet. A marine aquarium must therefore be filtered more efficiently and to a greater extent than a freshwater tank. Whatever the other important parameters are, pH or salinity for example, cloudy water should be changed immediately.

In the environment:
===============
Check of the clarity of natural waters can be more useful than the fun of just being there, doing that and getting wet. 10 to 40 NTU in a lake can be tolerable by the creatures there, but over 60 might need attention of the stewards of the environment. A value of 60 NTU indicates water that is relatively clear to a depth of five inches in shadow. As a comparison, a value of close to zero NTU corresponds to a natural water with visibility to 5-8 feet vertical, in not direct sunlight.

Care of Nature:
Heavy rainfalls, strong winds, convection currents or industrial activity can greatly increase the turbid state of lakes, reservoirs and rivers. Floating particles absorb heat in the sunlight, thus raising water temperature, which in turn lowers dissolved oxygen levels. Warm weather can also add to the problem. In natural water bodies high turbidity can reduce the amount of light reaching lower depths, which can inhibit growth of submerged aquatic plants. This decreases the rate of photosynthesis, so less oxygen is produced by plants. and consequently affects fish and shellfish, which are dependent on them. These processes typically upset Mother Nature's balance and cause unintended consequences down the road. The sooner we can detect such situation, the better are our chances to prevent damage to the environment.

Monitoring natural processes:
For with warmer weather, microorganisms and aquatic plants renew their activity in clean water. As they grow and later decay, these plant and animal forms substantially add to the turbid state of that previously clean water.

Freshwater fishing:
From the fish's point of view, the clarity of water is important and the effects of turbidity on freshwater fishes are well known by the anglers. Among other physiological effects, they found that fishes eventually loose appetite in turbid water. So, success of fishery can be affected by the time integral of turbidity level for a natural water body. It is suggested that clay turbidity levels in earthen ponds should be kept below 100 mg/l (approx. 100 NTU). This, of course, depends on species. For example, unlike most fish, carp can grow well under conditions of high turbidity. In fact, carp feeding areas quickly become turbid, and all other fish species move out. Turbidity could act as a cover and protect small fish from predators such as large fish and birds. Fish show diel migration in lakes so as to avoid predators or to find food. Turbidity of water should be considering by anglers when planning on horizontal migrations and distribution of fish in shallow lakes.

Marine fishing:
Turbidity levels in marine systems are generally not as extreme as fresh water therefore, behavioral effects are more important than physiological effects. Fishers might be interested in those effects, though.

Ponds and Fish farms:
Parasites or bacteria are found in any pond, running water and in any fish, usually coexisting in peace with fish harmlessly as far as the water is in good conditions. In fish farms the feed intake drops and the conversion rate (feed to fish flesh) increases with increasing water turbidity. As viruses and bacteria may be shielded and protected by turbid particles. The outbreaks of many diseases can result from and associated with high water turbidity. Nothing is better than early detection of such unhealthy situation. If you are a fish farmer or pondist, you certainly keep a careful observation of turbidity every day.

Plan of a DIY laser turbidimeter/colorimeter with quad modes of detection
=================================================
The arrangement of the laser and the light sensors is shown in the 4th attachment. A TSL230 B&W sensor chip is used to measure the transmitted laser light, and B&W line sensors, TSL1401s at 5 - 25 degrees collect and integrate the faint beams of forward/backward scattered light. The glass of water acts as an optical lens and deflects the scattered photons toward the line sensors, which have a 90 degree field of view. The actual angle, where the measured signal is maximum depends on the diameter of the glass and the distance of the TSL1401 from the glass surface. The TSL230 point sensor works at high level of turbidity in 'transmittance mode' and the line detectors work at low level of turbidity in on/off phase sensitive integration mode, collecting and summing the intensity of the wide beam of the scattered laser light. On/off phase-sensitive integration is achieved by periodically switching the laser ON and OFF and adding the corresponding data of the TSL1401 to SIGNAL and BACKGROUND data arrays. The difference of these arrays, pixel by pixel, are summed finally to give a sensitive detector output figure for low turbidity samples. One can, of course sum then subtract to have the same result, but the distribution or the maximum of individual pixel difference might carry useful information. Exposing and integrating long enough, the TSL1401 sensors measure well below 5 NTU easily with low noise and with high sensitivity, according to plan.

This basic set of sensors is expanded with two diagonally spaced TCS230 Color Sensor Modules, oriented square to the laser beam. This 90 degree orientation is the so called nephelometryc angle. These color sensing modules measure the sideward scattered red laser light, using similar on/off phase-sensitive data integration mode of the red sensors as the line sensor. They are planned to work optimally in the medium turbidity range. Beside measuring turbidity from the nephelometric angle, the TCS230 modules can measure some color properties of the liquid. The modules can switch the white LEDs ON and OFF in an alternating manner. While one of the TSC230s exposes the sample with intensive light, the other TSC230 measures the RGB components of the transmitted flock of photons. An easy to apprehend front panel design of the planned combined instrument is shown in the last attachment.

For precise measurement of natural water samples, the measuring unit should be able to be placed on a standard chemical laboratory magnetic stirrer, I guess.

Attachments
============
The 1st attachment shows the laser pointer and the glass of water + milk used in the making of the 2nd and 3rd attachments. 1 ml of 3.5% (alpen)milk was diluted in 64 liter of water. Not in the bathroom, but with 6 1:2 dilution with tap water from the first (1 l water + 1ml milk) solution. So, 4 l of water was wasted, altogether. The resulting water + 16 ppm milk sample in the glass seemed clean for the naked eye. The 2nd attachment shows the forward scattered beam, viewed from ~25-30 deg angle, and the 3rd shows the beam from the nephelo 90 deg angle. A Fujitsu FP7000 was not able to autofocus onto the beam inside the glass, so I used manual focus in macro mode. With tapwater only, these beams are so faint that I was not able to make a good photo of them. And this is the point and the take home message!

The 4th attachment presents the planned arrangement of the sensors. The last one displays a suggested front panel design of the measuring unit. Please note, that presently I am only collecting parts, cutting metal of the optical pad and experiencing with the sensors. Comments, (link to) device drivers in SPIN/PASM if any, shared experience with such measurements, sensors, colorimetry or suggestions of constructions are useful and welcome.

I have a dream of finding a collection of working and stable SPIN/PASM device drivers for all those sensors...

Cheers,
Istvan

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔


Intentionally Left Blank
423 x 317 - 26K
423 x 317 - 21K
384 x 288 - 22K

Comments

  • AleAle Posts: 2,363
    edited 2010-07-03 17:29
    cessnapilot said...
    Light scattered by the particles allows particles to be detected in solution
    [noparse][[/noparse]pedantic]
    There is something very wrong in that statement: If it is a solution it is homogeneous, no particles. If there are particles it is a heterogeneous mixture [noparse]:)[/noparse].
    [noparse][[/noparse]/pedantic]

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Visit some of my articles at Propeller Wiki:
    MATH on the propeller propeller.wikispaces.com/MATH
    pPropQL: propeller.wikispaces.com/pPropQL
    pPropQL020: propeller.wikispaces.com/pPropQL020
    OMU for the pPropQL/020 propeller.wikispaces.com/OMU
    pPropellerSim - A propeller simulator for ASM development sourceforge.net/projects/ppropellersim
  • ElectricAyeElectricAye Posts: 4,561
    edited 2010-07-03 18:04
    Interesting post. I'm in the middle of working on an Optical Densitometer to measure the growth rate, etc. of cells suspended in solution. I plan on using the TSL230 in transmission mode and to play with how it might work when I allow the light to backscatter. I'm using different colored LEDs to see if I can detect any kind of wavelength-dependent signature that helps predict where the cells are in their growth process. Keep us posted on what you're doing here with your turbidity measurements.

    thanks,
    Mark
  • Tracy AllenTracy Allen Posts: 6,664
    edited 2010-07-03 18:12
    Nice project, and an especially thorough and thoughful write-up. I've attached a couple of photos of installations for stream monitoring, using the expensive sensors ("OBS3+" Optical Back Scatter) from D&A Instruments, now acquired by Campbell Scientific. You're right about the cost, these things run near $1000 each, but the internal principle is very simple. They use an LED, not a laser. Some other brands, such as McVan instruments of Australia, have a little windshield wiper to keep the window free of algae and crud. A lot of the cost is in the housing and the waterproof connectors and rugged cable, and low volume production.

    The biggest issue we face in those installations are the fast water hazards. When first installed, the cross-bars on the flume in the first photo were made of metal angle stock, and at high flow a tree trunk came down at high velocity and tore out the whole 1200 pound flume and carried it down stream and smashed it up against a bridge abutment. Now the cross-bars are wood, but at this point in the remediation there is much less danger. The OBS is the object mounted on the side of the flume, and the vertical columns are stilling wells for depth measurement for volume flow calculations.

    The other two photos are a location downsteam of a bridge. It is a great protected location for a measurement. However, every few years there is so much intense rain that the water crests over the top of that bridge. I've taken to mounting the OBS inside a length of stainless steel pipe with the end cut off at an angle.

    There are issues of data interpretation too, which you pointed out. Sandy silt has a much different backscatter than the same weight of fine clay mud. In some cases we have these hooked up to a comms channel, so that a person can go out and take a grab sample for analysis. In the case of the flume, it was a land line modem. In the case of the bridge, there is an online USGS gaging station not too far downstream.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Tracy Allen
    www.emesystems.com
    787 x 614 - 981K
    491 x 657 - 588K
    491 x 646 - 641K
  • heaterheater Posts: 3,370
    edited 2010-07-03 19:34
    Tracy: Sorry, that made me chuckle. Here we are discussing how to detect microscopic impurities in water and you have a socking great tree trunk in there !

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    For me, the past is not over yet.

    Post Edited (heater) : 7/3/2010 9:08:07 PM GMT
  • $WMc%$WMc% Posts: 1,884
    edited 2010-07-03 21:00
    cessnapilot


    Heres a couple of ideas that I think would be useful for your project.

    1st: a square beaker .... I've never seen a round beaker used in any opto water analyzer.
    ' sources: HACK, Honeywell, Rosemount, Chem-treat, etc.
    '
    2nd: a laser is really powerful and it could be destructive to the water sample. I would use another light source like and LED or a incandescent lamp,etc.
    ' sources: HACK, Honeywell, Rosemount, etc.
    '
    3rd: the temperature of the water sample is really important. Pure lab grade water will become cloudy just above freezing and just below boiling.
    sources: Parallax, etc.
    '
    The rest isn't for turbidity testing but its needed when well water test are done to determine if the water is OK to drink.
    '
    4th: condutivity... microsiemens/cm or micromhos/cm = specific conductance, pure lab grade water will be extremely low in conductance. Tap/well water,Not so much
    sources: HACK, Honeywell, Rosemount, etc.
    '
    5th: PH 0-14, acidic or alkaline. A turbidity test wont show the PH of the water sample. Turbidity test Will be close to or the same for a water sample with a PH of 7 and a water sample with a PH of 2, DON"T drink the 2 PH water sample.
    sources: HACK, Honeywell, Rosemount, Chem-treat,etc.
    '
    6th: REAGENTS, This helps to determine whats in the water and how to treat it or filter it.
    sources: HACK, Honeywell, Rosemount, etc.
    '
    Please note that Turbidity testing alone IS NOT sufficient to determine if the water sample is fit for consumption!

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    The Truth is out there············___$WMc%___···························· BoogerWoods, FL. USA


    want speed?·want·to use the Propeller?·want to stay with BASIC___www.propbasic.com___


    You can feel stupid by asking a stupid question or You can be really·stupid by not asking at all.

    ·
  • ElectricAyeElectricAye Posts: 4,561
    edited 2010-07-03 21:28
    $WMc% said...
    ....

    1st: a square beaker .... I've never seen a round beaker used in any opto water analyzer.....


    I think $WMc% brings up a good point about the beaker. The curved, imperfect surfaces might make your measurements prone to distortions of the light beam or misalignments with its surface. A narrow laser beam might "see" localized effects (warps in the glass, clumps in the water) much more than the overall state of the liquid. So you might need a container that is optically more consistent. See, for example, how cuvettes are used in spectrophotometers:


    www.carolina.com/product/spectronic+plastic+semi-micro+cuvette%2C+pack+of+25.do?keyword=cuvette&sortby=bestMatches

    Also, I'm guessing it might help to disperse your light source somewhat. By using a beam that is narrow, that beam becomes susceptible to single particles that might happen to cross the beam's path. A diffused light source might actually make for better measurements.

    just my 0.00 cents worth.
    smile.gif
  • cessnapilotcessnapilot Posts: 182
    edited 2010-07-04 08:49
    Well, a curved surface can be perfect... OK. Let's get back to our topic.

    @ElectricAye

    I have made my first experiment with a hefty 10 mW laser diode with 35 deg (10 deg) oval shaped dispersed red beam. The baker was a cylindrical (~12 cm diam.) glass bottle and the detector was a TSL230, and a Propeller counted the pulses per sec. The round shaped water 'lens' effectively focused the dispersed beam toward the detector. It gave higher counts with the bottle+liquid than without them. This was the case with different solvents. E.g. with ethanol(96%) and acetone(100%).

    The TSL230 driver worked in autoscale mode. (100 ms for scale, 900 ms to count.)

    I used a collimator (3 cm brass tube with 8/4 mm od/id) with black internal coating before the detector to reduce the effect of ambient light. The detector was mounted inside a black plastic box. The typical count with laser on was about 25-50 K/sec and with laser off in daylight 5-10 K/sec count. The scale factor was 1. I felt that counts with laser on were low when compared to the background. Although the system detected a drop of milk in 1 liter water by ~10% decrease of the counts.

    With a ~0.5 mW red laser pointer beam the detector counts went up to 10-20 times higher (scale 10) as compared with the dispersed beam laser. Relative increment to the background then was about 100. And all these were achieved with a much lower power laser. The shining line of the laser beam could be seen from different directions but with different intensity. And then came the idea : to detect a line use a line detector. (That might even yield the scattering profile.)

    To measure the overall state of the liquid sample, I am making the baseplate of the detector from aluminum. The whole thing can be put on top of a standard laboratory magnetic stirrer. Such a stirrer can be driven from a 12V car battery with a 400 W cheap inverter. (Without heating, of course.)

    @$WMc%

    Yes you are right, turbidity check is only one of the tests of water quality. It is important, though. Maybe the most relevant...?

    I even wrote that clean (non turbid) water can be poisonous.

    There are round shaped sample containers perfect enough to give less than 5% intensity variation when rotated. I tested that. For a DIY turbidimeter,·5% is tolerable. The key point here is the sensitivity. Design objective is to measure below 5 NTU with laser ON/OFF phase sensitive integration. Calibrated results like 0.6 +-0.1 NTU would be great and useful. Price per performance of the device would be attractive, too. (About 20-50 times better than those of the professional laboratory units.)

    Note that all 5 sensors can be controlled by a single Propeller.


    Cheers,

    Istvan··

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔


    Intentionally Left Blank
  • AleAle Posts: 2,363
    edited 2010-07-04 11:42
    And how about the test for solubles in diethylether ?.
    After all that the water cannot be declared drinkable, you have to look for bacteria, enterobacteria [noparse]:)[/noparse]. That takes a couple of days, you have to take the sample, make a couple of cultures leave them growing at 37 °C and see if there was growth, if there was growth and depending on the number of samples that have it the water could be declared drinkable, if all other parameters are within specifications [noparse]:)[/noparse] I do not remember the numbers or the media used for culture, it was some 10 years ago, but I'm sure it can be found online.
    A 0.2 um filter will remove the bacteria, water will be then sterile but could still be poisonous [noparse]:)[/noparse]

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Visit some of my articles at Propeller Wiki:
    MATH on the propeller propeller.wikispaces.com/MATH
    pPropQL: propeller.wikispaces.com/pPropQL
    pPropQL020: propeller.wikispaces.com/pPropQL020
    OMU for the pPropQL/020 propeller.wikispaces.com/OMU
    pPropellerSim - A propeller simulator for ASM development sourceforge.net/projects/ppropellersim
  • bdickensbdickens Posts: 110
    edited 2010-07-04 18:35
    I've been running some down and dirty versions of turbity off my dock and so far, my single biggest problem (other than oil) is that it is hard to keep the tubes clean that the water runs through. Algae and other fun things seem to like the inside of my sensor. So I need to take it out, recalibrate and clean it about every week. I almost built a tube brush to clean them before use, but that seemed a bit absurd.

    I have been wondering if a camera sensor and an automated secchi disk would be a better choice but I'm still hoping to come up with something better
  • Cluso99Cluso99 Posts: 18,069
    edited 2010-07-04 18:59
    cessnapilot: Congratulations! You have presented a well thoughtout plan of attack. I wish you luck in your project. Thanks for sharing.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Links to other interesting threads:

    · Home of the MultiBladeProps: TriBlade,·RamBlade,·SixBlade, website
    · Single Board Computer:·3 Propeller ICs·and a·TriBladeProp board (ZiCog Z80 Emulator)
    · Prop Tools under Development or Completed (Index)
    · Emulators: CPUs Z80 etc; Micros Altair etc;· Terminals·VT100 etc; (Index) ZiCog (Z80) , MoCog (6809)·
    · Prop OS: SphinxOS·, PropDos , PropCmd··· Search the Propeller forums·(uses advanced Google search)
    My cruising website is: ·www.bluemagic.biz·· MultiBlade Props: www.cluso.bluemagic.biz
  • RavenkallenRavenkallen Posts: 1,057
    edited 2010-07-05 04:01
    Could you bombard the water with U.V radiation? That would take care of most of the problems, right?

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Propeller + Picaxe = Romeo & Juliet
  • ElectricAyeElectricAye Posts: 4,561
    edited 2010-07-05 05:00
    Ravenkallen said...
    Could you bombard the water with U.V radiation? That would take care of most of the problems, right?

    It might kill bacteria but it wouldn't remove lead or cadmium. burger.gif
Sign In or Register to comment.