Shop OBEX P1 Docs P2 Docs Learn Events
Real world microcontroller challenge -- How to monitor water filter replacement. — Parallax Forums

Real world microcontroller challenge -- How to monitor water filter replacement.

LoopyBytelooseLoopyByteloose Posts: 12,537
edited 2015-10-24 11:39 in General Discussion
Consumer water filters are extremely popular in Taiwan and throughout Asia. But I can't seem to get a really good answer to how people know when to change their filters.

Of course, the easiest way is by the calendar. But measuring the volume of water processed seems more accurate to me. And of course, actually testing the water for quality and testing the water pressure for a clogged filter would seem to offer up some reasons for a change. Some filters need occassional backflow purging to retain a long life.

I don't even know if this is possible to properly monitor without using a chemical set to test samples of water for safety. But if a microcontroller monitoring system could monitor, accumulate data, and issue warnings, I strongly suspect that there would be a commercial demand for such a technology.

Here is a quick link to a site to give you something to think about.
http://www.advancedwaterfilters.com/buying-guide-whole-house-water-filtration/

Overall, I think the industry has a lot of users that mismanage filter changes and the current criteria for change might just be sales driven.
«1

Comments

  • Heater.Heater. Posts: 21,230
    Interesting problem.

    Presumably measuring the volume of water passed through a filter is easy enough.

    If you want to measure the actual effectiveness of the filter then we would have to know what it is in the input water that should not be in the output water.
  • LoopyBytelooseLoopyByteloose Posts: 12,537
    edited 2015-10-24 16:12
    The problem is not simple. There is an array of items that filters can be provided for. Some are quite simple; others require some knowledge of chemisty. And of course, different places have different water problems.

    But having lived for 20 years in Asia, there have been several water contamination crisises locally. And people are equally unsure of bottled water. Many think the softeners used in plastic bottles with cause sterility. Others think it is all just tap water. Everyone boils water before drinking.

    In today's news, Taipei is trying to sort out how to remove all its lead water pipe from service without actually telling the public about the specific locations and causing property values to fall. So filtering out lead is probably in high demand these days.. even though the government claims that even with the lead pipes being use the water is safe by international standards.

    In the past, we have had tank trucks caught dumping toxic solvents from the semiconductor industry waste directly into a stream that feeds into the water supply for Kaohsiung and everyone had to stop drinking city water for about a month. But at that time, a lot of people turned to home water filtration... even though they were unsure if the filters actually could remove such solvents.

    My feeling is that there is a tremendous commercial potential for someone that sells and manages replacement of water filters right. I suspect a lot of the vendors are not doing anything close to an adequate job.

    Meanwhile, I still drink the tap water after boiling it. So does the dog and he is doing well at 14 years of age. And the landlady put in a water machine in the hallway that delivers hot, cold, and warm filtered water that may never get a new set of filters.

  • ercoerco Posts: 20,256
    If it's just sensing a clogged particle filter, a few pressure sensors before & after the tap valve might give enough experimental data to correlate to sense excessive back pressure.
  • I have depended on a fifteen by five foot dug well for over tweny five years.We moved forteen year's ago to a new house on the same property. This brought about a short course in water system's. Starting with a new well pump, line's to the well, whole house filter, chlorine injection, presure tank, retention tank, carbon filtraiton system, water softener, reverse osmosis. I done everything but the ditch. think I would need to monitor both sides of the whole house filter. Like to watch this thead progress.
  • kwinnkwinn Posts: 8,697
    Measuring the differential pressure across the filter is a good start, but you also need to know the flow rate of the water for it to indicate if the filter needs to be changed. The last house I had with a well I used a 20micron pre-filter for large particulates, an activated carbon filter to remove contaminants, a 2micron filter for any particles that got by the first two filters, and a UV sterilizer to kill anything that got by the filters.
  • ercoerco Posts: 20,256
    edited 2015-10-25 02:44
    Speaking of differential pressure... One of my favorite camping devices of late is the emergency drinking straw filter. Supposedly it filters giardia & more. Suck til' you juss cain't suck no mo', then it's clogged and you toss it. http://www.ebay.com/itm/Aquamira-30-Gallon-Frontier-Water-Filter-Straw-Emergency-Survival-Camping-A67005-/252132418191

    s-l1600.jpg
  • LoopyBytelooseLoopyByteloose Posts: 12,537
    edited 2015-10-25 07:42
    Well, I started this thread because I have absolutely no idea what people are doing about filter replacement and I simply haven't gotten any sensible answers from the filter users that I personally know.

    A microcontroller could do a lot, but a microcomputer may end up being included or just take over due it its power. It seems that using an SDcard to accumulate data with a real time clock stamp would be a minimal requirement.

    A. one needs a real time clock.
    B. one needs to store history over long periods of time.
    C. filters could have bar-code labels or some other means of identifying that would track performance, and replacement.

    There does seem to be the beginnings of a standard configuration strategy in the above comments.

    I suppose much depends on how many filters one desires or can afford or must have -- two filter systems, three filter systems, four filter systems, etc. And then, there are strategies for rural and urban water that may be tied to specific issues. Rural well water may have no concern with lead ions; whereas city dweller might worry a lot about this.

    Some filters might fail by clogging, making replacement obvious and easy to determine. But other might fail without notice. I guess which filters fail toward a safe mode versus which fail toward a hazardous mode needs to be identified.

    I hadn't been aware that some people actually add chlorine. That is another dimension. And I know that many USA water districts have quit using chlorine due to carcinogenic concerns and create ozone to do the same job.

    So far.. I see just passing mention of water softening. This is a big deal in southern Taiwan. You can't make a decence cup of tea or coffee with tap water due to its hardness. And clothing rapidly turns gray when washed directly in city water. Not sure how to specifically monitor this kind of filter, but what it does is to swap out CA ions with NA ions. I don't think they clog.

    ++++++++++++++++
    I guess what I am trying to say is this is a research project, not a quickie applications. Some sort of comprehensive framework needs to evolve so that people might choose what to include or reject for their location.

    Some special cases do exist where removal of a arsenic or lead ions are a special issue.

    And I don't feel that water sources are constant in their quality. Flooding and other events can suddenly change filtering priorities.
  • LoopyBytelooseLoopyByteloose Posts: 12,537
    edited 2015-10-25 10:52
    Well, I did a bit of reading about ozone water treatment as I do know that municipalites have gone over to it from chlorine water treatment.

    The good news is it is cheap, and eliminates the need for a lot of other levels of filtration.

    The bad news is that a home using 600 gallons of water per day would need a 2500 gallon holding tank. The ozone treatment requires 400% of the daily turnover to work well. (748 gallons require 100 cubic feet of storage -- 62 pounds per cubic foot.)

    http://www.tripleo.com/sizing.htm
    http://www.watertanks.com/products/0003-035.asp

    This seems to point to an observation that a lot of these other filter systems are just too small to deal with sustainable household demand. They may only provide a separate safe drinking water supply and just a few gallons per day.

    I need to re-evaluate whether there is any merit to using these devices over an extended period of time.
  • I believe microbes release acids into water so you could have a litmus test using a colorpal which had been previously calibrated to the acidic content of the normal water, also perhaps a methane sensor as microbes give off methane too?
  • It wouldn't make sense to install a large, expensive home water treatment system when the vast majority of home water usage would be fine with whatever comes out of the tap. You don't need drinking safe water to flush a toilet or wash a load of clothes.
  • LoopyBytelooseLoopyByteloose Posts: 12,537
    edited 2015-10-25 11:00
    It wouldn't make sense to install a filter system that doesn't do the job.. just provides a false sense of security. Defining what that job is should certainly be considered. We have a lot of people paying for bottle water that costs more per gallon than gasoline, just because they are wary of perfectly good water that comes out of the tap.. in the USA.

    I am not so sure about what comes out of the tap in Taiwan. But big installations deny those of us that rent or that are travelling.

    So.... the ozone treatment removes odor as well as iron and turbidity. The water looks cleaner, doesn't smell, and doesn't stain clothing a rust color. Also, it will reduce the acidity of water and has some ability to soften water. The hard water in Kaohsiung coats the shower walls and floor, requires treatment with an acid cleaner to get the bathroom looking good. But the cleaner eats all the grout out between the tiles and that is another task added to the pile.

    I got to looking at it because adding chlorine was mentioned and than has been phased out in the USA in major cities.

    There may be a smaller capacity ozone generator that might do just for drinking and cooking water. But still requires holding tank for an optimal installation... at 30 gallons per day, the required tank is 120 gallons and still large. Water heaters in the US usually are 40 gallons or less.

    My impression is the new filter market is trying to justify elimination of holding tanks entirely as they are both large and heavy. But ozone seems to be more optimal in terms of costs and simplicity.

    Ozone treatment is a minimal maintence method and uses about 55 watts of electricity per day. Advertising claims $5/month or less on your electric bill and cleanings that take 15 minutes on a quarterly and annual basis after the system settles into use.

    +++++++
    There was some discussion of using the ozone treatment first and then a reverse osmosis process, that would be a 2-step process. There also might be a front end filter to clean up really dirts water before the ozone treatment. So it would become a 3-step process. The ozone reduces the maintence with the reverse-osmosi.

    Cost wise, having a cheaper process do most of the heavy cleaning first is ideal.

    I really need to consider how much drinking water per person per day or per household per day is reasonable. And then try to figure out if there is an alternative to ozone that doesn't require a holding tank.

    Of course, each variation requires its own approach to monitoring and maintenance. Just beware of tiny devices that attach to a sink facet. I see these in appliance stores here.

    =======
    Testing litmus paper? You would require a robot providing an endless roll of litmus paper. There may be other means to test pH directly. But sensors accumulating dirt and becoming inaccurate is an issue. The ozone will simply drive our pH toward neutral without monitoring.
  • The differential pressure and the flowrate together gives the flow resistance = clogged'ness. The viscosity of water also comes into the equation if temperatures vary significantly, but usually they would not do so for water supplies. If the microcontroller measures the diff pressure and the flowrate and calculates the clogged'ness from this, it can tell when a filter is done. If it trends this value, it may predict when the filter needs to be replaced, assuming use pattern and supply quality average doe not vary wildely. What pollutes - or constitutes - natural water is bacteria, parasites, humus, minerals, metals and other chemicals (and H2O). Boiling it only kills the bacteria and the parasites. Filtering only removes particles above a certain size - metals and other chemicals generally passes, some bacteria and parasites may too, depending on how fine (micron) the filter is. To be fairly safe the water needs to be sterilized, ususally by means of chlorination, UV-radiation, or boiling. Particles in the water may actually be good - such as minerals and humus - so they are not really the main problem. But, even if the water is sterilized it can be bad. Typically metals such as lead and cadmium are contaminants that are impossible to remove by other means that distillation. Some chemicals can be even harder to get rid of. Radiation pollution is the worst though.
    The effect of minerals in the water is generally know as 'soft' (low Ph value - acidic) and 'hard' (high Ph - alcalic). Hard water contains (too much) minerals and will ofted deposit these when subjected to pressure differentials and temperature gradients (typically in the washing machines, coffee makers etc.). The other main effect it has, is that detergents (soap) are prevented from being effective (in modifying the surface tension of the water). To make hard water soft, acid in some form or fashion is therefore sometimes added.
    My guess is that R=dP/f**2, where R is the clogged'ness, is fairly close to the truth.

    You may want to google around in the process industry, where such filter monitoring is quite common.

    Erlend
  • JohnR2010JohnR2010 Posts: 431
    edited 2015-10-26 11:27
    Loopy, last spring I developed a propeller based TDS sensor that employs a ZigBee radio and an atlas Scientific TDS (Total dissolved solids) sensor. If you put a probe on the in and out of your water filter you will see exactly how well you filter is working. I even made an analog display to hang on the wall that shows the water quality for the whole house. Is this what your looking for? This is not based on flow but quality instead.

    Over the summer Atlas Scientific came out with a new TDS sensor and my next step was to port the circuit and code to use their new circuit. I got side track on my chicken door controller coopboss.com and put this project on the shelf buy my prototype is still running connected to our home water system hourly reporting the water quality over ZigBee.


  • lots of cheap tds meters/checkers for water purity on ebay:

    http://www.ebay.co.uk/sch/TDS-Meters-/147826/i.html
  • In my case, outside faucets only pass through the whole house filter. Everything else has to go through the softener due to a high level of lime, very bad for the mechanics. Chlorine injection was installed because of concerns of animals doing their business near the well, we hardly ever have that plugged-in. I have a concern with the activated charcoal filter system that was installed for chlorine removal, it should of had a bypas on it, activated charcoal can be your best friend or your worst enemy.
    640 x 480 - 107K
  • LoopyBytelooseLoopyByteloose Posts: 12,537
    edited 2015-10-25 14:08
    The idea that one element should have had a by-pass is interesting. My guess would be than any good system would allow a by-pass for any one element or all of them.

    So far it seems we have a lot of different items.

    A. filters that remove particles above a certain size.
    B. water softening
    C. chlorine injection
    D. ozone injection
    E. reverse-osmosis filter
    F. ultra-violet sterilization
    G. activated carbon filter

    Total Dissolved Solids? Not sure how important that is.
    https://en.wikipedia.org/wiki/Total_dissolved_solids
    It is a rough indicator of most the aesthetic aspects of drinking water.

    I was hoping to resolve mostly true safety issues. The first three on the list below. But it seems that resolving the other issues may actually make it easier or cheaper to acquire safe drinking water.

    Some items that may affect safety are hard to remove - maybe require distallation. So some water is just too darned difficult.

    1. sterile water
    2. inorganic toxins removed
    3. organic toxins removed
    4. softness or hardness managed
    5. no odor
    6. nice and clear
  • The idea that one element should have had a by-pass is interesting. My guess would be than any good system would allow a by-pass for any one element or all of them.

    In hindsight it would have been a good idea to have a bypass on flter, I thought it was always to be an integral part of the system, easy fix.

    An RTC would be essential to time and date stamp pump cycles or water demand.

    Pressure monitoring, in and out.

    Flow monitoring at various points, if all of that we're graphed, you would see a downward trend to be associated with filter clogging.

    I want to investigate system JohnR2010 has working, could use an LCD display maybe.

    _Mike
  • Related to pressure, I'm curios how ultrasonic "listening" would play out in something like this... As a fluid or gas flows over or through something (a filter in this case) it creates turbulence. Often turbid air or fluid will emit ultrasonic frequencies. Depending on the flow rate which is related to pressure, the intensity of the frequency as well as a possible frequency shift might be a relatively simple way to detect if the filter needs to be changed or not..... Ask yourself ... " How does my refrigerator do it?" ... It can't be overly complex, from a manufactures point of view due to any filed servicing that might be required.

    As far as ultrasonic listening goes, you would need an ultrasonic microphone that had a broad listening range instead of a typical fixed frequency transducer ... last time I checked these went for about $5 from Digikey
  • Sort of like a Dr. listening to your heart, and making a diagnosis.
  • erco wrote: »
    If it's just sensing a clogged particle filter, a few pressure sensors before & after the tap valve might give enough experimental data to correlate to sense excessive back pressure.

    That's the first thing that came to my mind as well.

  • That's a lot of pressure sensors, if you wanted to diagnose the problem in real time, on more than one filter. The the thread would end if we solved it tonight.
  • kwinnkwinn Posts: 8,697
    The idea that one element should have had a by-pass is interesting. My guess would be than any good system would allow a by-pass for any one element or all of them.

    So far it seems we have a lot of different items.

    A. filters that remove particles above a certain size.
    B. water softening
    C. chlorine injection
    D. ozone injection
    E. reverse-osmosis filter
    F. ultra-violet sterilization
    G. activated carbon filter

    Total Dissolved Solids? Not sure how important that is.
    https://en.wikipedia.org/wiki/Total_dissolved_solids
    It is a rough indicator of most the aesthetic aspects of drinking water.

    I was hoping to resolve mostly true safety issues. The first three on the list below. But it seems that resolving the other issues may actually make it easier or cheaper to acquire safe drinking water.

    Some items that may affect safety are hard to remove - maybe require distallation. So some water is just too darned difficult.

    1. sterile water
    2. inorganic toxins removed
    3. organic toxins removed
    4. softness or hardness managed
    5. no odor
    6. nice and clear

    For reliability and lowest operating cost you need to address these problems in a specific order. Going from the water source to where it is used the problems would be addressed as:

    1- 20micron particulate filter to remove large particles. If not removed they can coat and reduce the effectiveness of later stages.

    2 - Activated carbon filter to remove odor, organic, and inorganic toxins. Any particulates can coat the carbon filter and reduce it's effectiveness.

    3 - 2micron particulate filter to remove small particles, including any particles from the carbon filter.

    4 - UV sterilizer to kill any bacteria or virus. Any particulates in the water at this point can accumulate on the UV tube and reduce effectiveness.

    5 - Water softener, salt based or electronic to remove dissolved minerals from the water. Particulates can reduce the effectiveness of the salt based system by coating the resin beads. Not sure if they would do the same for the electronic system.
  • LoopyBytelooseLoopyByteloose Posts: 12,537
    edited 2015-10-26 04:16
    Monitoring turbulence as audio noise is flow seems a very interesting approach.
    One can always add turbulence factor to an otherwise 'quiet' system in the form of rough pipes, jumps, or sharp turns. Allow for calibrated turbulence and it may be a very economical sensor scheme.

    Do we have a typical or ideal complete system for a given setting? Maybe a urban dweller family of four would be a good context (Or family of three with a pet).

    I did run into small capacity ozone generators for hot tubs for $99.00 USD or less, so don't rule out ozone if you are willing to include storage tanks in a scheme. But I am not sure how to deal with excess ozone monitoring and control. Can one just ignore the excess and drink the water?

    Another secondary market is water filer cleaning devices.
    Many filters can be washed and flushed with reverse flow. But I haven't seen anyone selling setups to make the routine chore easy. Up until now, I have been thinking most people just are using filters in a one-shot manner. It doesn't have to be that way.

    I would love to know more about the usefulness of activated carbon filters. I'll try to read up, but some of you may have insights that sellers would rather not mention.
  • Hmm, you might also want to check out some of the reef forums like reefcentral.
    Reef aquariums generally need pretty clean water as you move form soft to stony coral growing.
    There are quite a number of uC knowledgeable folks involved in it, and not just Arduino....

    Ozone is nice, however to make it work, you normally need low flow and high lumens, and they need replacing probably almost as often as good filters, expensive too.

    You could go with an RO/DI, however they are very low flow and that why you need water storage. There is also quite a bit of waste water, so more than a couple of strikes against it for normal home use.
    If you want to go that way though, you do pay for water in TW, right? Monthly bill?
    If not, shouldn't be too hard to estimate water useage as you probably have low flow toilets and high efficiency showers, etc.
    Do the math, however just running a small reef tank with occasional partial water changes of 25g/week with a good 3:1 ratio means 75g of waste water to either pump on the lawn, or drain.


    IIRC, the water softeners are likely going to be the most economical way to solve your hard water problems, with the benefit of not having to worry about pressure differentials, reverse flushing intervals, etcs.
    Something like this I guess, I've never had to use one though, http://www.lowes.com/ProductDisplay?partNumber=416876-43353-WHES44&langId=-1&storeId=10151&productId=3824565&catalogId=10051&cmRelshp=req&rel=nofollow&cId=PDIO1



  • JohnR2010JohnR2010 Posts: 431
    edited 2015-10-26 11:29
    kwinn wrote: »

    Total Dissolved Solids? Not sure how important that is.
    https://en.wikipedia.org/wiki/Total_dissolved_solids
    It is a rough indicator of most the aesthetic aspects of drinking water.

    Well I can tell you the TDS sensor that came with our water filter does a great job of telling us when we need to replace it. When our reverse osmosis filter is working correctly it reads around 10ppm. It needs to be replaced up around 70ppm. TDS is going to catch most of the "stuff" but not everything (wont catch a water softener). I think basing your solution on pressure is going to be challenging, pressure is going to constantly vary on both sides of your filter.
  • LoopyBytelooseLoopyByteloose Posts: 12,537
    edited 2015-10-26 12:05
    Well TDS may work reasonably well for wholesale replacement of all the filters. But I can't help but think the first filter should be replaced more often than the last.

    I could claim that I am just trying to be objective about this. But then the Internet pokes this quote in my face.

    “Objectivity works to repel the attacks of critics, like a kind of ethical pepper spray.”

    How does one maintain and verify the TDS meters calibration?

    What we all really want is safe drinking water.
    But I am beginning to think I have no idea what I am putting in my mouth and may never have. Trying to be informed about it isn't simple. I guess I will know the water is unsafe when people around me start dropping off in dozens.

    It is work to get facts that offer a clear insight in how to do something.

    What should I do with the following?
    Objectivity has about as much substance as the emperor's new clothes. ~Connie Miller
  • You calibrate the TDS sensor based on the parts per million of a known solution. There are standard references based on the type of water your going to monitor. I calibrate my probe with this solution set from Atlas. To calibrate the probe (and circuit) its a 3 step process. First you do a dry reading (probe out of water), then dip it in the 84uS (microsiemens) solution, and finally the high reading in the 1,413uS solution. You do this to both probes then stick one probe in the filter's inlet and the other in the out of your filter pack. With a reverse osmosis filter pack you see a huge difference between the two probes. Over time (~10 months in our case) the output prob's reading gets closer and closer to the input. Another thing that is worth mentioning, is the TDS value has to be temperature compensated based on the temperature of the water and that adds quite a bit to the calculations. But the atlas circuit takes care of all that. I just read the value through the async interface.

    There has been a lot of work around detecting what is in your water. If you really want to know, take a look at this kit for $787 from Atlas-scientific.com Its just one of many. It seems to me if your willing to base your detection on pressure difference between the input and output of a filter group your looking for a general indicator that something is wrong (and I think that is the way to go). All I'm saying is the TDS value of your water is as good of an indicator or better for that. Will TDS detect everything dangerous? Not by a long shot but it is a good over all indicator something has changed.

    I think your really on to something here with your quest!! I bet in the US your going to find people fall into one of two general categories; 1) they don't trust their home's drinking water and never drink it unless it is filtered or, 2) they trust the local water utility to take care of that and tell them when something is wrong. In both cases I would guess 90% or more do not actually monitor their water.

    Good stuff Loopy!!
  • kwinnkwinn Posts: 8,697
    JohnR2010 wrote: »
    kwinn wrote: »

    Total Dissolved Solids? Not sure how important that is.
    https://en.wikipedia.org/wiki/Total_dissolved_solids
    It is a rough indicator of most the aesthetic aspects of drinking water.

    Well I can tell you the TDS sensor that came with our water filter does a great job of telling us when we need to replace it. When our reverse osmosis filter is working correctly it reads around 10ppm. It needs to be replaced up around 70ppm. TDS is going to catch most of the "stuff" but not everything (wont catch a water softener). I think basing your solution on pressure is going to be challenging, pressure is going to constantly vary on both sides of your filter.

    No argument from me regarding the value of TDS sensors for ro filters, but they do not work for particulate filters since they have little or no effect on dissolved solids. For particulate filters you use a low range differential sensor that measures the difference in pressure between the inlet and outlet of the filter, not the absolute pressure on each side. They are used on both water filtration and air handling systems.
  • How much do these filters actually cost and how long are they normally expected to last?
  • kwinnkwinn Posts: 8,697
    @Loopy

    I don't know what the situation is in your area of the world but here I can take a sample of the well water and have it analyzed for all the items you listed in your post and more. Knowing what you are dealing with will be a great help in designing a suitable system and can save you a lot of money in the long run. I recommend that you start there if possible.
    Another consideration is where to put the filters. You may want to remove particulates for all uses, minerals for a few areas, and other contaminants for water you drink and cook with.
Sign In or Register to comment.