I was searching for Noah's first word on our blog (I found his first signed word: please, but I couldn't find his first spoken word) because Faith has started talking a little, and I don't know exactly what her first word is, probably "hi".  But, today, she has repeatedly said, "tada" when putting a bit of a puzzle together.

While searching for that, I came across an old post where I had posted some word problems for you to solve.  And today, I have another:What method is the best (considering cost) for purifying a drinking water system that occasionally gets coliform bacteria in it?  Your answer should be expressed in terms of:

G = desired gallons per minute (I can't decide between 6 or 8)

C = preference to not put chlorine in the system (which then needs to be filtered out later, with various disposable parts)

B = number of coliform bacteria colonies (I don't know if this number is relevant to your answer)

T = temperature of the water at the output of an ultraviolet solution (should you decide to go that route)

W = wattage of the ultraviolet light

H = hardness of the water, and consider whether the water is too hard for proper ultraviolet cleaning, or if our water, though hard (13-17 grains per gallon) seems to not build up scale very much

$ = cost of the ultraviolet system ($1000 - $1500)

S = Cost and value of an automatic shut off solenoid valve for power outages, to avoid contaminated water going past the light when it is off.

I = Cost and value of a light intensity meter which watches the light and effectively measures the scale build up to know how often it must be cleaned.

E = Cost and value of a nifty gadget that has an ethernet jack on it, which allows for connection to the network, for easier monitoring, graphing, etc.


For bonus credit you can add in other variables affecting available brain power for decisions such as a wife with a flu/possible sinus infection, a new wood stove that needs a chimney liner, and a pile of wood to be stacked outside in the rain.

Posted by Jon Daley on October 15, 2009, 1:22 pm | Read 17916 times
Category Reviews: [first] [previous] [next] [newest]

One thing I learned in math class is that if the number of unknowns exceeds the number of equations, a solution requires more advanced techniques than I ever learned.

For those who might panic at the word "coliform," as I understand it, the coliform bacteria are not harmful in themselves, but are an indicator that something might be happening with the well that could also let in dangerous microorganisms.

Consider also variable A1 = a dad with all these problems and decisions who takes all three kids-under-six to the store with him so his sick wife can have a few moments of peace and quiet.

And variable M = having your mother-in-law visit with everything else going on. I've told him he doesn't need to impress me with his ability to handle unexpected problems by having several of them occur every time I visit, but I don't think he heard me.

Posted by SursumCorda on October 15, 2009, 4:55 pm

How about checking it for bromine content and, if that is low enough, using ozone?

Posted by Stephan on October 16, 2009, 9:01 am

I haven't heard of that method. I'll check it out, but I ordered the ultraviolet light yesterday.

Posted by jondaley on October 16, 2009, 9:24 am

This page has a list of pros and cons for ozonation.

Posted by jondaley on October 16, 2009, 9:35 am

One other option might be solar disinfection, but that, as I understand it, isn't really practical for piped water - you'd have to have enough UV-transparent piping on the south side of your house to allow for, oh, probably two days of transit (for cloudy days), and wouldn't work on rainy days...

My solution is roughly as follows:
Likelihood of using chlorine
RCl = (1-C)*((40-H)/40)*($UV/$Cl)*(max$Cl/max$UV)*(G/10)/log(W/1000)

I would incorporate S, I, and E, but they are defined as being both something that favors the chlorine approach (cost) and something that favors the UV approach (value).

Posted by Stephan on October 17, 2009, 5:36 am

I had to write that down on paper to be able to start to understand it.

And yes, I think you can safely ignore the S,I,E variables, since they are biased, and I haven't even thought about any corresponding devices in the chlorination realm.

I think your 40-H part is incorrect, unless I am reading it incorrectly, your formula says that you are less likely to use chlorine the harder the water? Which is incorrect, at least when comparing to a ultraviolet solution.

And now that I think about it, I guess G isn't really relevant to deciding between which system, though G affects cost in both cases (wattage of the light, and size of the chlorine tank).

And I can't remember my logs well enough, so had to get out my TI-85 calculator to graph it, but it seems that number is always negative, and so I don't know what that means in your answer.

(Linda says that the trouble with writing math problems is then you have to grade them... She also says it isn't fair that I didn't give you all the variables:

F: flow restrictor, which limits the flow of water (thus ensuring the UV light gets enough time to irradiate the bacteria

SC: (boolean) internal scrubber brush that makes cleaning the inside of the UV light easier, thus (hopefully) allowing for harder water to be maintained

I ended up with a 6 GPM system, with both of the above add-ons, but not any of the fancier ones, and we'll see how well it works, and whether the extra solenoid valve, and/or the light intensity monitor is needed for good monitoring.

I'll probably end up with some sort of inline chlorinator at some point, that is used for temporary chlorination when the power goes out and some untreated water slips past the light, or for other maintenance reasons. It would also take care of the iron bacteria in the MetalMaster particle filter that we have, though my recent water tests say that the iron bacteria is gone.

There are a couple inline (as opposed to tank based, which is required for on-going bacteria removal) chlorinator systems, one nifty little one $80 for the connectors, and $20 for a one-use/disposal chlorine tube. A $300 solution with chlorine pellets, and then this morning I woke up thinking about making my own, since I would only use it in maintenance situations. It seems like there should be a way to take a 5 gallon bucket with a water-tight lid, and dump some bleach in it, and then redirect the water supply through the bucket - sort of use the bucket as a pressure tank of sorts, though without a diaphragm or anything. Haven't quite figured out how that would work.

Posted by jondaley on October 17, 2009, 7:24 am

You're right, I didn't think any farther than "Ooh, a log would be a good way to get rid of the orders of magnitude." It should have been more like (G/10)*(3-log10(1000/W))/3.

And you're right about H vs 40-H. Oh well, I made you work... ;-)

I'm not sure F or SC do any good, as I have no idea of what values they could take.

Posted by Stephan on October 17, 2009, 9:47 am
Add Comment
Add comment
E-mail me when comments occur on this article