Is my local water report enough? And did I enter them in correctly?

Homebrew Talk - Beer, Wine, Mead, & Cider Brewing Discussion Forum

Help Support Homebrew Talk - Beer, Wine, Mead, & Cider Brewing Discussion Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

mtnagel

Supporting Member
HBT Supporter
Joined
Apr 15, 2012
Messages
2,562
Reaction score
623
Location
Cincinnati
Here is what my local water report looks like. Is this good enough? I was planning on getting my water tested, but do I need to or is this good enough? I assume I should use the "Finished Water" section and the average for 2012, correct?

2aqM9Jc.jpg


And here is where I tried to enter it in the Bru'n water sheet. Did I enter them correctly? For the bicarbonate and carbonate, I used the section where I entered total alkalinity and pH and then entered the resulting numbers in the input table. But then it's tell me my cation/anion difference is too big. Did I do something wrong?

cVqu0HQ.jpg
 
It looks like there are pretty big swings that are generating the averages. Martin will have more specifics, but it seems doubtful that averages across the 5 years for a given value will create a balanced report. You may also find an issue with a Ward Labs report, in that, it seems that your municipal water will change (dramatically in some cases) between brews, and the report will be a single data point. That said, I would do it to find your baseline and compare to other reports available. It is useful to know specifically what is coming out of your tap.

At a minimum, you should filter and use campden tablets.

Without more specifics on bicarbonate and carbonate levels, it seems that it will not balance. I tried plugging in Min and Max profiles and they are even worse than the averages, I suspect they are running mins/maxes, and not specific reports.

On Brewer's Friend I was able to find two Cincinnati water profiles and they balanced fine. One says to be a 2010 Ward Labs report...
http://www.brewersfriend.com/homebrew/water-profiles/ - however, no idea on the accuracy or what area within the city. The values differ from your report above.
 
I would say a Wards lab report is going to be better for you. When you have averaged numbers from both very high and very low max/min numbers...what you're entering into the spreadsheet could be radically different then what's pouring out your tap, and that's likely why the spreadsheet is saying the water report is unbalanced (cation/anion).

The interim step (if you don't want to go with a Ward labs test just yet) is to call an engineer at the water department. I've found them to be quite helpful and maybe they'll be able to help you know more specifically where YOUR water falls in those min/max ranges.

Matt's right, at a minimum make sure you filter and use campden...but I would go easy on any additions you might have calculated in Bru'n Water until you know more specifically what your measurements are.
 
BTW OP, kudos on the very nice screenshots which make your question so much easier to understand. A pleasant relief from the "what did I do wrong (followed by no details) kind of advice that is frequently solicited.

Also kudos to MattC for that Bru'n water tutorial. The instructions are good, but it took me some time to discover all of that stuff for myself. Wish I had your tutorial when I started out!
 
When you have averaged numbers from both very high and very low max/min numbers...what you're entering into the spreadsheet could be radically different then what's pouring out your tap, and that's likely why the spreadsheet is saying the water report is unbalanced (cation/anion).

It doesn't appear that the averages are the means of the high and low numbers but that doesn't change the fact that they are averages and do not represent any real, particular water sample. It is, therefore, not surprising that the numbers do not balance though it is surprising that they imbalance is as large as it is.

The issue of the reference pH value for the alkalinity measurement (from which, with pH, the bicarbonate is calculated and from that the balance) is always there as no one ever (or rarely ever) tells you what the reference pH is though Standard Methods requires that the analyst do so. The imbalance is worst if the Methyl Red endpoint (4.3) is used, a bit better if the ISO endpoint (4.6) is used and better still if the equivalence endpoint (4.615) is chosen though it should be clear that at least with respect to the last two there isn't much difference. Also many spreadsheets do not account for the alkalinity of the water itself (i.e. the pure water in the physical sample) but, again, this is a small difference compared to the errors induced by using averages. So no, I don't think you did anything wrong. It's that the data you have to work with isn't very good from this POV.
 
I expect that this is Ohio River water. More than likely, the sodium value is skewed from road salting runoff during the winter. However, I don't see the range in chloride that should accompany the sodium excursions. While many of the ion concentrations are confined to small ranges, some are excessive. Matt's comments about other Cincy profiles are a good alternative for now. Another thing to do is call the water utility and ask to speak to someone in the water quality lab. They will be able to tell you what the normal or typical values are likely to be. If that fails to give you an acceptable result, send a sample to Ward.
 
It looks like there are pretty big swings that are generating the averages. Martin will have more specifics, but it seems doubtful that averages across the 5 years for a given value will create a balanced report. You may also find an issue with a Ward Labs report, in that, it seems that your municipal water will change (dramatically in some cases) between brews, and the report will be a single data point. That said, I would do it to find your baseline and compare to other reports available. It is useful to know specifically what is coming out of your tap.

At a minimum, you should filter and use campden tablets.

Without more specifics on bicarbonate and carbonate levels, it seems that it will not balance. I tried plugging in Min and Max profiles and they are even worse than the averages, I suspect they are running mins/maxes, and not specific reports.

On Brewer's Friend I was able to find two Cincinnati water profiles and they balanced fine. One says to be a 2010 Ward Labs report...
http://www.brewersfriend.com/homebrew/water-profiles/ - however, no idea on the accuracy or what area within the city. The values differ from your report above.
Yeah, Cincinnati is pretty big and we have 2 treatment plants and one suburb on their own water source.

As for filtering, it's my understanding that filtering only removes chlorine. Isn't chlorine removed in the boiling step in brewing?

As for campden tablets, doesn't that remove chlorine and chloramine? I will confirm with the person at my local water company, but I don't believe they are using chloramines here, so if I were to confirm that for sure and then filter the water, wouldn't that remove the chlorine? Please educate on my I would filter and use campden tablets.



BTW OP, kudos on the very nice screenshots which make your question so much easier to understand. A pleasant relief from the "what did I do wrong (followed by no details) kind of advice that is frequently solicited.
Thanks. I try to make it easy for people when I'm asking for their help.

It doesn't appear that the averages are the means of the high and low numbers but that doesn't change the fact that they are averages and do not represent any real, particular water sample. It is, therefore, not surprising that the numbers do not balance though it is surprising that they imbalance is as large as it is.

The issue of the reference pH value for the alkalinity measurement (from which, with pH, the bicarbonate is calculated and from that the balance) is always there as no one ever (or rarely ever) tells you what the reference pH is though Standard Methods requires that the analyst do so. The imbalance is worst if the Methyl Red endpoint (4.3) is used, a bit better if the ISO endpoint (4.6) is used and better still if the equivalence endpoint (4.615) is chosen though it should be clear that at least with respect to the last two there isn't much difference. Also many spreadsheets do not account for the alkalinity of the water itself (i.e. the pure water in the physical sample) but, again, this is a small difference compared to the errors induced by using averages. So no, I don't think you did anything wrong. It's that the data you have to work with isn't very good from this POV.
Thanks for the info on pH. And of course I shouldn't have realized that the averages weren't a single analysis, so they probably wouldn't balance the cations and anions.

I expect that this is Ohio River water. More than likely, the sodium value is skewed from road salting runoff during the winter. However, I don't see the range in chloride that should accompany the sodium excursions. While many of the ion concentrations are confined to small ranges, some are excessive. Matt's comments about other Cincy profiles are a good alternative for now. Another thing to do is call the water utility and ask to speak to someone in the water quality lab. They will be able to tell you what the normal or typical values are likely to be. If that fails to give you an acceptable result, send a sample to Ward.
Thank you very much. I will email the water company again and ask for a single analysis to see what it says.
 
I would say a Wards lab report is going to be better for you. When you have averaged numbers from both very high and very low max/min numbers...what you're entering into the spreadsheet could be radically different then what's pouring out your tap...
Isn't the average going to be closer to any given time point than one single analysis done at one time point? Even though the averages don't balance out the cations and anions, wouldn't it be better to use the average values over a long period of time than one single analysis at one point in time?

For a crazy example to illustrate my point. Let's say I get the Ward report and the numbers come out closer to the minimums in my city's report. Then I plan a brew around those numbers and go to brew in one week from when my water was tested and the water coming out of my tap is actually closer to the maximums on my city report. (I know it's probably not likely, but it is possible since the water varied between the minimum and the maximum last year). Therefore, wouldn't using the averages over a year actually get me closer to what is coming out of my tap at any given time?
 
Isn't the average going to be closer to any given time point than one single analysis done at one time point? Even though the averages don't balance out the cations and anions, wouldn't it be better to use the average values over a long period of time than one single analysis at one point in time?

For a crazy example to illustrate my point. Let's say I get the Ward report and the numbers come out closer to the minimums in my city's report. Then I plan a brew around those numbers and go to brew in one week from when my water was tested and the water coming out of my tap is actually closer to the maximums on my city report. (I know it's probably not likely, but it is possible since the water varied between the minimum and the maximum last year). Therefore, wouldn't using the averages over a year actually get me closer to what is coming out of my tap at any given time?

Well this is getting into the more esoteric side of statistics, but the answer is it depends on the likelihood and size of the 'extreme' events. Many things in nature that have a tendency to 'spike' occasionally are well modeled by power law distributions. Classical examples that are relevant to brewing water are maximum river flows and maximum daily rainfall. The tricky thing about power law distributions is that they can potentially have infinite means, so the central limit theorem no longer applies. Essentially what this means is that the probably of getting a very large number occasionally is big enough so that as you collect data over a longer and longer time period the mean you calculate from your data never converges and slowly creeps up over time. In cases like this, a single data point is more representative of 'typical' values than the estimated mean.

Anyway, power-law behavior usually only occurs when you're trying to estimate 'worst-cast' type scenarios, (I.E. you might have trouble trying to estimate how bad your water could possibly be), but most likely the day-to-day variations in water chemistry aren't following a power-law. But, that being said, even if you can reasonably calculate a mean, if the water does have large spikes in ion concentrations from time-to-time you probably want to get a TDS meter to make sure you're not brewing your beer with salt water.
 
Thanks for that. It just seems weird to me that people get their water analyzed once and then use that for who knows how long to adjust their water.

I like the TDS meter idea. That will at least tell you if there so crazy spike for some reason.
 
Apologies. Perhaps better stated, use campden in the water preparation. Filter if you you feel the water tastes better. Aj has a paper on his site about Chlorine and Chloramine treatments (http://www.wetnewf.org/pdfs/Brewing_articles/BT_Chlorine.pdf). Filtering and campden (see the sticky) are what I have found to be most effective. I understand that chlorophenols are formed in the mash when hypochlorite (chlorine treatment byproduct) interacts with organic compounds - which survives boil and fermentation through to the finished beer.

I am on a rural water supply - and I carbon filter all of our drinking water which dramatically improves it's flavor. It is not necessary from a safety perspective, but it does remove much of the chlorine, but also other occasional smells and flavors, and judging from the filter replacement cycles, a good deal of sediment. I don't want any of that in my beer. I use campden as a preventative because it seems that some chlorine makes it past my filter, probably when I filter with too high a flow rate. My beers made an immediate improvement when I carbon filtered my water. I also had a couple odd batches with chlorophenol off tastes (despite filtering). Then I used campden as a second step resulting in another large improvement. Then, because I had so many pH issues in the mash - I got the Ward Report (high sodium and alkalinity) and switched to R/O water. R/O gives me a consistent basis to build water to suite any beer style. I hesitate to recommend it because of the relatively high costs and maintenance requirements when there are cheaper solutions and tap water is ok for brewing. But I am very happy with my decision. If considering R/O, re-brew one of your common recipes with store bought R/O or distilled, add salts to match the recipe, and compare.

As per your later example, yes. It is possible that you would get a report that represents the minimal characteristics of your water supply, however, it is very important that the anion/ion ratio balance, otherwise, any spreadsheet/program will provide garbage results. You can take a few steps to track and understand the consistency of your water supply - to your tap. A pH meter and a TDS meter, can provide statistical information about your water. Using both and logging results from the source water on your brewday will help you to compare to a Ward Report (or data from your water utility) and determine if the results are too far off your baseline to be useful. I do this behind my R/O system to determine the status of the filters, and now do it at our tap as we have had some very wild swings lately in water quality. It is a useful tool in communicating to a water engineer that there maybe an issue. To be clear, you will get very little diagnostic information, rather data that says something has changed, and perhaps a correlation (as Martin suggests) with a specific season or activity, and then can take appropriate steps (delay brewing or use store bought filtered water). I am also very OCD about my brewing... so this might seem overkill to you.

I should add, homebrewers brew with inconsistent water sources all the time, with award winning results. I know there is a large brewing community in Cincinnati - and consultation with other local homebrewers may may be more helpful. Taken on face value - I suspect your water is ok for brewing with a chlorine treatment. You would have to accept some variability on occasion that may show up in your finished beer.

- jbaysurfer - thanks for the kind words. Martin should also get a lot of credit - he did some serious editing, and I learned a lot in the process.
 
Isn't the average going to be closer to any given time point than one single analysis done at one time point? Even though the averages don't balance out the cations and anions, wouldn't it be better to use the average values over a long period of time than one single analysis at one point in time?

That's a good question. The problem is that you don't know how the sampling from which the averages were drawn was done. If the world were ideal the lab would do a thorough, complete and perfectly accurate analysis on each of lots of samples. The sum of the electronic charges from each analysis would be 0 and the averages would balance exactly too. To see why this is so picture an array of vertical columns. The numbers in each vertical column are the charges on each of the ions in one sample. As a sample has to be neutral the sum of the numbers in a column is 0. If we write the sum (0) of each column at the bottom and then add those all up the sum of the sums is 0. It represents the sum of all the numbers in the matrix. It doesn't matter whether you sum over rows first or columns first. Summing along a row gives, for example, the sum of the calciums over all the tests. Dividing this by the number of the tests gives the average of the calcium levels. Thus adding all the averages gives 0 too i.e. perfect balance. Note that this result does not depend on the distribution of individual ion concentrations nor does it care about outliers. It depends on each analysis being complete and accurate.

The preceding does not describe what you have here. The averages exhibit a (surprisingly) large imbalance. There are many possible explanations for this. I believe the most likely is that the averages are not computed from an ensemble of complete measurements but rather from ensembles of individual measurements. For example, calcium could be measured at one location on Mondays, Wednesdays and Thursdays at one plant and sulfate at another on Tuesdays, Thursdays and Saturdays. If there were, under this scenario, a spike in sodium chloride (road application) that got detected on a chloride test but not the sodium test that could really throw things off.

There may be an unaccounted for element (e.g. strontium) that would need to be included as part of a complete analysis in order to achieve balance. Finally, many analysts don't seem to know how to compute bicarbonate from alkalinity and pH. I believe this to be responsible for the fact that Ward Labs reports almost never balance and that they report concentrations of bicarbonate and carbonate that are not consistent with the reported pH values (the ratio of the concentrations of these ions is 10^(pH - 10.38)).

None of these explains an imbalance as large as the one your data show (1.85 mVal). It is most intriguing that while the data you showed is clearly labeled, respectively, 'As Ca' and 'As Mg' for those two ions interpretation of the numbers 'as calcium carbonate' leads to a much smaller imbalance i.e. 0.18 mVal. An imbalance at that level is easily explained by lab error, sampling irregularity etc. This, of course, makes me very suspicious.

The real question perturbing you at this point is naturally as to what numbers you should use. Data with imbalances of nearly 2 mEq are pretty useless beyond giving a broad picture of water that is pretty hard but not too alkaline. To use water reports such as yours is you have to tweak something to make the report balance. But you have no guidance to indicate to you whether you should be adjusting calcium and magnesium down or alkalinity up. In this regard a single report (which, from Wards, usually is balanced to better than 0.5 mVal) is much better than the averages because it represents real water if at only one point in time. And the way around the snapshot problem is to measure more often. And the way around that is to remove almost all the variation by processing through an RO system.
 
Apologies. Perhaps better stated, use campden in the water preparation. Filter if you you feel the water tastes better. Aj has a paper on his site about Chlorine and Chloramine treatments (http://www.wetnewf.org/pdfs/Brewing_articles/BT_Chlorine.pdf). Filtering and campden (see the sticky) are what I have found to be most effective. I understand that chlorophenols are formed in the mash when hypochlorite (chlorine treatment byproduct) interacts with organic compounds - which survives boil and fermentation through to the finished beer.

I am on a rural water supply - and I carbon filter all of our drinking water which dramatically improves it's flavor. It is not necessary from a safety perspective, but it does remove much of the chlorine, but also other occasional smells and flavors, and judging from the filter replacement cycles, a good deal of sediment. I don't want any of that in my beer. I use campden as a preventative because it seems that some chlorine makes it past my filter, probably when I filter with too high a flow rate. My beers made an immediate improvement when I carbon filtered my water. I also had a couple odd batches with chlorophenol off tastes (despite filtering). Then I used campden as a second step resulting in another large improvement. Then, because I had so many pH issues in the mash - I got the Ward Report (high sodium and alkalinity) and switched to R/O water. R/O gives me a consistent basis to build water to suite any beer style. I hesitate to recommend it because of the relatively high costs and maintenance requirements when there are cheaper solutions and tap water is ok for brewing. But I am very happy with my decision. If considering R/O, re-brew one of your common recipes with store bought R/O or distilled, add salts to match the recipe, and compare.

As per your later example, yes. It is possible that you would get a report that represents the minimal characteristics of your water supply, however, it is very important that the anion/ion ratio balance, otherwise, any spreadsheet/program will provide garbage results. You can take a few steps to track and understand the consistency of your water supply - to your tap. A pH meter and a TDS meter, can provide statistical information about your water. Using both and logging results from the source water on your brewday will help you to compare to a Ward Report (or data from your water utility) and determine if the results are too far off your baseline to be useful. I do this behind my R/O system to determine the status of the filters, and now do it at our tap as we have had some very wild swings lately in water quality. It is a useful tool in communicating to a water engineer that there maybe an issue. To be clear, you will get very little diagnostic information, rather data that says something has changed, and perhaps a correlation (as Martin suggests) with a specific season or activity, and then can take appropriate steps (delay brewing or use store bought filtered water). I am also very OCD about my brewing... so this might seem overkill to you.

I should add, homebrewers brew with inconsistent water sources all the time, with award winning results. I know there is a large brewing community in Cincinnati - and consultation with other local homebrewers may may be more helpful. Taken on face value - I suspect your water is ok for brewing with a chlorine treatment. You would have to accept some variability on occasion that may show up in your finished beer.

- jbaysurfer - thanks for the kind words. Martin should also get a lot of credit - he did some serious editing, and I learned a lot in the process.
Thanks for the info. That paper was very interesting. You've convinced me to at least try the campden tablets as my water does have a strong chlorine odor (which I erroneously thought was removed when boiling the wort). I use a PuR pitcher for drinking, but haven't treated my brewing water at all, but I will start with the campden tablet and then see if I need to filter before that.

That's a good question. The problem is that you don't know how the sampling from which the averages were drawn was done. If the world were ideal the lab would do a thorough, complete and perfectly accurate analysis on each of lots of samples. The sum of the electronic charges from each analysis would be 0 and the averages would balance exactly too. To see why this is so picture an array of vertical columns. The numbers in each vertical column are the charges on each of the ions in one sample. As a sample has to be neutral the sum of the numbers in a column is 0. If we write the sum (0) of each column at the bottom and then add those all up the sum of the sums is 0. It represents the sum of all the numbers in the matrix. It doesn't matter whether you sum over rows first or columns first. Summing along a row gives, for example, the sum of the calciums over all the tests. Dividing this by the number of the tests gives the average of the calcium levels. Thus adding all the averages gives 0 too i.e. perfect balance. Note that this result does not depend on the distribution of individual ion concentrations nor does it care about outliers. It depends on each analysis being complete and accurate.

The preceding does not describe what you have here. The averages exhibit a (surprisingly) large imbalance. There are many possible explanations for this. I believe the most likely is that the averages are not computed from an ensemble of complete measurements but rather from ensembles of individual measurements. For example, calcium could be measured at one location on Mondays, Wednesdays and Thursdays at one plant and sulfate at another on Tuesdays, Thursdays and Saturdays. If there were, under this scenario, a spike in sodium chloride (road application) that got detected on a chloride test but not the sodium test that could really throw things off.

There may be an unaccounted for element (e.g. strontium) that would need to be included as part of a complete analysis in order to achieve balance. Finally, many analysts don't seem to know how to compute bicarbonate from alkalinity and pH. I believe this to be responsible for the fact that Ward Labs reports almost never balance and that they report concentrations of bicarbonate and carbonate that are not consistent with the reported pH values (the ratio of the concentrations of these ions is 10^(pH - 10.38)).

None of these explains an imbalance as large as the one your data show (1.85 mVal). It is most intriguing that while the data you showed is clearly labeled, respectively, 'As Ca' and 'As Mg' for those two ions interpretation of the numbers 'as calcium carbonate' leads to a much smaller imbalance i.e. 0.18 mVal. An imbalance at that level is easily explained by lab error, sampling irregularity etc. This, of course, makes me very suspicious.

The real question perturbing you at this point is naturally as to what numbers you should use. Data with imbalances of nearly 2 mEq are pretty useless beyond giving a broad picture of water that is pretty hard but not too alkaline. To use water reports such as yours is you have to tweak something to make the report balance. But you have no guidance to indicate to you whether you should be adjusting calcium and magnesium down or alkalinity up. In this regard a single report (which, from Wards, usually is balanced to better than 0.5 mVal) is much better than the averages because it represents real water if at only one point in time. And the way around the snapshot problem is to measure more often. And the way around that is to remove almost all the variation by processing through an RO system.
Very helpful! I seriously learn so much on here!

I have asked the person I've emailed at the local water company for the most recent single analysis they've done. We'll see what I get. They've been responsive a couple times now, so hopefully I'll get what I need. If not, I will definitely send off my water for analysis.
 
My local water authority is pretty cool. I asked for the most recent single sample and he responded (which confirms what was speculated by someone above):
I do not have this information for a single analysis (sample).
We collect samples for these compounds on different schedules.
Seems like a Ward report would be useful for me.

He also explained how the sulfate level changes:
The sulfate level in our treated water is determined in large part by the level of sulfate found in the Ohio River.
We have seen with many compounds that their concentration in the Ohio River correlates to some extent with the River Stage.
The concentration of compounds in the River decreases as the River stage increases.
Basically, there is a dilution effect.
If there is a lot of rainfall (and the river stage increases) the concentration of sulfate will be less.
There are other factors but that is the general pattern.

But he also said this about chloramines:
Chloramines are not used during our Treatment process.
Hmmm... Seems like I still want to remove the chlorine, so I just need to decide if I should go with a filter system or use campden tablets. I have the tablets for my brew day on Sunday. Seems like they would be cheaper than replacing filters. Unless there is something else that I would gain by using a filter.

Entering in the numbers he gave me into the Bru'n Water spreadsheet, the ion balance is a little better, but still off.

2YtgUOz.jpg
 
But he also said this about chloramines: No chloramine

Hmmm... Seems like I still want to remove the chlorine, so I just need to decide if I should go with a filter system or use campden tablets. I have the tablets for my brew day on Sunday. Seems like they would be cheaper than replacing filters. Unless there is something else that I would gain by using a filter

The good news about regular chlorine is that it flies off by itself when the water is heated or even allowed to stand overnight. Campden tablets will remove it and carbon filtration will remove it as well but if you draw it the night before you brew and agitate it a bit as you are heating it up you will be low enough in chlorine that you do not need to consider further treatment.
 
Matt, the water report balances almost exactly. Your mistake is that you entered 49.2 ppm for carbonate. Since the report did NOT report the 82 ppm carbonate as carbonate, it was incorrect to convert that value as 49 ppm carbonate. The correct carbonate value is 1.8 ppm as calculated in the alkalinity conversion calculator.

You're good to go.
 
^ sorry if this is a dumb question. But I don't follow. I don't see where you entered 49.2PPM for carbonate? Looks like you did enter that in the calculator, but the water report input section has 1.6 for carbonate. What am I missing?
 
Back
Top