TDS meter for measuring salt

[AUSpool]

Hi all, I'm after a bit of advise for the correct use of a TDS meter for measuring salt for a SWCG pool. Should I deduct my ppm results for CH, CYA, TA and borate from the meters ppm reading?


Having read a few posts around the subject I get the impression that most think that TDS is a bit of a useless parameter to use in a pool since it includes a measure of several different ionic species including calcium, phosphates, nitrates, sodium, potassium and chloride. My phosphates and nitrates are zero or close enough to it and my SWCG runs quite high at 6000ppm.


The test strips aren't particularly accurate. I can't get the Taylor K-1766 hear in Aussie and I've read that it's only accurate to about 200ppm. I have at least 10 pool shops that I know of within a 10km (6.25m) semi circle radius (I'm coastal) and they all use a TDS meter for salt.


I have a HM TDS-4TMA with a range of 0-9990ppm and an a accuracy of +/- 2%. I can only calibrate it to within +/- 3-4ppm. It has a resolution of 1ppm from 0-999ppm and 10ppm from 1000-9990ppm. I finally got myself some 6000ppm standard NaCl solution for calibrating it but at the higher resolution I can only calibrate it to +/- 30-40ppm which isn't the best.

 

[Isaac-1]

I am not sure why you have such concern about the absolute accuracy given it is a general matter of either the SWCG is happy, or it is not happy. And as long as it is happy who cares what the exact salt level is, afterall all SWCG's internally use the electrical conductivity of the water to measure salt levels which has its own flaws as a testing method.

 

[AUSpool]

Thanks Issac, your right, I do tend to over think things sometimes. But, and I love this, we share 95% of our DNA with the common fruit fly, it's the last 5% that makes all the difference! I think I kind of answered my own question just last week when I topped up on calcium chloride. I took a calibrated TDS reading then added enough CaCl to raise my pools CH by 54ppm as calculated from the PoolMath, 12 hours later my TDS had increased by 50ppm. And that got me thinking about it, I'd added 54ppm of calcium ions to raise the hardness but my SWCG can make use of the chloride ions that came with that same addition. Although it doesn't matter much I still think it's a valid question, should I deduct any of my normal readings from my TDS reading to get a best estimate salt reading?

Again though your right, I'm over thinking it and with a few readings using the same method I can track the change over time. It did surprise me a little as my meter was factory set to 342ppm but was quite a bit out when put in a 6000ppm standard.

 

[AUSpool]

I've been thinking some more about this and really think I'm on to something, I've been looking to upgrade my TDS meter and asked the question of one manufacturer and one supplier. I also got this from HM's frequently asked web site,


Frequently Asked Questions - HM Digital


I got this,

How do I use a TDS meter to test for salt in salt generator pools?

Any of HM Digital's TDS meters can be used to test for salt (up to the maximum range of the meter). Salt is a part of Total Dissolved Solids and therefore will be part or all of the reading. If you are first filtering the water, and then adding salt, simply use the meter as you would under any circumstances. If there is only salt in the water, and the reading is 2500 ppm, then the it is 2500 ppm (mg/L) of salt. If you are starting with tap water and filling a pool, for example, prior to adding salt, then first test the level of your tap water. Therefore, if your tap water is 200 ppm, and your pool needs to be 3500 ppm of salt, then add 3300 ppm of salt. (A small portion of the tap water TDS may be salt.).

My tap water has a TDS of around 100ppm. My pool is not new but over the years there has been a lot of tap water added and serious amount of dilution from heavy rain. For a goal of 6000ppm of salt, a TDS of 100ppm for tap water is probably not worth worrying about.


But how about the other known ingredients, TA, CH, CYA and BA (borate)? So I thought of a little experiment. The TDS of a similar pool with no salt. Based roughly on my pool at 30,000L I can use a convenient division factor of 1000 to start with a 30L water sample. Using my target goals (TA-70ppm, CH-350ppm, CYA-70ppm and BA-50ppm) I've used the PoolMath calculator to get the amounts required to go from 0ppm to my target goals and then divide by 1000. That gives me 3.7g of sodium bicarbonate, 11.6g calcium chloride, 2.1g of cyanuric acid, and 13.3g of boric acid.


I have all my ingredients all I need is a clean 30L bucket.

 

[AUSpool]

I did my little experiment just for the 'fun' of it, it took longer than I thought, I made a few experimental design faults and it didn't really give me any definitive answer. I received my answers from from the supplier and manufacturer. The supplier wrote;

In terms of determining different ionic species a TDS meter is incapable of doing this.

If you take your sample to a lab for analysis they can give you a TDS factor that you can input into the meter that will more closely correlate what TDS actually is.


However, if you wish to determine specific NaCl content an Ion meter that measures this would be better suited.

The B-721 is an Ion meter designed to determine the NaCl level and this can be viewed here;
Compact Salt Meter LAQUA twin - B-721

An ion meter is beyond my budget and I'm not sure a measurement as % salt by weight is suitable. The manufacturer wrote;

A TDS meter is an electrical conductivity meter that converts the electrical conductivity reading to an approximate TDS value, based on a user determined multiplication factor of the conductivity value. When this factor = 1.0, the conductivity = TDS value.

All ionic species in water will contribute to the conductivity reading (and therefore the TDS reading). When there are multiple species present, it is impossible to know the concentration/contribution of just one species.

When CaCl is added to a pool, the conductivity/TDS will increase. It is impossible to determine a NaCl & CaCl ratio based on the conductivity/TDS reading alone.

You will be able to determine things like—is it higher than it was yesterday? It is going up? Is it going down? How much is it going up? Which pool has a higher level of dissolved salts? Also, if you take a reading before you add only CaCl, you can assume that the increase was only due to CaCl addition.

I hope this confirms your understanding of TDS. If you have additional questions please let me know.

Which from both responses it is pretty much what my understanding of a TDS meter was and similar to other descriptions within the forum so no big surprise. I can happily use my TDS meter to monitor how my salt level changes over time. Both responses seam to indicate that a correction or adjustment of the multiplication factor can be used to correct for known ionic species. I can assume any increase between TDS reading before and after a CaCl addition is due to the CaCl but I can't differentiate between calcium and chloride ions.

After a bit more searching through the forum I found this;

Nearly all of the TDS in a saltwater pool is from the sodium chloride salt. A fresh non-salt pool that has saturated calcium carbonate from adding calcium chloride for calcium and sodium bicarbonate for bicarbonate will have a TDS of around 525 ppm and have around 350 ppm salt (chloride as ppm sodium chloride). That is, the TDS of calcium and bicarbonate is not very high...

So I guess it's around 200ppm for the sodium, calcium and bicarbonate ions.

Anyway, for what it's worth here's my results for the TDS of a balanced swimming pool in an esky. I used a clean esky and an Ehiem 1048 hobby pump (10L/min) for circulation. My TDS meter is a salt meter with the result given as mg/L, a range of 0 - 9999mg/L, resolution of 1mg/L, an accuracy of salt ± 5mg/L +1%FS and calibrated in a 6440ppm solution where 1mg/L = 1.001142303ppm.

Results:

6000ppm standard solution. - 6040mg/L (6047ppm) (+0.666666%)
30L Tap water - 135mg/L (135ppm)
+ 2.1g cyanuric acid - 135mg/L (135ppm) +0ppm
+ 8.6g boric acid - 135mg/L (135ppm) +0ppm
+ 11.6g CaCl - 565mg/L (566ppm) +431ppm
+ 3.7g NaCO3 - 630mg/L (631ppm) +65ppm
+ 180g Salt - 6705mg/L (6712ppm) +6081ppm
+ 6 hours - 6820mg/L (6829ppm) +117ppm
+ 18 hours - 7040mg/L (7048ppm) +220ppm
+ 8 hours - 7055mg/L (7063ppm) +15ppm

Each reading was taken half an hour after the addition, just before the next addition. Sodium carbonate accounted for 65ppm which could be deducted from a TDS reading when evaluating for salt. Calcium chloride increased the TDS reading by 431ppm but a good proportion of that is due to the chloride ion that is accounted for in a salt reading. The salt raised the TDS by 6081ppm which is very close to what was predicted from PoolMath for that amount of salt. Both CaCl and NaCO3 additions where also close to the PoolMath predictions for both.

The TDS continued to rise over the 32 hours by 352ppm. Clearly I should have waited at least 24 hours between additions but I was worried that evaporation would affect the outcome.

I was surprised that I found no TDS reading for both cyanuric acid or boric acid, I was sure both were ionic species in water. Each one cold take longer to completely dissolve or disassociate and could be the reason for the climb in TDS. To test this I have set up to solutions, 8g of boric acid and 2.2g of cyanuric acid, each in 2L of tap water. Those concentration represent about 15 times the recommended levels for a pool. After 24 hours both showed no increase in TDS.

So at the end of it I think I could comfortably reduce my TDS reading by ~100ppm or mg/L to more accurately read for the salt. I found no direct comparison between a TDS meter and a Taylor K-1766 within the forums and have bought one to compare them. The postage from the US cost me more than the kit.

 

[Isaac-1]

Interesting, you might want to repeat part of it and give the CYA a few days to fully dissolve as it takes about a week to fully register on the conventional CYA turbidity test. I also potentially disagree that salt might only be a minor item in tap water, a lot of that will depend on where you live. For example areas that have very low calcium hardness, may still have high salt. For example the area where I live in Louisiana about 100 miles inland from the coast has 3 large aquifers below the land, only the upper 2 contain fresh water, and that is only partially fresh on the upper levels of the second aquifer and there are regions where the sands of all three allow some intermixing so it makes sense that there is some varying level of salt in the different wells in the area. In addition we have very low CH, I measure 50-65 ppm in my fill water with some seasonal variation, so I find it easy to believe that people in my area may have salt as the majority TDS item in their tap water.

Ike

 

[AUSpool]

Thanks Ike,
when talking about the salt level in my tap water I wasn't trying to evaluate it as a component of the tap water but rather compare the total tap water TDS to the TDS of a SWCG pool. For me that's 135ppm for my tap water compared to 6000ppm being the ideal level for my chlorinator. One of the articals I found said that I could deduct the tap water TDS reading when calculating how much salt to add for a newly filled pool. My pool is not new but I do add a lot of tap water.

I was thinking that the CYA and potentially the boric acid too may take a few days to register on a TDS meter. That's why I made the two solutions of boric acid and CYA - after 24 hours, still nothing. I would have thought if I'm going to get a TDS reading for either I would have had something after 24 hours. I'll see what I get after 5 days.

I might do the whole thing again after I get the Taylor salt kit. This time leaving more time between each addition and probably leaving out the salt.

 

[JoyfulNoise]

The conductivity contribution of boric acid at swimming pool concentrations (50ppm) will be in the noise of your meter. You can see from this paper on page 10 that the specific conductance of an aqueous solution of boric acid at 25C and 15.50 gm/kg (15,500ppm or roughly 310X normal pool water borate concentration) only increases above that of pure water by ~7.7 micro-siemens/cm.

You would simply do well to calibrate your TDS meter to a known salt concentration and simply assume that the conductivity is largely due to NaCl. You SWG is designed to operate over a very large range of salt concentrations and is mostly unaffected on the high end (still produces chlorine when the salt is high) and protected by the electronics from operating at the low end.

 

[AUSpool]

Thanks Matt, that's so cool, it's great to find literature that backs even something like my simple experiment. Borate in nuclear power steam circuits to prevent stress corrosion cracking due to alkaline contaminates. The coal fired plant we visited for our field trips added ammonia to raise the pH of the pure water in the steam circuit to prevent corrosion to the mild steel turbines.


Anyway after a week my solution of 8.0g of boric acid in 2L of tap water (15X normal pool concentration) had a TDS of 135mg/L (salt). The TDS of the tap water that I started with was 135mg/L. That borate conductance report suggested that any increase in conductance is most likely to impurities in the borate.


I found the same for cyanuric acid, 2.2g in 2L of tap water (15X normal pool concentration) had a TDS of 120mg/L (salt) after a week. The starting tap water again had a TDS of 135mg/L so the TDS actually decreased. FC in my tap water is given as 1.4mg/L?


I found that cyanuric acid and borate does not contribute to TDS - just nice to know.


I'm over thinking this and should just calibrate my TDS salt meter to 6000ppm and go from there and to be honest that's exactly what I do. It's nice to know however that any reading is likely to read high by as much as 70ppm or maybe even as high as 200ppm. I can't wait to compare my TDS meter with the K-1766 when it finally arrives.

 

[Griswald]

I use a TDS meter, but I tested my tap water at 150 ppm and guessed at the other "stuff" in my pool were around 50 ppm, so whatever reading I see, I simply subtract 200 from it and call that my salt ppm. It falls in line with my SWCG display.

 

[AUSpool]

TDS (salt) meter, Talor K-1766 comparison.

My K-2006 and K-1766 arrived today so I compared results from my TDS (salt) meter and the K-1766 (sodium chloride) kit. The K-1766 has a resolution of 200ppm, 1 drop = 200ppm.

First up is a 6000ppm calibration solution. The manufacturer/supplier said that this is exact with no variance.
TDS(salt) meter. - 6005mg/L.
K-1766. - 6600ppm.

Pool sample.
TDS(salt) meter. - 5685mg/L.
K-1766. - 6000ppm.

I can only assume that since that is an exact 6000ppm calibration solution that the Taylor kit is reading 600ppm high. Correcting for the variance and since the Taylor kit measures only sodium chloride via the chloride ion I assume that my pool salt level is 5400ppm and the TDS (salt) meter is reading high by 285mg/L which is kind of ball park where I had assumed it to be.

My last two store salt readings were 5040ppm and 3500ppm, that last one is out by 1900ppm and in that same test they under estimated my CH by 80ppm. It's a good thing that I've finished with the store testing.

 

[JoyfulNoise]

We haven't fully squared the circle on what I'm about to say with Taylor themselves, but typically speaking from what was described in the past to us, the error tolerance of any of their titration tests is not an absolute value. It is a +/- absolute number of drops when you are below a specific drop count number but above that number, it's a percentage of the final answer.

So, for example, what we've been told in the past is this for the FAS-DPD chlorine titration test -

When the drop count is 10 drops or less, the tolerance of the test is +/- 1 drop. When the drop count exceeds 10 drops, the tolerance of the test is +/-10% of the final answer.

The idea here being that the dropper bottles and solution vials are not perfect. It's not like we're using micro-liter titration burets made out of precision manufactured glassware. We're using plastic squeeze dropper bottles so there is some variance to the droplet volume. As you add more and more drops, the tolerance goes up because the chance of miscounting a drop or squeezing a little too hard gets worse with each droplet.

SO in your specific test, your water is titrating at about 30 drops. Therefore I would expect the error to be closer to 10% of 6000ppm or +/-600ppm which would put the Taylor test pretty darn close to your TDS meter.

Again, this is what has been described in the past to us as the proper way to determine the tolerance of a titration test. We have not checked recently with Taylor if that is true across all tests or if the tolerance is some other fraction. You could call up their customer service technical support line and inquire with them about the tolerance of the salt test.

 

[AUSpool]

Matt, again, that's awesome, thank you. I've been playing with test kits for arround 25 years and have had a little lab time, I was very careful but was left with a little doubt with that result. I'm good with it now and pritty the store and their unsuspecting customers.

I'm going to do the same as Griswald, calibrated TDS meter - 200.

My my first impression of the Taylor kits is that they are good quality, the silver nitrate leaves black stains on your fingers. Although the indicator is the same I found the old blue devil comparator easier to read. Just something to get used to I guess.

 

[JoyfulNoise]

First up is a 6000ppm calibration solution. The manufacturer/supplier said that this is exact with no variance.
TDS(salt) meter. - 6005mg/L.
K-1766. - 6600ppm.

Also, that statement leaves me a bit puzzled...it's simply not true. Well, it could be that their standard solution is quite accurate, but to say there is no variance is simply untrue to anyone that has worked anywhere near a chemistry lab. In the United States, the best insignia of quality and accuracy you can get for a manufactured, analytical reagent would be a NIST Certification (National Institute of Standards and Technology). Even with a NIST Certification, you would have to follow a predetermined protocol for making up the stock solution and then follow a specific method for determining the exact concentration. In the end, you'd still have to report a tolerance for the value quoted.

So for that manufacturer to make a statement like that absent any documentation that authenticates the accurate composition of their standard solution is a bit suspect in and of itself...


But hey, enough nit-picking, I think you've got your answer - trust your test kit and avoid the pool store.

 

[AUSpool]

I've been a bit nieve, that came from a phone conversation with the supplier where I was asking for the variance as it is not stated on the bottle. If the supplier can't tell me what do you think the closest guess would be, +/-5ppm? +/-10ppm? +/-50ppm? The 4000ppm salt solution I got previously was +/-100ppm which I thought was a bit much but it did prompt my call about the 6000ppm solution.

 

[pooldv]

Most salt tests and SWG cells have a variance of +/- several hundred ppm. I've heard as low as 200 ppm and as high as 400 ppm.

 

[AUSpool]

It's come to my attention that my CYA TDS experiment was flawed. With a relative high concentration of cyanuric acid in tap water I've lowered the pH, shifting the equilibrium from the cyanurate ion to cyanuric acid which is a neutral species. I should have checked and adjusted the pH.


I also think that the reported end point from the Taylor salt kit needs clarification too for how I've used it as a comparison to the TDS meter. Normally the drops would be added up and X 200ppm to get an answer but in reality the end point could be anywhere within the range of the last drop so my answer could more accurately be given as 5900ppm +/-100ppm. Which means the difference between the salt test and the TDS meter would be 185ppm +/-100ppm. But I don't think it really changes much for me, I'm still going to go with the TDS meter reading -200ppm.

 

[duraleigh]

With a relative high concentration of cyanuric acid in tap water I've lowered the pH

Is this a tipo tiepo typo or do they really put CYA in your Tap water "down under"

 

[AUSpool]

Is this a tipo tiepo typo or do they really put CYA in your Tap water "down under"

No Dave, going back a few posts I was trying to see if CYA contributes to conductance as measured on a TDS meter so I devised a little experiment and put 2.2g of cyanuric acid in 2 litres of tap water (~15 X the normal concentration in a pool), left it for a week and then tested for TDS. I should have adjusted the pH back up to 7.6 and probable raised the FC to 5 with a little bleach.