Repeatability of TF100 chlorine test

BowserB

Silver Supporter
Jul 29, 2018
343
Katy, Texas
Another thread which I can't find right now suggested a high variability in the FC/CC test. That piqued my curiosity ("OK, so you've got piqued curiosity. So what?"--Barney Miller 1975.) Well, I wondered about that and decided to see what happened when I did the same test with the same water again and again. In this case, six times. So to avoid being accused of "burying the lead", my conclusion is that the amount of R-0870 powder is a key to accuracy and repeatability. The laminated card that came with the TF100 kit says to add "one heaping scoop...(the exact amount is not important)...it may not all dissolve." And there IMO lies the key.

My method and procedure. I use the SpeedStir--VERY IMPORTANT to avoid manual swirl-induced insanity. I rinse the cylinder in 5-stage reverse osmosis filtered water, then again in sample pool water. I do the same with an eye dropper that I use to adjust sample level in the cylinder to exactly at the 10ml line on a level counter on a white paper towel under bright LED light. I put in the scoop of 870 powder, put the cylinder on the SpeedStir and drop in the stirrer "pill" to mix the solution. With the SpeedStir still stirring, I drop from the 0871 one drop at a time with the bottle vertical or nearly so, with a precision that can only be achieved with the SpeedStir doing the swirling. I watch for the color to look clear (kindof grayish), and then one more drop--which under my light is the real "clear" as the solution takes on the color of the SpeedStir case, a sort of light blue-green-gray. That tells me it's at the end, and I did that six times.

My results of six tests
FC6.55.55.07.04.57.0
CC0.51.0.51.0.5
TC6.56.06.07.55.57.5
The two results showing 7.0/.5 were both with very heaping scoops. In both of those, I could see a few crumbs of 0870 in the bottom of the cylinder, AND the solution was deeper pink than other times. The 4.5/1.0 test was from a scoop of 870 that was barely heaping. So my conclusion is that size (of the scoop of R-0870) matters. Had I done only my test #5, I would have rushed out to add 12.5% liquid chlorine to the pool, as 4.5 is a bit low for my CYA of 40, and CC of 1.0 seems like I've got biologics in the pool.

After test number six, which repeated the earlier 7.0/.5 result, I did one more with TWO scoops of R-0870. I wanted to be sure that more R-0870 would report more chlorine. With two scoops, there were quite a few crumbs of the powder in the bottom of the cylinder. However, my results were the same as the earlier "generous" scoop results: FC 7.0 and CC 0.5. I'm not a scientist, and I don't even play one on TV, but I think from my six tests, plus 7th extreme test, I have determined that amount of R-0870 used in the TF-100 chlorine test is important and should be generous.
 
  • Like
Reactions: mknauss

tim5055

Mod Squad
TFP Expert
LifeTime Supporter
May 11, 2014
11,021
Franklin, NC
You are correct, the R-0870 used in the TF-100 chlorine test is important and should be generous. In the Taylor instruction they actually say to use two scoops and we have found one heaping scoop to be sufficient.

The biggest problem with repeatability is the amount of water and the drop size. In addition to the speedstir, I use a sample sizer so that the amount of water in the tube is exactly the same every time. Additionally, when adding the drops, the reagent should not be held at an angle, but should be vertical allowing drops to form on their own and fall off the tip, not "squirted" out.

Following these methods my testing has alwasy been repeatable to teh exact or one drop difference.
 

BowserB

Silver Supporter
Jul 29, 2018
343
Katy, Texas
I haven't seen the Taylor instructions, but I know about the vertical bottle deal, and I have confidence in my 10ml accuracy--or at least consistency. I make sure the water is right on the line, not above or below, so in my six plus one tests, the amount of pool water and condition of the cylinder were absolutely the same. Also my drop technique was the same, as was my visual conclusion of when each was done. That left only the quantity of R-0870, which from now on will be very generous. Since I posted my test results, I did a couple more and found that as long as I used enough R-0870, the answer was the same. Thinking back over the last year or so, I've probably added chlorine to the pool a few times unnecessarily.
 

duraleigh

Admin
Mod Squad
TFP Expert
Platinum Supporter
LifeTime Supporter
In The Industry
Apr 1, 2007
34,363
Sebring, Florida
my conclusion is that the amount of R-0870 powder is a key to accuracy and repeatability.
I stopped reading your post right there.

Within reason, the exact amount of powder used in this test is not really important. If the solution turns pink, you have enough powder to complete the test. Residual, undissolved powder in the solution doesn't matter, either.
The amount should be "generous" because it gives you a somewhat deeper pink but the exact amount is just not significant. The Taylor video demonstrates two generous scoops but they (TAYLOR) emphasize the EXACT amount is not necessary
 
Last edited:
  • Like
Reactions: proavia

BowserB

Silver Supporter
Jul 29, 2018
343
Katy, Texas
I stopped reading your post right there.
Within reason, the amount of powder used in this test is essentially irrelevant. If the solution turns pink, you have enough powder to complete the test. Residual, undissolved powder in the solution doesn,t matter, either. Sorry to be abrupt and harsh but many of us have been doing this test for 15 years........successfully. Please consult with the Taylor chemists for further explanation.
Wow, that's harsh. What I found was that you can use too little R-0870 and get an erroneous low FC result, even though the solution turns pink. If you hadn't "stopped reading" you might have seen that my tests agreed with you...so long as you use PLENTY of R-0870. Tim's comment was enlightening on the quantity of R-0870 that Taylor recommends. However, I will send an inquiry to Taylor's chemists about that, as you suggested. Dave, you must be having a bad morning. I understand that your industry is busy these days, with people reverting to "staycations" and a pool being part of that. Busy and overworked usually is compensated by more green in the bank account, so don't worry.
 

duraleigh

Admin
Mod Squad
TFP Expert
Platinum Supporter
LifeTime Supporter
In The Industry
Apr 1, 2007
34,363
Sebring, Florida
Your thread title should have been "You can use too little powder and invalidate the FAS/DPD test" Ya' think? Your observation has little to do with the repeatability of the test. The instructions are really quite clear, put in the correct amount of DPD powder and the test is repeatable. How much did you use when you got your bogus results? Why did you use so little?

I will choose to ignore your personal attack but you really shouldn't do that on this or any other forum.
 

PoolNewb2020

Well-known member
May 7, 2020
300
Lathrop, CA
In my opinion this was not a repeatability test, it was a test of how the amount of powder affects the reading. And you found out correctly that too much is OK, but too little will affect the reading. The instructions clearly say "a heaping scoop".

A repeatability test would be for you to put in "a heaping scoop" each and every time and see what kind of reading you would get. I'm pretty sure you would be within 0.5ppm on each test. So it is quite repeatable.
 

Jeff J.

Well-known member
Aug 6, 2019
222
Staten Island NY
Your thread title should have been "You can use too little powder and invalidate the FAS/DPD test" Ya' think? Your observation has little to do with the repeatability of the test. The instructions are really quite clear, put in the correct amount of DPD powder and the test is repeatable. How much did you use when you got your bogus results? Why did you use so little?

I will choose to ignore your personal attack but you really shouldn't do that on this or any other forum.
Is it fair to say that, to be safe, you should see some undissolved powder at the bottom?
 

duraleigh

Admin
Mod Squad
TFP Expert
Platinum Supporter
LifeTime Supporter
In The Industry
Apr 1, 2007
34,363
Sebring, Florida
Is it fair to say that, to be safe, you should see some undissolved powder at the bottom?
Not a bad idea, but that's not a requirement. If you stick with the idea of 2 level scoops or one heaping scoop you will have enough powder each time you perform the test.
 

mguzzy

Gold Supporter
Jul 8, 2015
2,336
OV, CA
Perhaps a discussion of the mechanics of what is happening chemically, sans the equations.

There is a fixed amount of Chlorine in the sample. And the R-0870 powder reacts with it. Once all the chlorine is reacted that's it, no more R-0870 will make a difference. So you cannot put too much in, but you can put not enough in to react with all the Cl. The purple/pink just indicates a reaction with the Cl. The result of the reaction of the R-0870 and Chlorine creates another chemical that the R-0871 drops will react with. When all the combined R-0870/Cl product has reacted with the R-0871 in the solution, it turns clear. And that is the end point of the test.

So the variables become:
-Not enough R-0870 to react with all the Cl.
-Size of the initial sample. If you can measure the sample graduations line to the bottom of the meniscus its pretty darn accurate. But the test vial needs to be level and many people eye ball it by holding it up. put it on a level surface and get down to its level. My chem teacher always yelled at us if caught us eyeballing the meniscus while holding up graduated cylinder.. "What! are we in the movies!" he would scream.
-Size of the drops... if the drops are not the same consistent size the result will vary. Hold the bottle vertical and let them form and drop under their own weight. no squirting. In a lab one uses titrating columns.. but drops created like this are within the tolerance of the test.
-mixing, In consistent mixing is a bigger variable than most people realize. I'm pretty spastic so I know. use of a speed stir removes this variable.
-How did you prepare your test vial. I could see a scenario where they could be rinsed out with tap water that is chlorinated.. if that is the case then rinse them out with distilled water instead. I hate seeing the pool store prep... there the test cells are rinsed with the last guys sample, eeewe.
-There were no droplets of sample water stuck to the sides of the vial that dripped in.
-There are probably other sources of contamination that can happen, a stray Cl laced dust particle but I doubt it.
Try your battery of tests on a standard sample and see how consistent you are. They sell them at TFTestkits. That is the best way to fine tune your technique
 

Jeff J.

Well-known member
Aug 6, 2019
222
Staten Island NY
Perhaps a discussion of the mechanics of what is happening chemically, sans the equations.

There is a fixed amount of Chlorine in the sample. And the R-0870 powder reacts with it. Once all the chlorine is reacted that's it, no more R-0870 will make a difference. So you cannot put too much in, but you can put not enough in to react with all the Cl. The purple/pink just indicates a reaction with the Cl. The result of the reaction of the R-0870 and Chlorine creates another chemical that the R-0871 drops will react with. When all the combined R-0870/Cl product has reacted with the R-0871 in the solution, it turns clear. And that is the end point of the test.

So the variables become:
-Not enough R-0870 to react with all the Cl.
-Size of the initial sample. If you can measure the sample graduations line to the bottom of the meniscus its pretty darn accurate. But the test vial needs to be level and many people eye ball it by holding it up. put it on a level surface and get down to its level. My chem teacher always yelled at us if caught us eyeballing the meniscus while holding up graduated cylinder.. "What! are we in the movies!" he would scream.
-Size of the drops... if the drops are not the same consistent size the result will vary. Hold the bottle vertical and let them form and drop under their own weight. no squirting. In a lab one uses titrating columns.. but drops created like this are within the tolerance of the test.
-mixing, In consistent mixing is a bigger variable than most people realize. I'm pretty spastic so I know. use of a speed stir removes this variable.
-How did you prepare your test vial. I could see a scenario where they could be rinsed out with tap water that is chlorinated.. if that is the case then rinse them out with distilled water instead. I hate seeing the pool store prep... there the test cells are rinsed with the last guys sample, eeewe.
-There were no droplets of sample water stuck to the sides of the vial that dripped in.
-There are probably other sources of contamination that can happen, a stray Cl laced dust particle but I doubt it.
Try your battery of tests on a standard sample and see how consistent you are. They sell them at TFTestkits. That is the best way to fine tune your technique
All of those possible errors add to how much potential for error in the test?
 

mguzzy

Gold Supporter
Jul 8, 2015
2,336
OV, CA
All of those combined could easily explain the range of values the OP found. .. That's why its so important to do the tests correctly to reduce the variability that can lead to the wrong conclusions.

I've proven to myself that my stirring technique sux... and depending on the test, I get a result that is 5-10% different with and without the speed stir. That is just to much variability.. the speed stir makes it a lot more consistent.
 
Last edited:

PoolNewb2020

Well-known member
May 7, 2020
300
Lathrop, CA
The way I look at this is that high accuracy and repeatability is not the goal of these tests. That is why we have ranges for all chemical levels. With FAS-DPD test the worst case that can happen if too little powered is used, the reading will be falsely low. The owner will then add enough to bring it up to the proper level, but in reality it will be over by 1 or 2 ppm, but still below the SLAM level. When it comes to FC, being on the higher end is not a problem, but being on the lower end is.

To me it looks like this test was designed very cleverly with the error rate being directed towards reading low instead of high. This way we don't have a situation where too little chlorine is being added.