FAS-DPD and Speedstir -- strange results

[tpspeier]

I switched our pool over to the BBB method about two weeks ago. I purchased a Basic OTO kit and a Taylor K-2006 kit for testing. I have been very regular about testing and the chroline level has been very consistent (around 11 every night). Over the weekend I picked up a SpeedStir kit. I used the SpeedStir for the first time last night, and the chlorine level indicated was 8.5 (17 drops using 0.5 ppm per drop). I repeated the test without the SpeedStir (by manually swirling) and the level indicated was 11 (22 drops). I tried again and saw the same results. I then used the OTO test and the yellow was darker than the yellow for the 5-10 range (indicating the chlorine was above 10).

The delta between these tests seems too high to be accounted for in the margin of error. Does the SpeedStir somehow skew the measurement of the FAS-DPD test? I found another thread where chem geek indicated that the amount of swirling can lower the result slightly (http://www.poolforum.com/pf2/showthread.php?6427-Fas-dpd&), so I'm not sure if the SpeedStir "over-swirls" the test.

Thanks!

[PoolDoc]

I'd recommend calling Taylor, and asking - 800-837-8548. If you are not getting clear answers, ask to speak

Ivusich, Wayne J.
Manager, Pool & Spa Sales
410-472-4340 Ext. 121
or
O'Grady, Jody
V.P. Sales Operations and Support
410-472-4340 Ext. 122

However, I think if you want to run this down, you're going to need to do some experimenting on your own. In fact, you'll be likely to get a more thoughtful response from Taylor (and especially, from Wayne) if you've done the process below, first.

Do this:

1. Get 2 gallons of distilled water @ Walmart (NOT spring, bottle, artesian, etc: "distilled"!)
2. Get a fresh jug of bleach.
3. Soak your test tube and your speed-stir bar overnight in 2-3 cups of distilled water + 1 cup of bleach
4. TRIPLE rinse your gear in distilled water.

5. Using your cleaned test gear, run the pool water test, side by side again.
(This eliminates stir bar or other 'carry-over' as the source of the difference.)

6. Re-immerse all gear in distilled water. Use the Speed stir in the distilled water, to clean the stir-bar

7. Empty your 1st jug of distilled water, and fill it with pool water. Take it into the house, and cover it with a towel so that it's NOT exposed to any light. Wait 24 hours.
8. Using your cleaned test gear, run the pool water test, side by side again.
(This eliminates oxidizables in your pool as the source of the difference -- the improved Speedstir mixing could result in the chlorine oxidizing goo that's already in your pool. If THIS proves to be the case, it would be a VERY interesting result, with significant implications for pool care!)

9. Re-immerse all gear in distilled water. Use the Speed stir in the distilled water, to clean the stir-bar.

10.Using your cleaned test gear, run the pool water test, side by side again. BUT, use the Speed stir to pre-mix BOTH samples before testing. WHILE testing, use the Speed stir on only ONE sample.
(This distinguishes oxidizable POOL goo, from oxidizable DPD reagent materials as the source of the difference)


11. Re-immerse all gear in distilled water. Use the Speed stir in the distilled water, to clean the stir-bar.

12. Using your cleaned test gear, run the pool water test, side by side again. BUT, add 1/2 of your FAS dose, swirl, and then wait before stirring. Let the non-Speed stir sample sit for 5 minutes, plus however long you use the Speed Stir. Let the Speed-stir sit 5 minutes before stirring.
(This checks to see if different timing due to stirring is the cause of the difference.)


(I wrote this quickly, and haven't proofed it carefully, so if something doesn't make sense -- I probably screwed up.)

[PoolDoc]

(membership updated . . . AND test process edited, since you rec'd notification email)

[tpspeier]

Ben, Thanks for the thorough response. I will work on these experiments over the next few days and report back before calling Taylor.

[PoolDoc]

I'll look forward to it: I'm hoping to learn something.

And, it's just curiosity, but what is your technical background? Not many people understand, much less use, the word "delta" in that sense.

[tpspeier]

My day job is a computer engineer, so I am quite particular about details and precision. This is one of the reasons I was so disturbed about the different results from the different tests.

I was re-reading Taylor's test kit instructions last night and realized I was doing one thing wrong. I was taking my samples from near the surface of the water (about 3-4" below the surface) instead of the recommended 18" deep. I took the samples at "elbow-depth" and observed the same results, but at least this rules out one possible source of variation.

[PoolDoc]

Hm-mh.

My own, very informal testing, has never shown much of a difference between surface and 6""+ down. But, if the surface is covered in goop, including suntan lotion slime, I always go deeper. If you want to go a step further, it would be interesting to know if there actually is a detectable difference between surface and immediate sub-surface.

One other note: on pools under full sun you would expect to see a difference in chlorine levels between the surface and 12" down, if the CYA level is zero. Half-life at the surface can be as little as 15 minutes.

[tpspeier]

Our pool is in full sun, but the CYA level is around 110-120 (I used the distilled water trick mentioned in a thread on this forum to measure since it was above 100) so I would suspect little difference in chlorine levels between the surface and below the surface. I may try the depth experiment once I resolve the Speedstir vs. manual-swirl variation.

[tpspeier]

I finally did through step #5. The results were about the same: 24 drops for the manual swirl and 18 drops for the Speedstir. I now have the gallon of pool water covered to test tomorrow.

[PoolDoc]

Thanks.

You're going about it carefully enough, so that no matter what result you get, it will be helpful info for many.

[tpspeier]

Results of step #7 are about the same: 24 drops for manual swirl and 17 for the Speedstir.

For steps #10 and #12, does it matter if I use yesterday's pool water from the gallon jug or if I take a fresh sample?

[tpspeier]

I tried the first step #10 from Ben's list with the same results. I tried several other experiments involving varying the drop rate on the two samples, but all of those showed the same results too. I forgot to try the second step #10.

On the bright side, the pool is crystal clear and we've really been enjoying it.

[chem geek]

Have you tried slowing down adding the drops to the SpeedStir so that you take as long for the two tests? Also, if you let the SpeedStir water set still after the test, does it turn pink again after a minute or so? I'm just stumped.

[BigDave]

Two thoughts:
1) Have you compared the volume of the sample in the speed stir to the other test tube you are using?
2) Do you put the stirring magnet in after measuring the sample or before? I'm sorry for asking the obvious, but sometimes...

[tpspeier]

Thank you for asking the obvious. Often times things like this end up being something very simple and we spend a lot of time over-thinking the problem. Here are the steps I tried last night. Hopefully one of you can find something obvious I've done wrong.

1) Obtained 1 cup of water sample from the pool in a measuring cup. The sample came from the deep end at elbow-depth about 12 inches in from the wall near one of the returns.

2) Filled both test tubes exactly to the 10 mL mark with water from the measuring cup.

3) Dropped the magnetic bar into the Speedstir sample.

4) Added exactly two scoops of the DPD powder to each sample. I used a fresh container of powder since I had almost exhausted the container that came with my K-2206 kit.

5) Started the Speedstir and manually swirled the other sample at the same time.

6) Alternated adding drops of the titrating agent to the two samples (one drop to Speedstir, one drop to manual). I held the bottle vertically above the sample with the tip of the bottle at about the same height as the top of the test tube. I squeezed until the drop seemed to hold on the end of the tip and then allowed the drop to fall into the sample.

7) The Speedstir was stirring the whole time I was adding drops to both samples. I was only manually stirring after each drop added to the manual stir sample.

The results still showed a difference of 2 ppm less for the Speedstir than the manual swirl. As best I can tell, the fact that the Speedstir was constantly stirring was the only difference between the two tests. I guess I could try turning the Speedstir on and off between drops to see if that makes a difference.

As a sanity check, I also tried the TA test using both methods and got the same TA results from both methods.

This may be an obvious question to the forum, but has anyone else tried performing the FC test using the Speedstir and manual swirl side-by-side like this? Nobody has replied indicating they have seen the same behavior, but neither has anyone replied confirming that they get the same results both ways. I've assumed that it is generally proven that the two methods show the same results, but I wanted to ask to be sure.

[PoolDoc]

@tpspeier : I've converted the procedure to a 3-tab Google Drive spread sheet. Could you enter your data there. You should have an email, in the account you used to register, giving you edit access. This is becoming enough of a mystery that I need to ask Taylor to take a look, BUT I want to have the data in a form they can access and understand clearly.

@Chem_Geek : I've also given you access, so you can make comments in the sheet.

@everybody : I'll take this up directly with Taylor, once the spread sheet is completed. You can view the spread sheet here:
https://docs.google.com/spreadsheet/ccc?key=0Ahjo2iDF0aJgdEVOYnBOcVAxOW5JT055cDdEWE91Y 1E&usp=sharing

[tpspeier]

I've filled in the data from the for the first two rows. When I did steps #10 and #12, I incorrectly used fresh pool water instead of aged pool water. I'll refill my jug and perform those tests on the aged water tomorrow.

[chem geek]

Maybe the stirring is causing the chlorine to outgas. Yes, your suggestion of trying turning off/on Speedstir between drops to better simulate the amount of manual swirling would be a good test. If having the Speedstir not swirl as much is making a difference, then swirling is the culprit, though I would have the expected that if you took longer with the Speedstir (so longer time swirling) you'd have an even lower result than you've been seeing.

If swirling was causing more outgassing, I'd expect the pH to rise so one other test you can try (you already tried TA and found them to be the same) is the pH test. See if the longer you swirl the higher the measured pH using the Speedstir.

[JimK]

Maybe the stirring is causing the chlorine to outgas. ......

I was wondering the same thing. I had a friend that had several large aquariums and would aerate tap water to remove CL before using for the fish (she prefered not to use chems if possible).

[PoolDoc]

Maybe the stirring is causing the chlorine to outgas.

How, unless the pH is below 6.0?

I checked; the R0871 FAS solution shows a pH of 2.71. It's possible that when FC > 10 ppm (20 drops or 1.0 ml) that the acidity of the FAS overcomes the potassium phosphate buffer, and pushes the sample solution pH to 5.0 or below. If so, then Cl2 will be present in solution, and could be off-gassed under agitation.

I'll add a NEW test to the spread sheet, using 1 drop of R-0006 Base Demand Reagent, add to the water sample after adding DPD powder. I don't know if this will change the accuracy of the test, but it should reduce any off-gassing of Cl2.

http://www.taylortechnologies.com/MSDSS/0870.PDF
DPD powder => 60% potassium phosphate (pH ~7.2?)

http://www.taylortechnologies.com/MSDSS/0871.PDF
pH = 2.71

http://www.taylortechnologies.com/MSDSS/0006.PDF
Soda ash solution, pH => ~1.0

[chem geek]

Chlorine at pool/spa pH (i.e. in the 7-8 range) does not outgas as molecular chlorine (Cl2), but rather as hypochlorous acid (HOCl). The Henry's Law constant for Cl2 is 0.093 M/atm while for HOCl it is 660 M/atm so chlorine gas is 7000 times more volatile (from an equilibrium perspective). However, in pool water with an FC/CYA ratio of around 10% and pH near 7.5 and 350 ppm chloride, the HOCl concentration is over 3 million times higher than Cl2. At these concentrations, the equilibrium pressure of HOCl is roughly 500 times higher than Cl2. It is HOCl that outgasses at pool/spa pH, not Cl2. (A saltwater chlorine generator pool will have the Cl2 outgas 8.5 times faster than my example, but that's still over 50 times slower than HOCl).

Now that said, the aeration to remove chlorine from tap water is with chlorine that has no CYA in it so is roughly 10 or more times higher in concentration and it usually takes quite some time to lower the chlorine level. Even so, it's a possibility and one that could readily be tested on its own by using just using the Speedstir on its own before doing the test and doing so for different lengths of time. If the measured chlorine level declines as a function of the stirring time, then the loss from outgassing during stirring may our culprit.

[tpspeier]

I was able to get to the bottom of this issue, and I am relieved and embarrassed to admit that this is a case of user error. The short explanation is that I was not manually swirling the samples enough after each drop of titrating agent.

Here is the long explanation ....

My 8-year old daughter has taken an interest in helping me take care of the pool (yay!). She wanted to perform the FC test by herself. She did a really good job getting exactly 10 mL of water, and adding the DPD powder. However, the number of titrating agent drops was way higher than I expected it to be based on the amount of bleach I had added the night before. I had watched her perform the test, so I knew that she swirled between each drop. I then repeated the test myself and it resulted in several drops less than her test. It occurred to me that she gave a very light swirl between adding drops. I repeated my test, this time swirling more aggressively (almost shaking) between drops. This resulted in even fewer drops than my test before. I then got out the Speedstir and tried again. The Speedstir showed the same results as the "aggressive" manual swirling. I have been able to reproduce the same results between both methods the last two days.

I see how often many of you post of these boards, and I know your are busy. Thank you for your help with this issue. I am sorry that it ended up being user error. I will definitely always use the Speedstir from now on, and would recommend it to all new pool owners to help eliminate inconsistent results.

[singingpond]

I've been following this thread with interest. Glad you figured out a reason for the discrepancy!

However, the reason you discovered (a difference on how hard you swirl the sample) makes me wonder which result is actually the more 'correct' one. If the aggressive stirring result gives the truer report on the actual chlorine levels, that would suggest that a lot of us doing the manual swirl are getting results that are too high. I certainly don't 'almost shake' between drops.

I suppose as long as one uses the same technique every time, the differences between readings from day to day are still meaningful. However, one really wants correct readings, not just repeatable readings.

I'm going to try your experiment (doing the FAS-DPD manual measurement on two identical samples, just varying the amount of swirling between drops), to see how large a difference I can get.

edited to add:

It does seem worthwhile to look at the other thread you linked earlier:
http://www.poolforum.com/pf2/showthread.php?6427-Fas-dpd/page2

A comment from chem geek in that earlier thread:
> It's not important to be precisely accurate when it comes to chlorine. You want to be in the rough ballpark, but even if you are off by 20-30% it's not terrible assuming you maintain conservative chlorine levels (if you're off by 50% or more, then that could be a problem). <

It still leaves the question, though, as to whether the Speedstir result is closer to correct than the typical manual result, or whether it's the other way around.

[PoolDoc]

It would be worth knowing how 'swirling' affects the test . . . and which result is the valid one. So, please continue!

Thanks!

[JimK]

I've been following this thread with interest. Glad you figured out a reason for the discrepancy!

However, the reason you discovered (a difference on how hard you swirl the sample) makes me wonder which result is actually the more 'correct' one. If the aggressive stirring result gives the truer report on the actual chlorine levels, that would suggest that a lot of us doing the manual swirl are getting results that are too high. I certainly don't 'almost shake' between drops.

I suppose as long as one uses the same technique every time, the differences between readings from day to day are still meaningful. However, one really wants correct readings, not just repeatable readings.

I'm going to try your experiment (doing the FAS-DPD manual measurement on two identical samples, just varying the amount of swirling between drops), to see how large a difference I can get.

I was wondering the exact same thing. I also don't swirl/shake agressively. I would think if my results were inaccurate that I would have noticed an algae issue (I have not) since I tend to keep my CL level at the lower end of the Best Guess recommended level.

I guess the way to find out is to add a known level of CL to water known to be CL free, then test using both methods.

[nefretrameses]

I have also followed your test with interest. Thank you for doing all that work. As others have stated, most of us are manually stirring, but not aggressively. I suppose the technique shown in the Taylor demo videos would be "correct". It would be nice to know which is more accurate, typical manual stirring or speedstir/aggressive stirring.

[chem geek]

tpspier, please look at the Taylor video at this link (http://www.taylortechnologies.com/products_choose_slideshow.asp) where you should click on the "Pool / Spa" under "Liquid Kits" on the left and then click on the "To Test (Free and Combined) Chlorine using FAS-DPD [updated 3/5/10]" under the "K-2006 Complete™ Kit with FAS-DPD:" under the video area. Look at this video and notice how they stir. Do you call that "a very light swirl between adding drops" or "swirling more aggressively (almost shaking) between drops" or something in between. One should definitely not be doing just a light swirl. It should be a fast back-and-forth to cause a thorough swirl and mixing. It shouldn't be so vigorous as to cause splashing or getting air bubbles into the sample.

I still have a concern that too vigorous a stirring might be outgassing chlorine. A simple way to verify that is to take two samples where for one you do vigorous stirring BEFORE adding any DPD powder of titrating drops while for the other you do not. Then measure the chlorine in each of these using a consistent technique -- say, moderate swirling. If you measure a difference, then vigorous swirling may in fact be changing the chlorine level so would not be accurate. If there is no difference, then it says that thorough mixing during the test is important so swirling should not be light.

[JimK]

Wow, they use alot of powder in that demo!

I've found the suggestion here to use one scoop works just as well as long as CL isn't really high.

[singingpond]

I still have a concern that too vigorous a stirring might be outgassing chlorine. A simple way to verify that is to take two samples where for one you do vigorous stirring BEFORE adding any DPD powder of titrating drops while for the other you do not. Then measure the chlorine in each of these using a consistent technique -- say, moderate swirling. If you measure a difference, then vigorous swirling may in fact be changing the chlorine level so would not be accurate. If there is no difference, then it says that thorough mixing during the test is important so swirling should not be light.
I tried measuring 3 slightly different ways this morning, all with 10 ml water sample:

1) the way I usually do, with moderate swirling between drops -- 4.5 ppm FC

2) agitating the water sample, before adding powder, for 2 minutes (sloshing the water around as much as I could without actually losing any out of the open tube); then adding powder; then doing the habitual moderate swirling between drops -- 4.0 ppm FC

3) no pre-agitation of the water, but doing more aggressive swirling after each added drop -- 4.0 ppm FC

So, at these relatively modest chlorine levels, both of the approaches that involved more agitation of the water (either pre-test or during test) gave somewhat lower results. It wasn't a big difference, but maybe that difference would be relatively greater if the total chlorine levels were higher (as I think they were in the original post in this thread)?

[chem geek]

The chlorine loss rate would be a percentage so 0.5 out of 4.5 is somewhat over 10% though your resolution is only 0.5 ppm when using a 10 ml sample so this isn't really a fair test. With the higher levels of FC you were seeing 17 vs. 22 drops, a difference of more than 20%. Still possible, however. I just find it surprising that the swirling of the Speedstir is that vigorous. It's certainly very helpful at getting a better reading in the CH test where it requires much more thorough mixing and where doing it manually can take a much longer time having to swirl between each drop.

[tpspeier]

chem_geek: I watched the Taylor video, and I would say that my original swirling (not the aggressive style) was about the same as in the video. One difference between my swirling and the video is that I am not yet able to swirl and drop at the same time (my coordination isn't there yet). I would add one drop, then swirl, then another drop, then swirl, etc. I noticed in the video that the presenter was manually swirling the entire time while adding drops (which is similar in behavior to the Speedstir). When I use the aggressive style swirling, the water keeps moving between swirls while adding the next drop.

Last night, I performed test #10 from PoolDoc's list. This involved using the Speedstir to mix the DPD powder in each of two samples, and using the Speedstir while dropping one sample and manually swirling the other sample (with my original style swirling). The Speedstir took 16 drops and the manual took 20 drops. I also tried chem_geek's suggestion of agitating the water before adding any reagents. I ran the Speedstir on the water sample for about two minutes before adding the DPD powder. I mixed the DPD manually and swirled manually after each drop. This took 20 drops. At least with this test, it appears as if the difference in results is due to the stirring method while adding drops. I'll perform tests #12 and #14 from the PoolDoc's list, but I need to order more reagent.

[JimK]

chem_geek: I watched the Taylor video, and I would say that my original swirling (not the aggressive style) was about the same as in the video. One difference between my swirling and the video is that I am not yet able to swirl and drop at the same time (my coordination isn't there yet). I would add one drop, then swirl, then another drop, then swirl, etc. I noticed in the video that the presenter was manually swirling the entire time while adding drops (which is similar in behavior to the Speedstir). When I use the aggressive style swirling, the water keeps moving between swirls while adding the next drop.

Last night, I performed test #10 from PoolDoc's list. This involved using the Speedstir to mix the DPD powder in each of two samples, and using the Speedstir while dropping one sample and manually swirling the other sample (with my original style swirling). The Speedstir took 16 drops and the manual took 20 drops. I also tried chem_geek's suggestion of agitating the water before adding any reagents. I ran the Speedstir on the water sample for about two minutes before adding the DPD powder. I mixed the DPD manually and swirled manually after each drop. This took 20 drops. At least with this test, it appears as if the difference in results is due to the stirring method while adding drops. I'll perform tests #12 and #14 from the PoolDoc's list, but I need to order more reagent.

This is interesting.

I've always added drops while continuously swirling much like in the Taylor video.

[nefretrameses]

Doc or Chem geek: Just wondering if there was ever any resolution to this.

Santa brought me a speedstir and with pollen starting to fall in my pool, it's time to ramp up my testing program.

[chem geek]

No resolution as far as I know. It's very strange because your experiments would seem to indicate that it's the amount of stirring while adding drops that is making the difference. Agitation beforehand didn't affect the results. Did you ever contact Taylor about this? I know Ben mentioned you call them, but I didn't see any response about that.

[tpspeier]

Sorry I never replied back to this thread. Our pool developed a pretty serious leak around the time of my last post and we ended up losing (and replacing) a lot of water. As a result, the CYA level came way down (to about 40 ppm), and I brought the FC level down with it. After this drop in FC, I was no longer able to reproduce the differing results between the SpeedStir and the manual stir methods.

[nefretrameses]

Thanks for the update tpspeier. (I think Chem geek thought I was the OP). So maybe the differing results are related to the absolute CYA or FC levels or their ratios?

Your original CYA was 110 - 120 and now is 40. Original FC was 8.5 or 11 (without or with speedstir respectively). What's you current FC at CYA = 40?

[tpspeier]

Prior to shutting down the pool for the winter, I was keeping the FC around 5.