Test TA in 5ppm increments

[mgtfp]

🙋‍♂️

Newb question. (Username checks out )

Does diluting the sample also double the testing error variance like it does for the CYA test ?

I'm not judging, Lord knows I love overkill, I'm just concerned that trying too hard may be counterproductive and put you right back at square one, or worse.

That's why I suggested to use undiluted drops first, and only switch to the diluted drops for the last few drops. With a little experience you'll know when you're only 1-2 drops away from the endpoint. That minimises error accumulation, and you should at least be able to judge if TA is for example closer to 90 or to 100.

Let's take an example. Assume the real TA is 95. With a standard test, you would use 10 drops and call the TA 100. With a perfectly 50:50 diluted reagent you would need 19 drops, and call the TA 95. With a slightly 45:55 over-diluted reagent you would need 22 drops and would call the TA 110 - you would have made things worse. But if you first added 8 drops of undiluted reagent until the colour starts fading then you would have to add another 4 drops of diluted reagent and call the TA 100 - you are back to standard testing error.

I believe that you can get the dilution sufficiently accurate when working carefully. You could for example use syringes to measure the volumes.

Easier, and less error prone would of course be to use a 50ml sample. But one needs a large enough (including stirrer bean) test vial that still fits the speed stir. And you need double the amount of reagent of course.

It can be done. Everyone needs to decide for themselves if it's worth the effort.

 

[Newdude]

I believe that you can get the dilution sufficiently accurate when working carefully

You knocked all of the above post out of the park, as always. THANKS for breaking it down.

With pool friendly fill water, lots of rain and a vinyl pool, TA testing is more of a theory for me than a regular practice. 🤣

 

[JeanZ1]

That's why I suggested to use undiluted drops first, and only switch to the diluted drops for the last few drops. With a little experience you'll know when you're only 1-2 drops away from the endpoint. That minimises error accumulation, and you should at least be able to judge if TA is for example closer to 90 or to 100.

Just not sure how that is easier than just starting with twice the sample volume🤷‍♀️

 

[mgtfp]

Just not sure how that is easier than just starting with twice the sample volume🤷‍♀️

Easier, and less error prone would of course be to use a 50ml sample. But one needs a large enough (including stirrer bean) test vial that still fits the speed stir. And you need double the amount of reagent of course.