Here's a quick look at some data for those who want to see what the hanna P test does vs calculated stock additions, and how it compares to hanna uncertainty statements.
Hanna's accuracy statement for the ULR P checker is "+-5ppb +-5% of reading" without any clarity as to whether you should take the larger of the two uncertainties or add them.
First, here's one I did with new Instant Ocean spiked with Seachem Flourish Phosphorus. This was measured a day after the spike.
The error bars shown are +-5ppb, and the black line is the Ideal case if I make perfect solutions, the stock product is completely accurate, and there no loss in the sample.
Here's me spiking my tank water with known additions of monopotassium phosphate (KH2PO4) - blue data.
The yellow data points are what happened when I tried to run it in straight distilled water (spiked with P) - it seems to fail completely.
(error bars here are +-5%)
And after seeing the failure with straight distilled water, I wanted to know for sure if I could get away with diluting saltwater with distilled (which has seemed to work for me before). Tagging
@Dan_P and
@Rick Mathew on this one.
Tank water, spiked with P, diluted from 100% to 80%, 60% etc with distilled water. Seems to work fine, and you can likely get away with diluting an out of range tank water sample by 1/2 or 1/4 with distilled water, and the hanna test will still give pretty much expected results. Just don't run it in straight distilled water.
(error bars +-5ppb)
If you take the common interpretation of of hanna's accuracy statement: "+-5ppb" or "+-5%" whichever is larger, then the accuracy statement covers almost all the data above, as seen by the error bars. If you take the more conservative interpretation to add +-5ppb and +-5%, then all points shown above fall inside, even with whatever unknown errors are hiding in my sample mixing, or stock purity, or loss in the sample.
Overall, the hanna ULR P kit is the most impressive chemical test in the hobby, IMO.