I suspect that even in the worst case drowning is not more painful than what many women experienced during childbirth, especially before usage of analgesia/anesthesia became common practice. Somehow the fear of pain didn't stop them from having sex. And the majority of women probably don't hurry to do medical abortion after discovering their pregnancy out of concerns about painful childbirth. If you find out what complications can arise when giving a birth, you may be terrified. Some women go through hours or days of torture. What is 2 - 5 minutes of conscious drowning compared to this?
Whether classic drowning is painful presents mostly a theoretical interest. From the practical point of view, it's not so important, because
1) some people are not as sensitive and afraid of possible pain/discomfort as you can be, and they might consider your level of tolerance to discomfort shamefully weak; what seems "extremely painful" to you may seem like an easy walk for them;
2) those people who are genuinely interested in pain relief most likely can find good means to mitigate discomfort from drowning to bare minimum, while those who are interested in complaining about disadvantages of some method which they dislike can find their own set of reasons explaining why SN is utter crap as well.
You have created so many similar threads about drowning and SN that I have some doubts regarding whether you're looking for answers here. But I could try to address your questions nevertheless in hope that the provided info could be interesting for others at least.
IDK what means "reliable" to you. There is a small chance that someone on a boat will spot you and pay attention to your well-being, because you would be alone, without any watercraft, and obviously too far from the seashore. So your plan doesn't look absolutely perfect. If the possibility of being discovered is acceptable for you and you're pretty sure that you're motivated enough, so you won't change your mind at the last moment, then CTB in the described way may be deemed sufficiently reliable, I presume.
If you're so concerned about purity of SN, you can assess the % of useful nitrite using chemical reaction between NaNO2 and NH4Cl (ammonium chloride) or (NH4)2SO4 (ammonium sulfate)
NaNO₂ + NH₄Cl ⇄ NaCl + NH₄NO₂
2 NaNO₂ + (NH₄)₂SO₄ ⇄ Na₂SO₄ + 2 NH₄NO₂
The produced ammonium nitrite (NH4NO2) is an unstable salt which decomposes to water and nitrogen at temperatures above 60 - 70 °C:
NH₄NO₂ → 2 H₂O + N₂↑
Now suppose you can precisely measure 13.8 g of sodium nitrite. This is 0.2 mol. At 20°C and pressure of 760 mm Hg, the maximum theoretical amount of produced nitrogen would be
0.2 mol * 24 L/mol = 4.8 L
If your conditions differ from the aforementioned, you can calculate the molar volume using this online calculator
https://planetcalc.com/7918/ and then use the resulting value instead of 24 L/mol.
4.8 L can be contained in a sphere having diameter of approximately 20.9 cm. If you collect produced nitrogen in a latex balloon whose shape is very close to spherical, let it cool down to the room temperature, and measure the diameter of the balloon, you can estimate the real yield of nitrogen (there are online calculators like this
https://www.sensorsone.com/sphere-diameter-to-volume-calculator/ that can be helpful with volume calculations)
The method of measuring the diameter of an inflated balloon is briefly shown on the following video at timestamp 7:30 (the guy produced nitrous oxide from a mixture of sodium nitrite and hydroxylamine hydrochloride)
For example, if the measured diameter is at least 19.5 cm, this means 80%+ yield at 20°C & 760 mm Hg:
(19.5 cm)³ * π/6 / (4.8 * 1000 cm³) * 100% ≈ 81%
Since the SN protocol assumes that NaNO2 is used in excess, there should be no big difference between doses of 20 g and 25 g. Both are close to LD100, so the 80%+ purity should suffice.
It may be somewhat difficult to measure the sphere diameter precisely enough (because even a small bias in the measured diameter value leads to a noticeably greater bias in volume due to the cubic dependency), so if you want a higher precision, you can use the water displacement method
Since SN poisoning is a CTB method for the lazy, I've never actually seen any users of this forum who used this method of checking SN purity and told about the results. So you could be the first person doing it here ) Only one man reported a successful quality test of his SN using the reaction with NH4Cl (he saw intense bubbling in the hot mixture), but he didn't perform quantitative analysis.
Formation of a different gas than nitrogen that would not be easily detectable is very unlikely. Sodium nitrite potentially may be mixed with a limited set of chemical compounds:
1) sodium chloride (NaCl) - doesn't react with NH4Cl; with (NH4)2SO4 there is just a simple ion exchange without production of gases;
2) sodium nitrate (NaNO3) - some amount of NH4NO3 is formed due to ion exchange, but this compound is pretty stable in hot water solutions and doesn't undergo decomposition into gaseous nitrous oxide there;
3) sodium hydroxide - NH4OH is formed, which may decompose into water and ammonia (NH3), but NH3 has a very pungent smell with low odor threshold, so its presence is easy to notice;
4) sodium carbonate or sodium bicarbonate - a mixture of CO2 and NH3 may be released, but again, the smell of NH3 is very easy to notice.
NH4Cl powders may contain salts of metals or sulfates as impurities, but they won't really form any gases in remarkable amounts upon reacting with the aforementioned compounds.
Formation of NH3 would be a bad sign, because it may indicate the presence of impurities that can cause irritation of the throat when ingesting SN. If you're ready to experience such perceptions and still want to measure the amount of nitrogen without dealing with NH3 and CO2, you can neutralize the alkali/soda with some acid before performing the test.
NH4Cl or (NH4)2SO4 should be taken in excess. For example, 0.2 mol (13.8 g) of NaNO2 reacts with 0.2 mol (10.7 g) of NH4Cl or 0.1 mol (13.2 g) of (NH4)2SO4, and these amounts of NH4Cl or (NH4)2SO4 can be multiplied by 2 - 5 in order to cover possible impurities and speed up the chemical reaction. The unused amount of the reagent will just remain dissolved in the solution and won't affect the precision of the test.