Of course, you do not use a hydrometer in an aquarium, but in, for example, a measuring cylinder that you fill with water from the aquarium.
Surface tension of the water ensures that you can read a float very neatly.
You can do all kinds of things wrong, especially with the calibration of a refractometer. Not every refractometer can be calibrated correctly with osmosis water, it depends on the type. An incorrectly calibrated refractometer gives an incorrect value.
Incorrect use of a refractometer during measurement will give an incorrect value, no matter how well it is calibrated.
A refractometer has many more possible failure points than a hydrometer.
I did not suggest to calibrate the refractometer with osmosis water or saltwater, I suggested to
test the readings with osmosis and freshly prepared water
.
This means that if you test osmosis and the reading is different than 0, then you know for sure the refractometer is not properly calibrated.
If you do read 0, this of course does not mean that it is properly calibrated, so the information is not as useful for you then when it is not zero, but overall you have a chance at obtaining certainty with very little effort and that is always worth a try.
To put it in statistical inference terms, we want to test the hypothesis that the refractometer is not properly calibrated.
.If osmosis water reads 0, you cannot reject the hypothesis.
.If osmosis water does not read 0, you accept the hypothesis and move on to buying calibration fluid / recalibrating or I don't know, believing what the hydrometer says (I would never believe a hydrometer).
On that subject, I disagree with the statement that "A refractometer has many more possible failure points than a hydrometer."
You get the calibration fluid and you calibrate the refractometer by moving a screw (very simple procedure).
After that, there really is very little that you can do wrong, since the measurement procedure is very simple (just put 3 drops of water) and there are almost no moving parts in the device that can get broken (unlike a swing arm hydrometer, for example). It is also very easy to read the measurement, unlike with a hydrometer, where I personally have difficulties gauging the exact line where the water sits. With swing arm hydrometers the problem is even worse.
.Also , as far as I know, not all hydrometers have temperature compensation: I think the swinging arm type do have it (or claim to have it), but hydrometers like the one below do not have it. And this is one of the most expensive of this type, even more expensive than most refractometers.
https://coralandfishstore.nl/nl/measuring-regulating/hydrometers/tropic-marin-hydrometer-incl-kunststof-meetcilinder.html
Whereas all refractometers for the saltwater hobby do have temperature compensation.