Why NOT to trust a Refractometer against a profiLux
We recently completed a case study testing various brand refractometers against electronic conductivity devises, including ProfiLux.
The findings were very interesting and I thought I would sahre this with you as so many users question why does the $300 probe of profiLux read less than my bargian refractometer. Well here is the REAL answer!
The first test was as follows.
calibrate 3 devises per manufacturers guidelines
Refractometer
Digital salinity pen
Digital conductivity monitor
PROFILUX SALINITY PROBE ON A PROFILUX PLUS II EX
The next step was to do a test using only the guidelines of each devise.
The refractometers read using GHL calibration solution around 1.018ssg
All other devises read 1.024ssg
We then swapped the calibration fluid fluid with another brand and the readings did not change as expected.
Next step was to make sure all devises other than the Auto temperature compensated (built in feature) refractometers were at the same temperature of 22oC ALL units stayed the same other than the refractometer which reading crept up to 1.020ssg.
At this point I had already worked out that the temperature compensation of refractometers was not really doing its job on all ones tested so passed the evaluation to Brent of Big Al's in Montreal Canada.
Here is his findings. (copied text from email)
The digital units seem to work really well, and really doesn't agree very well with the refractometer. The digital pens are consistent with our conductivity meter, and the conductivity measurement on the Profilux, and consistent with various salinity calibration solutions that we have. That it agrees with the conductivity meter is not surprising to me, as it is also, well, a conductivity meter.
All this tells my pea-sized chemical engineer mind is that conductivity meters, calibrated using a standard solution (which may or may not contain all the minor 'noise' level ions such as Mg, Ca, etc), provide consistent conductivity measurements of saltwater solutions... From a technical standpoint, given the dominant nature of the Na and Cl ions in saltwater, conductivity should translate very well to salinity. And temperature compensates very easily. Yet the basic engineering behind a refractometer is also solid. Although refractive index (as a proxy for density as a proxy for salinity) is temperature dependent, it is not strongly so (heat water in a pot on your stove from 25 Celsius to 75 Celsius, and the water level does not change significantly). And it does not account for the difference in measurement that you found in your testing a week or two ago.
All of which brings to rise a few questions:
• How strong (or poor) is the temperature compensation on refractometers?
• Given that the heat capacity of a drop of water is MUCH less than the heat capacity of the prizm on the refractometer, do we need to heat a refractometer up to the exact temperature of the solution we are measuring in order to get a proper reading?
This is the point we are all missing, in my humble opinion. It is practically inconsequential what the temperature of the drop of water was prior to the application on the refractometer; it is only important the temperature of the solution on the glass. And the glass temperature has a huge impact on that measurement.
I am coming around to the fact that conductivity is a better proxy for salinity than is refractive index, at least within the context of the marine aquarium hobby.
Back to me -
To sum up - irrelevant of your refractometer stating it has ATC (auto temp compensation) if you want an accurate reading with a refractometer make sure the glass is the same temp as the testing solution, but as this is quite impossible then better still buy the accurate Profilux salinity probe and get true readings every time!
Class dissmissed ;)
