Salinity, Specific Gravity, Temperature, refractometers, hydrometers CONFUSION

[threebuoys]

Let me apologize up front for raising some questions that have likely been answered many times.

I have spent hours reading articles and threads on this website and results from google searches on this subject. Every time I think I've got it straight in my mind, something comes up that takes me back a couple of steps. Perhaps I'm spending too much time trying to understand the details when the answers are simple. These issues come in to play for me when I'm mixing salt and water at different temperatures and then adding the mix to an aquarium which may be at a different temperature. My guess is that in most cases the differences in salinity I encounter are not material enough to warrant concern. But, I still want to understand.

Scenario 1: I have a refractometer that indicates thru the eyepiece that it is calibrated at 20 degrees C (68 degrees F). The instructions say the ATC (automatic temperature compensation) to function correctly, the ambient temperature of the room must be 20 degrees C when calibrated. When calibrated at this temperature, the accuracy of the device at other temperatures between 10C and 30C will not be affected due to ATC. Question: If I calibrate at 77degrees F using a solution with a known specific gravity at that temperature, will the accuracy of the device for samples from my aquariums at the same temperature (77) be accurate. Getting the room temperature to 20 degrees C to recalibrate does not happen very often, particularly in the summer. I assume the ATC ability of the device will be compromised since it was calibrated at 77 degrees and therefore temperatures +/- 5 or 10 degrees would produce erroneous readings. I also assume that specific gravity/temperature/salinity conversion tables provide no value or need for this scenario?


Scenario 2: I have a hydrometer that was calibrated at 20 degrees C (68 degrees F). My water temperature in my tank is 77 degrees. To determine the actual salinity, I should use a conversion table that was calibrated at 20 degrees C using the measured hydrometer specific gravity cross referenced to 77 degrees F (25 degrees C) t? A conversion table calibrated at 25 degrees C would be useless?

Scenario 3: I have a swing arm meter. Generally, swing arm meters are not temperature dependent? If I have a sample with a known specific gravity or known salinity, I can confirm the swing arm is accurately, or not accurately calibrated? Conversion tables are not needed?

Thanks for your help!

 

[The_Paradox]

It’s not that critical over all, but the best practice is to calibrate to a known reference sample under the conditions you will be testing. After that it’s consistently in how you test. Ie test the exact same way every time and recalibrate periodically.

 

[threebuoys]

@Randy Holmes-Farley I would really appreciate your thoughts on the questions I posed.

Thanks

 

[Randy Holmes-Farley]

Salinity does not ever change with temperature.

Specific gravity has only a very small change with temperature over the ranges we encounter. No table that I’ve ever seen shows this small change. I show some values in one of mg articles. 77 F and 25 C are the same.

Glass hydrometer reading change a lot with temp and always need corrections.

Swing arm hydrometers do this correction mechanically and seem pretty good at the correction Adie r, even if the readings themselves are not always accurate.

Refractive index also changes a lot with temp, but most refractometers we use have ATC (automatic temperature compensation). If ATC is working correctly, and if you calibrate with a 35 ppt standard, it will not matter what temp you calibrate at or measure tank water at as long as both values are within the compensation temperature range. Statements by manufacturers that you must calibrate at a fix temp are incorrect (how could it be true if the device automatically adjusts the reading?). I also believe this last statement is true if you calibrate with RO/DI, but several errors creep in when using RO/DI.

Conductivity readings also need to be adjusted for temp. All devices we use do it, but some allow that function to be turned off or misadjusted to an incorrect correction.

 

[330Bob]

Refractometers are confusing so don't worry.

1. Salinity of the water is a constant it does not change
2. You can get SG reads of 1.0233 to 1.0277 yet the sample is 35ppt. All due to calibration temps.
3. The refractometer has two scales SG and Salinity. Both are only accurate at the calibration temp.

Example look at the solution of 35ppt it was calibrated at 25C and will read 1.0264SG and 35PPT on that particular refractometer. It will read 1.0252 on one calibrated at 20c but read at 25c. So you can get a bunch of different SG number all describing the same 35ppt.

If I were to calibrate that solution using a refractometer that was calibrated at 20c I would cool the temp to 20c calibrate it so the line passes through 35ppt and 1.0266 now the SG and Fluid PPT are in sync.

But... If it warms up to 25c my refractometer will show 1.0252 which on the chart is 33 ppt. This is wrong as we know the solution is 35 ppt. you would have to move the right side of grid/chart up so the 35 ppt matches the sg of reading of 1.025.

I think its probably best to take the known sample of 35 ptt calibration and zero using the right side of the chart and ignore the SG completely and only think and talk in terms of Salinity as that is the bottom line and does not change with temps. Do this at the room temp of your testing area or the refractometers temperature.
So
Keep the refractometer the same temp each time you test and only use the PPT scale and set it to 35ptt using your calibration fluid. Lets say your tank water is 35 ppt (same as calibration sol) Then it will read dead on 35ptt.
Tanks water is really 33 ppt its will read lower then 35. It might not read exactly 33 due to you not testing at the calibration temp but it will be close.

Don't calibrate with RO or Distilled. I checked mine and when im calibrated at 35ppt and I check distilled it will not read zero it will be above the zero line. If i set it on zero then my ppt wont be 35ppt. The scale is off or the distilled water has impurites in it. Use the known sample or two of them if you can. The salinity is all that matters not the SG reading.

I also set mine to 35ppt and took it outside and watched the sality reading drop from 35ppt down to 33ppt as the warmed to outside temps of 82. (I calibrated at maybe 72F) So the ATC did not work properly.


1. Calibrate using a 35 ppt seawater standard (not RO/DI water, unless it is a digital device that physically doesn't permit this). Set the reading to 35 ppt dont bother with SG scale. read directly off the Sality side.

2. Calibrate at the same room and refractometer temperature as you intend to make measurements.

3. Calibrate and measure using the same lighting source.

The issue with this method is the lines on the grid are calibrated to one refractive index (which can change w temp) The further away from the sweet spot of 35ppt line on the grid (ie 20 ptt) might read 22 if you are above calibration temp of the instrument to much. I think if you are within plus/minu 5 ppt from 35ppt is close enough.

not sure if this was any less confusing. Let me know if you see any other issues with just using the salinity scale instead of SG.

 

[Reefahholic]

Don't calibrate with RO or Distilled. I checked mine and when im calibrated at 35ppt and I check distilled it will not read zero it will be above the zero line. If i set it on zero then my ppt wont be 35ppt. The scale is off or the distilled water has impurites in it. Use the known sample or two of them if you can. The salinity is all that matters not the SG reading.

I’ve seen 3-4 different calibration solutions all read differently. I had more luck calibrating to RODI, but I know it’s better to calibrate at whatever you’re keeping your system at.

I finally switched back to a hydrometer. It was reading exactly 1.026. ICP came back at 34.3 psu. This was 6 days apart and I only sold 1-2 corals during that time so it should have been closer.

 

[330Bob]

Funny I was just thinking the same thing. A hydrometer might work better. Not sure why my tank maintenance fellow uses but he is on vacation so I looked at this Refractometer I have purchased 10+ years ago. My undergrad was physics chem minor and post grad Optical Physics. I enjoyed seeing the simplicity yet complexity of the manufacturing of these rafractometers. I have a feeling ones made in the 1960s/70s might have been better quality with less variance. These are like a cheap rifle scope, where you can't trust the turrets or mil-dots. Many areas that can be out of spec with the grid, alingment, abberations, the set screw and bimetal ATC etc.

I would expect most of them to have various errors that will add up to incorrect readings. That is why I have concluded (As many of you with more aquarium experience) have done that its best to get a calibration solution that is near your target salinity. And I don't know what the tolerances of those solutions are in the real world. Maybe the solution I purchased is not 35ppt. Its a a in-exact science with a sprinkle of "Close enough". I can see people going crazy chasing numbers with a never will be in spec Refractometer. Would need a bunch of calibration solutions and a few refractometers.

BTW I love the hydrometer in your video. Nice work!

 

[Randy Holmes-Farley]

not sure if this was any less confusing. Let me know if you see any other issues with just using the salinity scale instead of SG.

Most reftractometers sold to hobbyists have ATC, automatic temperature compensation. If you use one, you need not worry about temperature at all, either for calibration or for use, as long as you are in the range of temps the ATC covers (and assuming it was made correctly).