Please help me understand these surprise salinity readings.

[Ben's Pico Reefing]

Tank A 10 gallon standard is 20x10x12 inches. This equals 2400 inches3. Which equals 10.3896103896104 Gallons.

Tank B 40 gallon breeder is 36x18x16 inches. This equals 44.88311688311688 gallons.

These measurements are if you fill to the very top where adding would spill over and zero displacement from anything in tank.

This can also very based on manufacturing and process.

 

[drolmaeye]

These are two temporary tanks. They are bare bottom. For the purpose of my post, the tanks have 10 gallons and 40 gallons.

Again, I added 1 pound to each tank and received the posted results.

Since posting, I've added another pound to each. My expectation:

The 40 gallon tank will double to 10ppt and the 10 gallon will double to 20ppt.

There is nothing wrong with adding salt directly to a new tank. Mixing outside the tank is something you do for a water change. A person who owns a 500 gallon tank doesn't mix in a 500 gallon container and then pump it into the tank. They mix directly in the tank.

If they are bare-bottom then that minimizes error in the amount of water in the tank as well as whether it was well-mixed. For my scenario that only leaves the amount of salt that was added to the tank (assuming your salinity measurement was more or less accurate). Based on your starting salinity, the volumes of water, and the amount of salt you added the second time, it seems like the 10 gallon would indeed go up to 20 ppt, but the 40 gallon should only go up to about 7-8 ppt. Very curious to hear the outcome.

I did not say that mixing in a tank was wrong. If a tank is completely empty except for water, it is essentially the same as mixing in a separate bin. If one has rock or sand, then there may be added complication of properly estimating water volume or proper mixing.

 

[rhitee93]

I don’t actually know how good typical refractometers are at very low salinity.

The Milwaukee digital only claims +/- 2 ppt.

If the measurements are at low ppt values, these errors can become large fractions of the total.

I think this is your answer. The reason we calibrate refractometers with a 35ppt solution rather than pure RODI water is that these devices have more error as you get further away from the calibration point. Add that instrument error to a few % error for water volume, and other human error and you can end up with results like you got.

If you want to try an experiment: recalibrate your refractometer using fresh RO/DI water instead of a calibration standard. Then measure the 35ppt calibration standard and see how far off the reading is. That might help explain the error you are seeing. (Or, it might deepen the mystery if I am wrong. Fun to be had either way)

 

[Dom]

I think this is your answer. The reason we calibrate refractometers with a 35ppt solution rather than pure RODI water is that these devices have more error as you get further away from the calibration point.

I did not know this.

Add that instrument error to a few % error for water volume, and other human error and you can end up with results like you got.

I can understand this, but the discrepancy seemed very large. I wouldn't expect it to be that much.

 

[Hubert J. Farnsworth]

I think this is your answer. The reason we calibrate refractometers with a 35ppt solution rather than pure RODI water is that these devices have more error as you get further away from the calibration point. Add that instrument error to a few % error for water volume, and other human error and you can end up with results like you got.

If you want to try an experiment: recalibrate your refractometer using fresh RO/DI water instead of a calibration standard. Then measure the 35ppt calibration standard and see how far off the reading is. That might help explain the error you are seeing. (Or, it might deepen the mystery if I am wrong. Fun to be had either way)

yep, i think randy's on to it too. and if he recalibrates to rodi water, he should try taking the salinity of the two samples again to see how far off that moved the results.

 

[Dom]

yep, i think randy's on to it too. and if he recalibrates to rodi water, he should try taking the salinity of the two samples again to see how far off that moved the results.

I recalibrated using RODI. It was off by 2ppt.

 

[Reefering1]

Have you measured both tanks to be the same temperature?

 

[Ben's Pico Reefing]

I don't know how accurate this is but may help a bit. I typed in what you had and got quite a bit odd results around lol.
https://www.bulkreefsupply.com/reef-calculator/salt-mix-calculator

 

[Dom]

Have you measured both tanks to be the same temperature?

Yes; both at 76 degrees.

 

[Reefering1]

Add another pound. Let's see what you get(w/ a hour of circulation after it's mixed). Verify temperature and clarity of water before testing. It sounds like margin of error, as was said, especially so far from target range. I'm betting it gets closer the closer you are too 1.0264

 

[Dburr1014]

Tank A 10 gallon standard is 20x10x12 inches. This equals 2400 inches3. Which equals 10.3896103896104 Gallons.

Tank B 40 gallon breeder is 36x18x16 inches. This equals 44.88311688311688 gallons.

These measurements are if you fill to the very top where adding would spill over and zero displacement from anything in tank.

This can also very based on manufacturing and process.

Also consider these are outside dimensions not "between the glass" dimensions that actually occupy water.

Outside dimensions you have to subtract the glass thickness X2.

 

[Ben's Pico Reefing]

Also consider these are outside dimensions not "between the glass" dimensions that actually occupy water.

Outside dimensions you have to subtract the glass thickness X2.

Correct, but the glass is made differently in sizes as well so this adds further. If you measure from bottom of tank internally to top water line and then inside of glass for width and length, may even be able to do externally. But basically it will be a few mm. This was just a quick reference was all as how much water is an unknown variable. Plus there is something in the tanks circulating which is displacement. So was it before or after the circulation was added. Once these are solved we can plug those numbers into several different salinity calculators and see where it should be. It appears he used a scale as well to weigh the salt. The refractometer also appears to be within a good tolerance and calibrated well.

The math all around isn't matching for me lol. According to the calculators...if I remember correctly, the amount added vs salinity is off. Making me think currently if it is exactly 10 and 40 exactly the salt is not correct. But even with what was added, it doesn't come out the same. Unless I went crosseyed lol. This is a fun thread about salinity and interesting. Would love to see what the exact issue and outcome is.

 

[BeanAnimal]

@Dom and everyone else

1 - the water volumes are not likely exactly 4:1
2 - the salt (~ 1 pound) in each tank is not likely exactly 1:1 and is a small amount.
3 - the magnitude of error for the refractometer is not going to be the same for the two different readings.

Yes - it is a simple math issue. There is error in the volume, error in the total salt and error in both readings. It really can't be anything else unless your salt was that badly mixed and that is rather unlikely.

 

[Dom]

@Dom and everyone else

1 - the water volumes are not likely exactly 4:1
2 - the salt (~ 1 pound) in each tank is not likely exactly 1:1 and is a small amount.
3 - the magnitude of error for the refractometer is not going to be the same for the two different readings.

Yes - it is a simple math issue. There is error in the volume, error in the total salt and error in both readings. It really can't be anything else unless your salt was that badly mixed and that is rather unlikely.

Sometimes, I think too much. And this thread is an example of that.

That is exactly right; I saw the two tanks as a 4:1 ratio. And because of that, I expected to see 4 times the amount of salt used to achieve 35ppt in the 40 gallon as opposed to the 10. And when I wasn't seeing that, I was curious why.

I used a digital food scale to measure out the 1 pound of salt for each tank. I had no reason to believe that each tank received anything other than the 1 pound reported by the scale. But I accept that there are tolerances to be considered, and the 1 pound reported by the scale may be off.

I calibrated using RODI. But from this thread, I have come to understand and accept that this isn't the best idea, and that the calibration should be done via a solution that will allow you to calibrate at the target range. I've also read that it is likely the 4:1 ratio between the two tanks in question would be closer to that as I approach 35ppt.

In the end, when I consider variables such as inside tank diameters, scale and refractometer tolerances, individually, they didn't seem to be so great as to skew the number. But when you combine them, I can see how a larger discrepancy is possible.

 

[Dom]

Correct, but the glass is made differently in sizes as well so this adds further. If you measure from bottom of tank internally to top water line and then inside of glass for width and length, may even be able to do externally. But basically it will be a few mm. This was just a quick reference was all as how much water is an unknown variable. Plus there is something in the tanks circulating which is displacement. So was it before or after the circulation was added. Once these are solved we can plug those numbers into several different salinity calculators and see where it should be. It appears he used a scale as well to weigh the salt. The refractometer also appears to be within a good tolerance and calibrated well.

The math all around isn't matching for me lol. According to the calculators...if I remember correctly, the amount added vs salinity is off. Making me think currently if it is exactly 10 and 40 exactly the salt is not correct. But even with what was added, it doesn't come out the same. Unless I went crosseyed lol. This is a fun thread about salinity and interesting. Would love to see what the exact issue and outcome is.

At a 4:1 ratio, As an EXAMPLE: I would have expected that if the 10 gallon tank needed 1 pound of salt for 35ppt, the 40 gallon would require 4 pounds. I wasn't seeing that, which is how my OP came about.

 

[BeanAnimal]

Sometimes, I think too much. And this thread is an example of that.

That is exactly right; I saw the two tanks as a 4:1 ratio. And because of that, I expected to see 4 times the amount of salt used to achieve 35ppt in the 40 gallon as opposed to the 10. And when I wasn't seeing that, I was curious why.

I used a digital food scale to measure out the 1 pound of salt for each tank. I had no reason to believe that each tank received anything other than the 1 pound reported by the scale. But I accept that there are tolerances to be considered, and the 1 pound reported by the scale may be off.

I calibrated using RODI. But from this thread, I have come to understand and accept that this isn't the best idea, and that the calibration should be done via a solution that will allow you to calibrate at the target range. I've also read that it is likely the 4:1 ratio between the two tanks in question would be closer to that as I approach 35ppt.

In the end, when I consider variables such as inside tank diameters, scale and refractometer tolerances, individually, they didn't seem to be so great as to skew the number. But when you combine them, I can see how a larger discrepancy is possible.

Yep - sometimes when you step back and analyze what you are actually looking it things become clearer even though you thought they were simple to begin with.

Me... I would waste time doing the math and figuring out where and what error... just for the sake of figuring it out.

 

[Dom]

Yep - sometimes when you step back and analyze what you are actually looking it things become clearer even though you thought they were simple to begin with.

Me... I would waste time doing the math and figuring out where and what error... just for the sake of figuring it out.

I dig math.

 

[Ben's Pico Reefing]

There is so many variables and not a 1 to 1 or 4 to 1 lol. I learned a bit researching as well lol. But this will help others down the road as well. I have done a few experiments with things where I remember I didn't consider variables and made results skewed. But awesome investigation and wanting to figure this out. Need more of this. Or I'm just missing these posts lol.

 

[Dan_P]

Currently, I am setting up two temporary tanks which will house the contents of my main display, so that I can drain and relocate the tank to different wall of the room.

The two temporary tanks are a 40 gallon and a 10 gallon. Currently sitting at 76 degrees, I added one pound of IORC. After 12 hours, I tested salinity.

My results:

The 40 gallon tank has a reading of 5ppt at 76 degrees.
The 10 gallon tank has a reading of 10ppt at 76 degrees.

Since the 10 gallon tank is 25% of the volume of the 40 gallon, and since the same amount of salt was added to both tanks, it makes sense to me that the salinity in the 10 gallon should be 4 times that of the 40 gallon, or 20ppt. It isn't.

Thoughts?

Thank you,
Dom

I think you might have goofed up somewhere.

If you didn’t measure the water, that could explain some of the difference between expected and obtained results. Why don’t you try repeating the results on a smaller volume of water, say 1 cup and four cups of water? If you can’t reproduce the weird results, move on. If you can, report back.