Salinity goes up when temperature goes down

[Bramzor]

I noticed today that my salinity went up (Apex conductivity probe) while nothing changed. I expected that there was something wrong with my ATO, but apparently that was not the case.
Further investigation showed that salinity probe is affected by the temperature. I do have a TC of 2.2% configured on my Apex which should correct the salinity measurement based on the current temperature of the tank. But for some reason, the jump in salinity is too high for my liking: (and for being able to do anything useful with it like protect against ATO issues etc)

You can clearly see on the graph that if temperature drops a bit more than 0.5 degrees Celsius (which is about 1.8 degrees Fahrenheit) salinity raises with 0.5.
Does this mean that the general advised TC of 2.2% is not correct?

 

[lapin]

I have read a lot of posts that the probe is not very reliable.
Disclaimer: I have one but never used it.

 

[Bramzor]

I have read a lot of posts that the probe is not very reliable.
Disclaimer: I have one but never used it.

The reliability is more related to the difficulty to calibrate this (because you have to do a few steps very strictly). Haven't calibrated it anymore in years (and haven't seen any reason for it) and to be honest, the measurement is always right on target in my case. With the exception of the temperature compensation which is off but this is something that can be modified in Apex. But if you get it calibrated once, its in my experience, very reliable. You should however keep in mind that the probe is an analog measurement meaning it measures milivolts which can be an issue if you run power cables across.

 

[Larry L]

I don't know if the pattern that's showing on your graph means that the compensation is too small, or too large, so I'd probably make a small increase and watch over time to see if you get less of a temperature artifact in the salinity graph, or more of one. From there you should be able to gradually tweak it to minimize the effect of temperature on the reading.

 

[Miller535]

I noticed today that my salinity went up (Apex conductivity probe) while nothing changed. I expected that there was something wrong with my ATO, but apparently that was not the case.
Further investigation showed that salinity probe is affected by the temperature. I do have a TC of 2.2% configured on my Apex which should correct the salinity measurement based on the current temperature of the tank. But for some reason, the jump in salinity is too high for my liking: (and for being able to do anything useful with it like protect against ATO issues etc)

You can clearly see on the graph that if temperature drops a bit more than 0.5 degrees Celsius (which is about 1.8 degrees Fahrenheit) salinity raises with 0.5.
Does this mean that the general advised TC of 2.2% is not correct?

First off, while it's called a salinity probe, it's actually measuring conductivity, not salinity. Then based off of the conductivity the apex uses a formula that converts the conductivity to salinity. So there is a slight difference from that.

The salinity will move a little every time your heater turns on. It shouldn't be much. Mine fluctuates from 33.2-33.3-33.4. All all very close (1.0250-1.0252). Most people who complain about the probe being inaccurate did not calibrate them correctly.

 

[Bramzor]

First off, while it's called a salinity probe, it's actually measuring conductivity, not salinity. Then based off of the conductivity the apex uses a formula that converts the conductivity to salinity. So there is a slight difference from that.

The salinity will move a little every time your heater turns on. It shouldn't be much. Mine fluctuates from 33.2-33.3-33.4. All all very close (1.0250-1.0252). Most people who complain about the probe being inaccurate did not calibrate them correctly.

What TC are you using? Mine fluctuates double that amount as you can see on the graph.

Fluctuation has nothing to do with calibration though because you only calibrate on one value, one calibration fluid? So if fluctuation is too high, this has to do with something else, which might be the TC factor, or the probe quality or anything like that.

 

[Miller535]

What TC are you using? Mine fluctuates double that amount as you can see on the graph.

Fluctuation has nothing to do with calibration though because you only calibrate on one value, one calibration fluid? So if fluctuation is too high, this has to do with something else, which might be the TC factor, or the probe quality or anything like that.

Mine is set at 2.2. (even though I played with this early on and found it made little to no difference)

When I said most people complain about their probe being inaccurate, I was being general, not specific to your fluctuating problem.

One thing I found out early on, even where you put the probes in your sump matters. The salinity probe especially needs to be in a calmer part of the sump, not getting any bubbles or micro bubbles.

 

[Bramzor]

I don't know if the pattern that's showing on your graph means that the compensation is too small, or too large, so I'd probably make a small increase and watch over time to see if you get less of a temperature artifact in the salinity graph, or more of one. From there you should be able to gradually tweak it to minimize the effect of temperature on the reading.

Standard calibration fluids, for example 53ms, are for a specific salinity value. This calibration fluid equals about 35ppt (ocean levels) so if the value is lower than that, that value gets multiplied by the TC factor, which is 2.2% currently. So I assume I need to increase the TC?

 

[Bramzor]

One thing I found out early on, even where you put the probes in your sump matters. The salinity probe especially needs to be in a calmer part of the sump, not getting any bubbles or micro bubbles.

Correct but I assume micro bubbles would create a much higher measurement fluctuation? But I have a stable value overall so I think it's rather the TC offset as it is a bit related to the temperature in the graph? So maybe it is a bad calibration compared to the temperature doubling the offset?

 

[Sinibotia]

Any chance you can get an actual conductivity reading, and an independent salinity reading from a refractometer or something?

 

[Sinibotia]

Conductivity isn't a simple conversion to salinity; it's an awkward function of conductivity, temperature and pressure. So I'm going to guess that your problem is that outside of a certain range your conversion factor breaks down.

 

[Miller535]

Here's what mine looks like

 

[Miller535]

Conductivity isn't a simple conversion to salinity; it's an awkward function of conductivity, temperature and pressure. So I'm going to guess that your problem is that outside of a certain range your conversion factor breaks down.

It's not simple, but it is just a conversion that the apex uses an algorithm to do. With that said, it's not completely accurate. And is only meant to be a reference. If you look at mine, it's very stable. If it suddenly swung I would know something was up. Otherwise I use my refractometer still.

 

[Bramzor]

Here's what mine looks like

Thats stable... Mine is only 40g (140l), that obviously does not help for salinity stability when using ATO.

 

[Sinibotia]

It's not simple, but it is just a conversion that the apex uses an algorithm to do. With that said, it's not completely accurate. And is only meant to be a reference. If you look at mine, it's very stable. If it suddenly swung I would know something was up. Otherwise I use my refractometer still.

Ahh ok. Not familiar with the apex, just going off of programming my own temp/conductivity/salinity monitor and controller. If temperature going down makes salinity go up, that sounds like it's overcompensating for temperature.

 

[ajhudson15]

when I ahd my apex and my temp would go up my salinity probe chart would rise. and it was actually quite a bit. I think one time I went from 78 degrees to about 81(really hot day) and I went from 1.025 to 1.029

 

[Miller535]

Ahh ok. Not familiar with the apex, just going off of programming my own temp/conductivity/salinity monitor and controller. If temperature going down makes salinity go up, that sounds like it's overcompensating for temperature.

I am not really sure what about the temp makes it move. Mine moves even when I turn then temp compensation off even. I truly think the temp compensation does little to nothing to the salinity readings. Ph on the other hand is very temp compensated,.

 

[Miller535]

when I ahd my apex and my temp would go up my salinity probe chart would rise. and it was actually quite a bit. I think one time I went from 78 degrees to about 81(really hot day) and I went from 1.025 to 1.029

I have only ever noticed mine going up or down when the heater actually turns on or off. If my house just gets hot, or the water from lights, it does not effect my salinity reading.

 

[Sinibotia]

I am not really sure what about the temp makes it move. Mine moves even when I turn then temp compensation off even. I truly think the temp compensation does little to nothing to the salinity readings. Ph on the other hand is very temp compensated,.

I mean it's not the world's most complicated algorithm to write, so I don't know how they'd get it wrong. Wish there was a way I could look at the code.

 

[Miller535]

I mean it's not the world's most complicated algorithm to write, so I don't know how they'd get it wrong. Wish there was a way I could look at the code.

I don't know. Like I said, I am happy with mine. Ranging from 1.0249-1.0251. That's splitting hairs. The op may have a issue. I don't know. Like I said, sump placement matters a lot. They are very sensitive to bubbles.