I run my aquarium with help of one Profilux P4 and some other GHL equipment.
The profilux have one salinity meter that you with some tricks can have to run rather stable and showing nearly the right values.
1. Place it in high flow compartment. My is placed so the return water (2000 l/h around) just flush it and flush away the tiny gas bubbles that can be formed
2. Do the calibration very carefully. I have done a special calibration equipment that i have floating in the sump. One sample tube for the standard and one for the temperature probe (filled with water)
3. rinse the electrode once a week (tooth brush)
When this done - I always get a stable reading but the result could be out of chart
In my equipment - i have a function that allow me to have a temperature compensation that follow the temperature probe. I have never have this to work good. Instead I use manual temperature compensation. Still I often get an off track reading. Because my aquarium only differ with ± 0.1 degree C I do the next trick. I take a newly calibrated refractometer (showing right both at the calibration fluid and RO water) and analyze my salinity. Let us say - it shows 34.6 PSU. Now i test different manually temperature compensation temperatures in order to find that one that give me a reading of 34.6 with my salinity probe. For the probe I use now - it is 17.8 !!
I have done this for half a year - and I get rather stable readings - I trust the probe just now
The profilux allow a dose pump to react to a probe through a connection with a switch channel. I configure the probe to have a nominal value in mS - for me 52.8 ms which is for my probe 34.6 psu and lowes possible hysteresis (0.3 mS). Profilux computers have a possibility to react on on/off mode (in my case reading 52.95 - switch on and reading 52.65 - switch off (if mode later on is descending). This will cause the salinity to vary around ±0.2 PSU. But there is a better option - you can have the controller to react in a pulse variable/paus fixed mode. It works like a PID regulation function and let the real value come closer to the nominal value compared with on/off mode. I hope I will have ± 0.1 PSU at most (I run fixed 5 minutes - pulse max 10 minutes
This is the basic but I need to have safety too - otherwise it ends up with water on the floor. I have floater that will react when it is to much water in the receiving jar and block the function -> send an alarm.
I have also a software calculated max filling in the jar -> alarm. And all shout down at an alarm. I can do this with help of the PL language in the profilux computer. If this work - I will place the jar above the sump (have to redo a little) and have a overflow to the sump if the jar is filled to much
We will see how it will work out
More questions - just shout
Sincerely Lasse
1. Place it in high flow compartment. My is placed so the return water (2000 l/h around) just flush it and flush away the tiny gas bubbles that can be formed
2. Do the calibration very carefully. I have done a special calibration equipment that i have floating in the sump. One sample tube for the standard and one for the temperature probe (filled with water)
When this done - I always get a stable reading but the result could be out of chart
In my equipment - i have a function that allow me to have a temperature compensation that follow the temperature probe. I have never have this to work good. Instead I use manual temperature compensation. Still I often get an off track reading. Because my aquarium only differ with ± 0.1 degree C I do the next trick. I take a newly calibrated refractometer (showing right both at the calibration fluid and RO water) and analyze my salinity. Let us say - it shows 34.6 PSU. Now i test different manually temperature compensation temperatures in order to find that one that give me a reading of 34.6 with my salinity probe. For the probe I use now - it is 17.8 !!
I have done this for half a year - and I get rather stable readings - I trust the probe just now
The profilux allow a dose pump to react to a probe through a connection with a switch channel. I configure the probe to have a nominal value in mS - for me 52.8 ms which is for my probe 34.6 psu and lowes possible hysteresis (0.3 mS). Profilux computers have a possibility to react on on/off mode (in my case reading 52.95 - switch on and reading 52.65 - switch off (if mode later on is descending). This will cause the salinity to vary around ±0.2 PSU. But there is a better option - you can have the controller to react in a pulse variable/paus fixed mode. It works like a PID regulation function and let the real value come closer to the nominal value compared with on/off mode. I hope I will have ± 0.1 PSU at most (I run fixed 5 minutes - pulse max 10 minutes
This is the basic but I need to have safety too - otherwise it ends up with water on the floor. I have floater that will react when it is to much water in the receiving jar and block the function -> send an alarm.
I have also a software calculated max filling in the jar -> alarm. And all shout down at an alarm. I can do this with help of the PL language in the profilux computer. If this work - I will place the jar above the sump (have to redo a little) and have a overflow to the sump if the jar is filled to much
We will see how it will work out
More questions - just shout
Sincerely Lasse