Would You Rely on your Conductivity Probe to Control ATO?

[Dr. Jim]

I was planning on working on Programmable Logic to allow my ATO to be controlled by my Conductivity Probe (where low S.G. would trigger the addition of saltwater when called for by the ATO) but then started to wonder if this would be too risky. Of course all probes will eventually drift and even "go bad" so if this happens without my knowledge, it could lead to disastrous results.

Although my Conductivity Probe is "spot on" (when compared to a new, calibrated refractometer AND a new Milwaukee digital refractometer) my GHL equipment is only 2 months old, so I am wondering what others think of the reliability of the probe. Specifically:

-How long have you had your probe?
-Have you had problems with it drifting or "going bad," and if so, how long did it take?
-Would you trust it to regulate saltwater ATO additions when your S.G. drops?

I should mention that I also just started doing daily AWC (for the first time in my 50+ years in the hobby). Previously, when doing monthly water changes, I would only check my S.G. once a month at the time of a water change and make corrections at that time. But with daily automatic water changes I can see how the S.G. might drift. (I am using the method where the "In" and "Out" work simultaneously.)

At the moment I am thinking that it might be better to just pump in a little extra saltwater when the S.G. is low or pump out a little tank water if it is high (although it would be a shame to not be able to automate the process).

Thanks for your answers and opinions!

 

[ScottB]

It is tempting, but I would not do it.

I've had APEX for a couple years now, and of all the probes, the conductivity probe is the least reliable/stable for me.

For me -- with a few exceptions like the doser -- I don't let my controller automatically "do" a whole lot while I am away. I just want it to "tell" me when parameters are changing. This allows me to figure out what I might have left on/off, or up/down, or in/out, open/closed.

The one major thing I allow it to do for me is to shut the return pumps if there is water on the floor.

 

[Mhart032]

It is tempting, but I would not do it.

I've had APEX for a couple years now, and of all the probes, the conductivity probe is the least reliable/stable for me.

For me -- with a few exceptions like the doser -- I don't let my controller automatically "do" a whole lot while I am away. I just want it to "tell" me when parameters are changing. This allows me to figure out what I might have left on/off, or up/down, or in/out, open/closed.

The one major thing I allow it to do for me is to shut the return pumps if there is water on the floor.

Thats exactly what i was going to say. the conductivity prob is rarely correct, I cant actually rely on it to give me accurate info, i normally just look for trends in the numbers without focusing on the number itself due to the issues with reliability. its really a constant second guess scenario.

 

[ReefAddiction34]

Wouldn’t do it either. Technology is great at times, but too much automation will get you in trouble at some point. Let the ATO do its thing. More important is to have a general sense of how much ATO water you go through per day/week and if that’s suddenly changes look into what might be causing a flow issue in your tank... clogged filters, problem with a pump etc..

 

[KrisReef]

I have a swing arm to measure salinity because that method does not rely on electricity to measure salinity. To many variables to control for with a conductivity probe, imo a finger dip taste test has the same error margin as an e probe in saltwater.
But it’s a good question!

 

[griff500]

I'll go against the previous comments to a certain degree by saying that I've used Apex and Profilux and I found the GHL salinity readings to be far more stable and reliable than those provided by the Apex.

That doesn't mean you'll get the same result on your tank of course!

I would make sure you have a baseline where your manual measurement matches your probe measurement and then check from time to time. Once you are confident then let it take control of the water changes but perhaps make sure that the amount of time spent pumping water is limited, much like you would with an ATO to try and avoid having issues. Check manually from time to time regardless.

 

[AZMSGT]

Yes, I would.. since that’s one thing I’m trying to do. However I‘m considering adding a second probe. Perhaps use the two to make an average/virtual probe.

 

[Dr. Jim]

I'll go against the previous comments to a certain degree by saying that I've used Apex and Profilux and I found the GHL salinity readings to be far more stable and reliable than those provided by the Apex.

That doesn't mean you'll get the same result on your tank of course!

I would make sure you have a baseline where your manual measurement matches your probe measurement and then check from time to time. Once you are confident then let it take control of the water changes but perhaps make sure that the amount of time spent pumping water is limited, much like you would with an ATO to try and avoid having issues. Check manually from time to time regardless.

I'm so glad to hear you say this.... because it is exactly what I have been thinking. Every decision I make with the GHL Controller system is based on my "concerns and worries" when I go away for a week or so. Even though probes can "drift", it is not real likely that they will suddenly drift drastically in just one week (although one still needs to be prepared for this possibility). The idea of limiting the running time of the saltwater ATO pump seems to be the key/failsafe. I may initially set it to only 5 seconds. (A slight S.G. drop doesn't have to be fully corrected immediately, and should be gradual anyway, of course.)

I just checked my S.G. with my two refractometers (one, the digital Milwaukee). In 2 months, there has been a drift of 0.001, with the probe reading higher than it should. I think allowing the probe to trigger saltwater additions when needed will be safe as long as it is monitored (which is no problem when I'm not traveling, which isn't often) and when I do travel, be sure to calibrate the probe before leaving.

So, my next Programmable Logic projects:
Presently, ATO comes from the "kalk pump" when pH is below 8.3 and from the "non-kalk pump" when pH is 8.3 or greater. I need to figure out if it is possible to have "saltwater pump" activation have PRIORITY if the Conductivity Probe reads below 52.7 (assuming 77.5F) and disregard pH commands.
AND, it would be nice to also have tank water removed (with my AWC "Out Pump") if the Probe reads above 52.7 (and have this action also take priority over pH control).

Got some thinking to do!!

AZMSGT: Great idea about averaging 2 conductivity probes. Something for me to consider down the road.

 

[AZMSGT]

Yes, I’m equally trying to automate so I can take longer trips.

 

[griff500]

I'm not sure I would calibrate the probe before leaving for a trip - if it's all working then I hate to do anything to the system immediately before going away...

I liked this idea of maintaining salinity years ago but I was using Apex then and wouldn't have dared to attempt it with that equipment.

 

[Jaebster]

I'd like to know @Ditto @Lasse @Vinny@GHLUSA take on this. Maybe they know of others doing it this way. The GHL user group in Europe is much larger and I beat theres others that have actual experience in this and might even have some sort of redundancy control. Would be interested in doing same when my tank is set up.

 

[ingchr1]

While my conductivity probe has been stable for a majority of the time, it has drifted down suddenly then drifted back to normal on a few occasions. The drift was 3 PSU or so and took several minutes down then back up.

The only reason I knew it happened was because I had the probe set to alarm. Woke me up a couple of times, so I turned the alarm off.

 

[arking_mark]

I'm so glad to hear you say this.... because it is exactly what I have been thinking. Every decision I make with the GHL Controller system is based on my "concerns and worries" when I go away for a week or so. Even though probes can "drift", it is not real likely that they will suddenly drift drastically in just one week (although one still needs to be prepared for this possibility). The idea of limiting the running time of the saltwater ATO pump seems to be the key/failsafe. I may initially set it to only 5 seconds. (A slight S.G. drop doesn't have to be fully corrected immediately, and should be gradual anyway, of course.)

I just checked my S.G. with my two refractometers (one, the digital Milwaukee). In 2 months, there has been a drift of 0.001, with the probe reading higher than it should. I think allowing the probe to trigger saltwater additions when needed will be safe as long as it is monitored (which is no problem when I'm not traveling, which isn't often) and when I do travel, be sure to calibrate the probe before leaving.

So, my next Programmable Logic projects:
Presently, ATO comes from the "kalk pump" when pH is below 8.3 and from the "non-kalk pump" when pH is 8.3 or greater. I need to figure out if it is possible to have "saltwater pump" activation have PRIORITY if the Conductivity Probe reads below 52.7 (assuming 77.5F) and disregard pH commands.
AND, it would be nice to also have tank water removed (with my AWC "Out Pump") if the Probe reads above 52.7 (and have this action also take priority over pH control).

Got some thinking to do!!

AZMSGT: Great idea about averaging 2 conductivity probes. Something for me to consider down the road.

Interested in how you set this up with GHL. I just installed my probe and wanted to maintain salinity. Can you share?

 

[Dr. Jim]

Interested in how you set this up with GHL. I just installed my probe and wanted to maintain salinity. Can you share?

Unfortunately, I never got around to working on the salinity control....but thanks for reminding me. I still want to try this, but too many other GHL and tank problems going on that take precedence.

The things I have done with pH probe:
1) Used it to dose Kalk or Non-Kalk ATO water depending on pH.
2) Tried using motorized ball valves to maintain a constant pH.

I've abandoned both of these however. The first one worked (and I have a post with all the P.L.) but stopped it because my tank was so "young" and didn't really need much Kalk. I may go back to it. Some say it is better to control Kalk additions by ALK measurements, but Randy suggests considering using pH for this.

The second one I stopped because:
a) Who knows what pH value is the right one to maintain, it is not likely that all corals will like the same pH, and, a constant pH doesn't "happen" in the ocean.
b) Although the motorized ball valves did a pretty good job maintaining the constant pH, because of the hysteresis, there was always a zig-zagging up and down which bothered me.

Sorry I couldn't help with your salinity inquiry.

 

[arking_mark]

Unfortunately, I never got around to working on the salinity control....but thanks for reminding me. I still want to try this, but too many other GHL and tank problems going on that take precedence.

The things I have done with pH probe:
1) Used it to dose Kalk or Non-Kalk ATO water depending on pH.
2) Tried using motorized ball valves to maintain a constant pH.

I've abandoned both of these however. The first one worked (and I have a post with all the P.L.) but stopped it because my tank was so "young" and didn't really need much Kalk. I may go back to it. Some say it is better to control Kalk additions by ALK measurements, but Randy suggests considering using pH for this.

The second one I stopped because:
a) Who knows what pH value is the right one to maintain, it is not likely that all corals will like the same pH, and, a constant pH doesn't "happen" in the ocean.
b) Although the motorized ball valves did a pretty good job maintaining the constant pH, because of the hysteresis, there was always a zig-zagging up and down which bothered me.

Sorry I couldn't help with your salinity inquiry.

I'm definitely going to control salinity... GHL does make these things more complicated then it needs to be.

 

[AZMSGT]

I also haven’t been able to get around to this. I’ve had multiple negative things happen to my tank and reef and have been forced into several major steps backwards so once I get back I’ll see if I can get this figured out.

FYI, ICH breakout, then tank seem failure followed by setting up a new system, where my set backs.

 

[Jon's Reef]

Issue is that you are trying to control 2 variables with 1 sensor; Volume and salinity.

Salinity may stay on, but no guarantee that the volume stays consistent.

Also, there is no guarantee that you will only see a salinity increase, what about effects like salt creep, skimming and manual water removal (tank testing, coral maintenance, macro removal)

 

[Brett S]

For what it’s worth I attempted to do this with my Neptune Apex several years ago and eventually gave up. I have an auto water change system that uses dosing pumps to remove small amounts of tank water and add small amounts of new salt water every few minutes throughout the day. So my plan was to vary the amount of salt water that was added based on the salinity probe reading.

If the probe showed that the salinity was high then the system would dial back on the amount of saltwater that was added for the AWC. The same amount of tank water would still be drained, so this would cause my ATO to replace some of that with freshwater to bring the salinity down.

If the probe showed that the salinity was low then the system would add extra saltwater for the AWC. Again, the same amount of tank water would be drained, but this time the extra saltwater would make up for some water that was lost due to evaporation, so this would cause the ATO to add less freshwater and the salinity would come up.

I put a couple of safeguards in the system to prevent huge changes that could cause problems in the tank. If the salinity probe reading was too high or too low the salinity adjustment wouldn’t happen and it would send me an email alert instead. This way if there was a bad reading or the probe got way out of calibration it wouldn’t be attempting to change the salinity of the tank.

Additionally I configured it so that the amount I was adjusting the AWC would only allow it to change the tank‘s salinity level by .2ppt per day. This way there wouldn’t be any big salinity swings and even if the probe was slowly drifting out of calibration it wouldn’t make big changes in the tank’s salinity level.

However, even after letting it run for several months and trying to make tweaks to it I realized that it really just didn’t do anything beneficial. I found that while the salinity probe readings tended to be reasonably consistent they would still bounce around a little. So often I would have a situation where half the day the system thought the salinity was low and was trying to raise it and the other half of the day it thought the salinity was high and it was trying to lower it, so over the whole day the net result was no change.

Additionally, I discovered that even without something like this my salinity level tends to be pretty stable... certainly stable enough to leave the tank for a week or two on vacation. It seemed like a great idea on paper, but in practice it really wasn’t useful.

 

[Brett S]

Issue is that you are trying to control 2 variables with 1 sensor; Volume and salinity.

Salinity may stay on, but no guarantee that the volume stays consistent.

I don’t think this is accurate. As long as you have an ATO then the volume should absolutely stay consistent. As I said above, you can adjust the amount of saltwater you are adding to adjust the salinity, and between evaporation and the ATO the volume of water in your system will stay consistent.

 

[Lasse]

If GHL´s probe is situated in a high flow area and the flow is able to pass through the two electrodes - my probe (and I have test many) give a stable reading. However it is difficult to get a accurate value. My probes , typically, shows around 2 - 4 PSU lower than actual value. It is important that you know the calibration fluids real temperature and note that during the calibration.

If you chose to use PSU as measurement value - you cant chose any offset - however if you use kg/L you can use an offset value and have it to work. However the nominal value can only be expressed as mS and it complicate the thing much more. I prefer PSU - not kg/L. I have done an workaround. I have a very stable temperature in my aquarium - around 25 degree C - see last month below

I analyse my salinity with a magnetometer (both 0 and 35 promille calibrated. I put my salinity probe to respond to manual temperature- adjust down the temperature to the level my probe show the same salinity as my refractometer - in my case 19.6 degree C

​

I changed my display to show mS (blue) . Read the measured value in mS (read) and put a nominal value - (black) just below

Hysteresis to 0.3 ms (black) The operation mode controller (blue) to pulse variable and pulse 8 min and pause 8 min (in my case - with my system it was best with 8:8 - other with faster or slower turn over can need to change these figures) And back to PSU (red)

​

Now I connect this probe with a switch channel "virtual" with help of some PL commands in order to pump out water in a container if salinity is over 34,6 psu and at the same time sense an overflow sensor and not run if there is a general alarm

Fill water 2 respond to a level indicator in the container (10 L)

The switch channel

Now I knit a physical pump to this chanel. Alarm that stop the pump (PL8) if it is near the maximum and that the pump react on switch channel 61

With help of this I have succeed to take out salt water from the system and refill with RO water with my normal Top Off system - keeping the salinity rather stable. The figure show my salinity during last week and also (blue bars) when (and how much) saltwater have been taken out

Have run this setup for half a year now

Sincerely Lasse