Salt reading on SWG?

[keithpgdrb]

How dependable is the salt reading on the SWG? I’ve been getting some fluctuations recently having not done anything to the pool.

I picked up the salt read strips. I assumed those weren’t any more dependable than the typical pool chemical strips. Have t used them yet though. I’ll try them now. But I thought I’d ask.

 

[mas985]

How dependable is the salt reading on the SWG? I’ve been getting some fluctuations recently having not done anything to the pool.

I picked up the salt read strips. I assumed those weren’t any more dependable than the typical pool chemical strips. Have t used them yet though. I’ll try them now. But I thought I’d ask.

Based upon some anecdotal evidence, the Pentair SWGs seem to be a little worse at displaying accurate salt levels. But I would not say the salt strips are that much better. You are better off with a Taylor salt drop kit.

https://www.amazon.com/TAYLOR-TECHNOLOGIES-INC-K-1766-CHLORIDE/dp/B001DO35EU/ref=asc_df_B001DO35EU/?tag=hyprod-20&linkCode=df0&hvadid=167151535350&hvpos=&hvnetw=g&hvrand=14513426031740345243&hvpone=&hvptwo=&hvqmt=&hvdev=c&hvdvcmdl=&hvlocint=&hvlocphy=9032030&hvtargid=pla-313230194306&psc=1

As an SWG cell ages, the surface becomes pitted and for a given salt level, the current will drop which means the displayed salt level will also drop relative to the test kit. So if the SWG is reading much less than the drop test kit (<25%), then the cell could be failing. But this also happens with dirty cells so you need to make sure there are no deposits on the cell plates.

I would only add salt if the drop kit says salt is below around 2900 ppm and only add enough to bring it up to 3200 ppm. If you go just by the SWG display and the cell is failing, there is the potential that you could add too much salt and when you replace the cell, the SWG will alarm because the salt level is too high.

In my experience, if the cell is failing, the unit will continue to operate even if the salt level readout is very low but I have a different unit than you. So you may be able to keep using it until you decide to replace it.

I am currently in this situation. My test kit shows a salt level of 3200 ppm while the SWG shows 2000 ppm. I know the cell failing and I will replace it soon but I am going to get as much life out of this one as possible. I am also interested in seeing how low it will go.

 

[JamesW]

My test kit shows a salt level of 3200 ppm while the SWG shows 2000 ppm. I know the cell failing and I will replace it soon but I am going to get as much life out of this one as possible. I am also interested in seeing how low it will go.

Maybe change the setting to T-9?

 

[mas985]

AquaLogic only has a T-5/T-15 setting and T-5 trips the current alarm. But in reality, it doesn't matter. There is a message that salt is too low but that does not stop production. As far as I can tell so far, it never will stop production but we will see.

 

[JamesW]

I thought it stopped below 2,400 PPM?

 

[keithpgdrb]

Mine read 3600 4 days ago. Today it read 3750. Goofy. It recommends 3500. It’s slightly high. But I figured I’d get a bunch of splash out over the summer anyway.

 

[JamesW]

http://www.haywardnet.com/pdfs/ProLogic_4.45_Troubleshooting_Guide.pdf

 

[mas985]

I thought it stopped below 2,400 PPM?

The instantaneous salt is at 2000 ppm. The average salt reading 2600 ppm. Maybe when average salt gets below 2400 ppm it will shut off but so far, it is still producing.

The AMPS vs time is very interesting and I did this measurement sometime last year:


It is worse now and dips sooner and lower after cycle start.

 

[JoyfulNoise]

Mine read 3600 4 days ago. Today it read 3750. Goofy. It recommends 3500. It’s slightly high. But I figured I’d get a bunch of splash out over the summer anyway.

That’s well within the temperature variation for an IC unit. My salinity bounces around by 200ppm all the time. Pentair tries to compensate for water temperature variations but whatever implementation that are using to do that is terrible. And the thermistor built into the flow switch goes bad and fails quite often. Once that happens, you will see big swings in salinity with temperature.

 

[JamesW]

The average salt reading 2600 ppm. Maybe when average salt gets below 2400 ppm it will shut off but so far, it is still producing.

Correct, it goes by the average salinity reading.

Once you get below 2,400 ppm, I am pretty sure it will stop.

 

[mas985]

Perhaps, but you can also spoof it by saving the instant salinity right after cycle start. The salt level always starts off high and then drops off. If you time it right, you get the unit to read higher salt levels. It then takes several cycles to drop again.

 

[JamesW]

Maybe wire in a 24 watt (24 volts DC x 1 amp) light in parallel to increase the amperage by 1 amp?

 

[mas985]

I have a experimented with an additional resistor across the amps detection resistors to decrease the measured current so I can use the T-5 setting and that does work. That doesn't require any board modifications, just a couple of alligator clips. But I may not need that once the average salt gets below 2400 ppm as the current may stay below 5 amps anyway.

 

[JoyfulNoise]

Do you plan to do a post-mortem dissection of the cell to see what the plates look like after you’ve fully exhausted the life of the cell? Would be interesting to see the guts of the cell. I admit, I still have my old IC40 sitting in my home office on the shelf with the intent one day of chopping it up to see what it all looks like on the inside.

 

[mas985]

I did that on my first cell and there really wasn't much to see. The cell plates were just a flat black without much detail. They were not as degraded as I thought that might be but I don't know what a new cell plate looks like either so it is hard to compare. Unfortunately, I did not take a picture of the plates before throwing them out since they were not all that interesting. I do have another failed cell (#2) and if get inspired I might cut that one open too.

 

[JoyfulNoise]

I did that on my first cell and there really wasn't much to see. The cell plates were just a flat black without much detail. They were not as degraded as I thought that might be but I don't know what a new cell plate looks like either so it is hard to compare. Unfortunately, I did not take a picture of the plates before throwing them out since they were not all that interesting. I do have another failed cell (#2) and if get inspired I might cut that one open too.

Should be lighter grey, not black. Black would be closer to titanium that has undergone some anodization. They also might use a titanium nitride coating before depositing the ruthenium inside the magnetron PVD batch coater and depending on the nitride thickness, it can look black.

 

[mas985]

Black would be closer to titanium that has undergone some anodization.

Isn't that consistent with a failed cell?

I am going from memory which is not always reliable. They may have be a dark grey or a dusty black.

 

[JoyfulNoise]

Isn't that consistent with a failed cell?

I am going from memory which is not always reliable. They may have be a dark grey or a dusty black.

Yes. My understanding of the failure mode is that the ruthenium slowly dissolves over time as the cell produces. Many ruthenium oxide compounds are soluble at normal pool water pH. Once the titanium is exposed it immediately oxidizes. Thin oxide films of titanium can conduct electricity and so you don’t necessarily see any significant drop in current until a large part of the plate starts to fail. TiN can also be tuned to be conductive and so neither surface produces any chlorine but can support a small amount of electrolysis of water into hydrogen and oxygen gas.