Where is the magnesium coming from?

[H2OPlayar]

I have an interesting phenomena going on right now. My Trident is testing my system, and my magnesium is consistently going up. This would be fine normally as I would turn down my dosing pump, or even turn it off. What is weird is that the mag dosing pump has been unplugged for the last 3 weeks and I am getting a very linear increase in magnesium. Any thoughts?

I was thinking maybe the dosing hose is pulling mag out of my dosing system, but it is a peristaltic pump, so it can't really create a syphon (I think). Next step might be take the hose out of the sump completely.

FYI, that "big" jump in alk was when I got home from vacation and saw a bunch of precipitate from my alk doser in my sump. I was lazy and broke it up and let it dissolve and we get 8.3 --> 9.3, with no noticeable effects on coral etc.

 

[Darkxerox]

Confirm with an offline test kit to see if the Trident is drifting?

 

[jhuynh]

get a 2nd opinion in testing your mag?

 

[thesassyindian]

What are you dosing for Alk and Ca?
Some of them have Mg as well.

 

[Coral reefer]

I’d say testing error

 

[JVU]

I also would assume the Trident is giving you wrong readings and confirm the readings and/or recalibrate the Trident.

 

[H2OPlayar]

I dose soda ash, and calcium carbonate, so unlikely from there.

I can bust out my Salifert and test, but am too lazy to do the 2x a day. I can test every few days and see what is happening.

I tend to trust the trend of the Trident as it is doing it the same way every day, but I know the best way is to validate the measurement.

 

[CaptainB]

Peristaltic pump can't siphon, the line is pinched shut.

Could be a reagent problem also. I'd get fresh reagents first and then recalibrate.

If you still have problems I can look into what exactly the trident is doing in there, I do analytical chemistry development/troubleshooting as my day job.

 

[H2OPlayar]

If you still have problems I can look into what exactly the trident is doing in there, I do analytical chemistry development/troubleshooting as my day job.

Sweet job! I wouldn't call it a "problem" yet, since it's mag and everything in the tank is happy. I also have tried literally nothing to debug/dive into it. Neptune has great customer support and has helped me out any time I couldn't debug on my own but I really appreciate your offer. I like poking the hive mind and seeing where the discussion goes here

 

[Darkxerox]

Same as @CaptainB I literally deal with multiparameter testing systems daily in a lab and this is a pretty common thing for a sensor to drift over time due to a bad reagent pack or dirty sensor. We always confirm with a QC standard supplied by the vendor, but in lieu of that, testing once with a Salifert to confirm will help the troubleshooting.

 

[H2OPlayar]

Good point of the reagents, I did just switch from the Trident ones to the ABC reagents to see how those work.

 

[thesassyindian]

Good point of the reagents, I did just switch from the Trident ones to the ABC reagents to see how those work.

Ooh! please let me know how those work (outside of the current issue). Would love to save some $$$!

 

[CaptainB]

If it's a sensor/detector problem I'd flush the whole thing with di water first. The sample and reagent lines.

If you are switching reagent vendors this probably a good practice when making the switch

 

[H2OPlayar]

Ooh! please let me know how those work (outside of the current issue). Would love to save some $$$!

So far so good, I think there was a little jump of the measurement when I swapped, but the trends all make sense.

If it's a sensor/detector problem I'd flush the whole thing with di water first. The sample and reagent lines.

If you are switching reagent vendors this probably a good practice when making the switch

Agree, but that sounds like work and I think diluting it with the new reagent will solve itself over time. If this were an alk or ca, I would care more. Definitely interesting to watch.

I should have more time next week and this will be on the top of my list. RODI flush and recal.

 

[JVU]

Peristaltic pump can't siphon, the line is pinched shut.

Could be a reagent problem also. I'd get fresh reagents first and then recalibrate.

If you still have problems I can look into what exactly the trident is doing in there, I do analytical chemistry development/troubleshooting as my day job.

I like your new profile pic from the swap

 

[JVU]

Just as a reality check, Mg doesn’t change significantly under any normal situations and is high enough concentration that any changes require adding or removing a lot. I can’t tell what Mg it was reading before, but for example to go from 1400 to 1500 requires your readings to be way off unless you dumped a whole bag of Mg in (and you didn’t). Seeing a trend doesn‘t make the readings more likely to be accurate.

 

[IOnceWasLegend]

Just as a reality check, Mg doesn’t change significantly under any normal situations and is high enough concentration that any changes require adding or removing a lot. I can’t tell what Mg it was reading before, but for example to go from 1400 to 1500 requires your readings to be way off unless you dumped a whole bag of Mg in (and you didn’t). Seeing a trend doesn‘t make the readings more likely to be accurate.

Thirding this. A few months ago I had my alkalinity start to drift after weeks of being constant. Checked with a Hanna checker after a week of this and realized it was still consistent; calibrated my Trident and that solved my problem.

 

[Qwiv]

FYI, mine is doing the same thing and I don't even dose Mg. Longer term graph shows constant increase every day. Only difference is the weird negative correlation with a Ca at the end of the shorter graph below (not sure what happened there). I figured Mg. and Ca. were just playing Musial Chairs together on the last bit, but the mg testing has been less useful given it goes up every day and I don't dose Mg at all. (Use Kalk + Sodium Chloride)

Last weird bit has a correlation with a PH spike. My dosing has been off for a few days due to a PH cap to the dosing schedule, but PH went to 8.5 during that strange spike/dip a full 24/48 hours after the last dose.

Was hard to correlate Mg with another test kit since the values are just plain off from each other every time I test anyways and I don't have any trending from the other test kit in the last few weeks, just a spot test. Salifert test kit always reads way off from Trident and not even a consistent + or - value that I could correct for. Validates that I have no confidence in the value of Mg Trident testing.

 

[H2OPlayar]

Was hard to correlate Mg with another test kit since the values are just plain off from each other every time I test anyways and I don't have any trending from the other test kit in the last few weeks, just a spot test. Salifert test kit always reads way off from Trident and not even a consistent + or - value that I could correct for.

Thank you for saying exactly what I was thinking. The salifert I couldn't get to be within 150ppm from one test to another and found the Trident to be way more accurate and repeatable. I am leaning towards a reagent/cal issue on the Trident which should be easy to clear up. Only wierd thing there is the mg readings matched up for the 3 weeks prior with a water change.

 

[Qwiv]

What is the scale of the variance on your small graph? Is it even meaningful? You also show a sharp drop before a slow rise. That seams more significant.