Poll on water testing (for ammonia only)

[Clown2020]

It’s intresting following along reading this thread, to start with I was very much a sceptical of api‘s amonia test but now am to the point of believing that it is accurate but probably not the best choice But best choice is a whole other topic

on the point of prime being a known cause of issues, it would be interesting to better understand why

I have always been of the understanding that this is because the api test kit raises the ph converting the amonia bound by prime back to nh3 . SEachem even put this warning on there website for prime and on there ammolock product bottle, maybe it would be good of seachem to add the same warning to there prime bottle although that would likely cause the to lose some ammolock sales.


to test this I ran the api test on my mixed salt water (left) and salt water with the 5x emergency dose of prime. and can see without the presence of amonia prime has no impact on the test result

would be interesting if someone had some prime and amonia to hand if the could dose up to say 1ppm amonia bind it with prime and see if prime alters the levels reported or if prime purely causes a misread due to lack of awareness about the warning above.

 

[Brian_68]

How do you even know the results are incorrect if you do not have an accurate reference to compare?

 

[brandon429]

Brian agreed, I use the tank pics as that basis. They're usually reporting toxic levels from the kit so pics of a perfect tank help to validate

 

[ScubaScott]

There is widespread commentary out there that certain (API) test kits for ammonia are notoriously inaccurate. 1. I thought it would be interesting to see how common this thought is, in general - and compared to other tests. 2. At the start I will list several testing errors that I have seen - that have relatively specific causes. You can choose one of the first 3 and 1 of the last 3. Comment Below

I just stood up a new FOWLR tank using Dr. Tim's Fishless Cycle method. I also used Caribsea Aragalive sand, so there was already some ammonia and nitrite from the sand itself starting to cycle in the couple of weeks before I started Dr. Tim's. Side note: the reason there was about two weeks between when I started the tank and when I started the Dr. Tim's cycle was because the Aragalive was extremely silty and it took me a long time to get the water clear. I will never use that filthy crap again. At least not the Fiji Pink. Maybe their coarser Select Grade isn't so dirty with fines that are hard to filter out.

Anyway, back on topic. I started with a Red Sea Kit, but it only reads up to 2ppm. The idea of diluting the sample to bring the reading within the kit's scale and then multiply the reading by the dilution ratio hadn't occured to me yet, so I bought an API kit because it appeared to go high enough for the levels anticipated during cycling. Here are the results during the whole cycle (Red Sea in first column, API in second). Note, Ammonia spikes on December 2 and 7 are because the Dr. Tim's method has you add Ammonium Chloride on those days. Those were the only two times during the cycle that the tank was dosed with AC.

2-Dec	>2	2.0
3-Dec	1	1.0
4-Dec		2 to 4
5-Dec		2 to 4
6-Dec		2 to 4
7-Dec	>2	1 to 2
8-Dec		1 to 2
9-Dec		0.5 to 1
10-Dec	1.2	0.5
11-Dec	0.4 to 0.8	0.5
12-Dec	0.2	0
13-Dec	0	0
14-Dec		0
15-Dec	0	0
16-Dec	0	0
17-Dec	0	0
18-Dec		0
19-Dec	0.2	0
20-Dec	0	0
21-Dec	0	0
22-Dec	0	0

 

[brandon429]

I always enjoy seeing the little clues in cycle logs that support the ammonia drop date from cycling charts. that's a big deal pattern used even when the test kits don't show the drop by day ten. when it doesnt hit zero like your logs show, that's when the stuck cycle fun begins (they're not really stuck, the fun is convincing them they're not stuck)

that trending downward is the gold standard of duration-based cycles. the system with active surface area doesn't just stall at toxic levels that downward trend continues into whatever cycled reefs run at, to be determined once we have more than one digital tester to sample. looks like you have good tests and procedure above.

 

[MnFish1]

I just stood up a new FOWLR tank using Dr. Tim's Fishless Cycle method. I also used Caribsea Aragalive sand, so there was already some ammonia and nitrite from the sand itself starting to cycle in the couple of weeks before I started Dr. Tim's. Side note: the reason there was about two weeks between when I started the tank and when I started the Dr. Tim's cycle was because the Aragalive was extremely silty and it took me a long time to get the water clear. I will never use that filthy crap again. At least not the Fiji Pink. Maybe their coarser Select Grade isn't so dirty with fines that are hard to filter out.

Anyway, back on topic. I started with a Red Sea Kit, but it only reads up to 2ppm. The idea of diluting the sample to bring the reading within the kit's scale and then multiply the reading by the dilution ratio hadn't occured to me yet, so I bought an API kit because it appeared to go high enough for the levels anticipated during cycling. Here are the results during the whole cycle (Red Sea in first column, API in second). Note, Ammonia spikes on December 2 and 7 are because the Dr. Tim's method has you add Ammonium Chloride on those days. Those were the only two times during the cycle that the tank was dosed with AC.

2-Dec	>2	2.0
3-Dec	1	1.0
4-Dec		2 to 4
5-Dec		2 to 4
6-Dec		2 to 4
7-Dec	>2	1 to 2
8-Dec		1 to 2
9-Dec		0.5 to 1
10-Dec	1.2	0.5
11-Dec	0.4 to 0.8	0.5
12-Dec	0.2	0
13-Dec	0	0
14-Dec		0
15-Dec	0	0
16-Dec	0	0
17-Dec	0	0
18-Dec		0
19-Dec	0.2	0
20-Dec	0	0
21-Dec	0	0
22-Dec	0	0

So your cycle - it looks like - took 9 days - give or take. That seems pretty good. Its too bad we don't have your pH - and Salinity/Temp - or we could make a chart of the free ammonia (at least an estimate)

 

[MnFish1]

It’s intresting following along reading this thread, to start with I was very much a sceptical of api‘s amonia test but now am to the point of believing that it is accurate but probably not the best choice But best choice is a whole other topic

on the point of prime being a known cause of issues, it would be interesting to better understand why

I have always been of the understanding that this is because the api test kit raises the ph converting the amonia bound by prime back to nh3 . SEachem even put this warning on there website for prime and on there ammolock product bottle, maybe it would be good of seachem to add the same warning to there prime bottle although that would likely cause the to lose some ammolock sales.


to test this I ran the api test on my mixed salt water (left) and salt water with the 5x emergency dose of prime. and can see without the presence of amonia prime has no impact on the test result

would be interesting if someone had some prime and amonia to hand if the could dose up to say 1ppm amonia bind it with prime and see if prime alters the levels reported or if prime purely causes a misread due to lack of awareness about the warning above.

This has been done - there is a whole thread designed to prove that Prime does not detoxify 'free ammonia'. The problem with the API test is that its not sensitive enough to determine whether prime is affecting free ammonia. PS - I would read both of your samples above as 'zero' especially the one on the right. I think if you had a light source directed directly at the card and the vials - it would be much less 'green looking'. PS - I plan to do an experiment with Prime and a seachem alert shortly *the Seachem alert does read 'free ammonia'

 

[MnFish1]

How do you even know the results are incorrect if you do not have an accurate reference to compare?

So - for example - when I did my experiments - I verified it using the 'Seachem alert badge'. If that was 'yellow' (0) and the API tube was 'yellow' (0) - they agreed and they did agree about 59/60 times. one time I made a mistake with the number of drops. At least with the kits I've used - I have not seen any variation.

 

[brandon429]

I honestly find non test kit factors to be excellent benchmarks for test accuracy. no cycling charts allow for persistent ammonia noncontrol past day ten, no seneyes (only digital we have so far accuracy tbd once someone benchmarks with a Hach I guess) show anyone having extended ammonia noncontrol at day ten and the way I assess those non testing benchmarks as variable on the low level is that there aren't pages of outliers for each one, and because literally hundreds and hundreds of cycles that use timing-based procedure over any testing are at 100% safe completion rate. we did not have any cycler input fish and the fish die in gray smelly ammonia water, the ratio of safety is 100%.


its to the point I now prefer the non test kit markers for cycling posts over the non digital test kits. I acknowledge MN and Taricha and Dan and Randy and several others can wield the kit effectively and get zeroes, but that's not the majority it seems.

when I search online for any reef tank claiming to have extended ammonia noncontrol issues, the return pattern we see is clear water, happy animals, open corals, and a non digital test kit (user error applies) as the sold claimant. those posts are totally symptomless for the most burning/pain-inducing parameter we encounter in reef metabolism-the test kit or mode of use causes the false alarm in all cases I've seen online.

If someone reading right now is past day ten on cycling and api shows ammonia not compliant, I would never, ever ever assume that reading or the relay of the reading is correct at all.

 

[brandon429]

For the vote option that indicates misreads, it seems fair some of those votes thought it was a misread just because it wasn't zero, but it makes sense some slight green on API is not a misread and indeed matches the fact that running reef tanks don't run at zero-does that seem accurate MN? Brian's question seems to apply there too-without a literal test to benchmark the kit at the time of reading folks we only have subjective guesses on accuracy.

some of the downvotes may have been on kits reporting well.


Is there a consensus yet on whether api should read obviously yellow on a running reef or is the very light green presentation still ok? If a brand new cycled tank well past the 60 day mark shows yellow before fish are included, that seems good in context.


I can now see where both instances of yellow and light green are totally accurate reads within context.

 

[threebuoys]

After confirming my tank had cycled by using the API ammonia and nitrate test kits, I stopped testing for ammonia.
I did find the API colors a little difficult to differentiate, but it was good enough for me to confirm the cycle was finished.

 

[MnFish1]

I honestly find non test kit factors to be excellent benchmarks for test accuracy. no cycling charts allow for persistent ammonia noncontrol past day ten, no seneyes (only digital we have so far accuracy tbd once someone benchmarks with a Hach I guess) show anyone having extended ammonia noncontrol at day ten and the way I assess those non testing benchmarks as variable on the low level is that there aren't pages of outliers for each one, and because literally hundreds and hundreds of cycles that use timing-based procedure over any testing are at 100% safe completion rate. we did not have any cycler input fish and the fish die in gray smelly ammonia water, the ratio of safety is 100%.

I might suggest that part of your issue relates to selection bias. There are a lot of people (myself included) - who never check ammonia - so scientifically, its kind of impossible to say there is no such thing as a 'stuck cycle'. Second - a lot of people here keep their tanks 'cycling' for months - and never check anything.
Third - with the comment on the 'Digital way'. There is no data for or against that a seneye is any more or less fallible than an API test - or a sachem alert badge for that matter. The last thing I would buy a Seneye for is 'continuous ammonia values' - unless perhaps I was doing a study - and for that Seneye is not recommended (by Seneye). - depending on the experiment to be done - and the amount of ammonia to be used

For the vote option that indicates misreads, it seems fair some of those votes thought it was a misread just because it wasn't zero, but it makes sense some slight green on API is not a misread and indeed matches the fact that running reef tanks don't run at zero-does that seem accurate MN? Brian's question seems to apply there too-without a literal test to benchmark the kit at the time of reading folks we only have subjective guesses on accuracy.

For the vote option that indicates misreads, it seems fair some of those votes thought it was a misread just because it wasn't zero, but it makes sense some slight green on API is not a misread and indeed matches the fact that running reef tanks don't run at zero-does that seem accurate MN? Brian's question seems to apply there too-without a literal test to benchmark the kit at the time of reading folks we only have subjective guesses on accuracy.

You might have missed it - but a slight amount of green does not mean 0.25. It means 0. (according to the instructions - it says the reading is the closest color not the highest. Another problem is people using an incorrect background and or incorrect lighting - both of which I found to be very difficult to photograph carefully - unless I go basically to a lit window - as compared to a soft white non-cool which bulb. So - if you see a picture of someone with a tinge of green - and they say 'my ammonia is reading 0.25'. You can safely tell them its 'zero'. (if its leaning towards the green side - its 0.25.

Is there a consensus yet on whether api should read obviously yellow on a running reef or is the very light green presentation still ok? If a brand new cycled tank well past the 60 day mark shows yellow before fish are included, that seems good in context.

Stunning yellow or slightly green are still '0' - using the instructions in the test booklet. So IMHO, both are 'ok'.

 

[brandon429]

I do not buy into the stretch claim of symptomless ammonia noncontrol in a running reef tank. The tank and life will show symptoms, and the levels will rise to lethal levels as lack of surface area continues to be insufficient to meet bioload demands. The reason there aren’t stalled cycles is because ammonia was controlled in all cases. This reading here isn’t zero, it’s a misread and represents the range of readings possible with the kit (from the water parameters example thread prior pages)

(watches dateline while making cycle points)


a tiny overfill isn't causing it to go falsely green, lots of readings online are this dark in running tanks


that reading passes .5 ppm and nears 1 ppm in a 100% past cycling reef producing new coralline algae growths keeping fish and corals in a very low bioload high surface area setting.


confirmation bias also includes believing api when it doesn’t align with cycling charts, digital samples on file, and complete tank layout when full tank pics are posted and accepting a wide color gradient as all zero.

 

[Spare time]

There is widespread commentary out there that certain (API) test kits for ammonia are notoriously inaccurate. 1. I thought it would be interesting to see how common this thought is, in general - and compared to other tests. 2. At the start I will list several testing errors that I have seen - that have relatively specific causes. You can choose one of the first 3 and 1 of the last 3. Comment Below

Funny enough, I have used api ammonia tests probably a thousand times and never once had a "false positive." I think the issue is that people don't know what is 0 and what is actually registered ammonia. This is understandable given how difficult it is to read API tests. A no ammonia result is clearly a yellower color. I also test nitrite when I test ammonia, so I can tell where along a cycle is at, as nitrite test act as a sort of double check on ammonia test.

 

[MnFish1]

What is interesting about the Poll (so far) is that :

1. about 3 times as many people 'think' that the API ammonia test has more errors than other tests.
2. A lot of people never check ammonia (why get a Seneye)
3. about 3 times as many people have had no error (or made a mistake causing the error) as people who think the test has random errors.

To me - it makes me wonder - is the API issue a 'fact' (that there are more errors) - or is it a perception that is out there - giving the idea that the test has more false positives? Becasue - these 2 summary statements (to me) - don't match each other.

 

[MnFish1]

I do not buy into the stretch claim of symptomless ammonia noncontrol in a running reef tank. The tank and life will show symptoms, and the levels will rise to lethal levels as lack of surface area continues to be insufficient to meet bioload demands. The reason there aren’t stalled cycles is because ammonia was controlled in all cases. This reading here isn’t zero, it’s a misread and represents the range of readings possible with the kit (from the water parameters example thread prior pages)

(watches dateline while making cycle points)



that reading passes .5 ppm and nears 1 ppm in a 100% past cycling reef producing new coralline algae growths keeping fish and corals in a very low bioload high surface area setting.


confirmation bias also includes believing api when it doesn’t align with cycling charts, digital samples on file, and complete tank layout when full tank pics are posted.

I never said that an Ammonia of (whatever that is in the picture) - was safe. First - its not the way you're supposed to even read the test. (its supposed to be standing up - against the card - with a light source facing it).

Second. I'm using the word 'stalled cycle' - but I'm not entirely sure that there is a clear definition (I mean an exact definition) - that if you asked 100 people 99 would say xxxxxxx. Everyone has their own definition - but it seems like you're using the term as if there is some kind of glossary somewhere.

Third. It MIGHT be a misread - or it might be a mistake in testing itself (which is why I listed the 10 or so possibilities). One thing I can 100% guarantee you - that photo is a misread - because its not how you're supposed to read or interpret the test. There is a nice YouTube video showing this.

I think we're both saying the same thing in a sense. If you use the calculator I provided yesterday - even a clear 0.5 ppm total ammonia - at pH 8 and 78 degrees at 35 salinity comes out to a free ammonia of 0.02. The LOW range of toxic ammonia per Seneye's website - is 10 times higher than that. The 'deadly range' is >25 times higher - for free ammonia.

BTW - One concerning thing is the apparent overdosing that happens with Dr. Tims. Repeatedly - it seems like using the 'drops/gallon' recommended gives 4 or even more PPM ammonia starting out. This may also be causing consternation - why is my ammonia not dropping? (and its not an error - but nor is it toxic). if you start with an ammonia of 8 you're likely going to have a slower cycle than if you started at 2 ppm (for a lot of reasons)

 

[Clown2020]

What is interesting about the Poll (so far) is that :

1. about 3 times as many people 'think' that the API ammonia test has more errors than other tests.
2. A lot of people never check ammonia (why get a Seneye)
3. about 3 times as many people have had no error (or made a mistake causing the error) as people who think the test has random errors.

To me - it makes me wonder - is the API issue a 'fact' (that there are more errors) - or is it a perception that is out there - giving the idea that the test has more false positives? Becasue - these 2 summary statements (to me) - don't match each other.

For me getting the seneye was about having a par meter and amonia testing was a bonus that would validate if the results of either the seachem badge or the api test (which I would have been misreading at the time)

 

[MnFish1]

Funny enough, I have used api ammonia tests probably a thousand times and never once had a "false positive." I think the issue is that people don't know what is 0 and what is actually registered ammonia. This is understandable given how difficult it is to read API tests. A no ammonia result is clearly a yellower color. I also test nitrite when I test ammonia, so I can tell where along a cycle is at, as nitrite test act as a sort of double check on ammonia test.

I have to admit (and I'm no API "fan" vs the others) - I find the Salifert and Seachem to perhaps give a more accurate 'result' But I also find the colors extremely difficult to see. I don't know why - I found myself paying more for tests that I had to end up squinting and then basically flipping a coin as to which color it was. Meaning that (to me/my eyes) - that the somewhat larger area between results (i.e. alkalinity - you can read 8 or 9. You cant read 8.6) - doesnt matter - because with the other tests I couldnt see it anyway.

EDIT - of course the Hanna checkers take over the eye part. I have not been a fan of them. (I sold mine on Ebay)

 

[MnFish1]

For me getting the seneye was about having a par meter and amonia testing was a bonus that would validate if the results of either the seachem badge or the api test (which I would have been misreading at the time)

Thanks - yes - I meant to say that I would consider one for some of the other functions - and if/when they add more testing capabilities.

 

[nereefpat]

What is interesting about the Poll (so far) is that :

1. about 3 times as many people 'think' that the API ammonia test has more errors than other tests.
2. A lot of people never check ammonia (why get a Seneye)
3. about 3 times as many people have had no error (or made a mistake causing the error) as people who think the test has random errors.

To me - it makes me wonder - is the API issue a 'fact' (that there are more errors) - or is it a perception that is out there - giving the idea that the test has more false positives? Becasue - these 2 summary statements (to me) - don't match each other.

I'm seeing 3 times as many people vote for Option E than Option C. That makes no sense, to say "I think the kit has false positives" while "I haven't found false positives."

What I make of it is there are people voting for Option E that haven't used the test kit themselves. In other words, "I think this test has far more false readings," because someone on the internet told me the API ammonia kit is junk.