Aquatic Plant Forum banner

Just wiped out a tank of fish

9K views 55 replies 15 participants last post by  diablocanine 
#1 ·
I have been following the sometimes heated debate on wether or not to run a CO2 system at night or to supliment O2 with an airstone or powerhead at night and wether or not we are putting our fish at risk. I realize that in a well balanced tank with everything running smoothly, a CO2 system running at night should not deplete O2 levels to a dangerous level or cause a major fluctuation in pH.

But, what if something goes wrong? I have a beautifuly planted 55g tank that had, I emphasize had, a great little population of micro and dwarf rasboras. It was filtered by a HOT magnum. Thursday night we had some storms go through and lost power for about an hour. The HOT magnum decided not to fire back up. Today, I found most of my cool little fish floating with a few staglers gasping for their final breaths. I may save 4 out of about 12 of the fish.

A small air pump with an airstone and a timer could have prevented this tradjedy. It also would have been much cheaper than replacing the fish.
 
#2 ·
Sorry to hear that! It's why I never run my CO2 at night because if the plants can't metabolize the carbon in the water, the CO2 turns into carbonic acid using the hydrogen and oxygen from the H2O.

Ahh....live and learn, man! We have all experienced a catastrophe every now and again. I accidently electrocuted all the fish in one of my tanks when a light fell into it....long story. And that was a marine tank, so the fish were all above $20!

Salvage what you can, and at least you still have your plants! But that sucks!
 
#3 ·
The CO2 turns into carbonic acid no matter what. That's why the pH lowers.

And I would hazard a guess that the fish did not die due to CO2 but lack of O2 caused by the failure of the filter to restart. With out the water movement the water went stagnant and you lost the gas exchange at the surface.

CO2 doesn't deplete O2, not at night, not during the day, not at any time.
 
#22 ·
CO2 doesn't deplete O2, not at night, not during the day, not at any time.
CO2 can displace O2, if you pump a bunch of it into the water column.
This can happen if 1) you're pumping excessive amounts of CO2, enough so that your filter's agitation of the water surface is the only thing keeping enough O2 in the water for the fish and 2) surface agitation stops long enough to stress the fish.

Lesson: Setup the tank w/ good filtration and very low surface agitation, then slowly bump up the CO2. This way, the fish don't get killed every time the filter craps out.

Related lesson/suggestion - Eheims are more reliable at restarting than Mags are.
 
#4 ·
Another reason not to OD with CO2 injection. 15 to 20 pmm should be more than adequate for many tanks. For best performance, the CO2 should be added to the intake of the filter. If the filter stops, then the CO2 level in the tank will also stop due to the lack of water circulation. Too much CO2 will cause a significant drop in pH when working with low KH water. This can kill fish.

If the tank is NOT OVERSTOCKED, then the top surface area of the tank should provide adequate means for the exchange of O2. An experience aquarist will never allow the tank's water chemistry to go hay-wire during a power blackout. Remember, a good tank should not rely on filter and airstone to keep the livestock healthy.

Since you only have 12 fish in a 55 gal tank, O2 depletion should not be an issue. I've kept 40 cardinals, neons, and otos in a 50 gal for 3/4 day without the benefit of filter/pump/airstone.
 
#6 ·
Sorry furballi. You fell into the trap. Low pH caused by CO2 doesn't kill fish. And kH has nothing to do with it either. Many people are running tanks with pressurized CO2 injection with low kH water.

High levels of CO2 in an aquarium make it impossible for the gills of the fish to do the CO2/O2 exchange. This is what kills fish. Not the low pH.

Low pH is the symptom and not the problem.
 
#7 ·
"Too much CO2 will cause a significant drop in pH when working with low KH water. This can kill fish." What part of that statement is unclear? If the pH drops from 7 to sub 5 in 15 minutes (high level of carbonic acid in water with very low buffering capability or KH), then this alone can affect weak fish. Anabantoids can extract oxygen directly from the air, but they are still susceptible to a very sharp drop in pH! Plus, we all know that the concentration of CO2 in air is MUCH higher than water.
 
#8 ·
I'd assume that too much CO2 will build up a layer of CO2 on top of the water (CO2 is heavier than air), especially in those cases where you have a lid or a significant rim above the water line. This will invert the partial pressures of O2 and CO2 dramatically and lead to a situation where you will actually have more dissolved CO2 in the water than O2. And your poor anabantoid will also gulp down CO2 when he tries to get some oxygen from above the water surface.
 
#11 ·
Wrong again furballi.

It's not the pH drop that kills the fish. It's the high levels of CO2. The pH drops because because of the CO2.

And kH has nothing to do with it. It takes 30 ppm of CO2 to drop the pH of the water 1° no matter what the kH. That assumes of course that you are starting with a tank with no CO2 injection. But that first degree of drop takes ~30 ppm of CO2.

What part of "Low pH is the symptom and not the problem." did you not understand?

I will say it again and type it slowly so you can understand. pH changes caused by CO2 have no effect on fish. Of course if you get so much CO2 into the water that their gills can no longer function then you will harm the fish. But it's NOT the pH that causes that problem.

I have covered the reasons why pH changes due to CO2 do not harm fish in my FAQ. www.theplantedtankfaq.com

Also Tom Barr has shown that pH changes due to CO2 do no harm to fish.

Using CO2 to lower pH is an artificial means to do so. It does not change the hardness or TDS of the water. Normally when you lower pH you are also required to change the hardness and TDS. If this change occurs quickly you can cause problems with the osmotic balance of the fish. When CO2 is used to lower pH there is no change in the hardness or pH.

Again.... and I know it's hard, but you have to separate the pH and the CO2 levels. High CO2 levels will cause low pH. But the low pH is the symptom, not the problem.
 
#18 ·
Hmmm...it requires a very high level of CO2 in the water column to block the uptake of oxygen. High CO2 levels may cause some inhibition of respiration in fish, but it will take an abnormally high level of CO2 in the water to impede diffusion from the gills. Of course the ability to tolerate high level of CO2 will vary from species and condition of the fish.

I was tinkering with CO2 injection many years ago and accidentally lowered the tank's pH from 8.2 to 6.6 over several hours. Since the local water has a KH of about 10, the amount of dissolved CO2 must be at least 80 ppm. There was no loss of inhabitants (neons/SAE/otos) in that tank. Also note that it is very difficult to raise the concentration of CO2 above 60 ppm (room temperature) unless there is a malfunction of the regulator/needle valve assembly.

You are naive if you believe that a very sharp change in temperature, pH, KH, etc do not kill fish.
 
#13 ·
The pessurized CO2 system is a new addition to a well established tank, one that has comfortably weathered power outages in the past. I agree that lack of 02 killed the fish as opposed to overdose of CO2 but excess CO2 could deminish the carrying capacity of gases in the tank causing the depleted O2 levels.
 
#15 ·
The best thing to prevent this in the future is a power fail (normally closed) solenoid. You can also minimize the threat of this occurring with altering your CO2 setup.

Sometimes the CO2 injection rate is dependent on the back pressure of the water pump. I notice that my bubble rate increases 2-3x when my filter turns off. I have an external reactor connected to the output of my ehiem. In my setup, this would only fill up the reactor and not discharge into the tank until my 18" reactor fills. This would take several hours. I would get a CO2 burst when the power finally does turn on and my reactor then sounds like a waterfall. I have a bleed valve on my reactor for such occasions to drain the excess CO2.
 
#20 ·
Well, as expected this thread turned out to be an exciting one. So far, its been spirited, but not abusive. I applaud the combatants.

Nobody is challenging that rapid changes in any number of parameters isn't bad for fish. The point is that very high levels of CO2 kill fish, not the low pH that is a consequence of high CO2. Independent versus dependent variables.

In the absence of a drop checker, we really don't know what our CO2 levels are. My KH is now about 3.4 (LaMotte test kit) and I have my calibrated pinpoint controller at 6.0 to 6.2. This yields a "CO2 concentration" of 102 to 64 ppm and the fish are fine. When the solenoid sticks and the pH drops to 5.5, a "CO2 concentration" of 320 ppm, the fish come to the surface. When the concentration drops to about 150 ppm as indicated by a pH of 5.8 everything is hunky dory.

Why do the fish gasp for air at the surface when the solenoid sticks open and then resume their normal fishy routines at mid and lower levels once CO2 levels drop?

By the way, its a great way to take inventory if you've lost track of what fish you have....
 
#21 ·
You are naive if you believe that a very sharp change in temperature, pH, KH, etc do not kill fish.
Define "very sharp". I watch the LFS do 50% water changes on tanks full of Africans by going along and draining the water. Then going along and dumping in the buffer salts (the water here is extremely soft). Then going along and filling the tanks. Now most of these fish are in 40 gallon tanks. And there is a kH change of around 5-8° when the buffer salts are dropped into the tank. Then that kH goes down again when the fill water is adding. And with their water change system the fill takes about 2-3 minutes. So these fish are going though a quick rise in kH and pH and then a drop in kH and pH in a short amount of time. Yet they don't lose fish due to this.

But again. And I can't type slower than this. A normal pH changed caused by CO2 doesn't kill fish. Are you willing to concede that? You can drop the pH in an aquarium 1° (which gives 30 ppm of CO2) in minutes and not harm the fish. It's actually less of a change in pH than many fish have in the natural in a small stream with a tropical rain storm.
 
#25 ·
The OP said that a "dead hot magnum" caused high level of CO2 in the tank which killed the fish. I do not dispute the fact that a VERY high level of CO2 in the water column can kill fish. Normal CO2 level in air is ~350 ppm. 80 to 100 ppm in the water column is probably uncomfortable but not lethal to many healthy fish. Unfortunately, there is no way the CO2 level in the tank can go from a nominal of 20 to 30 ppm to above 200 ppm when the hot magnum fail...unless the tank has an exposed surface area of one square inch!

It is possible that the level of CO2 in the water column is already at the dangerous level with water circulation. When the pump fails, then the CO2 level can reach lethal level. This is a common failure mode among noobs (myself included when I first started CO2 injection). Inaccurate pH/KH readings can occur without good/calibrated test equipments.

CO2 death is often the result of a failed regulator/needle valve assembly. Noobs will also tell story about end-of-tank dumping. While it is true that the regulator may not be able to maintain the desired 15 psig at the outlet port when the tank is almost empty, the pressure rise is limited to about 50 to 70 psi, MAX! A quality needle valve with tapered needle and low Cv (flow coefficient) should not have any problem with this small increase in CO2 inlet pressure.

I use a Victor single-stage regulator and a brass Ideal 52 series -1- needle valve with 0.019 Cv (full open). It takes 20 turns to go from full shut-off to full-open. The valve is rated up to 3000 psig. Cheaper valves do not have full shut-off capability. If one cannot afford the $75 price tag, then go with the Fabco NV-55 valve. It's around 15 to 20 bucks with FULL taper needle for a bubble tight shut-off.

So do I run the tank to zero psi? Absolutely. Never had any problem with excessive CO2 level in the water column since 1996.

http://www.idealvalve.com/brassvalves.htm

"Normal" CO2 injection should not cause the pH to vary more than 0.8 point. This is safe for all aquatic inhabitants. However, a drop of 2 points can cause problem for aquatic life unless there exist a period of acclimation.
 
#24 · (Edited)
I'm not here for a fight, just stating that, under the wrong setup conditions (lots of filter induced surface agitation requiring a ton of CO2 injection to maintain 30ppm), it's possible for CO2 to displace enough O2 to cause fish troubles (when the filter craps out and the CO2 continues to get pumped into the now ~still water).

I don't see much to fight over here.

PS. CO2 is heavier than air, so if there is very little air movement in a room and enough of a gap between the top of the water & the top of the tank, you can build up too much CO2 at the surface (especially if you have the setup type I mentioned above). Have you ever seen soap bubbles floating on a cushion of CO2 in an aquarium? I have.
 
#27 ·
I'm not here for a fight, just stating that, under the wrong setup conditions (lots of filter induced surface agitation requiring a ton of CO2 injection to maintain 30ppm), it's possible for CO2 to displace enough O2 to cause fish troubles (when the filter craps out and the CO2 continues to get pumped into the now ~still water).

I don't see much to fight over here.

PS. CO2 is heavier than air, so if there is very little air movement in a room and enough of a gap between the top of the water & the top of the tank, you can build up too much CO2 at the surface (especially if you have the setup type I mentioned above). Have you ever seen soap bubbles floating on a cushion of CO2 in an aquarium? I have.
Possible, but not achievable in the real world unless the tank has a surface area of only a few square inches. If the tank is not overstocked and there is a small opening at the top of the tank for gas exchange, then there will be sufficient O2 for the fish. Again, we're assuming that the initial CO2 level in the tank is 30 ppm.
 
#30 ·
I think we've got the theory covered, now let's try to figure out what happened to Benjavan.

Benjavan, can you describe your CO2 setup? How is your CO2 injected into your tank? Do you have an open top or a lid? Please try to be specific about your CO2 setup.
 
#34 ·
There are many ways to setup a tank. An experienced aquarist will incorporate several safety features to maintain healthy livestock/plant.

1. Add a minimum amount of CO2 to the water to maintain healthy plants. 15 to 20 ppm is adequate for many plant tanks.

2. Leave a 0.5" minimum air gap along the entire length of the tank for surface ventilation.

3. Inject CO2 directly into the pump's intake tube.

4. Position the pump's outlet line within the first two inches below the water surface.

5. Use a single stage regulator and a high quality needle valve. Keep the CO2 lines as short as possible. Relays and other control devices can fail. A well-tuned CO2 rig should operate efficiently at 10 to 15 psig. KISS and you will have a lot more free time to enjoy the hobby.
 
#33 ·
If the CO2 is already at the nominal equilibrium of 20 to 30 ppm with water circulation, then a cut-off in water circulation ALONE cannot raise the CO2 concentration to lethal level unless the tank is sealed from the atmosphere!

No, I'm supposing that the only reason that the CO2 is at 30ppm is because of serious surface agitation (most of the CO2 is driven off by agitation at the surface, very short residence time).

If the pump is inactive, then the CO2 gas will be trapped in the pump. Remember that most needle valves require about 15 psig for proper operation (30 psig with a poorly configured setup). Therefore, the back pressure in the filter is more than adequate to prevent CO2 from escaping into the filter's outlet port.

No. I can blow the water out of my Eheim. There is not enough back pressure there to prevent even a DIY rig from doing the same. If there were enough back pressure, the CO2 would just fill the inlet tube, then bubble into the tank - all at well under 15 psig.

Let's go one step further with the pie-in-the-sky theory that raw CO2 gas will be dumped into the tank via the outlet port. Without the presence of a reactor or equivalent at the oultet port, the CO2 will quickly rise to the surface and dissipate into the atmosphere. Again, anyone with a CO2 injection rig and a small 10 gal tank can verify this statement.

Agreed - I think that feeding the CO2 into a canister filter inlet is a very good idea, provided it does an adequate job of getting the CO2 dissolved into the water. The safest and cheapest route to 30ppm CO2 is to establish water flow w/ just a little surface agitation, then bump CO2 up slowly.
 
#36 ·
I would focus my attention on the needle valve instead of the regulator. It is not necessary to maintain rock steady inlet pressure if the needle valve has sufficient resolution (Cv) to achieve the desire CO2 mass flow rate. A superbly engineered needle valve can handle a wide range of inlet pressure, from several psig to several thousand psig. I would personally avoid any needle valve that's not capable of bubble tight shutoff.

Almost all of the $35 to $40 single stage beer regulator will maintain adequate regulation...from 15 to 40 psig, +/-2. You'll need to spend an arm and a leg if you want to secure a commercial two stage CO2 regulator.
 
#37 · (Edited)
Only one thing I see wrong. A needle valve is a pressure drop device. That's how they work. Any needle valve is influenced by the high side pressure. So if the high side pressure goes up by a factor of 5X you will get 5X more CO2 though the valve.

To prevent this you must have a metering valve. A brass bodied metering valve runs around $55. Stainless steel is around $110 last I checked. A metering valve works totally different than a needle valve and is much less influenced by changes in the high side pressure.

I just checked on the valve that furballi is using and it's actually not a needle valve but a metering valve.
 
#38 · (Edited)
They all use some type of needle pin to control flow of gas/fluid across the orifice, hence the generic name needle valve. A metering valve is a very accurate needle valve with very low flow coefficient, especially during the first several turns. Its graph of flow rate vs number of turns is almost linear from full-close to full-open.

Let's take a look at the specs of the Ideal 52 series -1- with 0.019 Cv. The maximum flow of air at +20 turns is 15 CFH (cubic feet per hour) at 10 psig, and 77 CFH at 100 psig. Since this valve exhibits a near linear plot of flow vs. # of turns, we can safely state that a four-fold increase in inlet pressure (15 psig to 60 psig) will translate to twice the mass flow rate across the orifice at any number of turns. Note that this valve has a Cv value of only 0.0007 at +2 turns.

If you need one bubble per second to achieve 20 ppm of CO2 in the water column, then raising the inlet pressure by 4x will only increase the flow rate to two bubbles per second. This alone will NOT cause a sharp increase in CO2 level in the tank that could result in death to aquatic inhabitants. That's the beauty of selecting a very high quality needle valve with low flow coefficient to dose CO2. It takes a very large change in inlet pressure to significantly affect the mass flow rate across the orifice. Remember that this valve is rated up to 3000 psig! It is a steal at $75.

http://www.idealvalve.com/flowcal.htm

http://www.idealvalve.com/ftp/wwwbrass.pdf

The Fabco NV-55 is another low-cost needle valve with linear flow rate vs number of turns and bubble tight shut-off. It can easily absorb a change of 15 to 30 psig at the inlet side without causing the tank's pH to crash.
 
#39 ·
My regulator is a CONCOA it has a very fine adjustment on it so I dont use a needle valve although, I am rethinking that idea. It is attached to an ehiem bubble counter and then to the air inlet on a 301 Powerhead. Would I then add either a selenoid or a needle valve or both?
 
#40 ·
I would personally configure the rig with these four primary components:

1. Regulator
2. Needle valve
3. One way stop valve (prevent water from the tank from entering he needle valve and regulator)
4. Bubble counter/reactor assembly (inject CO2 to the intake of the filter)

Again, it's best to set the CO2 at the absolute minimum level to achieve healthy plant growth. Also keep all CO2 lines as short as possible. This way, you can run the CO2 24/7, even without water circulation, and not kill the fish with a lethal dose of CO2.
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top