[Chlorophile]

I would love it if someone could explain the details behind it - scientifically if possible, as I can make plenty of assumptions myself.
I've heard that the flow through an orifice is dependent on the pressure on both sides, but in my mind it seems like if tank pressure is low you would get a reduction in co2 output, not an increase >_<
Is the issue with the regulator or the needle valve? It would make more sense if it has something to do with the spring mechanism or the valve orifice or... something.
Please help!

Also, if your co2 regulator has say a max pressure of 50 psi, why not just turn it up to its max pressure, wouldn't this limit the working pressure from going up any further?
My pressure release valve starts to go around 47-48 psi so I run my regulator at 45 psi in hopes that I can prevent any increases in my bubble count.

 

[astonstreet]

I don't know the scientific explanation for this and I am keen to see someone come along and explain.

My assumption is that it happens when all the liquid in the bottle is lost and all that is left is gas. I think the problem there is not that there is too much pressure but not enough pressure for the regulator to maintain and the gas escapes.

 

[2in10]

I am not certain but it sounds like it has to do with the CO2 being in liquid form under pressure and when the pressure reaches a tipping point the remaining liquid reverts to gaseous form and that may be slamming the valves and causing the dump.

 

[DarkCobra]

The issue is primarily with the common single-stage regulators.

This provides regulator diagrams, and documents that the output pressure increases as input pressure decreases:

http://www.aquaticquotient.com/forum/showthread.php/77447-co2-regulator

Though it doesn't really explain why it does so, or why a single-stage regulator stops regulating entirely when input pressure drops below a certain threshold. I think I can see why by looking at three way interaction between forces on the spring, diaphragm, and poppet; but if anyone else knows for sure, explanation is better left to them.

However, I can explain what happens in the needle valve very well.

Needle valves aren't designed to regulate to a specific pressure, but to a specific flow rate, as long as the input/output pressure differential remains the same. If that differential changes, so does the flow, in a linear fashion.

But for our tanks we desire both very slow flow rates and good adjustability/stability at those rates; which requires a very thin and finely tapered needle. Hit that delicate component with higher than the acceptable pressure differential and it will begin to flex, and usually oscillate (audibly!), a non-linear effect that lets through a lot more than the pressure differential would dictate. EOTD pressure is enough to make that happen for some needle valves.

 

[Chlorophile]

I had a very long post I was going to make - but as I was proof-reading it I discovered that my theory was completely wrong...
There might be no point in a 2 stage regulator, and a single stage regulator might only cause EOTD if there is no resistance on the outlet, such as a lack of a needle valve, or going into a reactor or something...
I will post in a bit once I've figured it out... now wheres my thinking cap!

 

[Chlorophile]

Single Stage Regulator
In a single stage regulator pressure is regulated by the pressure of the gas that has already gone past the poppet and not by tank pressure.
The entire design of a pressure regulator is to maintain a specific pressure by the action that any increase or decrease in the pressure you are trying to maintain will cause the opposite to occur, keeping everything stable.
As your working pressure increases the diaphragm is lifted pulling the poppet up and restricting the orifice, preventing working pressure from increasing any more.
If the working pressure starts to decrease the diaphragm will have less force acting against it and the orifice will open, letting in more gas and keeping the pressure stable.
Any decrease in downstream pressure/resistance, such as by opening your needle valve more, will cause the orifice to open more, again keeping the pressure stable while increasing flow rate.

Dual Stage Regulator
The Dual Stage Regulator is really no different.
Pressure in the first stage is regulated by pressure in the first stage, if pressure decreases, the factory set spring(in some it is adjustable with a precision nut) will push the poppet and increase the orifice size, regulating pressure in this stage.
The main difference is that the pressure in the first stage is regulated by the pressure in the second stage, and not by what is downstream/restricting the outlet.
When something downstream increases the flowrate requested, the second stage pressure decreases which opens the orifice making first stage pressure drop... which opens the orifice.

Conclusion:
What this means is simply that there will be more fine control with a dual stage regulator, the pressure between the first and second stage is going to be pretty constant, so as flow rate requested changes the pressure will be more stable.
In a single stage regulator if the flow rate changes there will be greater fluctuations in the pressure.
Albeit they might not last for even an entire second but in a medical application it might be very important to have a constant 9.8 psi with a largely variable flow rate.

These kinds of variations are not a problem to us, as they only occur for a brief second after you adjust your needle valve.
When you open the needle valve the orifice opens briefly before it returns working pressure to its original value, causing a brief increase in bubble count before it balances out again.
Go ahead and try it.
Open your needle valve, this we can all conclude causes a brief decrease in working pressure as more gas is being let through the needle valve, diaphragm now has less pressure acting on it, orifice in the regulator opens wider, brief surge of pressure as everything balances out, temporary(.5-1.0 seconds) increase in bubble count, and then you see your stable bubble count.


So what can really cause EOTD?
My guess is that it is due to bad springs in cheap regulators.
The spring plays a large role in keeping things balanced and if it seizes or is old, possibly rusty, then there is not much regulating going on.
I thought it possible that if inlet pressure is less than working pressure, something odd could happen, but working pressure is whats regulating the orifice so it really shouldn't be possible for tank pressure to affect it in any way.

So how can we avoid EOTD?
I say other than buying quality regulators (I don't think we need dual stage regulators, although they are higher quality typically)
Occasionally check your regulator, see if the knob turns freely.
Also if your regulator has a pressure release valve, run the regulator as close to the pressure that opens said valve as you can - if pressure somehow increases the valve will open.
My regulator is set at 45 PSI working pressure, and my valve opens at 48.
Downside is that there may be leaks that you didn't know were there when running these pressures, but go ahead and check!


Critiques? Concerns? Corrections?:icon_surp
I wrote this, based on assumptions and observations, so if someone has info onto the more scientific aspects of it that might explain why we need a dual stage regulator then please, please chime in.

 

[DarkCobra]

As long as we've opened it up to speculation, here's my take.

The combination of opposing forces on the diaphragm (spring on one side, output pressure on the other) is what controls the poppet valve. That alone is fairly linear, and not prone to causing an EOTD.

But there is a third force. When the poppet valve is closed, the pressure from the tank also pushes directly on the poppet, tending to hold it closed. At a pressure differential of 800psi nominal to 10-50psi, that's quite a lot of force directly on the poppet.

The spring must exert extra force to overcome that. But as tank pressure drops, the extra force becomes increasingly becomes too much; resulting in higher output pressure. And at some point, the poppet will just stay open, resulting in EOTD.

In a dual stage regulator, the first stage reduces the tank pressure to 50-100psi, then the second stage reduces pressure to the final desired amount. As a result, the possible range of pressure differentials directly on the second stage poppet is greatly reduced, resulting in better linearity; and the second stage cannot dump even if the first stage does.

I have actually had my CO2 system dump twice. In both cases, the tanks CO2 level did increase, but not to the point of harm; because the diffusion systems I was using were simply incapable of diffusing most of the dump.

 

[Bettatail]

OUT PUT PRESSURE RISE

every single stage regulator has output pressure rise.
As in picture.

when the Force of spring one, Force of sping two are constant, the only variables are force of the pressure/area1 and force of pressure/area2.

If the input pressure lower(end of co2), the Force of pressure(input)/area2 lower also, which means the force of pressure(output)/area1 increase, basic physics, both forces are same direction.

area1 and area2 are constant also, so input pressure decrease, output pressure increase.

there is no strict definition of End of Tank Dump, please refer to Output pressure rise.

excess CO2 injection is a problem if the excess co2 volume is big enough to cause trouble for your fish tank, but if you set the output pressure of the single stage regulator higher, the output pressure rise is a fairly small number compare to the original regulator output setting, so EOTD can be avoided.

 

[OverStocked]

I think EOTD is like the Yetti. Lots of people talk about it, but basically 3 people have seen it.

 

[Bettatail]

I think EOTD is like the Yetti. Lots of people talk about it, but basically 3 people have seen it.

that is right, for single stage regulator, the output pressure rise is always exist, but EOTD, which we refer to as excess co2 that cause trouble for our fish tank, is supper rare.

0.1psi output pressure rise for every 100 psi input pressure drop, is 0.8psi total rise when a co2 tank go empty, if you set your regulator output pressure at 5 psi, that 0.8psi rise is a large number compare to the 5 psi output setting, so the excess co2 is a large number also.
but if the regulator output pressure set at 30 psi, that 0.8 psi rise will cause excess co2 injection but is a really small volume compare to the setting volume.


the 0.1psi output pressure rise for every 100 pis input pressure drop refers to the Airproducts and Victor single stage regulators.
avoid low quality single stage regulator, some of the regulators have output pressure rise as high as 5psi rise per 100psi drop.

 

[mott]

I think EOTD is like the Yetti. Lots of people talk about it, but basically 3 people have seen it.

My ma957 will dump every time unless I check it and adjust as soon as the tank starts dropping.

 

[Hoppy]

Isn't it fun how we debate this issue about every 2 months?

The problem with "end of tank dump" is that it isn't a dump. The phrase brings to mind a sudden rush of all of the remaining CO2 in the tank in a few seconds. That never happens. What does happen is that some of us run or ran our CO2 right up close to the maximum that the fish could live with, which was not a problem until the tank pressure started dropping after the liquid CO2 was used up. And, one characteristic of cheap single stage regulators is that they are not perfect. They produce a higher output pressure as the input pressure drops, and vice versa. When you have the bubble rate set with the regulator outlet pressure at 20 psi, to give you almost enough CO2 to start the fish dieing, you have a serious problem when that outlet pressure goes to 25 psi. That increases the bubble rate 25%, more than enough to exceed the maximum CO2 that fish can live with. Murphy's Law would cause most of these incidents to happen when no one is watching the tank. The tank owner walks in the room, sees his fish floating belly up, notices that the CO2 tank is empty or nearly so, and says "end of tank dump".

 

[Chlorophile]

Okay, interesting... I don't want to quote everyones posts, even though multiquote makes to very easy so I'll just address it as if we didn't have that technology.
First of all let me say I agree that EOTD is the wrong phrase, I will no longer use that phrase.

Hoppy - I see what you mean by running co2 to the near limit. A subtle variation in output pressure could easily push you over the edge killing your fish.
Does a dual stage regulator really prevent subtle variations in output pressure?
My understanding of pressure regulators as of today is that they keep the same pressure, but at any flow rate requested - EVEN dual stage regulators...
Doesn't that put the blame on the needle valve? Or perhaps on the person running co2 at the near limit?

Bettatail- I wasn't able to understand all of your english fully, but from what I gather I have just a couple questions.
You mention that Airproducts and Victor regulators have a .1 psi rise for every 100 psi decrease in inlet pressure.
Do you know what causes this to be different than any other regulator?
I still believe the springs have more to do with this than the inlet pressure, and I believe it has a lot to do with the targeted range the regulator was designed to work in.



DarkCobra - The poppet valve shouldn't be closed really, at any point, if it is closed pressure on it would just keep it closed. I don't think pressure on the poppet plays a large role, but what role I think it plays I will explain below.


Thanks to bettatail's diagram I see that pressure on the poppet would be taken into account when you first set your working pressure.
When the inlet pressure decreases there is less force holding the poppet closed, and the force in the regulated area is going to be stronger in comparison, and this will open it more - but by opening it more, doesn't the pressure in the regulated area increase?
If the pressure in the regulated area increases it should start to close the poppet again, even though there is less pressure on the tank side holding it closed.
I'm assuming this is the ratio of psi rise for tank psi decrease you were talking about.
This is logical.

But what about a dual stage regulator prevents all of that from happening?
The physics are the same, except now its in duplicate.
The only difference I can see is that the pressure is now being dealt with twice, but there are still all of the same issues to deal with.
Perhaps these still have a output pressure rise, but now the ratio of pressure rise to tank pressure decrease is even smaller.

I read a few PDF's from swagelok and parker, and one pdf about diving regulators and I couldn't wrap my brain around how there could be a pressure variation based on inlet pressure, since outlet pressure is what regulated the diaphragm!!!
I think I get it a bit better now, so thanks guys.

My apologies if, as Hoppy said, this is debated every 2 months.
I had questions that weren't being answered by reading other peoples posts on the matter.

 

[Chlorophile]

And hear I thought the only thing variable would be downstream FLOW RATES, and not pressure.
I guess the pressure really does change... which is odd because regulators are designed to prevent that, and they ARE designed to allow the flow rate to increase or decrease while maintaining constant pressure.

 

[Hoppy]

And hear I thought the only thing variable would be downstream FLOW RATES, and not pressure.
I guess the pressure really does change... which is odd because regulators are designed to prevent that, and they ARE designed to allow the flow rate to increase or decrease while maintaining constant pressure.

No mechanical device is ever perfect. Actually, no electronic device is perfect either. Regulators regulate the outlet pressure at a constant value with varying flow rate and with variations in inlet pressure. But, how well they do that job depends on how well they are designed. Really cheap regulators can hardly be expected to do the job as well as expensive regulators. Good quality regulators cost over $100 new, while typical aquarium regulators cost around $30 (without the needle valve, solenoid valve, and bubble counter.) Two stage regulators are used only because they do that job better, over a wider range of flow rates and inlet pressure variations than one stage regulators do it. Otherwise there would be no reason to waste money making a two stage instead of one stage regulator.

CO2 is a unique gas because it changes to a liquid at about 800 psi at normal room temperature. This causes the CO2 tank pressure to remain constant until all of the liquid CO2 has been used up, and only gas is left in the tank, allowing the pressure to drop below 800 psi (approximately). Because of that, a cheap regulator works great until the liquid is used up. Our flow rates are extremely low too, and don't vary much at all, making cheap regulators work even better. But, once the CO2 tank contains only gas, the pressure in the tank drops as CO2 is used up, and the cheap regulator no longer works great.

 

[DarkCobra]

DarkCobra - The poppet valve shouldn't be closed really, at any point, if it is closed pressure on it would just keep it closed. I don't think pressure on the poppet plays a large role, but what role I think it plays I will explain below.

The poppet valve definitely closes.

Imagine if you have a solenoid after the regulator, which is closed. If the poppet valve couldn't close completely as you propose, then the low pressure side of the regulator would eventually rise to 800psi in this situation; as the gas would continue flowing without anywhere to go. But that doesn't happen.

 

[btimmer92]

Chlorophile, I think I can answer your question of why 2 stages work better than one.

1st of all, 2 stage regulators maintain a constant output pressure. Single stage regulators DO NOT.

The reason for rise in output pressure on a single stage, is because when the pressure gets too low for the 1st stage seat to handle, it doesn't regulate it very well. On a 2 stage, this problem is solved because if the pressure rises after the 1 stage, it has the 2nd stage to handle the pressure change, and it will pretty much always be in range to handle at that point. Think of it like steps. Can you climb a 14 ft wall? probably not, can you climb two consecutive 7 ft walls? If you are in shape and normal height, then yes.

BTW, the poppet closes, that doesn't mean it stays closed. The Large spring in the regulator? yah that opens back up the poppet.

The spring will always open it back up. once the air after the poppet de-pressurizes, and the there is less pressure on the diaphragm, the opposing spring force on the diaphragm will open the poppet back up.

There are two opposing forces on the diaphragm: air pressure force, and spring force. They always equalize, due to the poppet connected to the diaphragm. The poppet opens and closes to let more air in to pressurize the output compartment. And this is why adjusting the spring adjusts the pressure.

What the poppet actually does is, once the output compartment comes to correct pressure, the poppet will open just enough so that is has the same flow as the flow that is going out of the regulator (or in our case out of the needle valve) This way, the same amount of air is going in as is coming out, and the output compartment pressure is maintained, and therefore the position of the poppet is also maintained. Fascinating how that works.

 

[bsmith]

This topic has been beaten to death but I will post what I always post in these threads.

A SSR isn't designed to handle constant outflow pressures. It just can't because when
Tank pressure drops (liquid co2 is gone and pressure drops below ~800psi) it cannot maintain the output pressure you set.

A DSR is normally used in medical environments and can cost well over $400 brand new. That's why everybody I know purchased one used. They have two regulating bodies that step down the pressure in order to keep the output you set no matter what the tank pressure is.

Often times when you get a freshly filled co2 tank the tank pressure reads 900-950psi. Now if you hook up your SSR on that cold tank, as it warms up to room temp your bubble rate will raise too.

A needle valve no matter what price will not help in an EOTD scenario again because it wasn't designed to do so. All a needle valve does is control outflow in relation to a specific input pressure. So say you have a working pressure (the pressure that the gas is leaving your regulator) of 15psi and that gives you 4 bubbles per second. If you have a working jump up to 60psi then accordingly your bubble rate will jump to 24bubbles per second. If you are like me and have your co2 rate set just a hair below causing stress in your fauna (as hoppy brought up) a jump like that can easily wipe out all the fish in the tank.

DSR is all I'll ever use. No fiddling with the NV to compensate for a cold new tank and no risk of pressure increase on a tank that has depleted enough co2 to be running on gas either.

 

[Chlorophile]

The poppet valve definitely closes.

Imagine if you have a solenoid after the regulator, which is closed. If the poppet valve couldn't close completely as you propose, then the low pressure side of the regulator would eventually rise to 800psi in this situation; as the gas would continue flowing without anywhere to go. But that doesn't happen.

Okay.. I suppose if working pressure is what has the ability to close the poppet, it would open again as the gas exited the system.
I'm dumb... haha i must have been thinking of tank pressure, because if tank pressure closed the poppet against the springs force, there would be nothing short of adjusting the spring/knob that would open it again

Chlorophile, I think I can answer your question of why 2 stages work better than one.

1st of all, 2 stage regulators maintain a constant output pressure. Single stage regulators DO NOT.

The reason for rise in output pressure on a single stage, is because when the pressure gets too low for the 1st stage seat to handle, it doesn't regulate it very well. On a 2 stage, this problem is solved because if the pressure rises after the 1 stage, it has the 2nd stage to handle the pressure change, and it will pretty much always be in range to handle at that point. Think of it like steps. Can you climb a 14 ft wall? probably not, can you climb two consecutive 7 ft walls? If you are in shape and normal height, then yes.

BTW, the poppet closes, that doesn't mean it stays closed. The Large spring in the regulator? yah that opens back up the poppet.

The spring will always open it back up. once the air after the poppet de-pressurizes, and the there is less pressure on the diaphragm, the opposing spring force on the diaphragm will open the poppet back up.

There are two opposing forces on the diaphragm: air pressure force, and spring force. They always equalize, due to the poppet connected to the diaphragm. The poppet opens and closes to let more air in to pressurize the output compartment. And this is why adjusting the spring adjusts the pressure.

What the poppet actually does is, once the output compartment comes to correct pressure, the poppet will open just enough so that is has the same flow as the flow that is going out of the regulator (or in our case out of the needle valve) This way, the same amount of air is going in as is coming out, and the output compartment pressure is maintained, and therefore the position of the poppet is also maintained. Fascinating how that works.

Your last paragraph makes perfect sense.
The pressure should, by design purposes, stay steady at any flow rate, or atleast I've seen nothing to prove otherwise.

I assumed the poppet would never completely close, I figured it would just close enough that the working pressure was greater than the pressure that could be let through the poppets open area.
My understanding of the forces at work would indicate that if the poppet was able to be closed by pressure alone, with the spring at a constant point, then the pressure would indefinitely keep it closed, the spring would not just become stronger all of the sudden to open the poppet.

I get though now, that the 2 stage is more steady because it ramps down pressure in 2 stages, but thats really it, no other mechanical or physical properties make it any better.
If stage one can lower the pressure from between 100-300psi for example, then the second stage will have a much lower range to deal with, that alone just means there will be less variation over the varied tank pressures.




I'm still not convinced a 2 stage regulator is worth it.
Thankfully I can see the benefits, but I've read countless excerpts by people who frequently run their tanks with single stage regulators to empty.
Some of whom may be running co2 at near the fish-safe limit.
Some of whom may not be.

But if you can get one for cheap on ebay, why not.

I got a cheap beveragefactory.com regulator
that I run at 45 psi working pressure.
I intend to run it to complete empty and I'll let you know if my tiger shrimp and oto's die, or if my bubble count raises from my easy to detect 1 per second, or if my working pressure gauge varies.

I recently ran my tank to the low 500psi range, and I saw no change in bubble count... but you have all indicated that the bubble count should only change when... uh actually I'm not sure, its been pretty nondescript - seems like bubble count changes whenever your fish die :icon_cool

 

[Chlorophile]

This topic has been beaten to death but I will post what I always post in these threads.

A SSR isn't designed to handle constant outflow pressures. It just can't because when
Tank pressure drops (liquid co2 is gone and pressure drops below ~800psi) it cannot maintain the output pressure you set.

A DSR is normally used in medical environments and can cost well over $400 brand new. That's why everybody I know purchased one used. They have two regulating bodies that step down the pressure in order to keep the output you set no matter what the tank pressure is.

Often times when you get a freshly filled co2 tank the tank pressure reads 900-950psi. Now if you hook up your SSR on that cold tank, as it warms up to room temp your bubble rate will raise too.

A needle valve no matter what price will not help in an EOTD scenario again because it wasn't designed to do so. All a needle valve does is control outflow in relation to a specific input pressure. So say you have a working pressure (the pressure that the gas is leaving your regulator) of 15psi and that gives you 4 bubbles per second. If you have a working jump up to 60psi then accordingly your bubble rate will jump to 24bubbles per second. If you are like me and have your co2 rate set just a hair below causing stress in your fauna (as hoppy brought up) a jump like that can easily wipe out all the fish in the tank.

DSR is all I'll ever use. No fiddling with the NV to compensate for a cold new tank and no risk of pressure increase on a tank that has depleted enough co2 to be running on gas either.

I've heard this all a thousand times...
I'm not asking what they are supposedly not designed to do.
I'm asking why people think they aren't designed to do so.
I've looked into this for three days and there is still no physics that indicates that a pressure regulator is NOT designed to maintain a constant pressure.
Even very detailed articles from parker and emerson have stated very clearly that a pressure regulator, no matter how many stages, can maintain pressure within 1psi at a very large variation of flow rates, and the pressure variations only occur for a brief period of time before they balance out.

Dual stage regs are typically used in some issue where a constant VERY specific pressure(maybe life support systems)is required, where very variable flow rates are common(breathing in/breathing out), and if something goes wrong you might rupture a lung, etc.



The information I've looked into as of late has used two main terms.
Downstream Pressure - something that has been described as constant within a reasonable range.
Downstream Flowrate - something that is as variable as need be, as per the design of a single or dual stage regulator.
Does a variation of .1 to .9 psi really not balance out?
The physics indicate that it should unless there is a rusty or cruddy spring involved.
The whole design of regulators is that more pressure causes a decrease in pressure and visa versa.
Thanks again.