[jccaclimber]

Experimenting with my new CO2 setup (used Milwaukee regulator/needle valve/solenoid/bubble counter) I decided to look into EOTD a bit. From what I've read here, there is a ~3 psi increase towards the end, and I was put under the impression that it did not depend on pressure. My thought was that 3 psi is a small change to a 50 psi system, while a large change to a 10 psi system. My discovery is that the pressure change is not constant The higher in the range I regulate at, the higher the pressure goes. In my case I regulated to 10-15 psi and had a 3 psi increase, then regulated at 70 psi and had a 10-15 psi increase.
My solution was to wire an a low pressure regulator downstream, which holds a nice constant pressure. More on that in another thread.
For my EOTD test I got the system pressurized, set the needle valve a couple bubbles/sec, and closed the tank. The two gauges shown in the video are input and output from the regulator showing the output pressure rise with the input pressure drop. Downstream is a low pressure regulator holding a constant 20 psi out. Note, the gauge labelled "Tank Pressure" is actually the output pressure as I have replaced that gauge.

Edit: I can't figure out how to post a video here, so I've put it up on youtube.
If anyone knows how I can post it directly, please PM me.
http://www.youtube.com/watch?v=M08hPs-J3SM

 

[Hoppy]

By adding the low pressure regulator downstream of the Milwaukee regulator, you converted it into a two stage regulator. This is a very good video to see. And, I hope it puts this issue to rest. Rex Grigg used to sell a low pressure regulator for this use, but I don't recall him ever giving a good explanation about why it was needed. I think he probably discovered just what you did, and that was why he recommended that extra regulator.

 

[jccaclimber]

Rex's explanation made me think that he was using it more like a needle valve rather than a regulator. It's possible that I read it wrong though as everything else he wrote makes sense.
A cheap 1 stage regulator from a beverage supply place is ~$50. A cheap low pressure regulator is $30, and the plumbing to fix things is <$20. Point being, for under $100 you have a low flow two stage regulator. You won't see this in industry because 1) it takes more space and 2) nearly all regulators need far more flow than a CO2 injector into an aquarium.

 

[Robert H]

To avoid end of tank dump all you need to do is add a check valve inline. It is such an easy no brainer solution I do not understand why this subject even gets brought up

 

[jccaclimber]

How do you figure adding an inline check valve solves the issue? Wouldn't you still get a change in the pressure output, and therefore bubble count?
Edit: If a check valve did solve this, what would be the point of dual stage regulators?

 

[Hoppy]

A check valve installed anywhere will not stop this problem. It isn't a real problem if you make a habit of checking your CO2 system daily, and refilling the CO2 bottle as soon as the pressure starts to drop, or readjusting the needle valve about twice a day as the pressure drops. Some of us, like me, are just too lazy to do that consistently.

 

[jccaclimber]

I'm of the opinion that checking something under my tank more often than I pull out a hose for a water change isn't a reasonable thing. I'm glad you agree that a check valve (and I presume needle valve) really doesn't stop EOTD. Now if only I could figure out where I'm leaking from. I've got a small leak somewhere before my solenoid, and my CGA 320 fitting is really tight.

 

[jccaclimber]

Edit, found my leak, it was at a fitting between the regulator and solenoid. It was one of the few fittings I didn't do myself, and somehow I missed the fact that it wasn't even hand tight. Leak testing again, but it seems to be holding pressure now. Also, what is a reasonable overnight pressure loss with the tank closed and the solenoid off? I'm aiming for zero, but not sure what is reasonable.

 

[Bettatail]

Rex's explanation made me think that he was using it more like a needle valve rather than a regulator. It's possible that I read it wrong though as everything else he wrote makes sense.
A cheap 1 stage regulator from a beverage supply place is ~$50. A cheap low pressure regulator is $30, and the plumbing to fix things is <$20. Point being, for under $100 you have a low flow two stage regulator. You won't see this in industry because 1) it takes more space and 2) nearly all regulators need far more flow than a CO2 injector into an aquarium.

Get a double stage in the first place, may cost less.

 

[Bettatail]

Edit, found my leak, it was at a fitting between the regulator and solenoid. It was one of the few fittings I didn't do myself, and somehow I missed the fact that it wasn't even hand tight. Leak testing again, but it seems to be holding pressure now. Also, what is a reasonable overnight pressure loss with the tank closed and the solenoid off? I'm aiming for zero, but not sure what is reasonable.

leak is the major problem if you don't see consistent bubble rate, or no bubble when you adjust the needle valve.

pressure can hold for weeks, or months, if there is no leak. I have a double stage regulator that locks the 50 psi co2 in second stage chamber for two month now.

 

[OverStocked]

To avoid end of tank dump all you need to do is add a check valve inline. It is such an easy no brainer solution I do not understand why this subject even gets brought up

I consider myself to be a pretty smart guy.... but apparently by brain isn't able to compute this?

How in the world does a check valve effect this?


I think the best thing you can do on a single stage reg, is to run in the 35-40 psi range. You seem to get much less rate wander over time. Dual stage regs on ebay are cheap though. Prob less than the 50 dollar reg from a bev site.

 

[Bettatail]

have some low pressure regulators available, if anyone interests, pm me, thinking about giving away for free, have no use of them...
Most of them are SMC(One of the leading players of Automation industry) low pressure regulators.

 

[jccaclimber]

leak is the major problem if you don't see consistent bubble rate, or no bubble when you adjust the needle valve.

pressure can hold for weeks, or months, if there is no leak. I have a double stage regulator that locks the 50 psi co2 in second stage chamber for two month now.

I'm talking pressure disconnected from the tank as I'm leak checking before hooking it in to the tank. I pressurized it, then closed the system. It held ~1100 PSI at the tank end with no noticeable change overnight, whereas before it was going to 0 within an hour. If my tank had been on my bubble rate would have been constant, but my tank would have drained out.

I think the best thing you can do on a single stage reg, is to run in the 35-40 psi range. You seem to get much less rate wander over time. Dual stage regs on ebay are cheap though. Prob less than the 50 dollar reg from a bev site.

This comment is part of why I posted this. On your regulator that may be true, but it isn't on mine. I see a lot of postings that imply "My regulator holds pressure x =/- 3psi". In my case, it is "my regulator holds pressure x =/-20%" Mind you that 30% is from full (1050-1100 psi) to empty (70 psi tank side).

Get a double stage in the first place, may cost less.

Apparently I looked in the wrong places. I found lots of dual gauge regulators, but not dual stage, I'll have to remember that if I ever set up another one of these.

have some low pressure regulators available, if anyone interests, pm me, thinking about giving away for free, have no use of them...
Most of them are SMC(One of the leading players of Automation industry) low pressure regulators.

I might just get you on that, my air tank is now short a regulator body.

 

[jerry1]

I consider myself to be a pretty smart guy.... but apparently by brain isn't able to compute this?

How in the world does a check valve effect this?


I think the best thing you can do on a single stage reg, is to run in the 35-40 psi range. You seem to get much less rate wander over time. Dual stage regs on ebay are cheap though. Prob less than the 50 dollar reg from a bev site.

My guess is that he thinks EOT dump means the water flows back down the air line when CO2 runs out?

 

[fresh.salty]

I watched the video and don't really understand what the problem is. Seems like the time line of the pressure change is so short that it wouldn't really matter? I realize that for this test you simulated an empty tank by shutting it off at the cylinder. But by doing that we can't see how long the real life event would last. I have a dual stage regulator already so it's not a biased post.

 

[JasonG75]

A check valve installed anywhere will not stop this problem. It isn't a real problem if you make a habit of checking your CO2 system daily, and refilling the CO2 bottle as soon as the pressure starts to drop, or readjusting the needle valve about twice a day as the pressure drops. Some of us, like me, are just too lazy to do that consistently.

I never mess with mine and have never had this "EOTD". I never look at the gauges, I know when I am out when there are no more bubbles.

 

[jccaclimber]

I've gotten the impression (haven't tested myself) that running an entire 20 lb tank down would take a while (days), while this was a much smaller volume. I also had the needle valve very open to keep the video short. While I'm not sure what the change in bubble rate is, I assume it would be significant with a 20% rise in pressure. Based on the fact that Tom has mentioned running CO2 at 60 ppm, and most of us aim for 30, I imagine the fish would live. However, for anyone running closer to the edge, or more likely people with an inaccurate reading, this could be helpful.

 

[houseofcards]

I never mess with mine and have never had this "EOTD". I never look at the gauges, I know when I am out when there are no more bubbles.

+1

This is just such a rare occurrence that I really hope it doesn't set an alarmist tone with people. There are much more likely ways to gas your fish. A few off the top of my head that happen much more frequently would be:

Running co2 at night
Not using a drop checker or other way to measure co2.
Cleaning a ceramic diffuser on a small tank.