Yes. It's not often done, but using a dP cell instead of a pressure transmitter is required. dP cells may cost a little more than common pressure transmitters.

 

To last MD post,

It is not normally done, if ever, but in theory it will work, you just need to increase the supply pressure to the bubbler to a pressure higher than the anticipated max pressure of the vessel pressure. That's the theory of it, but by doing that you are already making the instrument more sensitive than what is should be to start with.

If the vessel pressure changes the effect it will have on the system is the supply pressure to the bubbler will now become more or less, which will have the effect of making the system more or less sensitive. So the stability of he reading of the bubbler will be affected directly by how constant the vessel pressure can be controlled. So in this case you will have two things that can affect the bubler flow, head pressure as well as vessel pressure changes.

Pressure control in a live process plant is very fast and stable control difficult to achieve, so unless you have some sort of a very stable pressure controlled vessel an installation like this is not recommended.

This is one of those theoretical possible, but impractical issues.

 

For level measurement the pressure over the liquid is going to be your reference pressure. In a closed system, you might have liquid higher than the tank and you would be measuring the level of that as well. You could reference to the pressure at the top of the tank, but complications with where the measurement gas goes etc, etc. make things complicated. You have to really understand what's happening to make this work beyond the simple cases. And the DPs are low and easy to mess up. It's an attractive method where it fits, but there are easier alternatives for the places it doesn't

Regards
cww

 

Actually Dick we use bubblers on closed vessels, all you need to do is connect the LP to the head space. We also use bubblers for interface and density measurement in closed vessel.

In some situations bubblers are the only method we have found to be reliable.

 

>To last Sam post,

> "It is not normally done, if ever, but in theory it will work, you just need to increase the supply pressure to the bubbler to a pressure higher than the anticipated max pressure of the vessel pressure. That's the theory of it, but by doing that you are already making the instrument more sensitive than what is should be to start with" <

I don't understand that statement. The gas flow rate has no effect on the sensitivity, at the end of the day you are measuring the head of liquid you displace

> "If the vessel pressure changes the effect it will have on the system is the supply pressure to the bubbler will now become more or less, which will have the effect of making the system more or less sensitive. So the stability of he reading of the bubbler will be affected directly by how constant the vessel pressure can be controlled. So in this case you will have two things that can affect the bubler flow, head pressure as well as vessel pressure changes.
Pressure control in a live process plant is very fast and stable control difficult to achieve, so unless you have some sort of a very stable pressure controlled vessel an installation like this is not recommended". <

If you follow that reasoning you would never use a DP cell to measure flow

> "This is one of those theoretical possible, but impractical issues". <

Sorry, but as I stated several times we use it all the time on closed vessels

Several posters talked about wasting instrument air, I agree, instrument air is expensive that's why we use Nitrogen, a waste stream in most of the plants we deal with. The bubbler system gives the added advantage of purging the vessel of Oxygen thus minimizing the possibility of explosion.

Dick Caro mentioned that you need to use a DP cell not pressure transmitter, that's a no brainer you need a DP cell to get the low range required anyway. Of course you don't use a bubbler in a vessel with no means of venting overpressure, all vessels need that by code as far as I know.

Re Change in flow with Change of Pressure
Actually if you use the full line pressure rather than drop through a regulator the flow change will be insignificant. I'm not suggesting you use a bubble tube in a steam drum here, only closed vessels where the change in pressure is fairly minor, If your supply pressure for example is 100 psi a 5-10psi change in the vessel pressure will have very little effect.

 

To the last Roy post,

>To last Sam post,

> "It is not normally done, if ever, but in theory it will work, you just need to increase the supply pressure to the bubbler to a pressure higher than the anticipated max pressure of the vessel pressure. That's the theory of it, but by doing that you are already making the instrument more sensitive than what is should be to start with" <

I don't understand that statement. The gas flow rate has no effect on the sensitivity, at the end of the day you are measuring the head of liquid you displace

Roy, as per the manufacturer of bubbler regulators,

THE BEST RATE WILL DEPEND ON THE SIZE OF THE SYSTEM AND THE DESIRED SPEED OF RESPONSE TO LEVEL CHANGES. ORDINARILY, A RATE OF 0.9 SCFH IS CONSIDERED SATISFACTORY, ALTHOUGH THIS MAY HAVE TO BE INCREASED TO OBTAIN THE DESIRED SPEED OF RESPONSE, PARTICULARLY IF THE SYSTEM CONTAINS MUCH VOLUME OR IS SUBJECT TO RAPID PRESSURE OR LEVEL FLUCTUATIONS. THE FLOW RATE SHOULD BE CHECKED TO MAKE SURE IT IS NOT EXCESSIVE, BY NOTING THAT THE RECEIVING METER DOES NOT CHANGE BY AN OBJECTIONABLE AMOUNT WHEN THE FLOW RATE IS DOUBLED.

> "If the vessel pressure changes the effect it will have on the system is the supply pressure to the bubbler will now become more or less, which will have the effect of making the system more or less sensitive. So the stability of he reading of the bubbler will be affected directly by how constant the vessel pressure can be controlled. So in this case you will have two things that can affect the bubler flow, head pressure as well as vessel pressure changes.
Pressure control in a live process plant is very fast and stable control difficult to achieve, so unless you have some sort of a very stable pressure controlled vessel an installation like this is not recommended". <

If you follow that reasoning you would never use a DP cell to measure flow

THE BEST RATE WILL DEPEND ON THE SIZE OF THE SYSTEM AND THE DESIRED SPEED OF RESPONSE TO LEVEL CHANGES. ORDINARILY, A RATE OF 0.9 SCFH IS CONSIDERED SATISFACTORY, ALTHOUGH THIS MAY HAVE TO BE INCREASED TO OBTAIN THE DESIRED SPEED OF RESPONSE, PARTICULARLY IF THE SYSTEM CONTAINS MUCH VOLUME OR IS SUBJECT TO RAPID PRESSURE OR LEVEL FLUCTUATIONS. THE FLOW RATE SHOULD BE CHECKED TO MAKE SURE IT IS NOT EXCESSIVE, BY NOTING THAT THE RECEIVING METER DOES NOT CHANGE BY AN OBJECTIONABLE AMOUNT WHEN THE FLOW RATE IS DOUBLED.

> "This is one of those theoretical possible, but impractical issues". <

Sorry, but as I stated several times we use it all the time on closed vessels

Roy,
PLEASE LET US KNOW WHO MADE THE DECISION TO USE THESE BUBLERS ON YOUR PLANT IN CLOSED VESSELS AND WHY IT WAS DONE. I THINK A LOT OF US WOULD BE INTERESTED TO LEARN WHY THIS DECISION WAS MADE AND WHY YOU BELIEVE THAT THEY ARE THE BEST SOLUTION FOR YOUR APPLICATION SINCE IN 99% OF ALL OTHER LEVEL APPLICATIONS THE BUBLER ARE NOT EVEN CONSIDERED IN A CLOSED VESSEL LEVEL APPLICATION.I WAS GENUINELY NOT AWARE THAT THERE ARE ACTUALLY APPLICATIONS WHERE SOMEONE WOULD CONSIDER AN BUBLER INSTALLATION IN A CLOSED VESSEL AND IF YOU HAD NOT MENTION IT THAT YOU HAVE DONE IT ON SEVERAL OF YOUR VESSELS I WOULD HAVE THOUGHT THE PERSON WHO EVEN CONSIDER THIS IS, TO PUT IT MILDLY, NUTS.(THE MOD OF THIS FORUM WILL NOT ALLOW ME TO SAY EXACTLY WHAT I FEEL SO THIS IS THE BEST I CAN DO)

Several posters talked about wasting instrument air, I agree, instrument air is expensive that's why we use Nitrogen, a waste stream in most of the plants we deal with. The bubbler system gives the added advantage of purging the vessel of Oxygen thus minimizing the possibility of explosion.

Dick Caro mentioned that you need to use a DP cell not pressure transmitter, that's a no brainer you need a DP cell to get the low range required anyway. Of course you don't use a bubbler in a vessel with no means of venting overpressure, all vessels need that by code as far as I know.

Re Change in flow with Change of Pressure
Actually if you use the full line pressure rather than drop through a regulator the flow change will be insignificant. I'm not suggesting you use a bubble tube in a steam drum here, only closed vessels where the change in pressure is fairly minor, If your supply pressure for example is 100 psi a 5-10psi change in the vessel pressure will have very little effect.

 

We use a bubbler system to measure differential pressure across a traveling screen, the medium we are measuring is river water. So the high side of the system is actually river level, the low side being the downstream side of the screen and also the suction side of pumps. The pumps are several hundred feet away, but this area can be rather turbulent. The problem I have is that currently on the high side of our instrument, the river level is 4.2 feet above the high side bubbler tube, so effectively 1.82 PSI of head. we were using 38 psi air supply, I reduced it to 15 psi, which should be more than enough to defeat the max head pressure of approx 10.5 psi when the river is at its highest level. The low side today was 2.8 feet above the bubbler tube, so 1.2 PSI of head. If I set the flow indication by the rotometer to 2.0 SCFH or above the differential would actually would go negative, indicating the pump side of the screen was higher than the river level, not the case, I measured the water levels. If I adjusted the flow to .5 SCFH, the indication would match my measured results. If flow does not matter, why did I see the results I saw today? Another item to note is that both the high and low sides are 1/4" SS tubing with approx 15 90 degree bends in approx 70 feet of pipe. Help me if you can, I am trying to understand why the instrument read a negative differential when I increase the flow to the bubbler. The instrument is a Photohelic differential pressure switch.

 

1/4" tubing is much to small for the distances with at 2 scfh with several inches wc pressure losses. In raw water, you have to be concerned about algae and other types of build up at the bubbler tip end.

Ideally, you should have at least 3/4" bubbler piping in the water with properly cut ends for steady bubbling. In some cases stilling wells are also used. The 1/4" air lines are fine, but like you pointed out, flow regulation is needed for reliable readings.

 

> you have to be concerned about algae and other types of build up at the bubbler tip end. <

We do on occasion blow out these lines, in the past they have been clogged bad enough that people have taken a small bottle of nitrogen (1500 psi) to clear out the line. I cannot find a straight forward answer on how to design a functional bubbling system. Here are some of my assumptions: have the least about of line loss after the bubbler air supply, have enough air pressure to overcome the highest possible head pressure. Set the flow rate to 60 bubbles per minute...This will be hard to do since our lines are in the river. Original design from the 60's had 1/2" pipe (1/2" ID). At some point someone decided the 1/4" could be OK. I can have assumptions, but the place I work at wants facts before we put any $ into making this system that "has worked for years" a better operating system that will require less maintenance.

> Ideally, you should have at least 3/4" bubbler piping in the water with properly cut ends for steady bubbling. In some cases stilling wells are also used. The 1/4" air lines are fine, but like you pointed out, flow regulation is needed for reliable readings. <

Hard to maintain a steady flow with 70 feet of 1/4" tubing with 17 90 degree bends (right?) Or at least hard to detect changes in back pressure with this installation.

 

I really enjoyed this discussion and am looking to install a basic bubble system for a vertical turbine pump install that is coming up later this year and I am over using pressure transducer and need to know the static in the well and as well the pumping level. The issue is that it looks like Siemans is the answer. My range or length of the 1/4 stainless tubing will be 300 feet, which will be a continuous roll no fitting, attaching to the end near the water lube pump. Actually 10 feet above the pump. Static is about 100 feet and the pumping level about 235. Depending time of year and how many other pumps may be running in the area. I would like to hear any thoughts that may want me to change my mind on this basic operation. I do get it.

justin-

 

Great posts on bubblers here.

I have always used the diff regulator and rotameter with needle valve for flow adjust

I just saw my first application where they don't have diff reg, only rotameter. I am curious as to what the results are to be.

Engineer says that they are constantly adjusting the rotameter flow after they empty/fill the large tanks

System is currently offline so i cannot observe.
We may add the diffs, anyone seen what the effect of no constant pressure, it seems obvious that it would not be accurate.

Also as for a bubbler on a sealed tank, sounds crazy!
I have observed a pressure tank spray out a bolted manhole seal when a bubbler was purged and that wasnt even a closed tank be careful or things will blowup, please do not correct me here because i know it is low p but still

If some here are considering the standard diff reg, rotameter and xmittter, remember that air quality is critical, poor air and environmental conditions will ravage your assembly and result in unreliable system