Ball valves

[monkey spanners]

Been thinking again Ball valve in liquid line, chap comes along and thinks i'll change the drier, shuts the valve. Fine so far. we've now got the centre of the ball full of liquid, with no expansion room, any body else see a problem with this or am i missing something? Could this be why the stem seals go?

Cheers Jon

[US Iceman]

Been thinking again

That's a dangerous habit.

Seriously though, this is something that should occur more frequently. And, your question about ball valves is a very good one.

Any ball valve used for the purpose of isolating ANY cold liquid, needs to have a vented ball. All the manufacturer has to do is drill a small hole in the ball to vent the cavity when the valve is closed.

It does not take a very big hole, just large enough to not plug easily.

Worse case scenario: If the ball cavity has cold liquid in it, and the cavity is isolated from the line pressure the pressure within the cavity can increase dramatically due to a VERY slight increase in temperature.

The pressure rise is on the order of 1000's of psi (or hundered's of bars) with only a one degree increase (F or C).

This is a serious matter and could lead to stem leaks, or again, the stems blowing out if they are not captive.

Another problem with ball valves is they are only a quarter-turn valve. The valve handles should have a locking mechanism to prevent accidental movement of the valve handle.

You can well appreciate this after participating in an accidental ammonia vapor release in your face.

It would be even worse if the valve happened to ANY liquid refrigerant. So, my message is... be very careful with ball valves and ask the manufacturer is the ball cavities are vented and do they have locking devices.

If the valve is not available with either of these two features, I would look elsewhere for a suitable valve.

[NH3LVR]

US Iceman is entirely correct
I have never had a problem with ball valves, although the danger is real. If I do install a new valve I make sure there is a hole in the ball. It just takes a moment to look.
Ball valves are commonly used on Screw Compressors in the liquid injection line. I am not fond of them for several reasons, most of them given by Iceman.
In a emergency I have modified a valve by drilling a hole in the ball. (I am NOT recommending this, it was a small ***** system in a remote location where parts were days away at best.)
In addition, if the valve is not designed for refrigeration duty, you may have incompatibilities with the refrigerant or the lubricant.
If I needed to replace a valve in a screw, I would go to FES or other manufacturer for the proper valve.
There is not a problem in purchasing a part from one manufacturer to install on another manufacturers machine, as long as it is done intelligently. For instance I take care of one plant with screws that are no longer being manufactured. If I needed a load/unload solenoid I would go to FES to secure the part. I could go to the local hydraulic shop and buy one, but I would not be certain that that it was compatible with the application.

[chemi-cool]

I was wondering what do the ball valves manufacturers will have to say to this idea.

Frightening thought

Chemi

[US Iceman]

Here is some additional info.

http://www.xomox.com/Xomox/htdocs/no...eID=CL300BV1PC

http://www.conbraco.com/iom/pdf/engl...lish_79700.pdf

http://www.conbraco.com/iom/pdf/english/caution.pdf

The first link is from an industrial valve manufacturer so you can see the difference in construction. They use a sealing system designed to prevent "fugitive emissions" in chemical plants.

Does the "fugitive emissions" sound familiar?

It's a special sealing system to prevent small leaks around the valve stems.

[NH3LVR]

In a emergency I have modified a valve by drilling a hole in the ball. (I am NOT recommending this, it was a small ***** system in a remote location where parts were days away at best.)

Let me clarify this statement as I seem to have painted myself as a reckless Engineer.
This valve was modified in a remote area of Alaska. Telephones were non existent, radio communications unreliable at best. The only way in or out was by airplane or a long boat ride. Neither was practical as the winds were blowing 50 MPH across our improvised runway.
We had a half a million pounds of Salmon on the ground and no way to keep them cold.
I am sure the manufacturer, had I been able to contact him would not have approved.

[US Iceman]

I am sure the manufacturer, had I been able to contact him would not have approved.

A reasonable approach to solving a problem and done for the proper reasons should not attract negative attention.

I would have done the same thing myself. Going ahead with the modification was certainly better than the option of leaving at it was, and creating a potential problem later.

[Dan]

It's funny, I have never thought about the problem. We have Superior and Mueller ball valves in the thousands in liquid lines. Where I see most early blowby is past the valve, moreso than through the packing. But I must admit that I see a lot of packing issues with ball valves. I have always thought it was a teflon or plastic seal issue. I am not sure how trapped liquid within the port of a closed ball valve would cause a packing problem.

Interesting, indeed.

[US Iceman]

I am not sure how trapped liquid within the port of a closed ball valve would cause a packing problem.

Hi Dan,

Take a look at the example shaft construction in the following link:

http://www.xomox.com/Xomox/htdocs/no...eID=CL300BV1PC

[monkey spanners]

What made me think about it was i fitted a 1/2" valve in the liquid line of an old fixed orifice milk tank system so i could pump it down if need be. The valve was supplied in the shut position and i've always been told to braze things in in the open possition so if i cook them i can stll run the system, anyway opened valve and the two plastic bungs fired out of the inlets, so it had held that pressure from when it was tested at the factory.

NH3LVR to get someone out of trouble once, i used a piece of silicone milk hose and a jubilee clip to seal a fractured pipe Its getting things running again without the correct parts that make the difference between a fitter and an engineer.

Cheers Jon

[chemi-cool]

I was thinking about it today.

I came to another thought,

Inside the ball there only liquid, therefore the pressure will not rise. Only when there is vapor, the pressure rises.

Chemi

[NH3LVR]

NH3LVR to get someone out of trouble once, i used a piece of silicone milk hose and a jubilee clip to seal a fractured pipe

You sound like my kind of problem solver. Perhaps you can come over here and help me figure out a way to the water/ammonia mixture out of the vent piping on a three screw 1200 Horsepower system. The owner open the valves to the water filled blow off tank and filled the piping with water and ammonia mixture. (Saving money on decommissioning)
The fun part is that the engine room is in the plant itself and there can be absolutely no smell. Drained out 3 Gallons Friday and it was strong enough to frost the bucket.
By the way, what is a jubilee clip?
Never mind I googled it.
Further proof that we are separated by a common language.

[NH3LVR]

I was thinking about it today. I came to another thought, Inside the ball there only liquid, therefore the pressure will not rise. Only when there is vapor, the pressure rises.
Chemi

Chemi;
The fact that there is liquid in the ball is the problem. When vapor and liquid are present the pressure follows the temperature pressure relationship you have in the card in your wallet.
When there is only liquid the expansion that takes place has no volume to fill.
This can generate enormous pressures, which are nearly impossible to contain.
Usually we think of this as something that occurs when liquid is trapped in between valves and allowed to warm.

[US Iceman]

Here is what we are dealing with:

Volumetric thermal expansion coefficient

http://en.wikipedia.org/wiki/Thermal_expansion

This is a very dangerous condition when dealing with cold liquid refrigerant that can be isolated by a ball valve cavity or between two service valves that are closed (or a service valve and check valve).

Not to be confused with linear expansion due to thermal changes. This one is where the material shrinks or grows in length when the temperature is increased or decreased.

[NH3LVR]

Iceman;
I have always wondered about this.
If we are able to contain the pressure of liquid expansion in this situation, what happens to the contained liquid? Does it stay at the original temperature if it cannot expand? Is there a upper limit to the forces that can be generated by the expansion?
Physics is not my field, although I have some small knowledge.

[US Iceman]

I have always wondered about this.
If we are able to contain the pressure of liquid expansion in this situation, what happens to the contained liquid? Does it stay at the original temperature if it cannot expand? Is there a upper limit to the forces that can be generated by the expansion?
Physics is not my field, although I have some small knowledge.

This is one of those subjects I'm still trying to figure out. From what I have been able to learn (which is not as much as I would like ) you have several different scenarios.

One where the pipe is considered rigid (or non-elastic) and another where the pipe is condsidered flexible (or elastic).

If the pipe is treated as rigid, then pressure simply increases (depending on the actual temperature rise) and the pipe would eventually rupture at some pressure. In this case, the volume of the pipe is constant.

In the elastic scenario, the pipe walls are actually elastic and stretch as the pressure increases. This also continues until the pipe ruptures. So, as the pipe stretches the pipe volume slightly increases, which keeps the pressure a little lower (than if the pipe volume were constant)

I believe the upper limit is determined by the ultimate strength of the container (pipe in this case, or a valve body).

In both cases the temperature and pressure would continue to increase until something burst.

Now you have the limit of my understanding on this subject. I hope someone can shed some additional light on this subject (I suspect TXiceman or Andy_P could do I nice job ).

Until I find a more detailed explanation I tend to do some research until I find enough information to point me in the right direction. After that, if it looks like a duck, I can at least recognize it for what it is.

[NH3LVR]

You have a better grasp of this than I do. I would like to understand the Physics a bit better. All I understand is the effects. (Valve it off and it may go bang?)
Am going on the road tomorrow to see the Customer from Hell, along with the Lawyers and environmental people. I hope to see some good commentary on this subject to keep me amused. Thank god for my laptop and Motels with Internet access.

[US Iceman]

Am going on the road tomorrow to see the Customer from Hell, along with the Lawyers and environmental people.

Now that sounds like so much fun. I bet that ranks right up there with surgery without the general anesthetic.

[monkey spanners]

1200Hp, thats only 200 times bigger than most of the stuff i work on! LOL
Ive spent all day makeing and fitting a chemical dosing system for an automatic milktank wash system ( needed hypochlorite rinse added due to washing with borehole water) Farmer had kindly left a dead calf by the door which i had to walk past to get to the van

Cheers Jon

[NH3LVR]

1200Hp, thats only 200 times bigger than most of the stuff i work on! LOL

This is a area where size does not matter!
I can assure you that I struggle when I have to work on the smaller units. If you can do one well the other comes easily, after a bit of experience. Unfortunatley I only work on the smaller units when I have to fill in for the ***** guy. I used to be able to do them but am not up on the newer technoligies.

[US Iceman]

Unfortunatley I only work on the smaller units when I have to fill in for the ***** guy. I used to be able to do them but am not up on the newer technoligies.

I have a similar problem. It seems I have forgotten a lot of what I used to take for granted.

Unfortunately, I think this falls into the category of ; use it, or loose it.

I hope that only applies to certain things.

[SteveDixey]

Fine so far. we've now got the centre of the ball full of liquid, with no expansion room, any body else see a problem with this or am i missing something? Could this be why the stem seals go?

Cheers Jon

Ball valves with vented "balls"

http://www.rff-france.com/indexgb.htm

Steve

[binman54]

What US Iceman says is true about the rigid and elastic systems. For our purposes piping systems are rigid. You could argue the point if you are a scientist and require a rigorous explanation; metal expands with temperature, etc. but we can ignore all of that.
Trapped liquid will turn into trapped vapour. This I believe is what NH3LVR is saying.
Liquid is incompressible that's why it smashes compressors. Vapour is compressible and obeys the "ideal gas law". This law states that the pressure and volume of a gas is related to the temperature. This is written as PV=nRT.
P is the pressure, V is the volume and T is the temperature of the gas. The other two letters n and R are constants depending on the type of gas. We can ignore them for our purposes.
Using NH3 at 75F the saturated volume of gas is about 2.125ft^3/lb, the saturated volume of liquid is about 0.0265ft^3/lb (the inverse of density). When the trapped liquid changes to a gas the volume wants to increase about 80 times (2.125/0.0265). The system, piping or the inside of a ball valve, is rigid and will not balloon out to allow that kind of increase so the pressure must increase according to the ideal gas law.
Using the gas law the new pressure is 80 times the pressure in the system when you had liquid.

Binman54

[NH3LVR]

[quote=binman54;65751]Trapped liquid will turn into trapped vapour. This I believe is what NH3LVR is saying.
Binman54;
You may be correct, my knowledge of Physics is limited.
My point is that if you have the interior of the ball filled with liquid, there will be no chance for vapor to form. The liquid expands (similar to when waters changes to ice) The pressure developed is virtually unlimited. Correct me if my understanding is incorrect.

[US Iceman]

Correct me if my understanding is incorrect.

You are not wrong. This is very much like ice expanding. It is very destructive.

A simple increase of 1 degree F can cause a rapid pressure spike that can generate very high pressures.

Trapped liquid is extremely dangerous without turning
to vapor.

[goodguy]

While talking about this topic I think it might be worth mentioning poor soldering practices.

Putting a smaller size pipe into a larger size then brazing it can create voids where liquid refrigerant or oil can settle. A joint like this on a suction line or perhaps after a liquid line soleniod that will see constant temperature changes will collapse under hydraulic pressure, possibly causing a restriction.
Has anyone ever had this trouble?
I haven't actually had this happen to me and i would be lying if I sad I never jammed to pieces of pipe together!! A proper reducer can go a long way.

Cheers, Goodguy

[Tycho]

Just for good measure, if I remember correctly.

Trapped R-22 liquid will raise the pressure with 4 bar for every 1 celcius/farenheit the temperature rises.

so if you happen to trap liquid at -40 C and the temperature rises to +20 C the pressure would be +/- 240 bar, or 3480 psi if you wish...


Co-worker of mine was doing some service on a platefreezer, closed the suction valve and ofv valve on one freezer while emptying another platefreezer and forgot to stop the ref. pump. after fifteen minutes we heard hissing of liquid ***** evaporation and noticed a tiny stream shooting out of the weld on one of the freezer plates (they are rated at 200 bar). he was hanging from his wrench, jumping up and down and only just managed to open the suctionvalve.

On a ship, with an ammonia plant, there had been service people onboard, they had their own vacuumpump and would take care of that themselves.
the ofv valve was never opened and the end result was three plates exploding and the rest had big bulges.
I'll see if I can dig up some pictures...

[The MG Pony]

and people wonder why I am so neurotic about valve sequencing and fail safes.

[US Iceman]

This liquid expansion problem does not have anything to do with valve sequencing or fail safes. It is a natural function of the liquid and temperature rise.

It happens in sections of pipe or any other volume that have been isolated, which contain 100% liquid refrigerant. If this volume of liquid is cold, the problem is actually worse because there is more potential to warm up the liquid.

[Andy P]

Hi all - sorry I am a bit late to the party. There's a lot of good useful stuff in this thread. We use the RFF ball valve with a hole in the ball to vent - no problems. We used a different type on an early CO2 job and had seriously deformed valve seats due to internal pressure of the ball. Under certain conditions trapped gas (always unpleasant ) in CO2 systems can also cause overpressure :0

The explanations by NH3LVR and Iceman are spot on! - as the temperature rises, if the fixed space is completely full of liquid then the pressure will rise faster than the saturation pressure is rising ie as you heat the liquid it gets more subcooled. If this is difficult to understand then look at a p-H chart. Your starting point is on the saturated liquid line and the rise in temperature at nearly constant volume takes you in an almost vertically upwards direction on the chart. This moves you further away from the saturated line into the subcooled liquid region, hence you get "subcooled" by heating!

We had a plant (pumped R-22) where gauges on the pumped liquid line were failing for no apparent reason. On inspecting the gauge it looked as if it was reading 0 bar absolute, until it was noticed that the needle was the wrong side of the little pin that acts as a stop! The problem was liquid trapped between the pump discharge check valves and the evaporator liquid line solenoids. If the whole plant was duty satisfied for any time then the liquid would warm up a bit and expand. Iceman is right that the pipe also expands for 2 reasons: thermal expansion (it is also warming up) and higher strain because the pressure is higher. The combination of these is still less than the rate of expansion of the liquid, so the pressure rises.

Binman54 - you have got caught in a common trap: "if I heat liquid it must evaporate......" Not if there is nowhere for the gas to go to! Once you have the misconception in your mind it is really hard to shake it off - here lies the way to premature insanity!

I hope that this helps

cheers

Andy P

[Andy]

On a ship, with an ammonia plant, there had been service people onboard, they had their own vacuumpump and would take care of that themselves.
the ofv valve was never opened and the end result was three plates exploding and the rest had big bulges.
I'll see if I can dig up some pictures...

I have worked on some Gram plants in the past, they use a safety valve on all valve stations which is a check valve with a spring, releasing at a pressure just above the normal defrost pressure. These valves are live to the mains and cannot be isolated meaning you can never over pressure an evaporator. If you are wondering how you change them, it involves pumping out the systems and holding a vacumn until the valve is swapped out (uses nut and tail fittings on the valve).

I alway thought this was a very clever way to go.

Kind Regards Andy

[NH3LVR]

(uses nut and tail fittings on the valve).Kind Regards Andy

Could you help out a American a bit? Just what is a "nut and tail" fitting?

[US Iceman]

I'm glad you asked that question NH3LVR.

I wanted to, but after being blasted for mentioning the Queen of England was visiting I thought I would wait for someone else to do the dirty work of asking about this.

[Tycho]

I have worked on some Gram plants in the past, they use a safety valve on all valve stations which is a check valve with a spring, releasing at a pressure just above the normal defrost pressure. These valves are live to the mains and cannot be isolated meaning you can never over pressure an evaporator. If you are wondering how you change them, it involves pumping out the systems and holding a vacumn until the valve is swapped out (uses nut and tail fittings on the valve).

I alway thought this was a very clever way to go.

Kind Regards Andy

We used to have this in our systems before (long before my time) still use it on the liquid line from pump pressure and back to the liquid separator (think they open at 5 or 10 bar, can't remember)

The ones I have seen on valve stations are mostly cut out and welded shut. probably after the second time they started leaking

the ones we use are Ermeto RHD 10L.

Now that stainless is being used more and more in the corrosive exposed parts of the plant, maybe it wouldnt be a bad idea to start using them again


Cheers