Relief Valves ???

[VGV]

Would appreciate clarification on proper application of relief devices.

What I think I know is:
1. For application on ASME vessels, relief devices must be certified by National Board.
2. Relief devices are certified for either vapor or liquid service and are not dual service rated.
3. Manufacturers of vapor service relief valves advise to "connect to vapor space".
4. A vapor service rated relief valve may self destruct if subject to alternate vapor/liquid exposure.
5. If both liquid and vapor are in close proximity to the relief device (as on oil pots and thermosyphon tube-side applications) there is a high probability
the device will be called upon to handle two phase flow and violate all above intelligence.

What am I missing here??

[Grizzly]

Welcome VGV.
I will have to rely on a more comprehensive answer form my fellow forum users.
Particularly the Guys from your side of the forum.
On a practical note, I am not used to seeing a relief valve that is piped so that it would normally vent liquid to atmosphere.
By sensible sighting usually well above the working liquid levels.
The likelihood of liquid being purged through a relief valve is greatly diminished.
Having said that if a catastrophic failure resulted in the venting of liquid through a relief valve.
Then the least of your problems would be the fact that you might have to replace it.
All the ones I know are pretty robust and basic and more than capable of handling a bit of liquid!
But hey! Just in case of Litigation lets cover our backsides eh!
Grizzly

[chilldis]

what frequency of relief valve replacement is required in your area? we are required to replace ours every 5 years.:off topic:

[nh3wizard]

As Grizzly has stated, normally when a relief goes vapor is what is venting, due to it being vented to atmosphere. I believe the 5 year requirement that chilldis is referring to is a US requirement.

[VGV]

My understanding:
It all boils down to the law.
In the US, Local Codes and /or OSHA mandate process safety, and may reference ANSI/ASHRAE-15, ANSI/IIAR-2 and perhaps ANSI/IIAR-110 "Start-up Inspection and Maintenance of Ammonia Mechanical Refrigeration Systems"

IIAR revision to Bulletin 110 (2-29-04) 6.6.3 states:
"Pressure relief devices shall be replaced or recertified in accord with one of these three options:
1) Every five years from date of installation......
2) .....documented in service life.....
3. .....manufacturers recommendations....."

Somewhere in the fine print of most codes there is a clause concerning suitability of equpment for duty intended, and if the manufacturers recommendations are "every five years", that pretty well tie it down.

Word of caution, OSHA PSM programs, contain "management of change" requirements, and any replacement is best accomplished with "like for like"...

(Guess that would be assuming the original "like" was up to code. huh?)

VGV

[nh3wizard]

I know we don't have to but if we change a relief out with a different manufacturer we go ahead and fill out a MOC, just in case... It all boils down to how the inspector interprets the code. Just like I do not believe a condenser is a confined space, but some believe it is.

[acrefconsultant]

As Grizzly has stated, normally when a relief goes vapor is what is venting, due to it being vented to atmosphere. I believe the 5 year requirement that chilldis is referring to is a US requirement.


I do not want to correct here but just fill in what Wizard has said under BS EN 378-4:2008 it states


"Safety devices are checked on site: annually for safety switching devices (see EN 378-2:2008,
6.3.5.3.3), emergency signals and alarm systems; every five years for external pressure relief devices."

It only states checking them does not indicate replacing them every five years. Repalcement shall only take place if refirgerant has been discharged and the valve is proven to be loose extact below.


"After a pressure relief valve, which discharges to atmosphere, has been actuated, it shall be replaced
if it is not tight."

Hope this helps

[US Iceman]

"Safety devices are checked on site: annually for safety switching devices (see EN 378-2:2008, 6.3.5.3.3), emergency signals and alarm systems; every five years for external pressure relief devices."

It only states checking them does not indicate replacing them every five years. Replacement shall only take place if refrigerant has been discharged and the valve is proven to be loose extact below.



"After a pressure relief valve, which discharges to atmosphere, has been actuated, it shall be replaced if it is not tight."





This language opens up an interesting situation...

How do you test relief valves every five years?

If you test it to see if it opens within the prescribed pressure tolerance, the valve has then been operated which implies the valve needs to be changed every five years then.

[CHIEF DELPAC]

Here in British Columbia the safety valves have to be replaced or tested every 5 years. The testing has to be done by a approved testing facility. This means that safety valves have to removed and sent to the approved tester. I have around 50 safety valves which are replaced every 5 years. It is easier to replace the valves than to send them out for testing. AS you can imagine it is fairly expensive to do. However as safety is the name of the game, it is done. C.D.

[US Iceman]

One interesting point that would come from testing the valves is: At what pressure do they pop open? And how far from the factory set pressure is this pop off pressure?

[CHIEF DELPAC]

US Iceman I have been informed that one of the reasons for changing the valves is that the springs weaken over time. This being true then the valves would pop before their rated pressure. C.D.

[US Iceman]

US Iceman I have been informed that one of the reasons for changing the valves is that the springs weaken over time. This being true then the valves would pop before their rated pressure. C.D.

I have heard some information where a firm started to test their valves as they were rotated from service. The results were all over the place. Some popped at low pressures, some popped at pressures substantially over their set pressure.

Neither is a good indication of what we might expect for safety considerations as obviously (based on this small data set) we would never know where they would pop at after being in service for some time.

I think we need to carefully investigate the relief valve vent line piping installation methods...

[Grizzly]

I do not want to correct here but just fill in what Wizard has said under BS EN 378-4:2008 it states


"Safety devices are checked on site: annually for safety switching devices (see EN 378-2:2008,
6.3.5.3.3), emergency signals and alarm systems; every five years for external pressure relief devices."


It only states checking them does not indicate replacing them every five years. Replacement shall only take place if refrigerant has been discharged and the valve is proven to be loose extact below.


"After a pressure relief valve, which discharges to atmosphere, has been actuated, it shall be replaced
if it is not tight."


Hope this helps


I rarely say this but The above is complete rubbish when refering to refrigerant PRV's.
However the onsite retesting of steam systems is relevant.
I can't quote the Pressure regulations chapter and verse.
But i know enough to state that under the above regs.
A pressure relief Valve will be tested, certificated and have the date of when it was last tested.
Stamped on a tag or engraved upon a label attached to it.
How many times it has lifted is irrelevant but whether or not it is gas tight IS NOT!
In UK and I suspect Europe, we have been fitting Dual Manifolds that house the PRV'S.
To facilitate easy removal whenever re-calibration is due. Oh! By the way the required timescale is every 5yrs.
Unless the Insurance Inspector decides otherwise.
We have some older sites where 3yr inspection and recalibration is required.
Grizzly

[CHIEF DELPAC]

USIceman, Interesting about the valves. I will soon be changing out the safety valves in my engine rooms I think I will test some of them. Vent lines issues. I was moving a vent line due to plant expansion and found at the valve there was enough rust and corruption that it was highly likely that the valve would not have been able to open at its set pressure. I dont recall how long the valve was in service. However with changing out the valves it a simple matter to prove that the vent lines are clear and are able to work. C.D.

[tab71]

What would be considered best practice for thermosiphon SRV's? Other then have the valves at the top of the return piping and having an equally inaccessible isolation valve I am lost as to keep the valves in vapor in case they lift?

[US Iceman]

I was moving a vent line due to plant expansion and found at the valve there was enough rust and corruption that it was highly likely that the valve would not have been able to open at its set pressure.


There you go. That is what I'm concerned about. All too often the vent lines are installed in any way to get the pipe termination at some point outside of the building.

I'm beginning to think stainless steel would be better than carbon steel pipe and that we also need some good rain prevention devices on the pipe termination and some water drains in the vent line headers. And... I also think we need to have the headers trapped so that water cannot reach the relief valve outlets.

Beyond that, I don't know what else we could do.

[VGV]

Am real interested in the results of any relief valve tests you do Chief.

[nh3wizard]

There you go. That is what I'm concerned about. All too often the vent lines are installed in any way to get the pipe termination at some point outside of the building.

I'm beginning to think stainless steel would be better than carbon steel pipe and that we also need some good rain prevention devices on the pipe termination and some water drains in the vent line headers. And... I also think we need to have the headers trapped so that water cannot reach the relief valve outlets.

Beyond that, I don't know what else we could do.

I agree with US Iceman, but before you know it someone will say "Thats a GOOD idea!!! Lets make it a requirement / regulation"

[US Iceman]

... before you know it someone will say "Thats a GOOD idea!!! Lets make it a requirement / regulation"


Unfortunately, that seems to be the only we get anything done in this business. If not for PSM/RMP we would still be doing things the same old way.

[CHIEF DELPAC]

VGV I wont be changing out the safety valves until the end of March. I will post the results that I get on the safety valves lifting pressures. C.D.

[RANGER1]

We basically follow US codes and require relief valves + vent lines have a visual check annually . This would check for blockages , restrictions and relief valve damage/deterioration .
Usually weather proof outlet to stop water getting into vent line.
5 year recalibration interval c/w certificates .
Stainless vent lines can be cheaper to instal as bracketing is lighter and you dont have to paint .

[VGV]

C.D. will be looking foreward to your findings..

The reason for this thread was to get some sort of understanding for the proper way to handle thermosyphon oil cooler relief when both oil separator and oil cooler are rated at 300psi.

IIAR piping manual handles tubside relief by placing TS stop valves at level of return header with a vapor relief on top of the return line from the cooler. If they address oil side relief, I haven't found it.

IIAR 2-2008 8.3.1.2 says when tubeside can be isolated "pressure relief protect per 11.2" ...
11.2.1 says "See 11.4" 11.4 says "trained technicians" OSHA says trained technicians are not enough....

Nobody wants to talk about the oil side...

Any good thoughts??

VGV

[Oregon Jim]

I change out all of my relief valves every five years, and all of the old relief valves are then tested to see at what point they would relieve. The tests have been alarming. Simply put, they cannot be trusted to be accurate in their settings, as some relieve early and some don't ever relieve. After seeing our results I am convinced that changing the valves is VERY SMART, even though it is a pain and a big expense. Our safety is worth the effort and expense.

[Samwitch]

What about the possibility of installing bursting disks into your dual manifold/prv assembly. Surely these should rupture if over pressured their stated level i.e. 14 bar.

I'm aware some insurance inspectors are moving this way.

[US Iceman]

What about the possibility of installing bursting disks into your dual manifold/prv assembly.


True, these can help to prevent premature release if the PRV let's go at a lower than designated set pressure.

However, a rupture/burst disc would not help much if the PRV does not relieve at it's designated set pressure.

[CHIEF DELPAC]

VGV In my plant their is no rhyme nor reason to the pressure relief valves on the oil side. Some have relief valves venting into the oil seperator. others have adjustible pressure regulating valves that vent to the suction side of the oil pump. And others have no relief valves atall. The date of install seems to be the determing factor on the relief valves or lack of same. C.D.

[CHIEF DELPAC]

Samwitch, USIceman; In my previous post when I mentioned the escape line that was blocked with rust and c... A sugestion was made to have the escape lines have a rupture disk after the safety valves, in them to prevent water from reaching the safety valves. Needless to say as it was not a requirment it was not done. I personally would rather have safety valves than rupture disk. One reason being when a rupture disk goes all the charge would be lost. Hopefully with a safety valve after the pressure is relieved the valve would close and the remaining charge would be contained in the pressure vessel. C.D.

[US Iceman]

A suggestion was made to have the escape lines have a rupture disk after the safety valves,


You cannot do this. It would restrict the PRV valve capacity by adding addition pressure loss to the PRV.

There a couple of things that bother me concerning PRV installations.


The vent line piping is usually carbon steel, so that introduces potential corrosion.
Water vapor migrates to the lowest vapor pressure. On cold vessels you could induce moisture flow to the relief valves. Again causing corrosion.
The piping installations are typically fairly poor with no overview of what happens up until a relief valve lets go.

I would be interested to see if the relief valve issues have been categorized by application (high temp receivers versus low temp vessels).

[VGV]

CD, ThanX for your input...
Re Oilside relief, maybe I'm reading the code wrong, but 15-2007, 9.4.3 says "pressure relief... shall be used...liquid containing parts...hydrostatic expansion due to temperature rise."

IF the vessels are ASME rated, relief to oil separators using an ASME Certified LIQUID rated valve set at 75 psi is OK if the oil cooler is rated at 400psi and the separator at 300psi.
My concern is proper relief for the oil side when both rated at 300psi....

There's probably a good 300# hydrostatic out there that could be relieved to a separate oil relief manifold....

With ASME vessels, relief via a pressure regulator may not be OK as the regulator may not be ASME certified. (Don't know that such exists).

On the NH3 side, most relief's I see are vapor reliefs on the head of the cooler.. Manufacturers say "vapor only".

This is only a problem when isolation valves are closed on either NH3 or oil, so maybe the smart thing to do is to get the local approval agency to sign off on that.

[US Iceman]

There's probably a good 300# hydrostatic out there that could be relieved to a separate oil relief manifold....


With a hydrostatic valve it's not the actual relief valve set pressure but the differential pressure. You have to consider the back-pressure on the relief valve outlet!

This also applies to vapor only valves, which is why it is necessary to design the vent line piping properly.

[Grease monkey]

Most US and UK standards I have seen state 5 years between Safety valve testing if all is ok. If at this interval the results are more than 5% out, the 5 years must be reduced. In the UK we also have a directive from the H S E (PM81) which states. That all safety valves on Ammonia systems must be replaced with a new or overhauled valve every 4 years. So just testing is not good enough. We also have to install dual valves on any vesel with an internal volume over 300Lts.

[IceMan_4000]

http://www.hantech.com/library/pages/products/press-relief/rup-disc-gau.html

Hansen make a pretty nice set up that covers all that has been brought up in this thread.

It is in the B-52 that pressure relief valves must be changed every 5 years. A manifold like this makes life so much easier. Pay up front to make service and future less expensive.

also real easy to see if you have a leak through a valve

[brian_chapin]

Hansen make a pretty nice set up that covers all that has been brought up in this thread.

It is in the B-52 that pressure relief valves must be changed every 5 years. A manifold like this makes life so much easier. Pay up front to make service and future less expensive.

also real easy to see if you have a leak through a valve

They do work well, but be warned that you need to isolated them from vibration as the gauges tend to be extremely unreliable under heavy vibration (such as that caused by a compressor) resulting in false alarms.

[sterl]

Interesting train.

I am not an advocate of Rupture Disks upstream of SRV's.