single PZV instead of multiple PZV SETUP

[Rakesh2560]

Hi all,
We are in process of providing single PZV instead of existing two PZV INSTALLATION .This is done as there is a requirement for isolation valve at inlet and we need two PZV functioning at a time in current situation .So we are thinking of resizing the PZV so that the combined relief of both PZV can be given through a single resized PZV.
In our case the Nozzle size at most of the vessels is higher then the pzv inlet and pzv are fitted directly on vessels so resizing the PZV gives me option of 1 operating and 1 standby.
I Needed to know in the first place why would the vendor have provided 2 relief valves ?
2) is it proper to resize the PZV and what are the changes I need to take care of ?
3) My PZV are complying to see 8 div 1 whereas current system is of as me sec 1.
The service is steam.
Both current pzv have different set pressure.

 

[TD2K]

< Needed to know in the first place why would the vendor have provided 2 relief valves >

It's not possible to definitely know "why" but there are some possibilities:
1. Delivery or cost issues for two smaller PSVs compared to 1 big PSV.
2. Inability of 1 large PSV to meet the inlet condition requirements. You don't give a set pressure so I'm just going to put it out there.

If the vendor provided these, is there anything in their documentation why they picked two PSVs? How old is the equipment, can you ask the vendor? Is it possible that originally they were going to provide 2, 100% PSVs (1 online, 1 offline) but later engineering confirmed they needed more relief area than a single valve could handle so they just converted the PSVs to 2, 50% PSVs.

If you are going to put a single PSV on it, other than the required area, you want to make sure the inlet line losses are within either ASME Section 8 requirements at least as a starting point. Exit losses has to be within the maximum allowable for the type of PSV selected (likely conventional as I would expect steam PSVs to relieve to atmosphere).

This is also an excellent time to review the PSV sizing basis per API 520 and ensure the PSVs are still adequate. If the plant is older, it's quite possible that changes over time have affected the relieving capacity such that you need more area than you currently have.

 

[Rakesh2560]

Hi Td2k thanks for reply mate both PZV are at different set pressure and yeah we don't have any documentation for the same as it's a very old plant.
My question is whether it is ok for resize the PZV instead of two I can have one PZV with relief quantify equal to the two current pzv.
I did send data to pzv vendor and he is offering me a resized pzv but I am confused whether shud I resize it or provide two additional pzv for isolation.

 

[TD2K]

Yes, you can have 1 PSV that has the required capacity instead of two smaller PSVs

 

[TD2K]

If you have to isolate the PSV for maintenance for some reason, you will have to decide if the unit has to be shut down or if you can provide protection against overpressure (since you won't have a PSV) but some other mean(s).

 

[georgeverghese]

Get the impression that the vendor has offered 2 psvs' arranged to operate in a staggered RV arrangement.

In a typical staggered arrangment:

PSV1 is set at 100% of MAWP and has a low capacity

PSV2 is set at 105% of MAWP and has a higher capacity

This arrangement is some times chosen for the case where the more frequent relief case is a low flow relief, and the worst case, lower frequency event, has a much higher relief rate. Both PSV1 and PSV2 will be operating to handle the worst case flowrate.

Usually you may find that when this is done, the relief rate ascribed to PSV1 (for the low flow relief rate) is less than approx 30% of that for the worst case relief rate.

Such an arrangement helps to better suit the plant relief rate with the RV capacity. Using a single oversized RV to handle the low flow case can result in excessive chatter and damage to the single RV, as it doesnt get to lift off its seat completely.

Maybe the vendor has more information on how your plant operates and what are the various relief scenarios in this plant. In any case, check the relief cases in your plant and see if this is the case.

As in all other cases, the inlet lines to this staggered arrangement should enable compliance to the 3% of set pressure upper limit being the max permissible dp to each of the PSVs for the RATED sum total flow from both psvs' when in operation.

 

[Rakesh2560]

Thanks mate for your guidance .You are right the two PSVS are at different set pressures.But the issue here is that I have to add isolation valve for maintenance feasibility and still have two valves functional at a time so I have two option's
1) add two new pzv on each of the existing valve so that two remain operational at a time.
2) To resize the valve for current total relief rate of both the valves and replace them with a two resized valves which can h able cumulative relief rate.
I just want to know about safety of second option is there any way to avoid chatter .This valves have never popped up in 30 yrs .

I also need one clarification there is a clause saying that we can provide multiple psvs on same branch if their total cross sectional area is equal to that of a inlet .
What does this clause mean for eg if my nozzle is 3 inch then I cannot provide tow pzv with 3 in inlet on this pzv?

 

[LittleInch]

You've said the two PSV's (why do you call them PZVs by the way?), but nothing about their size , rated flow or orifice size, all of which should be on the plate on the valve or able to be considered looking at the system. First work out your relief scenarios then design accordingly - don't guess or assume what was originally designed is still valid.

Only then can you see whether the intention is to initially protect against a small flow but then have a larger flow at a higher pressure or not.

The fact the valves don't pop and have not done so for a long time is good and shows good control over the incoming pressure. chatter is normally avoid by not having a continuous relief event at < 30% of rated flow.

The double valve thing only applies if both valves are "active" at the same time, but yes your working is correct.

If you have two current nozzles, why not combine them and then branch off that with two or three PSVs. If the capacity of the PSVs are different then one low flow and two high would cover you for all eventualities.

We can't see your system so it is difficult to judge, but if this arrangement has been there for 30 years, why are you changing it now? Spring PSVs are inherently very reliable and whilst you need to re-calibrate at some interval, don't you ever have a shutdown?

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[georgeverghese]

The other possible arrangement is to have
One of PSV1 and one of PSV2: each psv with isolation block valves as required by your company's isolation practiceand a spare piping run that will enable fitting either of PSV1 or PSV2
and
a spare piping run with isolation block valves on inlet side and on vent side that will enable fitting either a spare PSV1 or PSV2. So some pipe reducers will need to be fitted on when making this switchover.

Your suggestion (a) would be ideal if you can find the space and block valves for all these 4 valves. Suggestion (b)is not a good idea if you ask me for reasons explained earlier. Suggestion made by Little Inch is okay if you want to minimise expense, but beware that in the mode when 2 of PSV2 are in operation (ie when PSV1 is offline), the inlet line rated flow would be higher.

 

[Rakesh2560]

Thanks guys for the reply let me be more specific now We are doing this change at 16 locations I will pick up data for one of them
The vessel nozzle is 8 in
PZV inlet size is 4 in
currently there are two pzvs (We call the pzv in our company) each with set pressure of 120 psig and 123 psig Now the mawp is 120 psig
Now to replace this PZVs and avoid adding a third PZV as in the current situation there are no isolation valve in both pzv hence both pzv are operational at a time both have different relief rate (Nearly 50201 lb/hr) with total relief rate being 1010237 lbs/hr.
Now we requested vendor to resize the pzv with each pzv having a relief rate of 101237 lb/hr so that we do not need additional pzv and can also provide isolation valve with one operating and one stand by arrangement.
About orifice sizes the current sizes are 3.6 in2 each.
The vendor is offering us a resized pzv of 4X6 with total relief rate of 101237 lb/hr each revised orifice sizes of 7.2in2.
Now my question is should we go for this resizing ?
What are the chances that there will be chattering In the resized pzv?Can I take any steps to avoid chattering ?
I had a chat with process and they have agreed to revise the set pressure to 120 psig to both of them.

Also I need one more clarification code says that if we want to provide a multiple pzv on same inlet then the sum oftotal cross sectional area for both the pzv should be atleast equal to the inletcross sectional area .
As per this clause if I have an inlet nozzle of 3 in then can I provide two pzv of 3 X 4 as total crsoss sectional area of both 3 in inlet pzv will be more then inlet crossectional area for 3 in inlet

Hope I am specific with my case.

Thanks in advance guys

 

[LittleInch]

Two PSVs with set points that close and seemingly basically equal seems to make no sense.

At the risk of repeating myself, you need to determine what the realistic relief scenarios are. If it is basically all or nothing then sure, replace two valves with one.

If there is some sort of credible low volume continuous relief < 30% of max flow, then you still need two, but set more than 3 psi apart, to avoid chattering.

The multiple PSV on one nozzle thing only applies if both are connected and needed for the relief rate. If you have 2 x 100% PSVs then the total flow will still be the same so in that instance you can have two 3" valves off a 3" nozzle - I would assume that one would be locked off and one active until you switch them over.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[TD2K]

< Now my question is should we go for this resizing >

How do you expect us to answer this? This is your plant, your operational issues, your practices.

< What are the chances that there will be chattering >

What are the various relieving cases and how do these compare to the capacity of the PSV? Chattering is a real issue if it occurs but on the PSV files I've done and reviewed, there are usually many potential cases identified that "could" overpressure the vessel and result in the PSV having to relieve. Some/many of these cases are much less than design so if that case occurs, chattering is a potential issue. I'm not sure what to suggest. I have rarely seen 2 or more sizes of PSVs installed with the intent for one PSV to handle some cases and other PSV(s) to be able to handle higher capacity cases. There's usually not enough gap available in set pressures for me to feel very comfortable that what may look good on paper is going to happen in the field. The other option is a modulating action pilot operating PSV. Some places love these, other places will just about throw you out if you suggest a pilot operated PSV.

< Also I need one more clarification code says that if we want to provide a multiple pzv on same inlet then the sum oftotal cross sectional area for both the pzv should be atleast equal to the inletcross sectional area >

I'm not following you. You said you had an 8" nozzle and the new proposed PSVs were 4" by 6". On the basis of just that, you should be good. However, I would want to look at my inlet piping and ensure my inlet line losses at the capacity of the PSV did not exceed 3%.

 

[georgeverghese]

a) 3.6in2 orifice is familiar- this is M orifice
b) What is 7.2in2 orifice - no such designation? - P=6.38in2, next is Q=11.05in2. Does your company allow the use of this non standard 7.2in2 orifice?

From the way the current PSVs' are set up, it is clear this is not a typical staggered arrangement to handle large variation in relief rates (set pressures are too close, flows are the same - this is consistent with your process dept allowing the set pressure to be 120psig on both RVs'). So, your approach to go for a single PSV to handle 101237lb/hr at 120psig is consistent with this current setup.

But that doesnt mean the current setup has been chosen correctly. Once you have tabulated all your plant relief cases and examined them, you can see for yourself if there are relief cases of say 30 000lb/hr or less which constitute the larger proportion of these relief cases( say for example, more than 7 out of 10 relief cases have relief loads less than 30 000lb/hr, assuming of course each relief case has the same probability of occurrence as the others). The process engineer will then have addressed this chatter concern. This task is the responsibility of the plant process engineer and his supervisor.