15 replies to this topic

[Afshin445]

Dear Process Experts,

I am in middle of an existing separator adequacy check.When I try to check PSV adequacy, I found existing PSV is oversized. Two case can be considered for that vessel :
1-Block outlet
2-Fire case.
In block outlet case existing PSV orifice area is about 8 times more than required area and in fire case this ratio is about 73 times.My first question is that existing PSV is adequate for new condition?

My another question is as a I know an oversized PSV can be cause of chattering but I don't know how I can calculated chattering condition?
Is that any method we can predict chattering in oversized PSV. That means how much excess area acceptable to avoid chattering?

Regards
Afshin

Edited by Afshin, 18 January 2012 - 09:08 PM.

[fallah]

Dear Process Experts,

I am in middle of an existing separator adequacy check.When I try to check PSV adequacy, I found existing PSV is oversized. Two case can be considered for that vessel :
1-Block outlet
2-Fire case.
In block outlet case existing PSV orifice area is about 8 times more than required area and in fire case this ratio is about 73 times.My first question is that existing PSV is adequate for new condition?

My another question is as a I know an oversized PSV can be cause of chattering but I don't know how I can calculated chattering condition?
Is that any method we can predict chattering in oversized PSV. That means how much excess area acceptable to avoid chattering?

Regards
Afshin

Afshin,

Your question is not completely clear to me. There is no information about old /new conditions. Please submit needed information in detail.

Indded, usually when minimum relief passing a PSV would be lower than 25% of its rated flow, there would be chattering due to PSV oversizing at those low relief loads.

Fallah

[Afshin445]

Afshin,

Your question is not completely clear to me. There is no information about old /new conditions. Please submit needed information in detail.

Indded, usually when minimum relief passing a PSV would be lower than 25% of its rated flow, there would be chattering due to PSV oversizing at those low relief loads.

Fallah

Fallah,

The existing PSV size is 6"R8".The PSV orifice size is 16".
The new condition is because of production rate has been changed in platform. As I calculated the required orifice area for block outlet case is 2" and for fire case is 0.22".In term of flowrate the detail as below:
1- Block outlet: Rated capacity=78,078 kg/hr,Relieving rate=9,841 kg/hr (ratio=12.6%)
2- Fire case:Rated capacity=102,225 kg/hr, Releiving rate=1,400 kg/hr (ratio=1.4%)

In this regards, do you have any referece for your mentioned number(25%)?

Rgrds

Edited by Afshin, 19 January 2012 - 12:10 AM.

[fallah]

Fallah,

The existing PSV size is 6"R8".The PSV orifice size is 16".
The new condition is because of production rate has been changed in platform. As I calculated the required orifice area for block outlet case is 2" and for fire case is 0.22".In term of flowrate the detail as below:
1- Block outlet: Rated capacity=78,078 kg/hr,Relieving rate=9,841 kg/hr (ratio=12.6%)
2- Fire case:Rated capacity=102,225 kg/hr, Releiving rate=1,400 kg/hr (ratio=1.4%)

In this regards, do you have any referece for your mentioned number(25%)?

Rgrds

Afshin,

The orifice area of "R" designation PSV is 16 square in. (not 16") not important matter... may be wording error... just for correction. Also would be repeated for 2" and 0.22".....

One PSV has a unique rated capacity. How much is the rated capacity of existing 6R8 PSV? The values of 78,078 kg/hr and 102,225 kg/hr are rated capacity of which PSV(s)?

Please clarify, then i can help you more...

Indeed, one of the references for 25% is API 521 Ed. 2007, page 67. If you need inform me. I can attach that page in my next post.

Fallah

Edited by fallah, 19 January 2012 - 12:43 AM.

[Afshin445]

Fallah,

First of all I am sorry for my wrong typing in dimension of orifice area, absolutley you are right.

Second,both rated capcity refer to one PSV (6"R8") but as relief condition are different for both cases (Temp. and Gas properties) then I came with two different number for block outlet and fire case as PSV rated capacity.

In this regards, I didn't find you mentioned page in API521.I will appriciate if you will send me that page in your next post.

Rgrds
Afshin

Edited by Afshin, 19 January 2012 - 01:11 AM.

[fallah]

Fallah,

First of all I am sorry for my wrong typing in dimension of orifice area, absolutley you are right.

Second,both rated capcity refer to one PSV (6"R8") but as relief condition are different for both cases (Temp. and Gas properties) then I came with two different number for block outlet and fire case as PSV rated capacity.

In this regards, I didn't find you mentioned page in API521.I will appriciate if you will send me that page in your next post.

Rgrds
Afshin

Afshin,

As i mentioned a PSV has a unique rated capacity based on its sizing case (a unique scenario with specified relieving pressure, relieving temperature, relief composition and MW,...) will result in largest orifice area. Hence, a PSV can not have two different rated capacity in two different scenarios. It might those two values be required capacity for two scenarios in old conditions; blocked outlet and fire. Please recheck again and clarify.

I have attached the pages you requested from API 521 about minimum relief load of a PSV respect to its rated capacity.

Fallah

Attached Files

•  API_STD_521 25%.pdf   133.26KB   773 downloads

[Afshin445]

Fallah,

First of all I need to thank you for sending API521 standard.In follwing
I need to say in existing PSV datasheet there is no any information about rated capacity and sizing case.
The numbers I told you actually are estimation of rated capacity in two different cases in new condition based on existing PSV area(16 sq. inch).

In other hand, I went through highlight section in standard but honestly I didn't understand relation between it and chattering in PSV.Could you please give me more explanation in this regards.

Afshin

Edited by Afshin, 19 January 2012 - 02:51 AM.

[fallah]

Fallah,

First of all I need to thank you for sending API521 standard.In follwing
I need to say in existing PSV datasheet there is no any information about rated capacity and sizing case.
The numbers I told you actually are estimation of rated capacity in two different cases in new condition based on existing PSV area(16 sq. inch).

In other hand, I went through highlight section in standard but honestly I didn't understand relation between it and chattering in PSV.Could you please give me more explanation in this regards.

Afshin

Afshin,

You should review PSV vendor print to get needed information...

Anyway, as i understood from your above explanations those two values are requried capacities of the PSV in two scenarios and can not be nominated as PSV rated capacity.

Now relation between relief load lower than 25% of rated capacity and chattering:

As mentioned in API 521 (pages i attached) in a PSV, let say conventional type, once it has opened kinetic forces are sufficient to offset the spring force untill the flow has been reduced to approx. 25% of the valve's rated capacity; means: if flow is going to be lower than 25% of the rated capacity then the spring force ovecomes to kinetic forces and the valve would be closed. On the orther hand, because at the same time the upstream pressure of the PSV is still equal to (or higher than) the set pressure, then the PSV would be opened again and this cyclic operation will lead to chattering.

Fallah

[Afshin445]

Fallah,

As I mentioned there is no any rated capacity in existing PSV datasheet (only mentioned N/A in datasheet)

When I want to check PSV required size spreedsheet give me a required capacity based on required orifice area and also as I mentioned in before post I also calculated mentioned capacities based on existing orifice size and that was reason I called them rated capacity.

In this regards, as I understand from your explanations when the relief rate is lower than 25% of PSV rated capacity chattering will be occured because relief gas kinematic force should overcome to PSV spring force and let PSV start to opening and after PSV release if relief flowrate be lower than PSV discharge rate, PSV will be closed again and waiting for increasing of flowrate to minimum rate and after that cycling start again.

If you think my calculation method for estimation of PSV rated capacity and follwing require/rated capacity ratio is not correct.Then how I can find PSV rated capacity in this two scenario based on vessel new condition.

Afshin

Edited by Afshin, 19 January 2012 - 09:13 AM.

[Lowflo]

How much excess capacity is too much? That is a very good question. Intuitively, we regard excess capacity as an economic waste, but not a hazard. That's true for most thing but it's not true for safety valves.

Unfortunately, all we can say is that you need to avoid excess capacity in safety valves. All properly sized safety valves are going to cycle. The more excess capacity you have, the higher the cycling frequency. When the cycling frequency gets too high, it's indistinguishable from chatter.

If this valve is in vapor service, and the system volume is high, then the risk is probably acceptable. If the system volume is low, then you ought to consider replacing this with a smaller valve. If the service is liquid, then the risks of mechanical failure are much higher.

Readers are probably thinking...."There are no specifics here - how am I supposed to use this information?" This is something for which we just don't have a way to objectively evaluate. Understand the general concepts, and apply them to specific applications based on your best judgment. That's all we can say and do.

[Afshin445]

Lowflo

But I need someting quantitive then as per that we can decide need to replace existing PSV's or not.your expanaition also rise another qustion for me, if chattering happen in PSV that means we can't consider it as system protection?

Afshin

[Lowflo]

This is entirely a judgment decision - there's just no predictive way to quantify the risk of valve instability. The only way to get a quantitative answer is to do a field experiment on this specific valve and piping installation. Intentionally create the over-pressure scenarios and see what happens - was the valve stable or not? Of course, that's not a practical thing to do except for extraordinary cases.

In cases like this, where there's no clear answer, let your judgment be guided by the details of the particular case. What are the specific risks and consequences? What other layers of protection exist? What is the process fluid and what are the consequences of a loss-of-containment?

There's a known risk associated with oversized safety valves, but we don't have a lot of evidence of cases in which this led to a loss-of-containment. So, from a probability perspective, your installation is probably OK. What you need to do is to assess whether "it's probably OK" is good enough for this particular case. I can't answer that question for you.

[fallah]

If you think my calculation method for estimation of PSV rated capacity and follwing require/rated capacity ratio is not correct.Then how I can find PSV rated capacity in this two scenario based on vessel new condition.

Afshin

Afshin,

There is no information about old /new conditions, yet. Please submit needed information in detail. Will the new conditions lead to new relief load?

Fallah

[S.AHMAD]

1. Check back the basis of design.
2. What about other emergency scenarios such as gas breakthrough via stuck opened level controlvalve or inadvertent opening of control valve bypass? What aoout liquid overfill? There must be reasons or design basis for such a big PSV?
3. Based on the figures given, it is obvoius that chattering will occur since the required relieving capacity is very much lower then the rated capacity.
4. Assuming the scenarios are valid, you need to have two PSVs since the fire case is 15% of the block-discharge case. The guidline as mentioned by fallah is 25%.
5. However, since the PSV is existing and budget is tight or not available, then you need to justify based on risk assessment.
6. For example, if the probablity of PSV is very low (never happened before of 1 in 100 years) then the wise thing to do is to leave it as it is.
7. Chattering my cause damage to the disk/seat and results in gas leakage - loss of containment. Then you can estimate the consequence in term of monetary value. Compare with the cost of two new PSVs. Is it justfiable?

Edited by S.AHMAD, 24 January 2012 - 10:38 PM.

[Zubair Exclaim]

i agree this PSV may be sized for a blowby case etc search for other scenarios first

[Kishor]

One more point should be considered is about the pressure drop in inlet piping to relief valve. Chattering may take place due to long inlet pipe to relief device creating more frictional losses. That's why we keep inlet losses within 3% of set pressure for better operation of relief valve.
Chattering will take place irrespective of valve size if inlet pressure drop is higher than allowed.