9 replies to this topic

[ryn376]

If a PSV on the cold side of heat exchanger (filled with cooling water at no more than 140 ^oF) has two scenarios -- fire and thermal expansion -- where the controlling case is fire, is a lifting lever required?

[fallah]

ryn376,

 

In thermal expansion case, if the heat exchanger is an ASME Sec. VIII pressure vessel and the hot water temperature never goes higher than 140 F, the requirement of lifting lever for the PSV can be ignored. Can a PSV in your case cover a very small relief load of thermal expansion case and a higher relief load of fire case without any problem in operational veiwpoint?

[ryn376]

Fallah, 

 

The HEX is an ASME vessel and the hot water temperature never exceeds 140 F during normal operation. Please explain what you mean by "without any problem in operational viewpoint". 

[fallah]

ryn376,

 

I meant with a wide difference between relief loads in thermal expansion case and fire case which are handled by the same PSV, doesn't the PSV chatter in lower relief load or does it reseat properly after relieving? 

[ryn376]

In my experience widely varying flowrates between scenarios is the norm not the exception. I believe you are correct that it might chatter, but I might have seen one or two installations where a designer has specified two devices due to this issue (I think I am being generous saying "one or two", I actually do not believe I ever seen it. Note that I have only seen calculations for 50 -100 devices total so I may be in error) . From what I have seen, two devices are usually only added when the relieving capacity cannot be handled by a single device or on boilers. Varying flows are an issue on nearly every relief device installation and clients are not willing to pay for two devices nor pay for the additional engineering and maintenance required.

[fallah]

ryn376,

 

For the case you described there are two possible arrangement as follows:

 

1- Considering one PSV has been sized for fire case as governing case to handle both thermal expansion scenario (with very low relief load) and fire scenario (with high relief load) leading to possibility of PSV chattering in thermal expansion scenario...

 

2- Considering one small (3/4" * 1") for thermal expansion scenario and a supplemental PSV for fire scenario would pop up just in fire case...

 

Which one is better in enginering+safety+economical standpoint?

[Marc-Andre Leblanc]

Hello,

 

While it is not on the safe side and depend mostly on your application and sound judgment, you should be aware that there is a few common sence exclusion to fire protection for heat exchanger. While these are not stated in API, they are commonly used in the industry.

 

Some design and engineering practice, based on experience, will show that heat exchanger that have large bolted body flange (common shell and tube design) tend to self-relieve during fire exposure because the heat relaxes the large body flange bolts and for this reason, can be exempted from the addition of PRV for fire case.

 

An other exemple of exclusion is the double pipe exchanger, since these are generally treated as piping components, the addition of a PRV for fire case would not be required in genenral.

 

PRV will normally be added on heat exchanger with LNG, reactive hazard potential and for other mechanical consideration (fixed tubesheet ...). 

 

In your case, on the cold side (water), there is definitly a requirement for a Thermal PRV, this PRV do need to be properly sized and can be larger than a (3/4" * 1") for thermal expansion since your heat input come from the other fluid in the exchanger and not only environmental consideration.

 

In the case of fire, if the mechanical construction allow it (bolted body flanges), fire will normally relaxed the bolts on the flanges and relieve the water trougth the loosened flanges ... there is no requirement in this case for a PRV.

 

Finally, when it comes to the lifting lever, if you normally operate at temperature lower than the one set by ASME, you do not need a liftign lever. Be aware that this temperature setpoint and consideration where clarified in a errata from ASME because the wording in the original print was unclear and gave the impression that if at any point, including the relief event, the temperature could be higher than the one Set, a lifting leverl would be required. This is no longuer the case.

 

Regards,

 

Marc-Andre

Edited by Marc-Andre Leblanc, 11 September 2014 - 09:12 AM.

[ryn376]

Marc-Andre Leblanc,

 

Would you please point me to the clarification in the ASME errata for lifting levers?

 

fallah,

 

I agree on what best engineering practice is, but will chattering present a safety problem or will it just cause PSV seat and disk failure? I believe chattering will just ruining the PSV. I would appreciate if you would post any references to the contrary, so that I may present them to clients.

[Marc-Andre Leblanc]

Hello,

 

1) The specific passage that was clarified in the ASME sec. VIII div1 is as follow:

 

The original wording (2007) was :

 

UG-136(a)(3) Each pressure relief valve on air, water

over 140°F (60°C), or steam service shall have a substantial

lifting device which when activated will release the seating

force on the disk when the pressure relief valve is subjected

to a pressure of at least 75% of the set pressure of the

valve...

 

 

 This was changed in a addendum in 2009 to the following :

 

 

UG-136(a)(3) Each pressure relief valve on air, water

at the valve inlet that exceeds 140°F (60°C), excluding

overpressure or relief events, or steam service shall have

a substantial lifting device which when activated will

release the seating force on the disk when the pressure

relief valve is subjected to a pressure of at least 75% of

the set pressure of the valve...

 

So if your operating pressure is normally always under 140°F excluding event resulting in overpressure or relief, you do not need a lifting lever.

 

In general, on a cooling water heat exchanger, its mostly during event that will ultimatly lead to an overpressure or relief that your temperature will rise to a temperature higher than 140°F, such as thermal expansion or, to some extent, fire case.

 

2) Chaterring could ultimatly render your PSV inoperative. The most probably mode of failure being unability to reseat completly or misalignment that will result in continous leakage from the PSV with very low inlet pressure. It cannot be excluded that a PSV will stuck shut.

 

BUT!

 

I think the main points you can make to your client why chattering in this case is of very low consequence is  :

 

The only way your PSV will fail because of chattering is if you have a relief event. These are very easy to avoid when if come to water side of heat exhanger and procedure are most likely already in place to avoid such events... (because its a mismanipultation of valves that will most likely result in the thermal relief event)

 

If the relief event lead to the failure of the PSV (and it is unlikely to), the cost of replacement for a small thermal PSV is of no consequence ( material cost would be around 250 - 500 US$...).

 

Regards

 

Marc-Andre Leblanc

[fallah]

ryn376,

 

A PSV is a safety device, then if it is going to be damaged due to resonant chatter; the valve can self-destruct unless the pressure is removed from the valve inlet. PSV destruction will certainly lead to a critical safety issue which necessitates preventing the valve to be subject to such situation...