Drilling & tapping an elbow for relief valve

[gboone]

An outside entity is designing and constructing a new plant addition within the property of our existing plant. This new addition is supposed to be per B31.3 piping code. We have observed that several 4" piping elbows have been drilled and tapped in the field for 1/2" threaded thermal relief valves. I attached a photo. I've never seen this practice and suspect that it is not allowed by B31.3. Can someone confirm or deny this? If not allowed by code, where specifically is this stated in the code? Thanks for any help.

 

[MTPipeliner]

Well,

I don't have an answer for you, but that picture is not at all what I was suspecting.

Initially I was going to suggest an elbow-let, but after seeing the picture that won't work.

 

[LUDPEKA1]

I’m sorry that I can't help you when it comes to the code issue. That's the first time I've seen that configuration and I know it would not be allowed in the refinery I work at. It even gave my boss with 30 plus years of piping experience a chuckle.

 

[Zoobie]

Like PatrickR, I don't have an answer with respect to code. I have seen a lot of pipe and relief valves but I have never seen anything like that before. My engineering spidey-sense is telling me its wrong. Piping code aside, why on earth would you put the valve there? If you find out the answer, I and perhaps others would be interested to know what the design engineer was thinking.

I certainly hope that someone with the code knowledge chimes in. I flipped through my copy but nothing jumped out at me...though I am not even close to an expert on this code.

Good luck!

 

[pennpiper]

What company in what country did the engineering and design?

What is the commodity, at what operating pressure and temperature?

 

[gboone]

Engineeering and design was provided by a local engineering firm (obviously not well versed in industrial piping codes and standards) in the U.S.

The fluid services vary from diesel fuel to mineral oil, used as emulsifiers, close to ambient temperature and 25 psi operating pressures. The blocked-in pressure must far exceed 25 psi as the temperature rises, effecting the need for thermal relief valves.

 

[Latexman]

I've never seen that before either. It might be worthwhile to ask them what part of the Code allows this, and if none is forthcoming, ask them to modify it to meet Code.

Good luck,
Latexman

 

[zdas04]

Yeah, put it back on them. I have never seen NPT threads with the female threads 0.237 (S-40 4-inch) long. Typical is around 8 times that long and they still leak if not sealed properly. The end of the PSV nozzle has to protrude into the pipe far enough to perturb the flow--vortex shedding can create a powerful scouring force.

I'd require (even if the code does allow it, I wouldn't tolerate the corrosion and leak risks) an elbow-o-let on the outer radius.

David

 

[jte]

I'll agree that this is unusual and would not accept them for a plant that I was responsible for. I think David’s comments on thread engagement length and vortex induced vibrations are particularly relevant.

But I like to play Devil's Advocate at times... I can't find a quote in B16.9, but if these were flanged elbows per B16.5, the tapped holes would be allowed:

6.12.2 Pipe Thread Tapping. Holes may be tapped in the wall of a fitting if the metal is thick enough to allow the effective thread length specified in Fig. 3 (Fig. II-3 of Mandatory Appendix II). Where thread length is insufficient or the tapped hole needs reinforcement, a boss shall be added.

B31.3 allows threaded plugs as noted in Interpretation 19-48. So, I would say that it may not be that big a stretch to argue that it could be in conformance with the minimum requirements of B31.3. Keep in mind that the minimum thickness for internal pressure is probably less than 10 mils on a straight pipe.

Now, before I get jumped, let me remind y’all that I don’t agree that this is good practice… And I’d question where the process fluid is discharged when the PV burps.

jt

 

[zdas04]

Good points jte. I'd also like to point out that the code represents the minimum safe practice. They make no claim to be optimum (too many non-physical components of that calculation). I've done code calcs and found that pipe which is one schedule lighter than what is generally available would work, but the project delay's associated with waiting for the unusual (in our area) schedule just weren't worth the cost difference.

I think the same holds here.

David

 

[AllHandlesTaken]

No comment on the Code aspect or the best practices part - but I would guess they have done this for protection. The inside radius is safe from overzealous forklift drivers, or even someone catching their leg on it.

 

[MikeHalloran]

I've seen something similar done, on a 2" elbow, on the 'side', normal to the intersecting axes, for a gage valve. I can't remember whether the gage valve was 1/4" or 1/2".
Medium was fuel gas at ~10 inches water.

I nearly had a heart attack, but the 'old hands' said they did it all the time.

I can't believe any code would allow it.



Mike Halloran
Pembroke Pines, FL, USA

 

[ebbal]

Looking at the pic, it appears that the PRV is 1/4". Assuming that, I believe the would be suffient thread to achieve a good seal. I have seen PRV's put on fuel lines on the suction side of pumps for the exact reason you stated. If on the discharge side of the pump I would require an internal relief on the pump. with 25 PSI you must have a 50+ foot tank.