Back pressure valves for Chemical Injection Pumps

[vjr0512]

For an offshore project, we have ordered chemical injection pumps and during the course of detail engineering, it was identified that for some of the injection pumps we require back pressure valves in order to avoid siphoning effect from the downstream to upstream.

The concern now is that these back pressure valves are normally constructed with the screwed housings (for piping inlet and outlet of these valves). Since this valves are going to be used for North sea offshore, NORSOK compliance is a must that does not allow screwed piping .

Client have apprehension that as these valves are of very smaller sizes, it may have vibration problem and breakage of lines and screwed joints / piping are a deviation to NORSOK

We have checked with many suppliers of these Back pressure valves but none of them are providing with the flanged body construction.

Thanks to guide whether the apprehension of the client about the small bore and vibration problems are correct and whether there are BPRs with flanged body construction?

 

[LittleInch]

There are a few things that puzzle me here:

1) I can't work out why you want a back pressure valve in this instance. surely a non return valve would do the trick or are you actually concerned about chemicals leaking through the pump into the process when the downstream pressure is very low?

2) I've had a quick look through some relevant NORSOK standards and can't see anything about not allowing screwed pipework. Is this a misinterpretation or simply a paranoid client? Please post the spec number and clause number or ask whoever told you this to provide yu with the details. It is quite common for people to spout incorrect data they "heard from xyz" or that "they read it somewhere". If someone tells you something important like this, do a it of research yourself to see if they have actually read "Norsok" or not.

3) If the client is worried about vibration just fix the pipework down with a few more clamps. I can't see why this would vibrate anymore that f it wasn't there. dpending on your pump you could have vibration / pulsation issues, but you would need to deal with these in any case.

My motto: Learn something new every day

Also: There's usually a good reason why everyone does it that way

 

[vjr0512]

Thank you littleinch for your reply. Yes we are looking for regulating the downstream pressure depending upon the injection point pressure in the oil well system for some of the chemicals.

Yes we have also seen that there is no specific mentioning about prohibition of usage of screwed joints. But it is not listed in any of the pipe classes of NORSOK for the screwed piping / fittings. CPY has indicated that it is a deviation to L-001 as it is not listed.

But what we could see from many manufacturers of these back pressure valves, they have the screwed female connection inthe valve body and i did not see any forged flange connection ends. Should there be any specific reason for the same? may be for manufacturing and machining process?

 

[JJPellin]

Forward flow (siphoning) can be a very real concern with these pumps. In addition, the pumps accuracy will be much better if you maintain a certain minimum back pressure. The pump manufacturer should provide the required back pressure to assure accurate and consistent delivery. I seem to recall a need for 20 psi minimum for some of our pumps. if the pump can ever see a condition where the back pressure is lower than the suction pressure, you can over-deliver the chemical. This can occur even if the pump has spring loaded check valves.

Johnny Pellin

 

[DubMac]

Unfortunately, threaded connections are becoming "outlawed" more and more; many times for no reason other than lazy engineers trying to cover their butts, and costing their companies small fortunes.

Nonetheless, you will have to comply with whatever it is the OWNER wants; regardless of Norsok, API, or anything else. Just ask the owner if the threaded connections are permissible and let him know how hard it is to find, and how expensive, to use flanged. They should have their own set of specifications for chemical injection systems. Let them make the decision; guessing and supplying something they won't allow can become expensive for you.

 

[GHartmann]

Considering the small flow for these type of systems, it is very hard (if not impossible) to find in line devices with flanged connections. The small size of the devices and pipe or tubing make it near impossible to construct such a device. You may have even had trouble ordering your pumps with flanged fittings.

I would ask your pump vendor if they could be of assistance. Otherwise you probably need to ask for a waiver.


+

 

[LittleInch]

The key issue is the screwed piping. I had a look at L-001 and whilst the piping specs don't particularly mention it, they do list ASME B 31.3 as the design code, which I'm sure does allow screwed joints.

Also if you look hard enough, you can find in the valve specs some small valves with Female threaded ends (look for the "T" at the end of the valve spec. If you have threaded valves then you have to allow threaded pipe....

If CPY has a certain view about threaded joints then apply for a deviation noting that there is no such prohibition in the codes, it allows use of B31.3 and you can't find a supplier. Small components are not commercial if fitted with flanges and would weigh two or three times as much. There are issues with threaded joints on vibrating / pulsating pipework, but they can be overcome by jointing compounds, pulsation dampers and good fixing o tubing / pipework.

My motto: Learn something new every day

Also: There's usually a good reason why everyone does it that way

 

[bingopin]

I agree that threaded connections can be problematic and should be avoided if possible except for utility piping. Having said that, allowing zero threaded connections doesn't seem feasible. How do you make a tubing to pipe connection or pipe to pump seal gland connection?
Part of the problem with threaded connections is the reduction in pipe wall thickness caused by cutting the threads. This can be mitigated by requiring the use of xxs pipe at the connection.

 

[DubMac]

While not involved directly with API 610 anymore, what I heard at the TurboMachinery show last week in Houston was scary. Apparently the API 610 committee is looking into banning ALL threaded connections on seal flush piping. Yes, to the gland, and also to the pump case! Welded and flanged connections will be all that is allowed. Don't know how far they have gotten with this, but it would seem a huge overstretch for API 610 to implement. SOme engineers just love that paperwork!

 

[vjr0512]

Thanks a lot for all the replies and very useful suggestions for the way forward. I have decided to go to client with a request for waiver and specifically the points highlighted by litteinch about B31.3 and also the acceptance of the threaded valves.

Thanks once again