Backpressure on PSV 1

[RJB32482]

I have a PSV with a set pressure of 90 psig with the case being blocked outlet. I want to pipe the relief header back into the suction side of the pump. Would the backpressure on the valve be the net positive suction head on the pump? Any induced backpressure on the valve from it opening? Thanks.

 

[Latexman]

Would the backpressure on the valve be the net positive suction head on the pump?

No, the BP on the PSV discharge nozzle is not the NPSH.

Any induced backpressure on the valve from it opening?

Probably.

Backpressure needs to be calculated as the "pressure at a point" with the PSV [green]closed[/green] and using the "worst case" of your normal operating limits and parameters that may affect backpressure. The "point" is the PSV discharge nozzle.

Superimposed backpressure is bacpressure that comes from other sources, like a pad gas in a discharge header going to a flare or when a different PSV discharges into a header. You probably don't have any in your particular case.

Built-up backpressure needs to be calculated as the "pressure at a point" with the PSV [red]open[/red] and using the "worst case" of your normal operating limits and parameters that may affect backpressure.


Good luck,
Latexman

 

[RJB32482]

Thanks for the reply. Let me give an example for my understanding:
The relief valve is set at 90 psig. The discharge header is a 1" Sch 40 SS line with 15 ft of straight pipe and 1 90 degree elbow (10 ft vertical drop). The discharge header discharges into the suction side of the pump. The relief max flow is 9.8 GPM. Viscosity is 2300 cps and S.G. is 0.98.

How would I calculate the backpressure on the valve? Do I need to know anything about the inlet PSV header? (Let's say its 10 ft of 1" pipe Sch 40 pipe all straight vertical rise) or the suction pressure of the pump?

Thanks

 

[EGT01]

RJB32482,

As a general comment, I find your description of the relief valve inlet and outlet piping as a "header" to be a little confusing. A header is typically a pipe to which several other pipes might be connected but I don't get that impression from your example. From your example, it sounds like you may have a positive displacement pump with a relief valve on the pump discharge with the relief valve discharging back to the pump suction, is that your system?

Latexman is correct about having to account for the superimposed and builtup backpressures but if the system is a positive displacement pump with a relief valve discharging to the pump suction then superimposed backpressure can be a significant concern.

When talking about systems like the one I've described for a PD pump, it is more typical to think of the relief valve setting in terms of "psi" or "psid" where "d" indicates differential rather than psig (gauge). To be on the safe side, be sure and check what the relief valve cold differential test pressure is to see if there is a difference from what you find for the set pressure.

If for the moment, we neglect the builtup backpressure, then if the pump suction pressure is 10 psig and the relief valve is set at 90 psi then the relief valve will not open until the relief valve inlet pressure is 100 psig. It then becomes important to know the maximum pressure for the pump suction to make sure the valve is set properly to protect equipment and piping downstream of the pump. What's the source for the pump suction? Does the source vessel have a relief valve and what is its set pressure? Unless you have good reason not to use it, you may want to consider the pump suction pressure equal to the source vessel relief pressure plus any static head effects.

With the relief valve setting properly accounting for the superimposed backpressure, you only need to check that the builtup backpressure (psi) is less than or equal to the amount of overpressure (psi) used.

If this doesn't help, come back with more details.

 

[Latexman]

Let's keep it simple and only consider that someone inadvertantly closes a valve in the pump's discharge line AND that the PSV was set to pop at 90 psig with 0 backpressure. Nothing is going out, so by mass balance, nothing is coming in from the suction source. Therefore, at the point where the PSV discharge line tees into the suction line the pressure is equal to the pressure of the suction vessel + the liquid head imparted by the suction liquid. Let's call this P1. With this, you can calculate the built-up backpressure at the PSV outlet connection. It's P1 + the pressure drop at the flow rate of the pump through 15 ft of 1" Sch 40 SS line, one 90 degree elbow, and 1 Tee (Is it a branch or line run? Probably a branch run.) minus 10 ft of liquid head. Let's call this P2. At the PSV inlet connection, the pressure is P2 + 90.

Good luck,
Latexman

 

[ruwl]

Can you select a balance bellow type valve and forget about the backpressure in your calculation?

 

[RJB32482]

OK Thanks to everyone for this post. It has helped me a lot since I have not worked on designing a relief valve system (we do have outside contractors that will check my calcs, but I want to do it right the first time of course).

The vessel is atmospheric, so the pressure is 0 psig. The liquid head is around 10 feet. So for simplicity if its water, the head pressure is 4.33 psig from the vessel. At the worst case flow rate of 9.8 GPM, the pressure drop in the suction line to the tee where the PSV discharge is located is 1.2 psig. So in Latexman's example, P1 is 3.13 psig. Also, the pressure drop in the discharge line of the PSV at 9.8 GPM is 0.5 psig. So this means the backpressure is 3.63 psig, correct?

Or don't you take in account the frictional losses in the suction side from the vessel to where the discharge line enters? Or am I correct how I approached it?

Thanks again.

 

[Latexman]

Or don't you take in account the frictional losses in the suction side from the vessel to where the discharge line enters?

No. With a closed discharge scenario, there will be no flow from the vessel to the tee where the discharge enters, therefore there is no frictional losses. If the highest liquid level in the vessel is 10 feet above the centerline of the suction line where the diccharge tees in, P1 = 4.33 psig. P2 = 4.83 psig. PSV inlet = 94.83 psig.

Good luck,
Latexman

 

[Latexman]

Btw, these pressure drops don't feel right for a 2300 cP fluid. I don't have the means to check this at home easily. I hope these were just for illustration only.

Good luck,
Latexman

 

[RJB32482]

Yes, they were just an illustration. Thanks for the reply. So for a PRV discharging to atmosphere, would the backpressure just be the pressure drop in the discarge line?

 

[Latexman]

You really need to differentiate between the various types of backpressure. In this case:

Backpressure = 0 psig
Superimposed backpressure = 0 psig
Built-up backpressure = dP in the line

Also in this case, the set pressure would NOT be adjusted so it pops at the right pressure, but you do have to select a PSV that meets the built-up backpressure.

Good luck,
Latexman

 

[RJB32482]

So would I need a higher set pressure for my blocked outlet case if I had a backpressure of 4.13, and built up backpressure of 0.5 psig if the relief scenario is 90 psig (maybe 95 psig)or will I just need a valve to set at 90 then be able to sustain the backpressure

 

[Latexman]

By your example if you *must* have an opening pressure of 90 psig, you'd have to have a set pressure of 85.87 on the bench with 0 psig backpressure. Since a conventional PSV can usually handle up to 10% backpressure, that's what you'd need, unless you wanted a bellows just to keep stuff out of the spring area.

Good luck,
Latexman