PSV for dosing pumps

[shvet]

Good day, forum

We have dosing positive-displacing pumps and PSV located at pump discharge. PSVs discharge back to pump suction. The problem is that pumps have variable suction pressure and I have an issue concerning PSV type and differential pressure. For example:
- PSV set pressure 10.5 barg
- normal suction pressure 350 kPag
- max suction pressure 995 kPag

I specified balanced bellows PSVs with such design that can assure variable superimposed backpressure does not affect set pressure. Vendor can not provide this and asked to approve constant superimposed backpressure 195 kPag.

Dosign pumps are widely spread. Please comment - was I correct? Can a manufacturer porovide a balanced bellows spring PSV for a dosing pump with variable differential pressure? In case of my fault - please advice type of PSV.

 

[georgeverghese]

Balanced bellows PSV can maintain relief setpoint in services with up to 80% of set pressure as backpressure, which is well exceeded in this case when backpressure is 995kpag. If I'm not mistaken, a pilot op PSV also isnt good for this case. Check with the PSV vendor to confirm.
In any case, industry practice is not to exceed 50% of SP as backpressure for balanced bellows PSV, since the capacity correction factor falls off steeply after 50%. I've used 80% of SP for balanced bellows PSV only when desired relief flows are very small in comparison to estimated corrected RV capacity, as in thermal relief cases. This case you have is blocked discharge flow, much larger than thermal relief.

How does the max suction pressure rise up to 995kpag, can you limit this to 500kpag or so?

 

[don1980]

There's a mechanical limit to the amount backpressure a balanced PSV can tolerate. That's true regardless of whether it's a balanced bellows PSV or a balanced piston PSV. When the pressure limit is exceeded, the mechanical force on the bellows (or o-ring in the case of a balanced piston design) will cause the bellows or o-ring to fail. At that point, not only is the valve unbalanced, but you have an open path for process fluid to leak to the atm through the bonnet vent opening.

So, a balanced PSV is not an option for the pressure values cited for this particular application. You're options are to use a pilot operated PSV or an instrumented HIPS design. I would use a pilot operated PSV. The max backpressure in your application is 95% of the set pressure. Pilot valve vendors will cite a maximum backpressure limit for pilot valves too, but that limit is very close to 100% of set pressure. Theoretically, a pilot valve will operate just fine, all the way up to 100% backpressure. The only reason manufacturers cite a limit (~90-98%) is because of internal friction in the valve. So I wouldn't be concerned about the functionality of a pilot PSV in this application. The worst thing that can happen is that the system pressure may exceed the set pressure by a few percentage points before the internal friction is overcome, but from a safety perspective that's an insignificant risk.

 

[georgeverghese]

Engineering standards at one large multinational OpCo limit backpressure to a max of 70% of SP for pilot operated PSV.

 

[shvet]

How does the max suction pressure rise up to 995kpag, can you limit this to 500kpag or so?

I am a detailed design Contractor. Licensor specified a positive-displacing pump for periodic pumping a chemical from a vessel (auto on/off). Vessel works with variable pressure 300-400 kPag. Vessel's PSV set pressure is 950 kPag. In pump datasheet Licensor specified max suction pressure 995 kPag (950 kPag SP + 45 kPag liquid height). Licensor specified design pressure of downstream equipment 1050 kPag and specified that pump shall have PSV set pressure 1050 kPag.
As you can see I have nothing to do with this situation.

I will discuss a pilot PSV with my Vendor.

 

[LittleInch]

Or reduce the vessel psv set point?

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[georgeverghese]

If there is a PSHH on the source vessel set at say 500kpag, and there is also a PSHH on dosing pump discharge set at say 900kpag,( and either PSHH trips the dosing pump) then you may argue that it is very unlikely that the pump will running in this scenario where this PSV is triggered coincident with this very high pump suction pressure.

You may also try to find out why a vessel that operates at 400kpag max must have a PSV set at 950kpag. You may be surprised how some design specifications have no or a poor rational basis - all you have to do is ask.