Calculate the effect of suction stabilizer on NPSHa

[lewanj]

thread407-256219

Is it possible to determine by calculation how much acceleration head will be mitigated by the use of a suction stabilizer or is this something that really needs to be determined experimentally?

 

[ccfowler]

You can compute a reasonable estimate of the performance by setting up reasonable equations to model your specific system using a spreadsheet such as Excel. This will probably be a fairly tedious exercise, but be very careful in the idealizations and assumptions that you include in your scheme of computations. Inadequately considered simplifications can easily negate the benefits of this computational exercise.

As always, pumps and their associated devices are inherently an intergral part of the greater connected system. The actual performance characteristics of pumps in service are always entirely interactive with the system as a whole. Pulsation attenuation will always be highly dependent on the specific details of the piping connections.

Valuable advice from a professor many years ago: First, design for graceful failure. Everything we build will eventually fail, so we must strive to avoid injuries or secondary damage when that failure occurs. Only then can practicality and economics be properly considered.

 

[micalbrch]

Yes, basically you can do that. As a rule of thumb a residual suction side pulsation of less than 5 % leads to an negligible acceleration head.

 

[lewanj]

All,

Thank you for your responses.

I agree that "reasonable equations" can lead to a "reasonable estimate of the performance" however where does one find these reasonable equations? Based on one of the responses in thread407-256219 I have ordered a book which sounds like it might be useful (unfortunately I haven't received it yet).

I have already set up a spreadsheet which models the effects of acceleration head, pipe & fitting losses, and elevation head. The problem is that I'm uncertain if my model of the suction stabilizer is accurate enough. The other problem is that the vendor is less than forthcoming with information on this unit.

Once again thank you for your help.

 

[ccfowler]

lewanj,

If your supplier (or manufacturer) is obviously not being helpful or truthful (allowing for normal sales & marketing nonsense), it is time to find another source. If they are trouble before they get your money, you can count on them being much worse after they have your money.

Get information for multiple competitors' products and pay some close attention to the range of performance claims for seemingly similar devices. If the claims of one or two manufacturers seem seriously inconsistent with the claims of others, it is likely that the majority are likely to be more credible. You may still eventually get the device from a supplier with questionable performance claims for other valid reasons, but you may want to discount their performance claims and choose a more suitable product from their line.

As far as your computations are concerned, it is best to rely on known, proven equations (dust off your old fluid mechanics, thermodynamics, ... textbooks) and put those many hours of classwork and homework to some good use. Where you are unsure of information on particular components' performance, some well reasoned guesswork will be in order. Try some reasonable approximations and you should be able to set some reasonable upper and lower bounds for likely performance. If what you need is somewhere in that presumably credible range, then most likely, the results your wise & prudent guesses and estimates should guide you in determining whether the planned system can reasonably be expected to function adequately.

Doing some serious "playing" with parameters in your spreadsheet, you may find that some other elements may be more important than you initially realized. For example, there may some surprising benefit to changing the suction line size or revising some of the piping configuration.

Valuable advice from a professor many years ago: First, design for graceful failure. Everything we build will eventually fail, so we must strive to avoid injuries or secondary damage when that failure occurs. Only then can practicality and economics be properly considered.

 

[waterpipe]

I would like to mention that you can do a CFD modeling for your suction system. If you've dealt with CFD before and have a proper software this would take only few days to model your current system and then implement the modifications to see the result in term of streamlines and losses. Actually I've done it before for a suction of a pumpstation.
Consider giving it as a mini-project to a university around your place, where they have a software and, interested students. usually they appreciate this kind of colaboration with industry.

 

[stanier]

A good reference is The Reciprocationg Pump Theory Design & Use by Miller.