Pumps In Parallel 4

[Chris0164]

We are doing a big revamp job here in Anwterp and we have two existing vertical turbine pumps in the CW circuit. Now we have to expand the capacity of the CW system and want to purchase two additional vertical turbine pumps, however of course the original suplpier is not the cheapest. Has anyone experience with four pumps in parallel but not from the same brand ?? (means different curves)

 

[MortenA]

In this case i think you will both pro and contra.

When looking at the cost of the pump have you considered requirements for spare parts?

If the pumo e.g. require that you keep a "warehouse spare" then you must include this pump - but for the original pump you will allready have the item in the warehouse.

The flow between the various pumps will not be equally balanced. This may, however, allready be the case due to piping? The existing pumps will get a new operating point when the new pumps are due to this. This may not be as close to the optimum efficiency as before. Has this been included in the OPEX considerations?

Best regards

Morten

 

[katmar]

Chris,

This sort of thing is done often. It is not as good as having four identical pumps, but it can be made to work.

Remember that the four pumps will all see the same head and each will find its own flowrate from its own curve. It is important that you get a good balance between the pumps not only at the design point, but also as the demand varies. If the shape of the curves is different you can suddenly find one pump doing no work at all, or suddenly tripping on overload.

In my experience, you can often buy identical pumps from different suppliers under slightly different model numbers. If this is the case I would be happy to take the alternate (cheaper) brand.

Morten raises some very good points about the TRUE cost difference, and even if the pumps from the original supplier are more expensive after taking Morten's factors into account it might be worth paying a premium for the simplicity and reliability of all pumps being identical.

All of engineering is an economic compromise.

 

[stanier]

You need to model the system with the four pumps over the expected operating range. If the shape of the HQ curves is markedly different then you may experience problems in one or more operating scenarios.

I would also question why one pump is so much cheaper than another to allow you the luxury of introducing a new manufacturer into the equation. Have they same level of For example:- engineering staf, expertise as the former company?? Many studies have shown that the initial price of an item such as a pump only rperesents 10-12% of its life cycle cost. Have you done the life cycle numbers?

In Australia the market was skewed here for progreesive cavity pumps as some fly by nighters came in sold undersized high speed pumps at lower initial prices and then slugged the customers for spares. The more conservative and professional company offered slower running speed pumps that would outlast their competitors, thus requiring fewer spares and the price of spares was reasonable.

DONT TAKE FIRST PAST THE POST PRICES AS INDICATIVE OF THE BEST DEAL.

 

[rmw]

In todays business climate, OEM's change their way of manufacturer, sources of castings and foundry items, and optimize and/or cost reduce their designs so that there is no guarantee that the two new pumps would be 'drop in' identical to the older pumps.

That said, several good points have been made, but the most important one was to analyxe the curves throughout the range.

Also, analyze how the newer pumps will be adjusted for wear during their lifetimes. The way they wear and are adjusted to compensate for wear, as well as their repairability (ability to be brought back to their original curves after significant wear) should be a factor in your selection process.

But to answer your question, I see different branded pumps running in parallel quite often, some successfully, and others not.

Be careful.

rmw