Industrial water intake generalizations 1

[elaughlin]

I need to determine avg cost for each user for every one foot drop of pool given only normal pool elev., intake elev. and permitted withdrawal amount (and general energy costs) - no pump info, pipe size, or anything about the system above the normal pool. What generalizations can I use, which I will state as given or assumptions, to accomplish this? I also have a few sideline questions, though they are actually a part of this.

1. At what level of water above the intake should I consider cavitation a problem?

2. As pool level falls, the head increases and the pump output decreases. How is this normally handled for minor fluctuations? Can power be increased to compensate?

3. My experience is minimal in this area. Can I just use the given head (normal pool minus intake elev) and given withdrawal amount to determin WHP, then use that constant WHP to determine output changes per foot differences in head?

Any help will be greatly appreciated.

Thanks!

 

[Artisi]

I don't think there are any generalisations that can be made although you could possibly apply some engineering to the problem once more facts are available;

1. the pump type and installation configuration.
2. pump performance curve
3. the design parameters - static head - total head and flowrate at the design point.

Given this info we could make suggestions on possible cavitation, output in terms of flow at various inlet levels and power requirements.

 

[BigInch]

There arn't any generalizations.

As the pool level falls, you lose suction pressure and discharge head tends to experience a correspondingly similar loss. If flow reduces, the pump will back up on its curve, thereby reducing flow. Suction pressure (pool level) or pump speed will have to increase to maintain flow. You can do that as long as you don't overspeed or begin to lose suction pressure again. If your pump is constant speed, you have no increase of speed option and will be forced to allow pool level to increase, if it can do so, to regain flowrate.

Look at the pump curve and determine over what flow range it is operating based on the above variation in suction pressure, then look down at the power consumption between the hi and lo flow range. Figure the cost of power at somewhere near the average of the hi and low power consumption rates.

Be sure the intake is below low water level enough such that vortexing is avoided and NPSH is always above what is required.



If you were plowing a field, which would you rather use? Two strong oxen or 1024 chickens?" - Seymour Cray (1925-1996), father of supercomputing
***************

http://virtualpipeline.spaces.live.com/

 

[elaughlin]

Thanks to both of you. Of course, you are correct. The purpose of this work is to provide an economic impact analysis of 'loss of pool level' effects as a justification for maintenance and improvement work on the locks and dam. I have looked at potential damage to docks and loadout facilities, private residences, embankment slopes, bridge piers and others, each with some degree of generalizations appropriate to the scope and scale of the work. What I am trying to do here is present a conceptual cost for all of the intake users combined, using what little detail I am given. Withdrawals range from 0.5 MGD to 32 MGD. Perhaps I could assign a specific pump for each range of output, i.e. 0.5 to 5 MGD, 5 to 10 MGD, etc. and use the specs for the specific pumps, assume variable speed, and calculate the additional power required to maintain the given historical or max. intake, and using the NPSH curve for each pump. When these parameters are exceeded, then supplemental measures would be required. This is engineering at the outskirts, I understand. But I will have to state my assumptions and qualify my answers subject to actual conditions unknown to me. This is all I have, and all I'll be given because of post-9/11 security concerns. If you have any suggestions I would be grateful.

Thanks!

 

[BigInch]

It helps if you have specific pumps, but increases the work load too working with all those curves and efficiencies. At this stage of your game I'd just use some genaric constant speed pumps working at BEP with an average efficiency of around 0.725 and estimate the HP required for each flowrate in your range on that basis alone.

If you were plowing a field, which would you rather use? Two strong oxen or 1024 chickens?" - Seymour Cray (1925-1996), father of supercomputing
***************

http://virtualpipeline.spaces.live.com/

 

[Artisi]

Maybe I'm not seeing the picture and repeating myself - but it seems that what is being asked without any design data and pump/s performance curve is not possible.

To establish anything worthwhile - pump curve/s are required and 2 things must be know to approximate power input and change to power input from variable inlet levels.

1. the flow rate/s
2. the head - preferably as - nominal inlet head,static discharge head and friction head at maximum flow rate (32MGD).
With the above data a system curve can be drawn and changes in inlet head and change in friction head for the various intermittent flows and power input can be approximated.

As for NPSH, are the pumps installed with negative or positive inlet head, ie, "suction lift" or "flooded suction" - this also makes a big difference to likely outcomes in terms of NPSH.

I see the problem you are faced with as the same as going into a store to purchase an item, the sales person hands it to you but doesn't know the price - you hand then $100 - how much change do they give?

 

[BigInch]

Artisi, He has to start somewhere. Essentially what he must do now is construct a pump curve he will match later with a good selection of (possibly multiple) pumps and an optimized configuration. So its possible he is constructing a pump curve with multiple pumps. Pump details arn't important at this stage. If you only operate each one, or each configuration at BEP, curves are irrelevant at this point. Suction press is usually a small % of total head, so even if it varies, it can essentially be ignored for now. Presumedly the pipe lengths and static head requirements are known now and the frictional drops can be assumed to not be more than some allowance. The discharge piping will later be designed with a length and diameter that does not exceed that allowance and will not produce a large difference in frictional drops between the range of flows finally selected. If flowrates vary significantly, there may be some worthwhile variable speed options that could be considered too. We're just defining a series of reasonablye sized boxes to make up the system and later we'll fit the equipment we really need into each box. Until then, we'll have a good idea of what the power cost is.

If you were plowing a field, which would you rather use? Two strong oxen or 1024 chickens?" - Seymour Cray (1925-1996), father of supercomputing
***************

http://virtualpipeline.spaces.live.com/

 

[Artisi]

BigInch, I am reading this as an existing installation and the OP is undertaking a study of the effects of a reducing water level.
"The purpose of this work is to provide an economic impact analysis of 'loss of pool level' effects as a justification for maintenance and improvement work on the locks and dam."

Maybe I am way off the mark with this assumption.

 

[elaughlin]

Artisi, your assumption is correct. They are existing intake structures, and I know virtually nothing about them. In fact, the intake elevation won't even be provided to me. I have to develop a spreadsheet to provide to the Client so he, who has the proper security clearance, can enter the intake elevation and withdrawal amount for each intake. The spreadsheet will calculate an approximate additional cost for each structure per foot of pool loss, and sum the totals for different reaches of the length of the pool. The Client will then give back to me the $$ for each reach for each foot of drop, for use in my report. So the $$ amount will not be accurate for any one intake, and not very accurate even for the lot of them. To at least be in the same city, if not the ballpark, I thought I would try to find a 'typical' pump for different ranges of withdrawal amounts the Client will enter, and work those pump characteristics into the spreadsheet with 'if-then' statements. Are they centrifugal pumps? Vertical turbine pumps? I have no idea, and not even enough experience to hypothesize. Can one of you suggest a good on-line source of pump selections and curves/charts for a wide range of discharges? I'm having a hard time finding them.

Regardless, this is what I have to work with. It's not the kind of engineering we like to do, but it comes close to being appropriate for this small $$ component of the economic impacts I'm looking at. Thanks to both of you for your help.

 

[rmw]

Put a place on your spreadsheet for him to fill in "minimum submergence required" for whatever pumps he has to take under consideration. If his pumps are intake structure type they are typically submerged and any submerged pump has a value for minimum submergence below which it will begin to suck air due to vortexing at which time there will be some financial considerations.

That should be his limiting factor in my opinion. Put the field and let him chase his pump information for the limiting criteria.

rmw

 

[Artisi]

For an applicaton like this I would think vertical turbine pumps mounted in inlet structures but for your exercise it doesn't matter, other than NPSHr/a considerations because of the difference in vertical and horizontal configurations the pump hydraulics are the same .
The following link will put you into Gould pumps who have a range of large vertical and horizontal pumps suitable for these applications - not sure if you can get curves but worth a look.

http://www.gouldspumps.com

-

These pumps are from what was Allis Chalmers Custom Pumps - no contact names unfortunately as I have been out of the pump business for quite some time.

Good luck - I think you will need it.