Well Pump Efficiency

Not my field of expertise at all, and I don't know the size of these pumps so this is just my humble Catly opinion, and accept my oppologies if I'm waaaay off the mark.

The manufacturer makes them and tests them...
Do you really think that he or you can suddenly improve the efficiency of what is presumably a standard production item without vastly increasing the cost?
Things work as they are designed and made, if you want high performance/high precision you will have to pay for it.
Messing around with bits of paper, figures and specifications won't alter how the pumps actually perform.
Talk to the manufacturer ask for his recomendations/guidance...

To answer your questions in order.
1. No idea. Was the specified figure sensible in the first place, or was it just an ideal or wet finger guess? Have you got manufacturers production test figures to show what the pumps usually give?

2. There are two questions here, a) probably. b) I don't see how on earth you expect to increase efficiency on site short of race tuneing the pumps individually (joke).

3. I have no experience of this but I have a KrisDel^TM degree in Common Sense (available in a multitude of colours to match your decor)
Del

The well pumps are identical and and should have the same pump characteristics

BTW. The above statement show a poor undersatnging of engineering.
The pumps aren't 'identical', they are doubtless the same model and will be 'similar', but the parts may well be from different batches of castings, there will be small diferences in every single part.
They will perform similarly and their performance can be characterised and analysed by statistical methods...
Del

I assume that your firm placed the 80 percent mark on efficiency in the bid request. The manufacture then bid on this to supply pumps to meet the bid. If the manufacture in fact did this, then I would not accept the pumps based on the resulting performance test. There again it depends on who set the specifications on the project (assuming there were bid spec's) when it was designed.

Pumps are in most cases held to a pretty good tolerance. They may vary from one manufacture to another slightly, but all in all the test curves should hold up to the design.

Your other question on improving the efficiency on site is near impossible. the pumps will be what they will be. You should have a pump test curve supplied by the manufacture that shows what the unit will preform to.

I'm no expert either, but, whether it's water pumps or anything else, actual efficiency rarely matches advertised efficiency. The advertised efficiency may well reflect the best they have ever done with one of these pumps.

Efficiency is also going to drop in direct relation to the depth of the well. Is it okay to accept? Maybe.

1) Will they accomplish what you need done?

2) Will the factory stand behind their pumps with a written warranty?

Whether higher quality pumps come from this supplier or somewhere else, they will definitely cost you more money. Ultimately, this will be a judgement call on your part.

Guest - Del is right. You may have specified 80%, but the efficiency is in fact given to you by the pump manufacturer for his pump at a set of conditions. If you do not like what he has to offer then you look for another manufacturer. You do not give the head or flow so it is difficult to guess whether 80% is even a reasonable efficiency. The only efficiency that is to be believed is the one from the "witnessed" (I hope) factory tests. All the others are pretty intelligent guesses on sales documents. There is a very good reason why the API gives such a large tolerance on efficiency - see Del's post.

Note that for API 610 (which is a very stringent standard), the efficiency is not a rating value (it does not specify a tolerance).

1) Low efficiency can be caused by increased clearances, roughness of internal wetted parts etc.

2)Yes it is OK to accept them if you accept the efficiency, which you have to as it is within the tolerance specified by your standard. There are things that can be done to pumps to improve efficiency (like smoothing surfaces, reducing clearances of wear rings and or back clearances etc., but the answer is no, there is nothing (sensible and cost effective) that you can do to improve the efficiency on site.

3) Unless one knows flow and head this question can not be answered. Efficiency is very strongly associated with Ns (Specific Speed) as shown in the diagram below from Lobanoff and Ross.

But I can see it quite clearly from here. Its the dingleberry you have flapping over the deelybop that is causing the framitz to not efficiently squeeze the oooga of the water.

What are you up to? Reverse pumping smoke bubbles through your bong water perhaps? Dammit, I meant to check, off topic. Went back to do it and can't.

If it's true, at least you have those yummy sandwiches to eat.

Just having a little fun while trying to point out that on site troubleshooting can only be done by somebody on site.

On a more serious note, the use of the ambiguous term "efficiency" is often nothing more than a marketing ploy. This is because the efficiency comparison can be measured in a whole bunch of different locations. Of course efficiency will be a percentage ratio of similar units but which nodes will be chosen always gets to be the rub that permits marketing to choose their preferred values. (I will admit the cited standard may remove any ambiguity to this value. I'm not familiar with the standard.) To make my point here are a few ratios that might be chosen for a pump's efficiency (naturally all times 100%) :

• water velocity / impeller velocity
• water kinetic energy out / impeller shaft kinetic energy in
• water kinetic energy (out - in) / impeller shaft kinetic energy in
• water kinetic energy (out - in) / (electric motor real power * 1 second)
• water kinetic energy (out - in) / (electric motor VA power * 1 second)

I could go on, but I think I made my point.

_{Oh, and that was some nice mustard on that sandwich.}

One other point, if you decrease the clearances to improve efficiency, (such as replace the impellers with ones that run closer to the housing and smooth out the housings.) then you run the risk of having problems with solids handling causing the rotors to lock up on you if something gets jammed in the impeller clearance.

Have you considered that you might be conducting accepting testing and performance testing at different elevations relative to sea level and how this might impact the results?

Assuming these pumps are driven by electric motors, has anyone evaluated the quality of the incoming electrical power? Pump factory testing would be done based on some specified voltage and current and power factor. Insufficient wire feeder diameter could produce unacceptable line losses resulting in incorrect motor speeds and torque. On site acceptance trials must duplicate factory test conditions exactly or all bets are off and so is any claims of failing to perform to contract.

Hi, 1. Efficiency of pump is depend upon Head and Flow.Every pump is design for particular head and flow if any variation is there then efficiency will go down. Use this Eff = Head(m) x Flow(m3/hr)/3.6 x 102 x eff of motor x kW consumption and also see the pump curve it will give you better clarification. 2.increment of efficiency of pump its not possible but if you install correct size of pump according to your requirement you get better result. 3. Submersible pumps are design for 70-80 %.

1. Check the voltage supplied is as per the manufacturers voltage. You will get an exponential roll off in performance as the voltage drops.

2. Check the current as well [this could help you with voltage fluctuations]

3. The efficiency will be determined depending on the total head, which is the suction plus the output.

4. What sort of pumps are they? When you say 'electrical submersible' you mean, for example, they are multistage? If so, check your supply wire calculations and your connections to ensure there is no loss of voltage due to under-capacity cable (voltage loss in the cable) or doggy connections.

5. If they are multistage, make sure you install a water sensor at 1 meter below the pump so if the well goes dry, you will not fry the pump. A multistage pump requires water flowing through it to cool the motor, impellors, bearings etc. The sensor [Omron etc] will connect to a shut off relay on the power input side of the pump.

1 meter below the pump?????

that is just mean..

1-The pump is multistage or not you should check alwayes your wire size and connection carfully. (most of the time if it is a well pump you know that it is multistage)

2- you put you sensor cabel (high and low sensores ) alwayes above the pump. the motor is below the pump!!!

Guest,

I do not agree with the sensor placement you describe.

One sensor (cut off) is enough mounted on a 1 meter extension pipe on the intake side of the pump. Mount a non return valve with a rubber gasket (ie not a brass check valve) at the end of this pipe. The extension pipe should be the same size or larger than the intake diameter. Larger is better as the non return valve has a friction co efficient. Say the pump is 1 1/4" intake, go up to 2" using thick walled PVC or best, HDPE pipe.

Why? If you get air into the pump because the water level falls below the motor then placing the sensor 1 meter lower than the pump gives you a margin of error and you are less likely to loose prime and start cavitation which will effect the impellers in the stages.

Not all times is the pump multistage. Typically, because of the cost of multistage pumps, jets or venturis are used in Australasia. See www.davey.com.au/ApplicationandProducts/Deep_Well_Jet_Pumps.aspx

There is nothing more tedious than hauling up 60 or 100 or 200 meters of pipe and pump and cables out of the bore because someone rushed the install and forgot the basics.

Off topic but just as important: for multistage pumps installed in deep wells (over 30 meters) use 5 times the PN rating of HDPE pipe (not PVC or steel) and use a high quality compression fitting or welded fitting on the pump output and do it up as tight as a nun. Every time the pump starts it goes *wack* on the pipe from the startup inertia and tries to loosen the fitting. And use stainless steel wire rope onto the chassis of the pump so there is a way, even if the pump separates from the supply pump, you can always haul the pump out of the bore. Don't use galvanized wire etc. They just disintegrate over time.

My understanding of the original post is that the 72% was ascertained during the "factory" test.

If this is so then the comments with regard to elevation, voltage and all the other on site variables, though correct, do not apply, as all of these are taken into account in the test procedure. In service efficiency measurements are at best very rough indications of a pumps condition. It is impossible to get accurate enough readings from field instruments/set-ups to conduct a really meaningful efficiency test.

Efficiency with respect to a pump is the hydraulic power/shaft input power. (pghQ/P_input)

API RP11 is a very old "Recommended Practice" and has very large tolerances for acceptance. It is after all a recommendation. If more stringent requirements are desirable they should be specified or a standard such as API 610 used.

Pump test acceptance levels for API RP11:
Head - flow rate plus/minus 5% head AND plus / minus 5% flow rate.
BHP - flow rate - 8 % allowance
Pump efficiency - flow rate 90% of efficiency at rated flow!!!

All of this is pretty irrelevant.
It's down to 'are they fit for purpose'? yes or no.
If 'No' then reject them, if 'Yes' accept them but note your concerns and try to screw a little extra discount/extended waranty/free maintenance/crate of booze or some such from the manufacturer.

If you want a concise answer, ask a politician

Del

Hi

I install and design submersible pumps, so hopefully I'll be able to give you a few pointers,

You don't mention the gross production rate of the pumps or the make or type.

hopefully you are dealing with a reliable pump supplier.

Causes for lower efficiency:

• wear on the impeller, but it a new pump.. however it might have been used before
  

• the factory don't know how to run the test bench, who is the company?
  

• badly designed and machined impeller and defusers:- start here, most impellers and defusers are sand cast in China and machined somewhere else.
  

what type of pumps are they, compression, floating? The list is endless!!

If the pump curves produced from the test bench DO NOT match the manufacturers specs, its your call if you accept them, but... you cannot increase efficiency on site.

All you can do is use a VSD and increase the pump speed, that will change the pump curve to produce more fluid to surface

Pumps although they will have the SAME name plate stating that they can all produce X gallons/litres/barrels per day/minute/hour (take your pick there) are NEVER exactly the same, they would be close.

And finally in answer to Q3, its your call.

If it was me... I would take one pump and motor... run it in the hole and test it to see if it performs to requirements, at the end of the day its produced fluid you want.

If you are happy with one pump, install the others and you will have to adjust each one to produce at th same flow rate (if thats what you want), do not use a orfice plate, use a variable flow valve, you will get a better result and the pumps will last longer

regards Bryan

any other questions please feel free to contact me via CR4 (I consult as well for a nominal fee. HINT!!)

In water wells the efficiency of the pump can be better if there is an assist from Artesian aquifer characteristics. conversely, there can be a loss of efficiency if the pump "over-pumps" the well.

My old company installed lots of submersible water pumps. I have sold the business and am now retired. However, I have visited pump and motor manufacturing sites in Arkansas. One thing the pump manufacturer does provide is a complete set of dynamic curves for all pumps 6 inch or larger (usually greater 5 HP). These curves will have all the pressure v output at various energy inputs (line voltage does vary). Similar pumps will have different curves on these larger pumps. If you are using large subs, ask the manufacturer for the test curves. They usually keep them on file.

Fractional HP submersibles (5 hp and less) are usually stock items with rated efficiencies and curves based on averages of thousands of similar pumps. Because these tests are not dynamic and individual as in the larger pumps, there can be differences through motor or pump end fabricating. These are the curves provided in the catalogs. FYI