Well Pump Failure

Going on the assumption that the pump and VFD are operating properly, something has changed in the well or with the pump delivery piping.

Could any debris have fallen in, onto the pump within the well bore to prevent the normal availability of water to the pump?

Have you checked for blockage or kinking (if applicable) of the delivery piping?

If there an inline filter somewhere between the pump and where you are taking your flow-rate?

A new pump installation will sometimes cause excessive sediment to to be taken into the pump and pushed through the piping and filter for a few days which over time will cause loss of flow.

Check valve operating properly?

The only other thing I can think of is blockage at the pump intake screen which will require removal of the pump.

I rely on a well for my water and these are the first things I would check.

Thanks KJK ,

My concern is that assuming we have blockage at the suction of the pump or at the check valve which is installed 3m above the pump , can we have the same flowrate if the frequency increased , I assume that we should have increase in the flow as the RPM increase , BUT in our case we don't have any change , can the blockage cause constant flow value ???? The well pump is at deep setting , so we can't check this issue untill we pullout the pump ??

That is a good point, if you are changing the input value to the pump better than 20%, I do not feel that the blockage would cause the constant flow value as you described.

Are you checking flow at the nearest point to the well pump? In other words, before any filters, valves, regulators, ect.?

Is the entire run of delivery piping visible to the eye prior to running vertical into the well?

Did you drop the pump all the way to the bottom of the well?

How deep is the well and where does the water level sit?

The pump should be suspended in water, not resting on the bottom.

i agree. it's also possible that dirt has been knocked loose from the sides of the well and has covered the pump on the re-install and it's starving for water. the dirt may eventually settle around the pump, slowing or stopping it's output. sorry for your luck.

i'm not an electrician, but is reversing the pump a possibility? i"m not an electrian, but i have reversed compressor pumps.

I am assuming your pump is three phase, so reversing the phases will reverse the motor. All you have to do is interchange any two pump motor leads. After the pump has ran for a while check the flow with the pump running in reverse. If the flow is greater than you started the pump was initially wired wrong and leave the interchanged wires..interchanged. If it is much slower then put it back the way it was initially and recheck the flow. I have experienced pumps running for years in reverse and went in and changed leads. The reversing may help unclog it but not always.

If the water level in the well has changed that will also change the output of the pump. You will have to refer to your pump curves and figure out what head you pump was designed. If the water is lower in the well than initial level, the output will be lower (refer to curve). If you have increased the pressure you will also have increased the head and have a lower output. it is probably much easier to attempt and measure these possible changes before you try to pull a large pump. It is possible you have picked up a piece of gravel or debris that is interfering with an impeller and that can be a big job (removing the pump and tearing down the wet end).

reversing the pump will not achieve anything if, as is the case here, and together with the symptoms a check valve is fitted.

It might help if the pump tripped on stall or instant overload on starting.. then reversing might free the impellers.. I've done it many times, not always with positive results

Reversing the pump might help free the impellers. I don't know what the check valve has to do with it though. Reversing the impellers doesn't reverse the direction of flow it just reduces the speed the impellers turn. I've done this in the past and the different flow rate has freed up some crap which built up in the impellers and improved flow once the pump was wired up with the correct polarity again.

Reduces the speed the impellers turn. What bull I'm saying. It reverses the direction which the impellers turn. Thats what I should have said but I took a phone call half way through writing and hence spoke the proverbial sh*t.

I take it from your reply you don't work with ESP's.

check valves are a Non return valves that is placed in the tubing usually one or two joints above the discharge head. The reason for these CV's as they are known, is to prevent back spinning of the pump, motor and all the shafts when the pump trips or is switched off from "back spinning", so to prevent an operator from attempting to restart the pump while it is back spinning and breaking shafts and coupling or both. So if you can imagine 6000ft of fluid above the pump in a 3 1/2 inch tube, returning to its static level within the well because the pump has stopped, it will do it very quickly for at least 2 to 3 mins and rotate the pump etc backwards at a fair speed. That's the purpose of the check valve.

now many posts have said "reverse rotation".. good idea.. I would do it as well.. however the point of reversing it is to dislodge the crap, if there is any in the impellers, and hopefully when your reinstate the correct rotation the crap, which hopefully would have been broken up, will leave the pump to surface.

With a CV installed and when you switch off a pump you get no fluid going BACK thru the pump.. which would HELP keep it clear of crap.

As for reducing the speed.. you're wrong. The motor will turn at 3600rpm at 60Hz in either direction turning the pump at the same RPM, but you are right when you say reversing direction does not reverse flow.

hope you understand the purpose & pit falls of a CV now and why I mentioned it.

ps do you know anything about BSR's?

Ok, firstly I don't work with pumps any more, haven't done so for many years now.

Thank you for your description of NRVs but I already know what these are and why they are used after installing hundreds of them. I just couldn't see the relevance of why you mentioned them. I've never built a pumping system with centrifugal pumps without check valves on the discharge side and while I'm sure people will give me examples of times when you could I take it for granted that any centrifugal pump would have a check valve after it.

I agree with you on the reverse rotation to dislodge anything which might be stuck in the impellers. Your flow rate will reduce but the impeller will spin at the same speed.

Reducing the speed. I know I was wrong on that and I should probably not have been replying while at work and being interupted by various people. I had a few other things on and when I wrote it I wasn't paying attention hence my correction straight away.

I have to admit that the acronym BSR is unfamiliar to me. If you want to enlighten me I'd be more than happy to know.

Hi Maca

The use of check valves mainly depends on two things, one, what type of production operator is looking after the ESP and second, how old the switchboard is.

Production Operators.. they can kill an ESP in short order if they don't know or have not been taught what to do, and its usually all of the aforementioned. So they will switch off the ESP for a few minutes or find it just tripped and re-start, not being aware of the back spin (which can take some time to stop), breaking shafts etc.

The second is the age of the switch board... does the switchboard have a modern motor controller that has a BSR.. "Back Spin Relay". This little unit detects the voltage generated by the motor back spinning and until the volts drop below a certain set point to almost zero, it holds the re-start off, not allowing the motor controller to close the main contactor. Then there is the restart timer delay.. however that's another function of the modern day MC's that we'll not go into right now

As for ESP's.. and I'm only talking about ESP's as used in the oil industry here, on or off-shore, I have installed all types, with or without CV's, Y Tools, dual completions etc, etc, and there is no real standard as to what is used. Many application technicians are comfortable with designing completions with CV's and BV's, I think as "belt & bracers", as they STILL don't trust new technology... and the clients (oil companies) are the same, mainly because, and I hear this ALL THE TIME.. "We have always done it this way!"

On the other hand, certain producers will not install CV's as they want the option to back flush to clear the pumps as their wells produce a lot of solids.. whichever way its looked at, there are positives and negatives

But..so long as they want to pay for it, they can have what they want..

Hi Birch

Thanks for that. Like I say I haven't worked with pumps for many years now. I'm actually a mechanical engineer for an oil service provider so I deal with big lumps of metal for subsea applications. In the past I was a maintenance tech in the aquaculture industry and I have found that many acronyms used in one industry are different in another. Glad to hear that BSR doesn't stand for B*ll S*it Radar.

you are very welcome!

If the pump suction is clogged, the pump may be cavitating. In that case, speeding it up might even make matters worse.

If the water is brackish, as is likely to be the case, and with the growth of certain microbes in the well bore, hardness could be deposited as scale on the intake screen, thus starving the suction of the pump. I am surprised it hasn't completely shut the pump down. I like the idea of temporary reversal. Also, production could be reduced if any debris has come into contact with the impeller(s). I think ultimately you are simply going to need to pull the pump string, clean the well, and start over.

It is probably the suction that has clogged up. Depending on the amount of dirt inside the well what I did last time I summerged my pump was wrapping it in a stainless steel mesh to avoid dirt entering my pump and damaging it.

GA Brich. I missed that obvious potential problem and thanks for pointing it out. As this is a new install it would likely be at a connection and not in pipe (assuming new pipe). Perhaps the OP should test the outputs at the well head and do measure dynamic drawdowns. A leak anywhere on the discharge side could be problematic. Not always readily obvious on a Variable drive. Op could check the line pressure with pump off to see if it falls. A leak could still exist between the check and pump.

Dear Brich and all ,

We have conducted hydrostatic pressure test for the well column pipes at 12 bar from the wellhead , but there was no pressure drop after two hours and the pressure value stay the same. I don't know if we have any leak between the check valve and the pump which is 3m above the pump.

What I mean by the current was normal , is that after the decrease in the flow we have normal decrease with the current value, and there was no current rise up that may indicate we have any type of stuck.

But can we have this flow decrease if one of the impeller is damaged as I thought impeller damage will affect the head value only !!!!

The installed well pumps has three components , the pump , the seal (protector) , and the motor , with this type of failure (Flow decrease), I assume that we have mechanical problem , BUT can this problem be due to any type of failure with the motor ??????

The flow deacrese is not related to the well water level as we have level transducer and we checked the water level manuallay by level sound detector and the value was OK.

Thanks for all your inputs and your help ....

thank you for the info..

So.. run the pump at the frequency you had it at when pumping at 160cubes.. shut in the well head and watch the pressure. If it rises rapidly, you have another problem, if not you have a leak.

If one impeller is damaged it will not affect the flow rate, only the discharge pressure. One stage will pump the same volume at the same Hz as 100 stages, the lift is the only thing that is affected by the number of stages. So for those who don't understand what I've just said.. the deeper you set the pump the more stages you need to lift the fluid to surface.

Now as you've just said the running current dropped, not enough to trip on underload, this tells me you have a leak either at the pumps joints (blown O ring), or a hole in a pump body. Motor damage is not an issue here as if the motor was sticking, the seal bearings failing or the pump sticking or overheating then the current would rise.

So you have two options as I said in my previous answer. Let me know what you find!

addind to a comment aready made, i've have seen plastic discharge pipe leak from the pipe slapping against the sides of the well due to pump torque. i always use toque arresttors when installing a pump for this reason.

you're welcome.. however new install or not, tubing holes can appear as and when they like.. as we have no info from the OP.. not sure if steel tubing or plastic pipe work (coiled tubing) or any other type of tubing or the depth

Taking a fluid shot will not give you anything to go by unless the OP has fluid levels results taken prior to the problem as, if there is a hole in the tubing the circulating fluid will give a false reading, the same to be said if a DMT is fitted. but DMT, fluid shots are not the usual SOP for a water well

The only way to find out if there is a hole (is there a check & bleeder valve??) is to have a pressure gauge at the well head, shut in the well head and watch the pressure, as I said in my earlier reply.

As for the VSD, the frequency should have remained constant while pumping at the desired rate of 160cubes..so to the running current, from the settled RC you then set your underload on the drive.

What type of ESP is it? single vane/combined motor or multi-stage centrifugal?

Dear Brich and all,

Sorry for this delay to reply, but we have a decision to pull out the pump, at the meanwhile I wowld like to clarify some issues related to this case:

· The pump is submersible turbine well pump , multistage centrifugal (16 stages) , equipped with motor, seal, and there was SS shroud installed from the pump suction till the bottom of the motor

· The pump setting is about 500m depth and the total well depth is about 700m , the pump is set inside the upper hole were we have steel casing , while the lower hole is the source of water and it has stable and strong formation of rock

· The entire column pipe was inspected during the pump installation; the pipe is of steel and with threaded joints.

· We tried to reverse the phases but the flow was slower

· When we have the sudden decrease in flow from 160m3/hr to 120m3/hr, the current decreased from about 90 Amp to about 80 Amp. So I assumed that we don't have any leak as the current decreases while the flow decreases!!! And we don't have any high value of current which show that we have any type of stuck!!! I think something wrong occurred inside the pump as we may have damage for impellers, shaft … or we may have stuck at the suction side of the pump, or the check valve is not open totally!!!

· We don't have any sudden drop for the water level of the well , and after the flow decrease the water level start to recover (goes up)

Brich , you said something I need to understand , you said "the deeper you set the pump the more stages you need to lift the fluid to surface " , what I know is that the lift is related to the dynamic water level inside the well and not the pump setting , I mean if the pump setting is deeper but the dynamic water level is the same so I think we will have the same lift or the same required number of stages to lift the water , Please correct me if I am wrong ….

I will let you know what we will find after pulling out the pump !

Again... Super thanks for help

I think you have made the correct decision to pull.... but be sure someone who knows what they are looking at is there to witness the pull.. service hands have a habit of hiding things

To expand on my statement for you.. the deeper the pump the more stages you need etc...

The dynamic level is just that.. dynamic... meaning it moves up and down. Its level in the well depends on two things while the ESP is lifting fluid to surface:

1. the flow rate (pump flow rate at surface).. the flow rate/pump size at its BEP should be matched to the inflow of the well. This can be controlled by the "Choke" valve at the well head together with, in your case, the Hz/RPM of the motor/pump.

2. the inflow from the reservoir/BHP.

If for example you have a static level (SL) at 1000ft, dynamic (DL) at 2000ft and intake setting depth at 3000ft, that gives you 1000ft above intake (while running).. thats what you want providing all is balanced in the well.

So if you calculate your lift required based on your DL of 2000ft from surface with a pump that each stage gives a lift of 20ft (at its BEP), the number of stages required is 100.

What happens if the choke is opened to increase flow (and keeping the pump in its OR)? The DL drops, maybe 500ft, so now you have 100 stages that can lift 2000ft, and you need to lift 2500ft.. the result is no fluid to surface.

This is because as you increase the flow, (no Hz increase) the "Lift Head in Ft/Stg" drops as the flow rate increases.. check you pump curve.. the further to the right of the curve you travel the lower the "Lift Head in Ft/Stg", and it can be as much as 15ft per stage (look at a GN1600 curve)

So maybe you can increase the Hz to give you a new profile for the pump, that will lift to surface, but as you increase the pump RPM you increase the flow rate (Laws of Affinity) to surface which lowers the DL.. so now you have a fine balancing act between DL and flow rate at surface verses inflow/BHP, to ensure that you don't "pump off". In other words, you are producing that surface more than is flowing into the well from the reservoir.

Same problem remains if the BHP drops, the DL also drops; and no increase in pump speed (and flow rate) will help, as all you will do is "pump off".

So you calculate your lift from the setting depth of your intake or close to it, to allow for the fluctuations in the DL as you cannot 100% accurately guarantee your DL will stay at the same level, each and every time you install a pump or operate the pump at the same flow rate..... but hopefully you will get close to it.

Now, we all know that more stages require more HP well... so be it, all more stages will do is increase the WHP, but you will have fluid to surface if the DL drops for whatever reason.

Hope this explains. if you're not sure please come back to me