Offshore water pump. 6

[jamesss]

The project;

I am currently working on a pumping system to draw/pump water from the sea, to the access level of offshore wind turbines.The sea water will be used to wash off bird waste that accumulates in large quantities causing problems for offshore operations.
The pump will be a self priming centrifugal pump 230 v 160L/min 5 Bar.
The suction line is 7 meters maximum length, 1.5 inches diameter pipe.The discharge line is also 7 meters but has a diameter of 1 inch.The pipe diameters are matched to the inlets and outlets of the pump design.
There is a non return valve on the suction side 3 meters from the pump, this is to maintain water in the pump body and discharge line to assist with priming.
There will be a control circuit activating the pump at high tides only and during night time i.e. one high tide per day.

Testing of the self priming pump, suction hose,non return inline valve offshore.

During high tide the pump was situated on the resting platform (7 meters from sea level to pump inlet).the pump body was then filled with water ready for self priming.The suction hose was placed into the sea water and the discharge hose was run uptothe turbine access level 7 meters up.

When the pump was activated, i could see that the pump was extracting air from the suction hose through the pump housing and into the discharge hose, the air bubbles and water mixture would rise so far and then fall back into the pump housing. After 5 minutes i turned the pump off and could see the trapped air in the pump housing beginning to naturally rise up through the discharge pipe and out into the open.When reactivating the pump after all the air had vacated the pump casing, the pump performed as it should,displacing water from sea level upto the turbine access level at a rate of 80 liters per minute.

Each time i turned the pump off and then back on, it would activate and start pumping straight away however, when i turned the pump off and introduced air into the suction pipe to simulate the tide going in and out, the pump would not prime and would continue to collect a water/air mixture in the pump casing and fail to prime fully, leading to failure to pump.As mentioned, if i turned the pump off then on to allow the air to escape the casing then the pump would perform faultlessly.

After investigating further into the reason the pump was failing to prime first time, i came to the conclusion that the head of water on the discharge line was not allowing the air/water mixture to escape resulting in air binding thus loosing suction.

My questions,

Is this something anyone has seen before?
Is my conclusion plausable?
if so, would an automatic air breather installed on the pump casing be of use for my problem?

I don't really want to re-configure the control circuit so the pump activates twice as i don't see that as a sound solution to the problem at hand.

Any help will be much appreciated!

 

[JJPellin]

I have seen this before. Strangely, it was not as part of my role as a pump engineer. It was at my cabin. The shallow well pump there draws up from about 5 or 6 meters below the pump. With the pump flooded, it cannot prime the suction piping on the first run. Only by starting and stopping two or three times can I get it to fully prime. Once primed, it has stayed primed for years.

I am not sure that an automatic air breather (air release valve) will solve the problem. When it is failing to prime, on the first attempt, it may not build enough pressure to push the air out. I would consider different solutions that may not be practical for you. I would have placed the non-return valve further down in the suction piping. A foot valve near the very bottom would reduce the amount of air that needs to be expelled when priming. Or, you could drain down the discharge piping so that there was not so much head to overcome during priming.

Is there a second non-return (check) valve on the pump discharge line? If not, the head pressure in the discharge line may allow the automatic air release valve to expel the air during priming. I don't have experience with these valves, so I am not sure.

Johnny Pellin

 

[jamesss]

Thanks Jonny,
The problem i have is if i installed a foot valve, then the incoming and outgoing tides/waves would lap up against the valve and eventually release the water.Also, as this system will be getting fitted on many locations i really wanted a robust system that was able to prime any air in the system.Ideally,even if the non return valve 3 meters from the pump failed, i would like the body of water remaining in the pump housing to be sufficiant enough to prime the system even if it take a little longer as this is what the manufactures instructions says (if you are drawing water more than 6 meters it is advisable to use a foot valve).

[highlight #729FCF]Or, you could drain down the discharge piping so that there was not so much head to overcome during priming[/highlight], if the head of water definitely is what is causing my problem then i could easily get around it.
It is one of the first things i will try next time i am running a test.

Thanks again for reply.

 

[JJPellin]

I am not sure how lapping waves would release the water. With a foot valve at the bottom and the pump not running, it would have 14 meters of head pressure against it. I would expect that to keep the foot valve sealed tightly closed. Perhaps I am missing something.

Johnny Pellin

 

[jamesss]

It’s not 14 meters, It will only be 4 meters of head as I have a non return 3 meters from the pump on the suction side.
I don’t want to be in a position where 20+ turbines all need to be visited because a small amount of air has leaked into the suction pipe.I feel I am missing something fundamental here.

 

[EdStainless]

Why the NRV 3m up?
You need to keep all of the suction line full of water. A foot valve 0.5m up and then nothing between there and the pump.
What is the material of construction? Something immune to corrosion in warm seawater and also naturally resistant to biofouling I hope, such as 90/10 CuNi.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy

 

[jamesss]

Hi Ed,
I know ideally the suction line should be full of water but I just don’t think relying on one foot valve submersed around 3 meter waves is going to cut it for the long term.The slightest leak and the system doesn’t work.
Is it not normal to expel the air at the pump?
I think I’m going to try reducing the head height of the discharge line.
I need to figure out why it is not behaving according to the manufacturers data.
When I finally get the system running robustly I will think about the materials.At the moment I’m using hydraulic like flexible hose that has suction hose printed on the side, this I don’t see as a great hurdle at the moment.

 

[miningman]

Well like an wise young engineer said to one of his colleagues within my earshot " Pumps dont suck worth a s**t" If I understand your situation , at 7 metres suction you are extremely close to the theoretical maximum that any pump can suck.

 

[jamesss]

The theoretical maximum is 10+ metres however in my case the actual height I will be drawing water will be 6 .5 metres.
The manufacturers say the pump is more than capable of this.I have the system working fine,but it takes two start ups which is not good enough for what I need.

 

[miningman]

I stand corrected. 7 metres is the practical maximum any pump can suck. Many years ago I was running a number of large syphons and they all lost prime when the suction side increased to around 21.5 feet. The hydraulics here are identical. How much dissolved air is in the water on the suction side??? How much air is leaking into the suction side thru defective gaskets , porous materials etc??. Have you taken into account the SG of sea water being greater than clean water?? I suspect not........ this is the fundamental issue you are missing.

 

[JJPellin]

Some rental pumps I have used have an installed vacuum pump that pumps out air and keeps the pump constantly primed. The little vacuum pump shuts off automatically when water reaches it. It restarts if water runs down and air is present. This little vacuum pump is electric powered and small. Could this address your problem?

If you had a foot valve near the bottom and no other non-return (check) valves, there would be about 14 meters of head pressure on that foot valve. That should keep that valve closed tight.

Johnny Pellin

 

[TugboatEng]

The amount of pressure generated by a pump while self-priming is 1/1000th the rated discharge head. Self-priming centrifugal pumps cannot prime against check valves, trapped water loops, etc. Depending on the location of the pump we often just drill a small hole in the discharge pipe to facilitate self-priming.

 

[jamesss]

Hi Tug Boat,
This is the reason why i think an automatic breather mounted on the pump casing (as there seems to blank on there already,maybe to accommodate a breather?) will do the same job as drilling a hole on the discharge line, as i can visably see the water air mixture trying escape then rebounding back into the pump.

 

[LittleInch]

OK, a few things here.

I'm surprised you've gone for a self priming centrifugal rather than a submersible pumps or in pipe downhole submersible my self for exactly these reasons or indeed for low flow some sort of PD pump. This would IMHO, solve many of your issues and for a remote location would offer advantages in terms of having to go and fix it at regular intervals.

If you look at how self priming pumps work they basically re circualte some of the initial fill water and effectively pump a mixture of some water and air. Clearly here the pump is just not capable of pumping the initial head with this weird mixture of air and water.

SO if you want to go with this my recomandation would be just a simple discharge loop at or close to the pump back into the sea which is opened for a set period of time to allow the pump to fully prime ( say 30 seconds) The breather is difficult to mount and maintain and only really works well with smooth flow, not this maelstrom of water and liquid you have going on. Or accept a certain permanent flow going back to sea.

As said by a few others 6 to 7 m is about it in terms of pump suction lift. Your theoretical pure vacuum lift is just over 10m, but your pump needs probably about 3m minimum before it starts seriously cavitating and loosing the ability to pump properly. I would love to see the pump data sheet from the vendor if he says this sort of pump is "more than capable". I would say it's at maximum and you got a response from a salesman who knows not very much.

That doesn't look much like a self priming unit to me - care to share the details / vendor?

https://www.youtube.com/watch?v=2FA6t7wHtF0

for the basics (sorry if you know this but i's always useful to see again)
This guy calls it air blocking - see about 1:30 in the following

https://www.youtube.com/watch?v=GrpXdcSfbfA

The other thing to remember here is that this process uses some quite small gaps and holes inside the pump casing. any bit of fish / weed / dirt you suck in from the sea could very easily block these holes and stop the whole thing from working.

Is this pump supposed to be left in situ and then used automatically / remotely?? Doesn't look like a system that will last a longtime before needing maintenance.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[JJPellin]

I mentioned issues I have priming the shallow well pump at my cabin. The picture provided looks almost exactly like my well pump. It is considered self priming because a portion of the discharge is recirculated back to the suction through a venturi fitting that is supposed to act like an eductor and draw in air or water from the suction. The ports you have circled are the ones I use to dump water into my running pump to speed up the priming process.

Johnny Pellin

 

[Artisi]

jamesss,
Do you have access to a compressed air supply at the installation or could air be brought to the pump location,
If yes, I can shown you system to make the pump truly self-priming.

It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. (Sherlock Holmes - A Scandal in Bohemia.)

 

[jamesss]

Hi Little inch,
Thank you for your feedback.
The reason i have selected such a pump is because designing a system whereas the pump is submerged in the open sea with 4 metre swells and a 8 metre tide every 6 hours is just not a good idea.Severe corrosion will also be introduced to each and every pump whilst it is sat in the salt water.
That doesn't look much like a self priming unit to me - care to share the details / vendor?

hmmm The pump is 100 % self priming as i have already managed to draw water the 14 metres head i require at a significant flow rate and pressure as already mentioned?
JBR4

http://www.cempumps.com/site/cem_webprofessional_it/catalogo/catalogo/J-mini_J-INOX_MG-INOX_JBR.pdf

The 'weird mixture' of water and air is the air from the suction line mixing with the water in the casing of the pump...this is supposed to happen in self priming pumps a that is how they function.

Any pump can be operated remotely if you have a robust enough control system.

A continuous flow back to the sea may be a good idea.

I will try the automatic breather out.

 

[jamesss]

Hi Artisi,
No unfortunately, i could incorporate a compressor into the system however i was wanting to keep it as simple as possible.

If i have no luck with the automatic breather then i will have no choice other than to build a two start process into the system.

I would like to learn more about your system though.

 

[LittleInch]

Thanks for the information, but I must object to this statement

"The reason i have selected such a pump is because designing a system whereas the pump is submerged in the open sea with 4 metre swells and a 8 metre tide every 6 hours is just not a good idea.Severe corrosion will also be introduced to each and every pump whilst it is sat in the salt water."

A very simple search for "submersible pump sea water" shows the error of this. Just like your pump is made from Bronze, so submersible pumps designed for seawater use materials resistant to seawater corrosion. Many systems use a caisson approach to protect against direct damage / wave motion etc. But the issue is that you don't need to prime the thing. Ever.

It's not totally clear how the self priming works but seems to be some sort of venturi effect as JJ says, but I'm not sure if this is internal or external to the pump. But do note the data sheet is asking for liquid free from solid impurities. Not easy in seawater.

But you've clearly decided that your way is the best so please let us know what you do and how it works.

Good luck.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[jamesss]

Hi Little inch,

Not sure why you have some sort of a problem with everything I write.I thought this was a forum for assistance and ideas, everyone else seems to get that.

Marine growth is also a massive problem, barnacles and mussels would soon Take hold of anything submerged in the water, then there’s the pump head and pressure required.What about the maintenance on 25 offshore pumps that are submerged?
I understand a sump pump would not need priming but I don’t see how it would be suitable.
I haven’t ‘clearly decided that my way is the best’ the environment and the location has made that decision for me.

I have the control system functioning perfectly, I have the pump working perfectly, I just need the pump to self prime first time and that is it, problem solved.