calling all pipe/water/tank gurus ;-)

[Gabriël]

hi!

the problem is [please see image] that i do not seem to be able to ensure that there is enough water in the pipe sections to keep the pump primed, i.e. full of water - i want to prevent it from sucking air.

please note that the pump is not connected yet. at present i have pushed water into the pipe pump side in order to fill the sections with water to get a siphon effect, but to no avail; the water i pushed in via the municipal pressure just flows out without creating a siphon effect although the tank is full of water.

any suggestions?

in Christ

gabriel

[Chris Hobson]

Siphons operate to a height of about 10m so I doubt there is a breakdown of the siphon flow. There cannot be any leaks allowing air into the system. Air will leak into a system if there is sufficient vacuum without water leaking out.

1. ) Can the pipe not connected to the bottom of the tank - relying on the pressure from the column of water to prime the pump. I fear once the pump is running you are going to have cavitation if the siphon flows too slowly.
2. We have a small centrifugal pump mounted  over a partially buried rainwater tank mounted roughly where your ball tap is above the tank in your diagram. We have a non-return valve on the end of the pipe in the tank to keep the supply primed. The secret is that the pump has a piss-willy delivery and so it never cause cavitation.

[Gabriël]

tx chris, yes, cavitation, that has to be prevented - just googled some images and they look real bad...

although i believe there is no air which can enter the pipe between the bottom of the tank and the pump [i really über-cemented all joints

] , my primary problem is to get all the air out of there in order to 1st of all get it to siphon. if the pump is connected it must suck water, siphoning is just a test i want to make in order to see if an unbroken water column will exit the tank and in order to fill the pipe with water prior to shutting the valve pump side.

the ideal will be to connect the pipe at the bottom of the tank but the tank at present is full.

the pump has an automatic pressure control which will at least ensure that it will not run dry.

AAAAAAAAAAAHHHHHAaaaaaaaaaaaaaa, just read this in the pump manual which corroborates your 1. re siphon and air bubbles:

Suction Piping
• Fit the suction pipe for drawing water rising towards the
pump. Absolutely avoid fitting the suction pipe higher than
the pump (formation of air bubbles in the suction pipe).
• The suction and delivery piping must be fitted so as not
to be able to apply any mechanical pressure on the pump.
• The suction valve should be situated at least 30 cm.
below the bottom water level.
• Suction pipes that are not airtight suck in air obstructing
suction of the water.
 

which leaves me no option but to somehow attach the suction to the tank bottom - i am in the process of getting another tank so i will then transfer the water to it and make the necessary connections - BUT now i still sit with the water level of the tank being at least 2m higher than the pump which WILL "apply any mechanical pressure on the pump" [manual], any ideas re this?

phew - nearly blew that one!

thanks again!

 

[Chris Hobson]

There will be more pressure at the pump in the siphon method than if the pump was connected directly to the tank. The pressure being directly proportional to the height of the water column. 

I find the "apply any mechanical pressure on the pump" unusual but it appears that pump can only function in suction mode.  So it sounds like the pump needs to be above the level of the tank and you need to add a footvalve (non-return valve) at the base if the suction pipe. The system should then remain primed. Most pumps have a small bolt which you remove and this exposes a small hole through which you can add water to prime the pump.

[DeepBass9]

21 hours ago, Chris Hobson said:

There will be more pressure at the pump in the siphon method than if the pump was connected directly to the tank. The pressure being directly proportional to the height of the water column.

No, because the siphon is lifting the water at the other end. It is exactly the same in the two methods as it is the difference in height between the water in the tank and the pump.

 

[Gabriël]

but the chances for air in the direct coupling with the tank vs the 'up-and-over' pipe seem to be far less...

[maxomill]

 im surprised the tank does not have a outlet at the base

but try this

remove the ball valve at the top of the tank

swop the elbow at the corner of pipe b and c with a t piece 

using the t piece as an elbow connect the ball valve on the top of  new t piece

open valve and fill pipes with water until water enters the tank and most of the air is out

close ball valve while still filling

start pump

all the remaining air will be drawn out  of pipes

if the output of the pump is connected to a closed system then the pump should remain primed

if the output of the pump is open then switch of pump and close valve at pump

alternatively raise the output pipe of pump to same height of tank so ensuring water will always remain in pipes and pump

also you could place the valve at the pump on the outlet side  so that pump always has some water in it for the next start

[Gabriël]

hi @maxomill, 1st off let me state i'm sorry my drawing does not reflect the bottom outlet, it is there; it is an ordinary jojo/roto tank, new image attached.

we have established that the pressure on the pump inlet would be the same whether the pipes are as in the image or whether it is connected to the bottom of the tank as the present water level is above the level of the pump.

further, the presence of air in the suction side, in either latter configuration, must be prevented.

i did contact the manufacturer, speroni, the pump data is here http://www.speroni.it/public/en/prodotto_dett.php?id=32 and will let you know the response.

 

[Chris Hobson]

3 hours ago, DeepBass9 said:

No, because the siphon is lifting the water at the other end. It is exactly the same in the two methods as it is the difference in height between the water in the tank and the pump.

 

Yes I was not thinking properly.

@gabriel you have my sympathies I do not know how you are supposed to install that pump. Speak to your supplier and if he gets it wrong then you can go back to him and kick his shins. 

[Gabriël]

ok guys [and gals?] here is what the guy at agrinet says, they are distributors for the speroni pumps in sa.

the pumps can be placed next to a tank, i.e. under the water level.

a class 6 40mm must be on the suction side with a .........[forgot the name] to 25mm just before the inlet. thus my 40mm pvc is way too flimsy for this operation - back to the drawing board...

the suction has to be attached to the underside of the tank where there is a 40mm outlet.

thanks for all the tips!

[ghatikar]

On 04/09/2017 at 11:27 AM, Chris Hobson said:

There will be more pressure at the pump in the siphon method than if the pump was connected directly to the tank. The pressure being directly proportional to the height of the water column. 

I find the "apply any mechanical pressure on the pump" unusual but it appears that pump can only function in suction mode.  So it sounds like the pump needs to be above the level of the tank and you need to add a footvalve (non-return valve) at the base if the suction pipe. The system should then remain primed. Most pumps have a small bolt which you remove and this exposes a small hole through which you can add water to prime the pump.

2

apply any mechanical pressure on the pump - I think refers to mechanical load from joints/  from the pipes/misalignment etc.

[Gabriël]

26 minutes ago, ghatikar said:

I think refers to mechanical load from joints/  from the pipes/misalignment etc.

makes sense!

[maxomill]

you missed the part that says this is a SELF PRIMING . last I heard that means it can suck out the air until the water gets to the pump and then all is good .

a pipe full of air will not damage the pump

cavitation is like when the pools weir basket is blocked with debris and it cannot pull anymore water or hence the class 6 pipe to ensure that the pipe will not collapse under the suction

I have no doubt that you can place the pipe in the top of the tank and it will work

my biggest concern is what is on the outlet of the pump   is it open or does it connect to some other piping

as  long as you get some water in the suction pie the air is irrelevant

cheers

[maxomill]

It is important to understand that self-priming pumps cannot operate without water in the casing.

Here's how it works:

During the priming cycle, air enters the pump and mixes with water at the impeller. Water and air are discharged together by centrifugal action of the impeller into the water reservoir. The air naturally tends to rise, while the water tends to sink.



Air-free water, now heavier than air-laden water, flows by gravity back down into the impeller chamber, ready to mix with more air coming in the suction line. Once all air has been evacuated and a vacuum created in the suction line, atmospheric pressure forces water up into the suction line towards the impeller, and pumping begins.



Recirculation of water within the pump stops when pumping begins. The next time the pump is started, it will "self-prime" -- that is, it will be able to once again mix the water and air in the casing to create a pumpable fluid until the pump is fully primed again.



This type of pump differs from a standard centrifugal pump in that it has a water reservoir built into the unit which enables it to rid pump and suction line of air by recirculating water within the pump on priming cycle. This water reservoir may be above the impeller or in front of the impeller. In either case, the "self-priming" capability of the pump comes from the pump's ability to retain water after the very first prime.
 

So Can I Just Start a Self-Priming Pump Any Time, Even If It's Dry Inside?

No. A self-priming centrifugal pump must have water in the casing in order to operate. You cannot pull any self-priming pump right out of the box, turn it on, and expect it to pump. If it's full of air, it won't prime. "Self-priming" refers to the pump's ability to repeatedly turn an air/water mixture into a pumpable fluid -- NOT the ability to create a vacuum (literally) out of thin air. In fact, you should never try to run a self-priming pump without water in the casing. It's dangerous and will often lead to seal failure.

Nothing here is intended to replace or stand-in for proper training and/or the use of your pump's owner's and safety manuals. ALWAYS READ THE SAFETY AND OPERATION MANUALS BEFORE OPERATING ANY PIECE OF MACHINERY, INCLUDING PUMPS.

[maxomill]

but all things considered    can you not just connect to the bottom outlet then all problems are solved

 

[Gabriël]

so that is what the term self priming is all about, thank you very much @maxomill - there it is and i miss it, and yes,

9 minutes ago, maxomill said:

you can place the pipe in the top of the tank and it will work

now i agree with you. in this case i will replace the 40mm pvc going into the top with the class 6, easy peazy.

the outlet will go into class 6 22mm and run to the house. the auto pressure control on the pump will sort out any pressure related issues on that side, like if all taps in the house are closed, it will stop the pump; it seems to be quite a sophisticated piece, see here

the outlet seems to be a reasonable trouble free zone of a pump i read

[Gabriël]

3 minutes ago, maxomill said:

but all things considered can you not just connect to the bottom outlet then all problems are solved

my tanks are chock-a-block at this stage and i have the wrong fittings on the lower outlets, so i'd rather go for the 'over-the-top' pipe option until i can work on the bottom outlets of the tanks.

[PaulF007]

You could also consider to place a one way valve (Foot valve)

at the bottom in the suction pipe this should prevent the water from "leaving" the pipe. Keep in mind that I have been too lazy to read through all the comments so if it can not work , disregard 

[Gabriël]

thanks again for all the tips!

some feedback on the water tank etc operation

i went ahead keeping things more or less as in my latest drawing. although a guy warned me that the pumps would suck the ordinary 40mm pvc to pieces, the smaller pump did a great job yesterday without the pipes moaning. the big pump might squash them, but i doubt - if so i shall replace the suction side with heavy duty 40mm.

i still need to connect the pump 1 supply to the house mains, do external electricity supply and another suction side supply for tanks on order plus build a housing for the setup. in the meantime though herewith an image of the present situation.

[Chris Hobson]

Nice and neat

[Goedman]

A newbe question:

Why do you have two pumps?

Those yellow things on top of the pump, (guessing it might be digital pressure control valves), did you buy that separately? How are they working for you?

[Gabriël]

20 minutes ago, Goedman said:

Why do you have two pumps?

Those yellow things on top of the pump, (guessing it might be digital pressure control valves), did you buy that separately? How are they working for you?

hi @Goedman, i have two pumps because i am stupid - i should have gone for the bigger pump from the outset. it is not only stronger but is self-priming [ see link], which means it does NOT need to be 'flooded' with water prior to operation to prevent burnout. if you buy a pump go for a self-priming unit, it will last longer.

the yellow things are apc's as indicated - here is a good description

if you are tinkering with pipes and pumps consider azud filters, i use these filters and they do a great job

 

[Goedman]

Thanks @gabriel, don't feel bad, this is such a steep learning curve for most of us.

That is why a forum like this helps a newbie like me avoid some of the most expensive mistakes.

I catch water from the roof and balconies and up to now only used it in the garden. We have been spoiled with dependable clean water from the town council but lately there have been some unplanned disruptions in supply. Now I am doing some research to learn how to pump the rainwater into the house on those days that we don't have any water. When we get lots of rain (mostly in the winter for us), I might want to use that from my 10000 liter tank when it gets full enough. As a pensioner it is also important to save some money wherever I can.

I am still looking for a nice diagram to show how a system like that could work.

Thanks to you and everybody else on the forum that shares so generously for the benefit of others.

[Gabriël]

35 minutes ago, Goedman said:

I am still looking for a nice diagram to show how a system like that could work.

the guys at jojo tanks or roto tanks should be able to help.

basically it would be that your municipal water should go into a tank with a ball valve, like a wc cistern. from there it must go via a pressure pump into your house i enclose an idea

if you want to mix rain and municipal water in the same tank you will have to set the ball valve in a way to say limit municipal inflow to half the tank - i'll have to think about that...

[Goedman]

Thanks @gabriel,

I checked out some of your other diagrams as well (with the 2 x 10000L tanks). Thanks for all the good ideas. It is so much easier to see it from a diagram.

What do you suggest for filtration? Because of the rainwater from the roof (no first flush diverter yet).