Bilge pumps and airlocks

P

PembrokeshirePromise

Well-Known Member
Joined
7 Mar 2005
Messages
844
Location
Pembrokeshire !
Visit site
Hi,

I'm looking to fit an emergency bilge pump with an auto switch (something like JOHNSON L4000)

The problem is that the only location it will fit in an otherwise deep narrow bilge involves routing the pipe under a cross member below the level of the pump resulting in a potential airlock in the pump.

Any ideas to avoid an airlock ? I can't route the main pipe anywhere else (its 1.5 inch diameter). I was wondering whether I could route a tiny "breather" pipe another route - rejoining the main pipe after the dip - but I'd need some unequal T fittings and couldn't see any anywhere.

Or is there another method - or am I completely barking up the wrong tree.

Many Thanks
 
Fit a non return valve before the uphill bit, and then your pump will just keep shoving water through it and up the hill until it flows away. Expect quite a lot of performance loss.
 
Hmm - the book says the pipe should be continually uphill. I assumed that the problem with a low bit where water will sit (shortly after the pump) means that as the water rises it fails to properly fill the pump so the impellor is trying to push air.

Or have I got this all wrong ?

Thanks
 
As long as the depth of the water is above the pump outlet, it should not matter. But the non return valve will hold all the water its managed to shove past it.
You can adjust the float switch for that.
 
Have you thought about changing where the pump outlet is routing it mabe to the sidedeck or into the cockpit? Reducing the pipe length also inproves pump efficiency. Providing the pump housing fills I don't think you will hve a problem, the pump develops enough pressure to counteract more than 5m head (7psi) so should blow any air out of the pipe
 
I think you are quite correct. With a dip in the discharge hose there is a risk that the pump will not prime. The simplest way to avoid this is to drill a small hole in the highest point of the hose just after the pump. This will allow any air to be vented from the pump as the water level in the bilge rises. Water will leak back to the bilge when the pump runs, but if it is, say, 4 mm in a 38 mm hose it should not impair pump performance significantly. You may need to set up a deflector plate to prevent the water jet from striking the sole boards or whatever else seems important.
 
Thanks Vyv,

That was the conclusion I was getting round to - but was starting to doubt myself...

My thinking on the breather pipe was to avoid the small fountain - but this may be tricky to arrange.

As its an emergency bilge pump (and relatively high in terms of the deep bilge) - something has gone pretty badly wrong by the time it starts - so I think in retrospect a small fountain will be the least of my concerns at that point /forums/images/graemlins/shocked.gif

Thanks for the advice
 
Am I missing something? The L4000 is a centrifugal pump, which demands a flooded inlet. With this requirement satisfied, it will pump, irrespective of the layout of the discharge pipe, provided it remains within its delivery pressure spec. Trapped air will eventually be swept out of the discharge outlet along with the pumped water. I can't see how a breather hole would make any difference, apart from reducing the pump's effectiveness.
 
Well the manual that comes with it is quite specific that the outlet hose should rise steadily upwards - and certainly not dip below the level of the pump.

I assume this is so that the water in the pipe can drain back when the pump stops.

If you have a U below the level of the pump - water will sit in here presumably - and I think that the issue is that when the water rises in the bilge - it can fail to flood the pump inlet properly as you get an airlock formed between the water in the U bend and the pump itself.

The theory is that a hole in this section of pipe would prevent the airlock forming
 
Interesting discussion. I have a L2200 which has been somewhat irregular in performance, and it was not clear why. It was inherited with the boat, and as the wiring in much of the boat is rubbish, I suspected that. Mods to the wiring helped, but did not solve the problems. On some occasions the pump simply failed to pump any thing despite running vigorously. I do have a slight down ward run in the pipework after the pump. I hadn't considered air locks in the pump.

Now a have a new plan of attack!
 
I suggest that you don't have a lot of experience of pump priming problems! In order for the pump to deliver, its impeller and usually a fair part of the volute needs to be immersed. In this case the discharge hose descends from the pump discharge, to where there will be a water trap. In order for the bilge water to enter the pump it needs to displace some of the air that is trapped between the pump and the water trap. Until the bilge water has sufficient head to push the trapped water up the discharge hose a short way this will not happen, particularly with an automatic switch system that will probably start the pump before this head can be reached. Adding a breather hole allows the pump to prime fully.
 
Let me get this right. The Johnson L 4000 is a submersible pump?

The outlet hose has at some point to pass BELOW the level of the pump?
How can that be ? Why don't you place the pump at the lowest point? Is the lowest point (ie where the hose will loop) always full of water then?

Wherever you route the pipe, if a submersible pump is submerged it will prime. How can it not? Isn't that the point of having it submersible? If it can't push an airlock out of the way, how will it ever push a pipe full of water?

Surely airlocks are problems on the SUCTION side of pumps, which you don't have.
 
This is a larger capacity pump in case of catastrophic flooding - it won't fit into the bottom of the bilge "sump" - but will fit at a higher point - still in the bilge.

The hose has to pass under a cross member - and yes you are correct - if it were not for other (manual) pumps - I suppose this would be permanently under water. But in practice the manual pump emptied the bilge sump.

If a submersible pump always self primed when underwater - why would the manual feel the need to specify how the pipe was routed ?

I still think Vyv has it right - but can see it may need some real world testing to prove the point !
 
<<< Wherever you route the pipe, if a submersible pump is submerged it will prime. >>>

I suggest you read my post more carefully. If you like, try this experiment. Take a milk bottle. For the sake of the test, this is the pump with a discharge flooded with water, i.e. closed off. Push the milk bottle downwards into a bucket of water. Does it fill? No? That's what will happen to the pump.
 
But water does enter the milk bottle, and the deeper it is submerged the more will enter! More relevantly, the pump only requires a depth of, say, 2 or 3 cm. to fill the volute and build a further small pressure head. I find it difficult to visualise a typical pump, with its numerous vertical inlet slots, keeping the impeller/volute starved of water sufficiently to prevent the pump flow starting up, provided the inlet is adequately submerged; but if you say that this is an observed effect, then fair enough. Problems with bilge pumps can arise because they pump rapidly from restricted spaces and, often, with marginal submersion of the large inlet area. Hence the prime is poorly maintained, but, for myself, I have never seen it improved by modifying the outlet.
For the outlet pipe run, I would suggest that the main reason for avoiding lutes is to prevent settlement of solids.
 
[ QUOTE ]
Fit a non return valve before the uphill bit, and then your pump will just keep shoving water through it and up the hill until it flows away. Expect quite a lot of performance loss.

[/ QUOTE ]

I tried this. It doesn't work.

Whilst the pipe above the non-return valve was empty it was OK and water could be pumped out. The second time it is used, water is now sitting above the valve holding it closed, the pump couldn't push the water past.

I concluded that these centrifugal pumps rely on the velocity of the water coming out of the pump to have sufficient momentum to get to the top of the outfall. The stationary water above the NR valve slowed the pumped water too much. (I hope that makes sense)

And, yes, I did fi the valve the right way around!
 
You must log in or register to reply here.

Share this page

Top