Placement of electric pump in fresh water system

======
Quote:
However, as previous posters have advised, it will all be a lot easier, and your pump will last longer, if you put it below the tank.

Why do you think that the pump will last longer?
=======
On the assumption (admitted) that repeatedly having to reprime would shorten the pump life - though I guess it would depend on the pump design.
 
Power consumption wise it takes more power to push uphill, so what is the head that your pump needs pump to? I've installed my pump right under the sink and as the shower in the head is about 5ft away the pump feeds both with no problem at all.
 
[ QUOTE ]
Power consumption wise it takes more power to push uphill, so what is the head that your pump needs pump to? I've installed my pump right under the sink and as the shower in the head is about 5ft away the pump feeds both with no problem at all.

[/ QUOTE ]

OK - its a bigger scale ... but we use various types of pumps ... high pressure, low volume turbine, double diaphragm high pressure, high volume, etc. etc. in our Additive work.
We even have pumps based on same principle as the rotary engine ...

Having a +ve feed makes a lot of difference to delivery pressure and volume at outlet...... with all our pumps. We can "lift" with some of our pumps considerable heights before "pushing" as you call it ... But as we have to lift more - the delivery reduces. I would suggest that the pump is having to produce "vacuum" to suck the product up .... then pushing the product to outlet. We now try to avoid all "lift" of product wherever possible and get much better results. Some of our jobs we have to "push product over 15m height .... we found that if any significant "lift" involved before pump it often didn't work.

Just passing on observations ...
 
There is quite a bit of confusion between pump types in several posts in this thread. Centrifugal pumps will not self prime and their head/flow curve is greatly affected by throttling, either of the suction or discharge. A positive displacement pump (such as used in pressurised systems on boats) has a head/flow curve that is a straight, vertical line. It will develop as much head as it needs, or die in the attempt. The flow is determined by its dimensions. Throttling either the suction or discharge will not affect the head.

A positive displacement pump will draw water up from a lower reservoir. The theoretical limit is just under 10 metres, determined by the vapour pressure of water. In practice this is not attainable. At some height above the water level, affected strongly by suction pipework resistance, the water will begin to vapourise in the pump suction. These pumps are not of industrial quality so the actual head when cavitation occurs may be quite low, e.g not much more than a metre or two if there are some elbows or constrictions in the pipework. Suction filters have a strong negative effect.
 
Yes, I agree with the principle that you and sbc are making in that at some stage when the 'suck' gets to a certain level, a 'vacuum' is formed and, of course, no pumps of the type we are interested in can work. Neglecting vapour pressure, you still have a 'Toricelli vacuum' which happens with 760mm mercury or 10.3m of water. In other words, no way can you 'suck' a greater head of water than 10.3m and in any case, this is reduced a little bit by the saturated vapour pressure of water, but it is hardly relevant. The OP was discussing much lower lifts and even on a large yacht it is hard to imagine a configuration in which the pump was sucking more than 2 metres....a long, long, way from the point at which these effects become an issue. My yacht was factory-fitted by Nauticat in Finland, with a pump that sucks from tanks at least a metre below the pump - could be more, it is hard to measure these things on a boat unless you are very determined. Our Surflow pumps run perfectly and the pump that was installed in the boat when we purchased her in September 2004 has lasted until only last month - i.e. 2.6 years of 24/7 liveaboard and the pump was almost certainly well used when we bought her. I have the sort of engineering 'ear' that is sympathetic to mechanical things being run outside their preferred range and I can say that the pump is running perfectly, very sweet at all stages of fill. The OP really will not have a problem with his proposal.
 
[ QUOTE ]
Yes, I agree with the principle that you and sbc are making in that at some stage when the 'suck' gets to a certain level, a 'vacuum' is formed and, of course, no pumps of the type we are interested in can work. Neglecting vapour pressure, you still have a 'Toricelli vacuum' which happens with 760mm mercury or 10.3m of water. In other words, no way can you 'suck' a greater head of water than 10.3m and in any case, this is reduced a little bit by the saturated vapour pressure of water, but it is hardly relevant. The OP was discussing much lower lifts and even on a large yacht it is hard to imagine a configuration in which the pump was sucking more than 2 metres....a long, long, way from the point at which these effects become an issue. My yacht was factory-fitted by Nauticat in Finland, with a pump that sucks from tanks at least a metre below the pump - could be more, it is hard to measure these things on a boat unless you are very determined. Our Surflow pumps run perfectly and the pump that was installed in the boat when we purchased her in September 2004 has lasted until only last month - i.e. 2.6 years of 24/7 liveaboard and the pump was almost certainly well used when we bought her. I have the sort of engineering 'ear' that is sympathetic to mechanical things being run outside their preferred range and I can say that the pump is running perfectly, very sweet at all stages of fill. The OP really will not have a problem with his proposal.

[/ QUOTE ]

I don't doubt the pump will work as he wants ... in fact he quoted "box figures" that indicate the pump is designed for this. But I still am of opinion that having +ve pressure on a pump is kinder to the pump and delivery is better.

I applaud the posts by Vyv and Lemain ... no problem at all with the logic and physics ... In our additive pump work - we only get one shot at it ... get it wrong and a few million dollars worth of cargo is un-treated or lacking. So we err on safe side and even with our +ve displacement pumps - fit so that source is gravity fed to the pump ... then pump has all effort devoted to output.
We had one pump that had to be rebuilt after a few months ... a self-priming professional job of over £5000 value ........ reason - it had to self-prime each time we changed a container supplying it. Service agents showed us the internals .... We do not run any pump like that again ... (only one I have that has to "lift" is my house pump that feeds from the well. It has a +ve pressure reservoir on the pump in my cellar and non-return valve on the very end of the draw pipe in the well..... so in theory never runs dry. 3yrs and still going well.

OK - I know the above is industrial and not a yacht situation ... but you can imagine it tends to colour one's opinion of how to fit things ...
 
I'm sure you are correct, if we are talking about a high-quality, factory installation. My pump is very difficult to access about 1 metre above the tank. Hoses, wires, strainers, ducting and other things all compete for a very small space. From time to time it has happened that my suction hose has been kinked and there were definite signs of flow reduction and cavitation noise.

I also use a domestic water pump for my refrigerator heat exchanger and this is very sensitive to fouling. With a relatively small reduction of the cross-sectional area of the seacock, the discharge flow will stop completely, even though the total lift is less than a metre on the discharge side and zero on the suction.
 
The other point that hasn't been mentioned is that even the tiniest air source will cause cavitation - e.g. jabsco impeller pumps for raw water engine cooling often fail due to a small ingress of air and the wrong gap between impeller and wear plates.
 
You must log in or register to reply here.

Share this page

Top