Water flow issues

[Omatako]

I need some advice on hydrodynamics of a raw water system.

The genset that I am busy reconstructing used to have an external raw water pump, apparently electrically driven. Whilst I have the appropriate electric pump, I have a nervousness that it would be more likely to fail than a belt driven one.

So in my reconstruction, I have incorporated a belt driven pump of the same capacity but due to severe space limitations, I cannot place it into the system where it used to be without considerable re-engineering.

So my question is - in the sketch below (excuse the quality, pencil drawn), what would the implications to water flow be if the pump was located in a pull-orientation as opposed to a push-orientation?



I have a sneaky suspicion that the heat exchanger may lose some efficiency if the water is drawn through it under a vacuum as opposed to forced through it under a mild pressure.

As a possible variation, I could locate the pump in the circuit directly at the outlet of the heat exchanger but the position shown in the sketch is the line of my least resistance

Any ideas out there (or even better, previous experience)?

[Faster]

Whether or not you'd have an issue would be related to the actual pressure on the suction side of the pump in the heat exchanger itself.

And you'd need to have an idea of the actual raw water temperature there as well to know if you were at risk of dropping below the vapour pressure temp, which could cause flashing and pump cavitation. Few systems actually monitor the RWC temperature....

[tommays]

Pumps in general do a much better job of pushing than pulling and the amount of vacuum they create to self prime is limited and changes a LOT as the pump size increases


I work with a pump series for example that comes in size 15/30/60/130 etc

The size 15 will not self prime worth a poop were the 60 might have no issue pulling material up 6'

[jrd22]

I think in your proposed location which is after the anti-siphon you would lose the prime on the pump because you would be pulling a vacuum on the a-s. If you move it to the other side of the a-s by the exchanger I don't see why it wouldn't work. You always want to keep the run to the pump as short as possible but as long as there are no air leaks in the supply line and the pump is close to the water level it should prime OK and pull water through everything. The danger I suppose is that you risk losing the engine if a small air leak does develop in any of the components in the supply line and the pump loses it's prime.

[RichH]

You risk BOILING the water INSIDE the pump and also IN the Hx, therefore BAD application position for a pump. ---
Depending on the amount of suction head developed by the pump, it may cause the hot water to 'flash' or boil in the position youve located. Such pumps should always be placed on the 'cool' side and deliver PRESSURE to the hot side to prevent 'flashing' of the hot water (boiling at temperature below 212° when at less than atmospheric pressure).
Pulling a vacuum on hot water will depress the temperature at which it BOILS; so you need to be very careful so that you dont exceed the 'vapor pressure' of the hot water.
(I dont have my 'steam tables' with me but youre going to be close or exceed the vapor pressure (~6 psi @ 170°F) causing the water to boil with a single stage pump pulling up to 6" hg. or ~3+psi 'vacuum').
If the water 'flashes' or boils because of the vacuum, you will not have any water flow, just 'vapor' flow!!!!!! Thats why pumps always are used in 'pressure mode' in hot water applications ... the increased pressure RAISES the temp at which the water boils/flashes.
I would also surmise that youve omitted the cooling loop for the exhaust manifold as the Ex. Manifold will put as much heat into the cooling loop as does the engine block circulation. The exh. manifold (when not included in the FWC circuit) is always 'after' the Hx.

Other considerations - If this is a rubber vaned Oberdorfer type pump the rubber in the vanes probably will quickly deform as the 'softening point' will be around 170 deg. F.

With two such vulnerabilities ....... suggest you put the pump in 'normal position' in the circuit and 'suffer' the needed engineering to put it there.


;-)

[MacGyverRI]

What Rich said, all of it...

It needs to be on the thruhull side to cool the impeller/s also or they will burn up.

I would go w/ a quality replacement elec. pump like it had and a spare if it's feasible $$.

[Omatako]

Originally Posted by RichH

You risk BOILING the water INSIDE the pump and also IN the Hx, therefore BAD application position for a pump. ---

;-)

Thanks, that's the sort of positive information I needed. I had a feeling it was not clever and you've succintley confirmed that.

I will re-engineer as required and have the pump still belt driven but on the cold side of the heat exchanger. It's gonna be a tough fix but I believe it to be the best solution.

Originally Posted by MacGyverRI

I would go w/ a quality replacement elec. pump like it had and a spare if it's feasible $$.

Thanks but I have little faith in electric pumps so I'll pass on the suggestion. I have got one which I will use if an emergency presents itself but realistically, the genset should never constitute an emergency. I still have another 200amps of power generation on my normal alternators.

Thanks to all for your input.