Flow Rate for water not to freeze in pipe?

[Novice Pipe D]

I am a novice pipe designer and I have a project starting on Monday. I have to design a Filtered Water Supply system. The pipe is SS and the majority of the header is located outside. The engineer said no insulation is needed if we put a recirculation system in. He says by keeping the water flowing the water will not freeze... I am concerned about this. The pipe will be exposed to the MidWest winter elements. Is it actually possible to design the system without insulation? If so what is the flow rate of the water to create enough friction to keep it from freezing?

Thanks in advance!
NPD

 

[RocketSci5KN]

You need to calculate the heat loss through the entire length of SS pipe at the coldest ambient temperatures you'd encounter, and compare that to the heat input caused by the recirculation pump. If the latter is larger, the water can't freeze. I don't see how flow rate really matters, except in the fact that it would take more pump energy to make this happen. Also, don't forget that a thin layer of ice could form on the ID of the pipe, and affect the heat transfer. Insulation is probably still a good idea. Compare the long terms costs with and without.. consider too the costs for a failure of the recirculation pump, and the bursting of all external piping.. and downtime costs..

 

[russ_watters]

1. You're a designer, not an engineer, so these calculations/decisions are above your pay grade. You can certainly raise an objection (in writing, even...), but you'll have to go with what your engineer says...and you won't hold any of the blame if the design fails.

2. Uninsulated and unprotected water piping outside? That would never fly in my company. We'd insulate and use heat trace. There are several problems with the idea:
a. What if the pummp fails?
b. Can you be absolutely certain there will be no dead spots anywhere in the pipe where flow slows enough to start freezing (and once it starts, it will quickly freeze the whole pipe).
c. Since this isn't a thermodynamic system, there is no heat being added to the water (the heat added by the pump in this case is so small as to be negligible), so the water is just going to keep getting colder and colder. You may even end up with a situation where the water is subcooled and freezes as it is coming out of the tap!

Yes, there is a school of thought that says if you keep the water moving it won't freeze (and you can see this in streams), but I've never even tried to calculate it because it is unreliable both in theory (b) and practice (a).

 

[minger]

Yea, I have to agree with Russ. As a textbook problem, sure you could say that there exists a flow rate which a Nusselt number gives minimal heat transfer compared with the time the water is physically in the pipe cooling off.

However, I would be concerned about Russ's point b. There will always be, for a real piping situation, bends and couplers, etc. At these points, there will be flow separations, or points where the fluid will be essentially stagnant.

These areas will almost surely freeze, creating bigger flow separations, etc, etc.

 

[Robot B9]

If you knew the worst case thermal properties of the exposed pipe you could then calculate the heat flow (Watts) out from the water at say 34f to the outside air at its coldest expected temperature, a known delta T.

Based on this heat flow in Watts you could then convert Watts to Horse power. Use this HP number to select a water pump. Energy expended by the pump will convert to heat by the flow resistance in the pipe system. Probably be a very small pump!

But the slick way to do this would be to work out a thermo siphon system so there is no need for pumps. Water would sink in the exposed pipes, rise when it got inside where it was warm. All this would need would be watching out for the slope of the pipes to allow for natural circulation.

 

[sophiecentaur]

Sounds a very dodgy design philosophy to me. Insulation (enough of it) is fail safe. If you are relying on warm water to be circulating around then you are effectively heating up the surrounding countryside - that heat will appear on your fuel bill, in the end. You only pay once for installing insulation.
I have seen photographs of massive waterfall that have frozen solid - that could be your filter system!

 

[FloridaJo]

I think the thing to observe is that the pipe will be coming from underground. Being that at certain depths the ground remains at a certain temperature for a period of time, one can calculate amount of pipe in contact with 55 degree Earth (or less depending on depth and time and zone) amount of heat lose from the exposure and calculate flow rate needed to replenish lost heat coming from the water that was in the ground (ground heat sink). Lots of factors.
But I mention this because it is the heat from the ground that is keeping the fluid from freezing, not so much the friction of flow.

 

[sophiecentaur]

But I mention this because it is the heat from the ground that is keeping the fluid from freezing, not so much the friction of flow.

Absolutely. It takes 4200J to raise the temperature of 1kg of water by 1°C. That's a lot of mechanical energy - equivalent to raising a 1kg mass by about 420m. The 'mechanical equivalent of heat' (as it used to be called) is huge - so you need to do a lot of work (against friction) to have much effect on the temperature of water flowing through pipes.

I can't believe anyone seriously would suggest that it wouldn't be worthwhile providing insulation round the pipes.