Preventing Pipe Freeze in Cold Weather

[dyounger]

Hi Guys,

I am not an engineer and was hoping someone could help me with a question I had about piping in cold weather. I am setting up a temporary 4" Bauer water pipe that I need to keep from freezing. It will be outside (non insulated) in February in Ohio. I am expecting temperature ranges from 0-32 degrees and not many days above 32. I also expect my incoming water temperature to be around 40 degrees. I have a couple of questions about how to prevent the pipe from freezing:

1. How much flow(GPM) would be required to keep the pipe from freezing?
2. Worst case scenario- 0 degrees, 25 mph wind speed, 40 degree water temperature, how long would it take the water to freeze with no flow?

Heat tape and insulation is not really practical for our application due to the short period of time it will be in use (7-10 days). But it will be extremely critical that none of the pipes freeze.


THanks ahead of time

 

[Simon Bridge]

I am setting up a temporary 4" Bauer water pipe that I need to keep from freezing. It will be outside (non insulated) in February in Ohio. I am expecting temperature ranges from 0-32 degrees and not many days above 32.

Then you probably won't be able to keep it from freezing... but I guess if you are prepared keep the water flowing at high speed from a warm source (just shunting it around in a cycle won't help).

1. the flow needs to be faster than the heat flow out of the pipe to the surroundings. The buildup of ice around the inside diameter should help. Could you bury the pipe in dirt or snow?
http://en.wikipedia.org/wiki/Heat_transfer_coefficient

2. high winds don't cool much faster than slow winds - as long as the wind is fast enough to take the warm air away at least as fast as the heat arrives. The worst case scenario - cooling is proportional to the temperature difference: see "Newtonian cooling" to see what you are up against.

I have a feeling it will be more cost effective to rig up some sort of insulation.

 

[etudiant]

The length of the exposed pipe is not stated. If it is just a stub, water flow may be enough, but if it is longer, the water will freeze from the outside in within the pipe.
Surely a few rolls of insulation plus a few hundred feet of heat tape are the safer way to go if it is indeed critical the pipe not freeze. Belt and suspenders for critical tasks was always good policy.
That is especially true because no winter is average. If there is a severe winter, can one blame the statistics?

 

[dyounger]

The pipe will be approximately 700 LF .The flow in the pipe is going to be around 300GPM under normal operation, but we are would like the reduce the flow, if possible, during shut downs. Do you guys think it is possible to reduce it ?

 

[dyounger]

Another question:

Is anyone aware of some sort of pipe heating system that is not electric heat tape? We are going to be on a remote job site that will have power limitations.

 

[etudiant]

That is helpful input, the flow volume and the pipe length. I can see why heat tape might be difficult.
The flow rate is pretty decent, so even if you cannot prevent ice from forming within the pipe, I would be astonished if the pipe freezes solid.
Still, given it is critical that the pipe does not freeze, heat will be required. Maybe enclose the pipe in a polyethylene sausage inflated by a blower feeding air heated by a burner? It will help insulate the pipe in any case and the heating would give peace of mind. Heavy duty poly is pretty available in long sheets, so rigging up something that works should be cheap.

 

[Baluncore]

The pipe will be approximately 700 LF

Please pardon my ignorance, but what unit is an LF ?
What is 700 LF in metric or SI ?

 

[etudiant]

Not sure, but I think it means linear feet.

 

[dyounger]

Yes, linear feet.

 

[Baluncore]

Having operated traveling irrigation systems in winter, I found it best to irrigate overnight and move the pipeline during the day. That prevented freezing of the pipeline which was a difficult problem to overcome when it happened.

If you can keep the water running then all should be OK, but if it stops you should vent the pipeline immediately with compressed air, nitrogen or possibly exhaust gas. That could be done from either end, but must be done before ice begins to form. Using “lay-flat” hose might be a solution as it could be mechanically rolled up to remove water immediately it is not flowing. Maybe look at fire fighting hose handling technology.

If that fails then you will need to lay, or have laid, another empty pipeline in preparation.

If there was still a chance of freezing the pipeline then some method of recovery from the frozen situation should be prepared. I would consider running two small plastic pipes inside the main pipe, both also filled with water. In each of those small pipes I have would run a fence wire. When the main pipe was frozen I could run an electric current from a generator through the two wires, (there and back), to melt the ice. The inner pipes provide electrical insulation. It may be an agricultural solution but when all else has failed it might recover the situation economically.

Covering the pipeline with some commodity such as small straw bales or sawdust should be considered as any insulation will make a big difference.

 

[Chestermiller]

This is a fairly straightforward heat transfer problem. The idea is to get an estimate of the rate of heat loss from the pipe. Once you do that, you can calculate, for a given water flow rate what its temperature drop will be from entrance to exit. To get the rate of heat loss, you multiply the temperature driving force (max of 40 F) by the surface area of the pipe (nominally 760 ft²) and by the overall heat transfer coefficient. There are three resistances to heat transfer in this system which combine to give you the overall heat transfer coefficient: the convective heat transfer resistance outside the pipe (from the pipe to the air), the resistance through the wall of the pipe, and the resistance on the water side (from the mean bulk temperature of the water to the pipe wall). The dominant heat transfer resistance is probably going to be the one on the outside of the pipe. Using the appropriate heat transfer correlations (see e.g., Transport Phenomena, Bird, Stewart, and Lightfoot), you can calculate the three resistances using (1) the convective heat transfer correlation for flow in a pipe (water side resistance), the resistance for heat conduction through the wall, and the heat transfer correlation for flow over a cylinder (outside the pipe). I did a rough calculation of the heat transfer coefficient outside the pipe for the case of air flow across a cylinder at 25 mph, and got a heat transfer coefficient of 35 Btu/(hr-ft^2-F). I didn't know the thickness of the pipe wall, and didn't feel like doing all the work for you, so I didn't calculate (i.e., neglected) the heat transfer resistance on the water side. So, 35 would be a worst case number, but it would have to be redone and checked. With a heat transfer coefficient of 35 and a maximum driving force of 40 F, I calculated an upper bound rate of heat loss equal to about 1 million Btu/hr. At 300 gpm, the mass flow rate of the water is 150000 lb/hr, and, with a water heat capacity of 1 Btu/lb-F, this all translates into a temperature drop of about 7 degrees. So, with this worst case scenario, as long as you keep the water flowing at 300 gpm and keep its temperature at least 40F, it shouldn't freeze. Of course, these calculations have to be repeated to make sure that there weren't any glitches. I leave that up to you. I also leave the job or refining the calculation to include the other heat transfer resistances up to you. But so far, it looks like you're OK.

Chet

 

[dyounger]

Thanks for the response Chester. So the more surface area I have exposed to the elements is going to ultimately = higher GPM to keep it from freezing?

 

[Chestermiller]

Thanks for the response Chester. So the more surface area I have exposed to the elements is going to ultimately = higher GPM to keep it from freezing?

If the reason for the additional surface area were increased length of the pipe, then no question. If the reason for the additional surface area were increased pipe diameter, then the heat transfer coefficient would be lower, but the area would be higher. However, I think that the increased surface area would win out over the lower heat transfer coefficient, so the heat loss would increase. (Of course, I would want to check on this to be sure). So, pending such a check, increasing the surface area would require increased GPM to prevent freezing.

Chet

 

[256bits]

You also might want to think about it this way:
Approximitely 2 feet of you pipe contains a gallon. So with 700 feet of pipe and a flow of 300gpm, your pipe 'empties' its contents ( to be replaced with new water), about every 60 seconds. The gallon at the beginning will have to suffer a drop in temperature and chance of freezing in 60 seconds until it exits.