Exactly. The water feed into the house is somewhat above 0°C*. so if you keep feeding the pipes that slightly warmer water, it will not freeze solid… depending on temperature. Remember that the “heat of fusion” of water is about 80 cal/g. That’s a lot of heat the water has to lose before it turns to ice. The colder the temperature differential - the more likely the pipes freeze. In most house situations, you have a water pipe that runs for a short distance near a cold spot or poor insulation. keep water flowing past that spot and it never loses enough heat to freeze. How long the cold spot is determines how fats the water should trickle. If you find yourself in a “let’s go to Cancun” situation with no heat, better make it a noticeable flow.
People I knew who lived in older trailers (one place in town was notorious) usually had heat tape wrapped around the pipes for those nights when it hit -30°C. One fellow described how the trickle worked - his water pipes did not freeze - but his drain did. He solved the resulting problem by drilling a hole in the floor of his bathroom to drain the puddle. (Probably poor plumbing, a drain not inclined enough downhill so the water built up and flowed slower.)
Water stops contracting and starts expanding at about 4°C (39F) so most lakes and rivers, the surface cools and the water sinks as it contracts, until it reaches that 4C. Then the water on top expands, is less dense, so you get a cold layer on top that freezes. This can provide a cover to the water below. Even in the extreme north where -40C can happen, most lakes and rivers rarely freeze more than 3 feet of ice. I saw one locale where the rapids never froze, but the eddy currents in the pool below them created a 100’ giant disk that would spin slowly inside a slightly larger hole in the ice. Snowmobilers who are not careful risk going through the ice, particularly at corners on the rivers where currents create areas of rushing water hitting the far bank undercutting ice to form treacherous thin spots. Know where the faster water should be.
I live in an area that only occasionally drops below zero, and the utilities put out freeze warnings including advice to leave a faucet on a trickle/drip. I Am Not A Plumber, but as with @saje and @Sigene, it is also my understanding that while the additional flow can help prevent a freeze at the margins, it isn’t really much a factor in practical terms. It’s the pressure relief that keeps the pipes from bursting that is the main reason for the advice.
If there is a flow of water through the pipes, then the pipes haven’t frozen. Once they freeze, they will likely burst, so it’s a bit of a chicken and egg issue with which is the primary goal.
Heh. My Grandfather built a 230 lot Mobil Home park. My family ended up working there in one capacity or another. I started on weekends when I was about 10yo during the construction.
Anyway, especially for mobile homes, dripping the water can cause another problem. The drain lines can freeze. The tiny bit of water hits a long horizontal drain line run and freezes. Skirting on the MH helps, but all that plumbing is basically exposed. Sort of a dammed if you do dammed if you don’t situation.
I’m not convinced of the pressure relief argument. When pipes freeze they rip open on the side where the ice is. If relieving the pressure was the goal, wouldn’t a frozen pipe potentially cause a rupture anywhere downstream that’s not frozen since water is functionally incompressible? That doesn’t happen though. It also means letting the faucet drip does nothing until after the pipe has already frozen over, otherwise the pressure would be the same as it always is.
At a construction job a number of years ago we installed a snow melting system in the driveway with plastic pipes and a water/glycol antifreeze solution. As winter approached they hadn’t added the glycol yet so they just kept the pump running to circulate the water. This worked even below 20ºF but it was still a closed system so pressure wasn’t a factor.
Yeah, I’m not either. The water pressure in a dripping pipe is not substantially different than the water pressure in a totally closed off pipe. When pipes burst, it’s because the water in them freezes first in two cross sections of the pipe and then the water between those cross sections freezes and pushes out on the walls of the pipe. If the water gets cold enough, it will still freeze across the pipe and the pipe will burst. The reason the water doesn’t get cold enough is that the flow of water is constantly bringing new heat to the system. I guess another reason is that in a pipe with standing water, there will be relative “hot spots” and “cold spots” and the cold spots will freeze first. With flowing water (even a little), there’s mixing.
I believe the reason this works is that the pump is heating the water.
The colder the ambient temp, and the farther the pipe segment is from where the ground-temperature water supply pipe is, the more flow through the pipe you need to keep the water above freezing. For sufficiently cold ambient temps and sufficiently long pipes, yes, eventually you’ll get to a point where this trick won’t work.
The temperature of the incoming water varies depending on your location:
So during that ridiculous cold snap in Texas last year, it’s possible the trickling water solution might have helped some people avoid broken pipes, as the groundwater temperature there is much higher than, say, northern Minnesota.
After the Louisville/Superior Colorado wildfire a couple of weeks ago, the natural gas supply to the entire town of Louisville was shut off for several days to effect repairs. On the second night after the fire, the temp got down to 8F. Anyone whose house hadn’t burned down needed to drain their pipes and toilet tanks, and either empty their sink/toilet drains or add antifreeze.
I was trying to figure this out. We had a ‘deep freeze’ after Christmas, with temperatures as low as 8ºF. Our pipes under the house are insulated. (I have heat tape and one of those foam covers on the outside pipe.) The one pipe that froze was the hot water pipe to the utility sink in the laundry room. Since it’s the hot water pipe, it must be after the water heater. Why did it freeze, and the cold water pipe didn’t?
FWIW, we have a radiator-style space heater in the bedroom, next to the wall behind the utility sink. Given amperage restraints, it was not heating up the wall very well. I swapped out a different one, and the wall warmed up and the pipe unfroze. Of course temperatures up into the 20s helped.
If the pump is submerged then yes it will to some extent, but submersible pumps are generally only used in aquariums, wells, irrigation systems, and sump pumps. In most other setups the motor is external so they just heat up the air in the equipment room. The heat added to the pumped water from impeller friction and compression of gas bubbles appears to amount only to fractions of a degree.
The only point to letting the water run is that moving water is less likely to freeze. it helps if the water is a bit warmer or the pipe conducts heat, but the point is simply to keep water moving.
Something similar was used to keep ice from forming at a dam where I worked. They had pipes buried under the water that sent up a constant stream of bubbles, which kept the water roiling in motion. Even though the rest of the surface of the pond formed by the dam froze solid, the portion near the dam stayed liquid under the action of the bubbles.
If you want to know why they felt they had to do this, it was a deliberately built of wood dam that could have been damaged by expanding ice. The reason the dam was made of (rot-resistant) wood is that it was intended to fail and dump the contents of the pond if something went wrong. But they didn’t want it breaking if it didn’t have to.
If the pipe freezes, the drip will stop, and the pressure will no longer be relieved. (Plus which, you won’t have any water until it thaws again.) So unless leaving the water running keeps it from freezing for other reasons, the pressure release theory won’t work.
Pipes don’t always burst when they freeze, even without the water running. Depends on the material of the pipes and on other reasons, not all of which I know; but I do know a) that frozen pipes sometimes burst and b) that I have more than once had frozen pipes which didn’t burst; and generally the water hadn’t been running.
This.
However, if you’re on a house well, and if that well has limited water: be careful that you don’t run the well dry. One tap trickling isn’t likely to do it; but, with some wells (not all) multiple taps running significant streams might do so.
I took the info from what I thought was a good site, greatlakes.guide, but I guess I misinterpreted it.
Records show that Lake Michigan and Lake Ontario are the only lakes to have resisted freezing over since recording began in the early 1900s. This is due to their lower latitudes and large depths. The deep lakes provide massive heat storage and allow the lakes to better resist freezing.
That’s very interesting, so it’s not so much the freezing but the huge pressure in the pipes the expanding water causes. Opening the faucets slightly greatly reduces how much the pressure can build.
Sure, but pipes don’t freeze instantaneously, and are likely going to freeze gradually from the outside in. With an open spigot, as the outside layers start to freeze and expand, the liquid interior just gets pushed further along and outside the spigot (relieving the pressure from expansion). If a blockage starts to form inside the pipe, a water pressure is differential is going to form across it and automatically clear the blockage to keep everything moving. As you mention, once an entire section freezes solid, you’re likely not getting any more pressure relief (except back through the main) but up until that point you are.
A closed pipe is going to have no flow, and no ability to build up pressure behind/clear a forming ice blockage, and thus is much more likely to freeze in sections. 100% of the ice expansion is creating pressure that either has to go back towards the main, or expand/burst the pipes.
That’s fine, though. Simply having a cylinder of ice in your pipe is not a problem. The issue is that the volume expands during the freezing process.
Presumably there’s going to be one coldest spot where the ice forms first. Downstream of that, you want some pressure relief as more water freezes (and it will only drip as much as it needs to account for the volume difference). Upstream of that, presumably the water system can handle some degree of pushback.
There might be local temperature or convection variations that meant your hot water pipe was better cooled.
The other factor is that tap water typically contains dissolved solids that lower the freezing point. The chemistry of dissolved calcium carbonate is such that heating the water actually causes lime to precipitate out of solution in the water heater. The water leaving the water heater ends up with less dissolved solids and a higher freezing point than what’s in the cold-water pipes.
