What are the odds this will end up bad? Frozen pipes?

[jadnashua]

I own a townhouse in a condo. The unit next to me has been vacant for awhile (bank forclosure). The bank just had the place 'winterized' by shutting off the water, draining the pipes, and putting antifreeze in the traps, and shutting off power and heat. This is all well and good, but the unit is in the middle of a string of units. While their water may be off, the supply runs THROUGH the unit, in the ceiling of the basement (which is closed in and insulated) to units down the row (we don't have individual water meters, so there was no reason to run the supply in from outside for each unit). Last winter, a unit on the end had it's pipes freeze, split, and create a bunch of problems, but we were able to get in (it also was forclosed), and put a cap on the pipe near where it came in from the unit next door. Then we insulated it against the firewall between the units to try to trap some heat from the occupied unit. It made it through the rest of the winter without further problems. There is a common firewall (double-stud, insulated sound and fire wall) between each unit and right now, the units on either side of this forclosed unit are occupied and would be heated over the winter.

During the day, there'd be sporatic water use, and even on a really cold day, my feeling is it is unlikely the pipe in this unit might freeze. I'm somewhat worried that at the end of a long, cold night in the middle of winter, though, it could, since there'd be no flow.

How likely is this to happen? WHen it happened earlier this year on that end unit, it had been really cold for about 5-days, where it rarely got out of the single digits, even during the warmest part of the day. Trying to explain this to the bank, and forcing them to now pay to provide some minimum heat may be hard. Suggestions?

[Terry]

It would be rare that an interior unit would freeze pipes in a basement if the other units are heated.

The ground floor will be the warmest place in the unit. It becomes more of a worry on the upper floors away from the tempering of the ground.

For example, I've almost never seen a frozen pipe in a crawlspace.
Even uninsulated pipes in a crawl do well.

[jimbo]

I gather that there are supply stops in that basement where that one unit taps of the main run?? You are probably OK, but it is an issue I would raise with the HOA and ask them to take it up with the bank. Are you in like 10ºF climate or -20ºF? That would change my opinion. If it's a basemement, I gues the only damage is to the bank's unit, but you would lose your pressure, and the HOA would pay the water bill!!

Perhaps the bank would agree to put a temporary remote reading thermometer in the basement. I'm talking about something you can get a HD for $25. You could put the sensor in the basement and the digital readout in your place. You could monitor how things go as the weather gets colder.

[jadnashua]

When the unit froze earlier this year, it had been getting down to -8 to -10 during the night, and only up to about 10 max during the day and it did this for about a week. Unusual, but it happens. Since it was on the end, and the water was fed from the other end, no 'normal' flow into that unit. This is not the case on the one next to me, as there are heated spaces on both sides. But, for probably 6-8 hours in the coldest part of the night, there migth be no flow through there at all. That is when, if it was to occur, it might occur. Just trying to understand the likelyhood.

I'll see if I can talk the bank into springing for a remote sensing temperature device.

[Thatguy]

How likely is this to happen?

Google the exact phrase
"outside design temperature"
for your location.
In DC it's +14F; 99% or 95% of the time, temperatures here are warmer than that. There's also one for the expected high temp.
And Googling
"heating degree days"
will give you how far below 65F your temps have been. +10F all day long is 55 HDD.

If broken pipes costs $1000 and a sensor costs $50 you should get a sensor if your likelihood of breaking a pipe is more than 100x[$50/$1000] = 5%.
For five kilobucks worth of broken pipes buy the sensor for >1% likelihood.

[Dana]

The design temp isn't necessesarily going to tell you much, since construction details and not-insignifant issues like the deep soil temps & soil type make a large difference in the sustained or average outdoor temp at which the pipes will reach 32 F in an unheated space. Unheated spaces with large air infiltration factors will freeze up at outdoor temps a few 10s of degrees F above where it might freeze if the building was essentially air tight. Foundations insulated down to below the frost line in low leakage areas would be nearly immune from freezup below-grade, but with the supply pipe passing through the ceilings of above grade condos will be more susceptible. Infiltration rates dominate the equation here more than deep soil temps, but the surface area to heated space vs. to the exterior could be as big or bigger.

But the most-susceptible in a townhouse situtation would be the end units, since they have signficantly more surface area exposed to the exterior air. If only an unheated zero-flow end unit froze up, that's not much of an indicator of risk for a unit between two heated units even with minimal flow. The empty unit you're concerned about has 2x the parasitic heating from ajacent units that the end unit has, and only about 3/4 of the exterior exposure. I'm thinkin' you're not in trouble until the high for the day is well into negative numbers for a few days, which is even rarer than the cold-snap you described where the end unit froze. Twice the heat gain and a lower heat load buys a lot.

Terry- in New England crawlspace plumbing freezups are common, but cold snaps are 20-30F cooler than you'd ever see on the western slopes of the Cascades. (For me -20F with a high for the day of -5F is rare, but has happened within the past 20 years. In Seattle -5F LOWS are extremely rare, even though postitve single-digit lows happen often enough. I've seen freezups occur even in full basements (even when the house is occupied) with uninsulated field stone foundations, and I'm far warmer than cooler parts of New England.

[jadnashua]

Well, if this adds anything, one outside basement wall is only about 1/2 below ground level (it's poured concrete with NO insulation on either side), and the other side is stick construction with a garage door (drive in garage in about 2/3'rds of the basement space), so there's only a small portion that actually has any decent insulation. We've had kitchen sink supply lines freeze in occupied units if they turned the heat down low, but they drained those lines.

[Thatguy]

100' of half inch pipe holds one gallon. If a metal basin of water containing one gallon totally freezes in this environment in X hours then the pipes will freeze faster since they have more surface area.
A one foot section of copper pipe full of water and exposed to air would be a good test specimen, a "canary in the coal mine", and a thermistor on the copper registering less than +31F probably means it frozen solid.

It also depends on the thermal inertia of the building and the pipes and the water in them.

I think the gov. should give me a generous grant to study this problem!

[Dana]

Well, if this adds anything, one outside basement wall is only about 1/2 below ground level (it's poured concrete with NO insulation on either side), and the other side is stick construction with a garage door (drive in garage in about 2/3'rds of the basement space), so there's only a small portion that actually has any decent insulation. We've had kitchen sink supply lines freeze in occupied units if they turned the heat down low, but they drained those lines.

It's a crime to build without insulating foundations in New England (or at least it ought to be, IMHO), but I still see it all the time. Building codes may eventually catch up, but I'm not holding my breath. Between the plumbing protection issues and supporting the extra heat load R12-R15 foundations are as cost effective as R19 fully conditioned space exteriors.

If the feed line that 's still pressurized is running above the ceiling of the above ground portion the relevance of earth coupling at the foundation is probably an order of magnitude behind the heat gained from the adjacent units- freeze up risk is still much lower than on the end units.

[Thatguy]

The ice in the pipe probably has to be considerably below +32F for it to expand with enough force to break a pipe, and the burst strength depends on the pipe size & material.
Once this temp. is known, a thermal sensor on the coldest section of pipe would probably give fair warning and somebody probably makes a programmable temp. alarm with a thermocouple or thermistor at the end of a cable.

A household fridge/freezer is between 0F and 10F, if any DIYer wants to do some sample testing. A 24 hour cold soak with a small length of water-filled pipe crimped closed at each end should show a crack, if there is ever going to be a crack.

[jadnashua]

I think it's a 2" pipe with 3/4" branches off it for each unit - I know it starts out as 2", it may be smaller by the time it gets to the middle (it's hidden in the ceiling of the basement). And, it's about in the middle of about a 200' run. Not worried during the day when there's likely some water use, but at night (well, the end of night when it's the coldest) and before people start to use water again.

[Thatguy]

So the incoming water is at 40F and you've got 8 hours to drop the water temp to 32F and then freeze some part of the 60 pounds of water, then cool it further so it expands and bursts the pipe.
This is almost a problem for college physics. Lemme' dig up some old textbooks.

Another way is to ask local plumbers how often they have pipes burst per week, out of how many customers they see per week.
That's your odds per week of a pipe breaking.
Maybe 1 in 20?

[nukeman]

Are these copper pipes (makes a big difference)? I assume that the pipes are uninsulated, etc.

I ran a quick calc based on 2" type L copper, and basically if the air in the basement gets to 32F or less, you will certainly have to worry. Basically, since the walls are thin on the copper and the thermal conductivity is high, it takes a very small temperature difference (a fraction of 1 degree) to take 40* water to ice over an 8 hr period. I also considered just conduction (no air movement). If there were some convection (and there could certainly be), they would freeze up even faster. There will be some axial conduction of heat along the pipe/water from the heated units, but typically, this will only help over a very small distance. The water towards the center of the unheated unit will not even know that the heated units exist.

Where water expands when turning to ice, it is right near the freezing point. The max density for water is at about +4 C. The density drops (water expands) as you approach 0 C (especially right at 0 C). Once it turns to ice, the ice will then start to shrink as the temperature drops, so I would only consider what it takes to get it to freeze and nothing beyond that.

Your best bet would be to mount a temperature probe on the wall of the pipe in the center of the unit. It would be nice to have one with an alarm that could warn you at say 34-35F. There will always be uncertainty in the temperature measurement, so you want to keep on the high side of freezing. If you wait until it says 32F, the real temperature may have already fallen to 31F or less and you might be too late at that point.

[jimbo]

The ice in the pipe probably has to be considerably below +32F for it to expand with enough force to break a pipe, and the burst strength depends on the pipe size & material.
.

Just a couple of years ago, we had a rare cold snap here in San Diego. The overnight temp probably got down to 26º. We had a pipe burst on the solar water collector on the roof of the pool house! Temps below 40 are pretty much unheard of , so there is no attention whatever paid to having exposed water pipes. Meters area at ground level, main shut offs out on the side of the house.

[jadnashua]

I've measured the incoming water temp in the middle of a cold snap in my upstairs bathroom at 33-degrees, and below 40 often in the winter. That's after it has traveled about 100' or more 'inside' the building. The pipes are in the ceiling, and they're covered with drywall, so hopefully, there isn't a huge amount of convection, mostly still air, but no heat source.

We're going to talk about it at the board meeting, and may pay the utilities, set the heat to maybe 50 or so, and put a lien on the place for whomever ends up buying it that would have to be cleared...just don't want to have to hassle with it in the first place. Not sure if the bank actually understands that 'winterizing' the unit does not absolve them of risks of damage...maybe internally in the walls, but not the through pipe.