So far this winter has been alternating between periods of unseasonably warm and unseasonably cold. A few weeks ago it was above freezing and rained, but then it plunged into the minus temperatures about a week ago, with one daytime high of only minus 2.6 degrees.
This month we’ve now had three days where it got below -10 for lows, and ten days where the high temperatures have been above freezing. Needless to say 40 degree temperature swings are difficult to deal with even though it’s not that uncommon in Minnesota. (Have I mentioned lately that Minnesota weather sucks?…)
Normally I’d be grateful for the warmer temperatures in the middle of winter, but this year many of the warmer days have been followed by extremely cold weather. And it’s the bitterly cold weather that’s been causing problems.
I have a security camera inside the temporary enclosure around the radiant heat system in the new building. It points at a thermometer and the boiler so I can monitor the temperatures inside the enclosure and watch the boiler from afar. The first season we ran the boiler (two winters ago) it was suffering faults and shutting down (due to incorrect configuration settings) and when it does so an LED blinks on its front panel.
Once I figured out the proper configuration for the boiler it hasn’t had any problems but I still monitor the system and the temperature inside the enclosure, but only occasionally.
When it started entering one of the extremely cold spells I started getting concerned about the ability of the radiant boiler to handle the cold. Since our building is un-insulated (yes, still) I was concerned that the boiler would be unable to raise the water temperatures high enough to be able to prevent it from freezing.
It’s something that I wasn’t as concerned about in previous years, mostly because I simply hadn’t thought about it. Last year’s winter was so mild it didn’t matter, and two years ago was the first year we had the boiler running and with all the modifications and other work on it that thinking about the ability of the radiant system to keep up with the cold was the least of my concerns.
I thought I had frozen some of the radiant tubing the first winter we had the system running, but the problem magically fixed itself and I didn’t consider it further.
The boiler is only capable of raising water temperatures so many degrees based on the flow rates. In our radiant heat system for all three heating zones the total flow rate is over 7 gallons per minute. At that rate our boiler is only rated to be able to raise the water temperatures around 25 degrees.
The ideal would be that all three zones wouldn’t be running simultaneously, in which case the boiler could heat the water more. But when it’s cold all three zones in our system are running continuously.
In an insulated building the radiant system will progressively warm the building over time. But in the case of an un-insulated space when its extremely cold the boiler will be unable to make gains with the result that the slab will keep getting colder and colder.
Granted, I know very little about thermodynamics but it would seem that if the outside air temp remains around 0 degrees or colder that keeping the slab above freezing is going to require more than a 25 degree temperature differential, especially over an extended period of extreme cold.
Since moving water freezes at lower temperatures I tried to convince myself it would be OK for a short time, and warmer weather would eventually arrive at which time the heating system would be able to make some gains again.
In addition, normally (I had thought) the heat emitted by the radiant system was enough to heat the area inside the temporary enclosure around the boiler, pumps and manifolds. But a little over a week ago the temperature dropped to -11 degrees at night, and the next morning I noticed that the temperature inside the enclosure had dropped below freezing. With the boiler struggling to put out enough heat it just wasn’t generating enough residual heat to keep the area inside the enclosure above freezing.
Here’s a screen capture from the security DVR that day showing the below freezing temperature inside the enclosure:
Image may be NSFW.
Clik here to view.
I went out and placed a small electric heater inside the enclosure, but by that time in spite of the fact that all three heating zones/pumps were on, only one of the manifolds (my shop zone) was showing any flow. A single loop on one of the other manifolds was showing flow, and the big manifold had no loops with flow. To my dismay I realized that that meant I had likely frozen all but one loop on two of the three heating zones.
As you might recall, I opted to not add glycol antifreeze to the radiant system for two reasons. The first was cost, and the second was that glycol would reduce the efficiency of the system. Since we didn’t plan on shutting down the radiant system when it was cold there was no reason to add glycol.
I’m still not inclined to add glycol, but between insulating the building and/or adding glycol, it’s pretty clear now that we needed to do one or the other if we were going to run the system…
The next day I checked the floor temperature sensor and found it at 18 degrees. I had no doubt then that those zones had indeed frozen.
Considering that, I think it’s pretty likely that at least part of the system froze the first year as well, but then ignorance is (was) bliss.
The radiant PEX tubing is supposed to be able to handle accidental freezing, and the Uponor brand we paid extra for is supposed to cope with freezing better than cheaper brands.
But worst case the frozen radiant tubing could expand and crack the concrete floor. In our installation our PEX tubing is lying directly on top of extruded polystyrene insulation (with the concrete poured on top), so I convinced myself that the tubing would seek the path of least resistance and expand downwards into the polystyrene instead of cracking the concrete.
It didn’t make me any happier about the situation though because I couldn’t assure myself that no damage had been done, and it would remain frozen until we could heat the entire building with some other heat source (impossible until it was insulated regardless), or until spring and warmer weather arrived.
Warmer weather was forecast the next weekend, so I decided to leave the pumps on the frozen zones running, thinking that it would possibly help unfreeze the system. In the meantime all I could do is stress about the damage that had already (possibly) been done.
Then about a week after that happened I came home from work one night to see the alarm light for the septic system on. That night I tried to delude myself into thinking that it was a wiring fault of some kind – we had just had the system pumped out last fall, and the pump was replaced only two years before that.
In our septic system we have three tanks. A septic system is designed to let the solids settle out into the main tank(s), where they continue to decompose. The system is pumped out every few years, primarily to eliminate the solid waste accumulation. The liquid effluent runs out of the tank into the drain field where it percolates through the ground.
Our system has two holding tanks, and the third is a tank that holds a lift pump that pumps the liquid effluent up to the drain field which is several hundred feet away (and uphill). We had just replaced the pump ([1] [2] and its electrical wiring) two years ago.
Unfortunately the next day I found that the lift pump tank was overflowing, and although the pump was running the effluent in the lift tank wasn’t being pumped out the drain line to the drain field. I called the septic pumping service and they told me they had several cases this year of frozen drain lines, due to the extreme cold and lack of snow cover on the ground.
We would have to get the system pumped out, and then later have a heated jet sewer cleaning service come and try to unclog the frozen drain line.
So the cold weather had claimed both parts of our radiant system and our septic system.
By then though we were back into an above-freezing warm spell (with more rain/freezing rain), and when I checked the radiant system I noticed a small leak on the bottom-most circulator pump (on one of the frozen zones). I also found that the relatively new ball valve main water shutoff was leaking – presumably due to frost damage as well.
I had just replaced that same ball valve this last summer because it was cracked and believed that since it was inside the warm enclosure that it couldn’t have frozen. But it was clear now that my make-shift radiant enclosure didn’t keep the space above freezing in colder temperatures – that ball valve had most likely suffered frost damage too.
The same day they came and pumped out our septic tank I ran to the home improvement store after work to buy an oil filled electric heater to use in the radiant enclosure. More cold weather was forecast for the end of the week and I didn’t trust that the small electric heater I had put in there would ensure that the radiant enclosure stayed above freezing.
I also bought another replacement ball valve.
When I got home and went into the utility area of the new building I immediately realized something was amiss, as there was a smell of hot/burned electrical hardware in the air.
I found that the radiant system had no water pressure, and that the two previously frozen radiant zone pumps were running dry (a big no-no for water circulating pumps as the water cools the pumps). Obviously there had been a leak where much of the water drained out of the system, and it had happened some time earlier because the floors weren’t even wet (the prime suspect was the aforementioned leaking pump).
We had only one more day forecast to be above freezing, and then we were headed back to the teens and the single digit temperatures by the end of the week.
The decision was clear: there simply wasn’t time to fix and debug the inoperative/drained loops since I didn’t know where the leak had occurred and whether or not I had damaged the pumps. So I knew that I would have to try to get as much water out of that part of the radiant system as I could and wait until the building was insulated or for warmer weather to try and make repairs.
As luck would have it, I found that the two days of above-freezing weather had been enough to warm the slab and thaw the frozen tubing. But it would re-freeze in short order if I didn’t get the water out.
I spent half a day using compressed air to blow out as much water as I could out of those tubing loops, leaving only my shop zone filled/operative. I put the compressed air on the return side of the manifold, closed all the flow control valves, put a drain hose on the supply side, and one by one opened the flow valves to blow out the water in each loop.
I also replaced the damaged water main ball valve.
As I write this the temperatures outside are below freezing and still dropping. Hopefully I got enough water out of the radiant system to prevent any damage (I’m pretty confident I did – but then these exploits have demonstrated that I really don’t know what the hell I’m talking about…).
The good part is that we should save some money on the natural gas bill with two thirds of the radiant system now shut down.
As for the septic system, we probably have a few weeks before we need to decide whether we need to get the drain line unfrozen or simply have the system pumped out again. For the cost of the drain clean-out, we can have the system pumped out a couple of times. And if we’re really lucky, the drain line might unfreeze like the radiant tubing did.
Meanwhile I’ve made little progress on our flooring project in the house. I removed the old patio door between the house and the four season porch to find rotted sub-floor that needs to be replaced. The hole was big enough that I had to temporarily screw a piece of wood over it to keep our dog, Basil, from falling into it.
Image may be NSFW.
Clik here to view.
In that same area is the floor heat duct that needs to be relocated.
I thought I could simply remove/plug the heat duct temporarily and relocate it later, but it’s the only duct on that side of the house and with the cold weather I didn’t think it wise to eliminate it. So now I’ll have to move the duct; a project that alone will probably take the better part of a day.
Then I’ll have to cut out and replace the damaged sub-floor.
Additionally I found that the sub-floor elsewhere suffered more damage due to the leaking dishwasher than I had realized. The sub-floor has to be relatively level to be able to install the laminate flooring and the OSB sub-floor expanded at the seams due to the leaking water; I don’t know if simply sanding the seams is going to work or whether we’ll have to use floor self-leveling compound or replace the sub-floor there as well.
Overall, it’s been a pretty lousy month. But then it is 2013 now. Hopefully it’s not going to go like this all year…