Heating a Below-Grade Pump House

Seeking advice on a building project at Wheaton Labs. Photos for reference included below.

To sum up:
We want to know what sort of heating options can best help our plumbing endure freezing temperatures. Will electric light be sufficient to keep pipes from freezing?

More details:
The pump house up at the [off-grid] Lab is partially-underground. We're relying on thermal mass and some myco-insulation in the roof of the structure to keep the pipes from freezing. The structure is roughly 6ft x 6ft x 8ft tall on its highest side.

The past couple days had lows in the low 20s and upper teens Fahrenheit, and when checking the pump house thermometers today I noticed it was at 31 deg F. After consulting with Paul, we elected to add an incandescent bulb in there, aimed at a large rock and the plumbing lines. The battery connected to this will likely be depleted when I check on it again tomorrow morning. At that point however, I want to exchange it with another battery, and then include a smaller, DC power, USB-connected light instead of a conventional AC incandescent bulb. Refer to the photos below for more details.

We have a backup battery system installed that connects to solar power, though it's not hooked-up to these lights yet. It's a matter of finding an effective longer-term heating method. These bulbs are our current short-term solution.

Our thinking goes like this:
1. Heat the interior and the large rock with incandescent light overnight. The rock and nearby earth will absorb heat and will hold the temperature stable and above freezing until tomorrow.
2. Maintain the thermal mass with a smaller light. Even the power station, which diverts electricity through the USB port, will generate sufficient heat to keep interior temps above freezing. Based on typical power consumption, I'll only need to replace the battery after several days.
3. Double-check the air seals around the pump house roof/entry to prevent frequent air circulation and to promote still air.

Anyone with suggestions on how to create a low-maintenance emergency heat system for this scenario? Any additional questions to clarify our situation? Please post in this thread. Thanks in advance!

Here's the interior of the pump house, illuminated by the incandescent bulb. In this photo, the water pipes come from the back wall of the structure and across the floor to the other side, while the pump mechanical parts are to the right.



Here's the position of the second, smaller bulb.



Here's the exterior of the pump house, with additional insulation (burlap bags) placed around the edge of the opening. The roof will be resting on the burlap bags when closed, in an effort to reduce air circulation.

I would initially use a light bulb.  Put a thermometer inside and outside…compare the temps after a few hours.  That should tell you how much of a gain you get.

You don't actually want or need "light", you just need heat. It's hard for me to tell how much heat, but here are some thoughts.

Hubby's the Electrical Engineer, but he's away, so let's hope some of his bodging has rubbed off on me?
1. Is it possible to get a small "infrared" light. Chick brooding lights are probably bigger than what you're looking for, but that's the concept. Less "light" more "heat" for the electricity used, and it heats the objects it strikes - like that rock.
2. What's the budget? Hubby bought a nifty gizmo that looks like a plug, but it has a temperature sensor in it. So if the spot it's plugged in goes below a set temperature which is a few degrees above freezing, the plug allows electricity to the light bulb until the temperature goes back above the set point. If the concern is to not use too much battery power too soon, this might be helpful. (I can ask Hubby for the specifics if you're interested.)
3. Our back well shed has a very small electric baseboard heater in it. Normally, these run on the two sides of AC current using a pair of breakers. Remember that "Electrical Engineer" bit? He set it up with a plug that only accesses a single side. The heater therefore only gives off 1/2 the heat, but that's OK because we're just trying to keep everything enough about freezing that we don't have damaged pipes. The longer term plan is to use one of the above mentioned plugs so it will *only* run as much as needed to keep things from freezing. Again, if you want to try something like this, I can ask for specifics.
4. With all the wood in there, I hesitate to suggest any sort of combustion process, like candles or oil lamps. If the building's sealed well enough, a fire might be self-limiting. It also seems as if this  building isn't super easy to access and it would be a lot of work on a daily basis. That said, if you're having to swap out a battery on a daily basis, refilling an oil lamp might not be so bad either. One would have to choose the style carefully for a maximum of safety.

I'm a bit surprised that it's getting that cold in there.  I'd focus on leaks.  Getting it totally air leak-proof is essential.

I'm not sure about how to add heat off-grid but it rings true to me that a light bulb is wasting some fraction of it's energy making light instead of just heat.  And the things you plug a heater into that only come on if it gets cold are an option.  One style is called Winter Watchman but it runs from 35-45F.  I couldn't easily find one that ran below 35F only.  Maybe a bit more searching would reveal one.  I have a few milk house heaters and they often have a freezing setpoint.  They come on if it goes below that temp and turn off once their thermostat says it's happy.  milk house heater

Is the water system in use in the winter?  If not, drain it to prevent expensive freezing issues.  If it is, I'd trickle water out of a spigot somewhere so water is moving through the system during night.  That won't fix 25 degrees but it will help if it gets down to 30 in there.

Light is just radiation in the visible range - when it hits a surface inside it is reradiated as heat. No wastage.

As the pumphouse is mostly below grade and has some depth inside, I would put in an insulating blanket not far above the plumbing to keep the lower part of the space warmer than the top. An insulating blanket around the tank where it is above blanket level would be good too. Fasten the blanket along one side, and make flanges or something around the other walls so the blanket has a decent seal on those sides.

Just like the teepee a tarp ceiling just above the pressure tank would reduce the cold air dropping to the bottom and the heat rising to the top and getting cooled.

Short-term possible solution might be a PowerBlanket of the following size:  https://www.northerntool.com/products/powerblanket-rapid-thaw-ground-heater-3ft-l-x-4ft-w-model-eh0304-7002343#hotbar-keyspecs  .   These are construction grade for thawing ground for construction projects.

Optimum would be something that might give more coverage near floor-level than the one in the link, but at a top temperature of 140 degrees F., may be okay as-is.  Depending on how much you want to engage with rigid foam insulation, you may wish to consider cutting two 6 X 3 foot pieces (assuming inside space to be roughly 6 X 6 ft.) and add some supports so that these pieces will create a ceiling just above the ballast tank.  It sounds like you already have temperature sensors to ensure the space near the pipes is well monitored.  Place PowerBlanket in a 'best-fit' arrangement across the pipes, use thermostat to set desired temperature value of the blanket, then close up to retain heat.

Blanket appears to draw 400W:  Could a solar array drive this by day through an inverter to heat up the space which would stay warm enough through the night?  A standard 120V (AC) electric blanket for a bed draws ~120-150W....don't know the temperature range or safety aspects vs. PowerBlanket, but the bedroom blanket would be ~1/10th the price.  Maybe?...

I'm not sure how exactly these would perform, but when I sorted through the electrical area of the repair bay rolly shelf last winter, there were a number of ceramic heater elements that looked like this, only white: Ceramic E26 heater element

Where/how is the temperature monitoring being done?

bury the whole thing completely. if the pipes, pressure tank, pump, ect is 4 to 5' underground it will never freeze.

I built pump house below grade, almost exactly like in your pictures and buried it completely, the whole thing is underground. using lots of cement, about 6" thick on top, but logs, a membrane and dirt would do the same thing.
made a round mold about the size of a manhole, just big enough to get the pressure tank in or out and cast a rebar reinforced cement plug about 6" thick. the only way in or out is with a ladder and leverage or people power to move the plug off the entry hole.
the water has never froze no matter how low the temp.

Thanks everyone for your feedback and suggestions. A few responses here...

- Temperature is monitored by two digital thermometers, both inside the structure and one at a higher elevation than the other.
- A couple of the suggestions require typical AC power. To make these workable, I would need to devise a DC to AC inverter system. Limited supplies on hand for that, not to mention expertise (Jeff would normally handle these kinds of things but has headed on down the road).
- I like the suggestions for tarps, blankets, etc. as these are both passive elements and are things we already have on hand.
- We had tracked-down a thermostat-controlled outlet modifer, only to discover after purchase that it was AC only. Again, someone with legit electrical skills might be able to make it work but that's not an option right now.
- Not sure if Paul is interested in burying the entire thing (and yes we would opt for a membrane and soil cover in this case), as he's keen on the "hingeless hinge" design of the roof and wants to observe its effectiveness over time.

I will have a look today and see what can be done, then report back. I appreciate everyone's input!

Do you need to get into it over winter?  Could you bury the whole thing in a foot of hay/straw and then uncover in the spring?

Mike Haasl wrote:Do you need to get into it over winter?  Could you bury the whole thing in a foot of hay/straw and then uncover in the spring?

Organic hay or straw can be pretty pricey. Wilderness emergency shelters often use evergreen tree boughs to cover a frame. Are those any easier to get?

Both of these options may be popular with rats and mice, so I would want to be sure there weren't any holes for them to sneak into the building part.

We use a thermostatic outlet controller in our walk-in cooler when we switch over from cooling with a CoolBot controlled-A/C unit, to heating to prevent freezing temperatures. It is called InkBird (yes, AC) , it is a fully adjustable, differential thermostat, unlike another (Thermocube, also AC) which has fixed setpoints designed to prevent freezing, which is a bit ham-handed.

Maybe heating an inner enclosure, just around the pump and plumbing would allow for a bit of efficiency? Or a resistance heating pipe wrap? (A' la "heat the people not the space") Teens-twenties outside with 31*F inside, means you only need a few degrees of warmth...
Thicker earth burial on the outside might also help. As Mike said, strawbales, or maybe compost/leaf piles?

What's the water temperature from the well? And is there a use for water downhill from the well house? Here's an idea: barrel or stock tank in the well house with a float-fill-valve or float switch, and a metered drain house leading "somewhere" (wofati greenhouse, daylight, sump pit, ice making for wofati freezer?)... pump fills tank with groundwater at "X" unfrozen temperature, which acts as a thermal mass for the time it takes to trickle away, then is refilled... so the pump runs intermittently (possibly also thermostatically controlled) to refill the tank, ideally feeding some downhill use...

Maybe an alternate version using a large pressure tank for the thermal mass, and cycling the well on a timer, pumping to a storage tank or a pond at a higher elevation for gravity delivery later?

You could also do a drain-back solar hot water system... thermal storage tank in the well house, collector and solar PV (which you already have?) to run the pump outside. Run when there's sun, heats the thermal storage daily?

One thought, one question for Stephen.  

Thought:  I'm guessing straw/hay would insulate best, followed by piles of sawdust, then boughs.  Keeping them all dry would be helpful.

Question: I'm guessing that this well and pump is direct wired to the solar panels and it pumps when the sun is out and doesn't pump when it's dark.  So the only nighttime power you have is in batteries you bring to the site?  Or do the solar panels support a battery bank somewhere at the well also?

Mike Haasl wrote:So the only nighttime power you have is in batteries you bring to the site?  Or do the solar panels support a battery bank somewhere at the well also?

There's a battery bank inside the pump house, powered by the solar panels. This is what provides the well pump with power 24/7.

Our current (no pun intended) issue is that our resident electrical expert/tech is no longer here, and it's up to me to find a solution to (for example) powering a light with the DC battery power system in there. I don't know much of anything regarding electrical hookups. In a few hours' worth of research I ought to have a solution and can likely cobble something together with what we have here at Basecamp. But when I made my original post time was certainly of the essence and I was looking for ways to prevent freezing water in the pipes and the rest of the system.

We've since drained the pipes that lead to the rest of the hydrants on the Lab, so the only hydrant with water is the one directly off the cistern. Ideally, at the bare minimum we can keep this one hydrant running throughout the winter so that residents and guests up there can have water on-demand. I'm still terribly concerned about the pump house however, because for one thing if we need to do any repairs or replacements then it'll be me doing the digging. :)

Are those electric thermometers sending to a remote display outside the pump house? The short term battle here is being hampered by the need to essentially remove the entire roof and loose the whole house's heat to convection. I think the suggestions for a tepee-style layer(s) of fabric above the pipes would be a big aid for the immediate efforts.

Another short term stop-gap method that will bypass the need for electricity while working out a long term solution if you going in there anyway, heat water on the RMHs at base camp or the wofatis, put it in sealed containers, and place those in the pumphouse to provide more thermal mass. The less headspace you leave in the container, the more the water mass will retain the heat, in case it needs to be released over a longer period. Mason jars work great for this as I prefer to avoid plastic around hot fluids, but as large a container as you can handle while going down a ladder would likely be most effective.

Kenneth Elwell wrote:What's the water temperature from the well? And is there a use for water downhill from the well house? Here's an idea: barrel or stock tank in the well house with a float-fill-valve or float switch, and a metered drain house leading "somewhere" (wofati greenhouse, daylight, sump pit, ice making for wofati freezer?)... pump fills tank with groundwater at "X" unfrozen temperature, which acts as a thermal mass for the time it takes to trickle away, then is refilled... so the pump runs intermittently (possibly also thermostatically controlled) to refill the tank, ideally feeding some downhill use...

Oh, I wish we could do something like this. :)

An interesting feature about the well and cistern up at the Lab... Once the cistern is full, there's an overflow line that runs down a hill from the cistern - opposite from the pump house - to flow into a small pond. If the overflow was sent instead through the pump house, then we could attempt the solution you suggest here. As it stands, the best we might be able to do is leave a hydrant open at a trickle somewhere else in the Lab, down the line. Paul has stated that the water temperature there is a slightly-variable 45 to 47 degrees F, and a constant flow of some kind would provide the higher temperatures required.

Did some rearranging and data collection at the pump house today. Here's the current setup (although a blanket was added at a later point, before I wrapped work up there for the day).



In general, the temperature variance between interior and exterior is about 16 degrees F. This may increase if I can find a permanent light-heat solution that works with the battery system inside the pump house itself.

If I can manage to decipher the solar power station's inner workings I ought to be able to connect a light from there to provide a low-maintenance, constant heat source, and then turn water on for the Lab throughout the winter (which is ideal). I grabbed several photos of the main components today, and will spend time devising something to work with and hopefully apply tomorrow. The goal:

- interior light + heat,
- connected to the solar generator and battery,
- with water flow to the entire Lab.

Once something is done inside, then I want to start lining the exterior with pine boughs and other green debris to provide more insulation and keep the air still inside.

THANK YOU once again to everyone who provided suggestions and guidance, along with the useful questions that helped me think-through this new territory for me.

Stephen,   I skimmed the thread again and if the following was already answered, please direct me to the entry in the thread where it is noted:

A)  Does the mountainous region of Montana have average frost depth maps?  

B)  How deep below ground level is the pump and pipes and how tall above the floor is the top of the ballast tank?

I realize that this appears to be on a slope......and therefore the 'frost line' is going to run along a wavering plane that is modeled on the surface of the ground, thereby making the estimate a bit more difficult than normal.  The thinking here, dovetailing on that from many others already, is that you may already be at a depth where much of the system would not freeze....*if*....properly insulated above the pump and components.  Geothermal energy should be sufficient....and is the basis for the fact that hydrants at Wheaton Labs and elsewhere, plumbed to a valve depth below the frostline, can c0ontinue to operate in mid-winter.  If you are still considering some sort of light bulb as additional heat, then it may also be worth looking at ceramic flat or tubular heaters that fit into light bulb sockets (below....from Grainger.com).  I have no idea when it comes to resistance heating if there is some way (without an inverter) to run these off of DC current that is not 120V.  But a low heating option with good insulation layer above + depth near or below frost line should bring you to where you need to be in a Montana winter.  Even if you are undergoing, temporarily, one of those bone-crushing -30 F spells this season, you no doubt will be 'rescued' by one of the legendary sieges of Chinook wind, causing you to break out the khaki shorts, t-shirt, and sandals by noon the next day! ;-)     Good luck!

When considering long term solutions involving the electrical install, there are many items out there prefabricated to provide direct heat that include thermostats, like pads or wires to run along pipes and wrap the exterior of a tank with. Gutter clearing and seed starting are a couple of common ones that come to mind, if not something predesigned for this purpose. I'd think direct application would be much more energy efficient if there is any question of solar power supply...

You can hook a solar panel directly to a heating element with no need for a battery or controller.

I have connected the output of a 320 watt panel(36volts) directly to a water heater element put in a 55 gallon barrel of water.  Any time the sun shined the water absorbed the heat and released it throughout the evening.

I have also connected the same panel to a length of nichrome wire that was sandwiched between two scraps of marble counter top.  Any time the sun shined the marble heated up.  I liked this method better because it got hotter and then I could carry the top marble slab anywhere I wanted to in the evening for a bit of extra heat.

In both cases you have to use ohms law to roughly size your heating element to match your panels output so you use up all the power the panel makes.  It is much easier to do that with nichrome wire because you can cut it to any length.  But with a water heater element there are only so many options and you just have to pick one that is vaguely close to what you need.  I personally bought a roll of nichrome wire that was very close to 1 ohm per foot.  That way the calculations were easier.  If I recall correctly I needed 11 ohms so I used 11 feet of wire.  I think the wire cost $16 for 100 feet so it isn't that expensive.

Another option would be to make a solar heat panel and blow air from it into the pump house using a fan connected directly to a solar electric panel.  I find free fans from electric radiator fans from vehicles going to the junk yard.  A 12volt radiator fan can be connected directly to a 36 volt panel and run for over a year without burning out.  That way the fan only runs when the sun is at least partially shining.  The brighter the sun the faster the fan blows and the more heat the heat panel makes, the dimmer the sun the slower it blows and the less heat the panel makes.  You can set up the whole system so the air coming out of the panels is always the same temp regardless of how bright the sun is shining.

How deep is bedrock up there? You could combine J Hillman's direct solar to water battery concept with a large reservoir dug out as deep as possible to raise the resting temperature of the water battery, allowing the solar to get it a few degrees warmer when on, and possibly keeping the tank above freezing passively like geothermal.

Hi Stephen

Air temps don't correlate directly with water temps in a space. Ambient heat, whether in air, wood, stone or water has momentum (not literally, but it's a way to think about it). It takes time for it to change, whether up or down and thus, for example, 12 hours where the air temps are 20F. might lower a 40gal tank of water from 45F. to 35F. If, during night, the tank lost 10F. and ended up at 35F when the sun came up and then heated 12F. so it was 47F by sundown... That would make a sustainable system for the tank.

The above doesn't speak to individual pipes where the volume of water is much less and thus more subject to cooling quicker. But, it does present an important way to think about sustaining temps. And it may  indicate that the tank is not the problem and you can concentrate your thought on the pipes alone. There are just broad strokes of thought but they help prioritize and point out what's important and what's not. The _average_ temperature is very useful information.

Air leaks and most especially convective currents of air which carry the heat up and down (the pipes are "down", right?) are a _major_ problem when trying to maintain temps w/out heating. Major. So sealing all your air leaks would carry a huge ROI (ReturnOnInvestment) or BangForBuck. The ideas above about creating a tiny sealed space about the vulnerable pipes is spot on, too. First, stop that air from moving around those pipes - iow, seal your tiny pipe  enclosure (blankets, tarp, plastic sheeting...) tight to the walls. Gorilla tape comes to mind, but gravity using boards or such to capture the edges of your membrane against something would work well also.

But if you had more exact information you would be able to think about this more productively. Do you have an IR laser thermometer? They are relatively cheap, less than $100 and often much less. For your purposes, accuracy isn't critical - "close" is good enough. What are the temps of the walls? Floor? Ceiling (bottom of the roof)? When you "step in" first thing after opening it up, check immediately, starting with the ceiling (that's the one that will change fastest). If they are all above 30F, you may be fine and IAC, you're way better than if they're 20-25F. If they're low, as in outside air temp... They shouldn't' be. As was mentioned above, soil temps are normally well above air temps in winter, especially deeper below the surface. If the floor temp and lower wall temp is 25F, there is probably  really major air leakage through that space. And _that_ tells you where real effort is needed and will pay back quickly.

Eventually, a drain back solar "water" system may provide heat using a very small pump to move the fluid. A heat pump. But those require careful planning and some technical skills and so are not an immediate fix.

Best luck
Rufus

Paul released a pump house related youtube vid: