Holding tank that freezes- when should it be emptied?

[StevSmar]

At our cottage the sewage holding tank freezes over winter. I think it freezes solid since there is no insulation around the tank or on it’s neck and cap.

One fellow who pumps out holding tanks said that as long as they’re below 2/3rds full, freezing has a small chance of damaging the tank.

Since the pumping contractor is unlikely to be calculating volumes, I think by 2/3rds full he meant is when the tank is dipped.

Our tank is cylindrical with spherical ends.

Does anyone have an opinion on whether the 2/3rds rule of thumb is reasonable?

It seems to me that as the freezing penetrates deeper, 2/3rds full will start to create quite some forces on the tank. The tank doesn’t seem to lose liquid, so I doubt it’s been damaged over the 25 years it’s been installed- but I always get nervous if the tank is over half full at the end of the season.

 

[skippyelwell]

What is the tank made out of, plastic, concrete?
If the tank is completely covered by earth then there is likely the same freezing pressure from outside the tank as there is inside.

 

[Susquatch]

I'd think 1/3 is better than 2/3. Even if the earth outside the tank is dry and can't hold back the ice pressure, the inside ice block will simply slide up the surface as it expands. At 2/3 or even 1/2 the block is trapped.

And yes, I'm pretty sure dip is a better way to gauge the risk.

Maybe he meant 2/3 empty instead of 2/3 full..... LOL!

 

[trevj]

Uhhh..... I think the answer is "Before it gets below freezing weather!" no?

How deep is the frost line in your area?

And, why can you not provide some insulation to prevent the cap and neck being exposed. Whether that is a doghouse full of fiberglass or rockwool mat, or just a tarp and a half dozen bales of hay or straw?

Electricity on all year? Put in a heat source.

 

[historicalarms]

Let the grass grow all summer over it, creates am insulation matt with snow on it that will prob totally eliminate freezing

 

[Susquatch]

Let the grass grow all summer over it, creates am insulation matt with snow on it that will prob totally eliminate freezing

LMFAO! What part of Canada are you from? Colorado? 😉

 

[historicalarms]

Good ol' cold ass Alberta. Anybody that has been in the dirt moving bidness has noticed the lack of frozen ground under a good layer of grass and early snow. Usually you will find only a thin layer of frozen ground less than 4 inches thick even in mid winter, sometimes you can easily dig with a common spade shovel...the frost doesnt even reach the tanks to freeze them.
Now, if you drive continually over any spot the frost can /will go down several feet in that area.

 

[FatherWes]

@historicalarms is right. Only clear the snow or drive over the area you’re going to work TODAY. Otherwise, tomorrow it’ll be frozen solid. (The temperature under the snow rarely gets much below freezing - think igloo or snow hut). This is true even on the prairies in minus 40 temperatures.
Except on those years when we get minus 30 before we get snow cover.

 

[FatherWes]

Burying a layer of styrofoam above the tank, extending out past the frost depth should work.
If the tank is 2 feet down and the frost depth is 6 feet, the insulation has to cover the tank and extend 4 feet out past the tank. (Frost then has to extend 2 feet down and 4 feet sideways = 6 feet = frost depth).

 

[DPittman]

@historicalarms is right. Only clear the snow or drive over the area you’re going to work TODAY. Otherwise, tomorrow it’ll be frozen solid. (The temperature under the snow rarely gets much below freezing - think igloo or snow hut). This is true even on the prairies in minus 40 temperatures.
Except on those years when we get minus 30 before we get snow cover.

And the other exception is in the areas where we get little or no snow cover at times but very cold weather.

 

[Susquatch]

Doesn't sound like a gamble I'd want to take. According to the latest info, the frost line in Saskatchewan can vary between 6 and 12 feet.

Looking at the insulation factor for snow, it looks like the depth of snow cover reduces the depth of the frost line by the same amount. In other words, a foot of snow raises the frost line from 6ft to 5ft. So you would need to maintain 6 to 12 ft of snow over the tank to protect it.

Styrofoam insulation is better than snow, but not infinite. The same factors apply. It only has an equivalence.

As many have said, there are lots of ways to reduce the frost line. But if it were my sewage tank, I'd keep the level below 1/3, let it freeze, and simply not worry about it. That said, I suppose you have to put snow someplace, so why not on top of the tank for extra insurance!

 

[historicalarms]

Looking at the insulation factor for snow, it looks like the depth of snow cover reduces the depth of the frost line by the same amount. In other words, a foot of snow raises the frost line from 6ft to 5ft. So you would need to maintain 6 to 12 ft of snow over the tank to protect it.

I know for a fact that you dont want to test that theory with anything heavy such as Cat or farm machine in Muskeg...you will have a different opinion after the fact.
Their theory may have merit on ground that is regularly traveled on/packed down but in fluffy snow/grass areas it is proven absolutely wrong on hundreds of locations yearly

 

[FatherWes]

I think the kicker is the unreliability of snow cover.
A good layer of snow does dramatically reduce frost penetration.
A cold snap without snow cover and the tank could freeze.
Not an acceptable design strategy.
Especially since I don’t know where this tank is, what the soil conditions are, what the tank is composed of, how big the tank is, how deep it’s buried, nor do we know what the piping configuration is. Way too many unknowns to come up with an engineered solution IMHO.

So, I asked ChatGPT…
“How much water can I leave in a septic tank that will sit dormant over the winter and will likely freeze?”

Response:
Good question — septic tanks do need special care if left idle in freezing conditions. Here are the key points:

⸻
1. Freezing Risk

• If the tank is in a climate where the frost line goes deeper than the top of the tank, the liquid inside can freeze solid.

• A full or nearly full tank has more thermal mass, which actually makes it less likely to freeze completely than one with just a shallow layer.

• But if the tank is not being used (no warm wastewater inflow), even a full tank can eventually freeze at the surface and gradually down.

⸻
2. How Much Water to Leave

• Do not pump the tank completely dry. An empty or near-empty tank can:

• Float (“pop up”) if groundwater rises.

• Allow soil pressure to stress or crack the walls.

• Best practice in most cold climates is to leave the liquid level at the normal operating level (about 85–90% full, just below the outlet).

• If you’re shutting down for the whole winter and want to reduce freeze risk, some installers recommend pumping it down to about 25–30% of capacity — enough to keep the tank weighed down, but less water to freeze.
⸻

3. Other Protective Measures


• Ensure the lid and risers are insulated (some people add rigid foam insulation or mulch over the cover).


• Make sure there are no leaks allowing continuous trickle-inflow (which could form ice dams).


• In very cold regions, some people drain tanks completely and ballast them with sandbags or water/antifreeze mix, but this is less common and requires professional guidance.

⸻

✅ Rule of thumb: Leave at least 1/4 full to prevent floating or structural stress. More is fine if the tank is well-insulated and deep enough below frost.
⸻
Would you like me to check your specific region’s septic regulations or climate guidelines for winterizing? Some local health departments publish exact recommendations.

Seems reasonable to me…

 

[Susquatch]

Seems reasonable to me…

Me too.

FWIW, a neighbour of mine had his tank pop up out of the ground when he emptied it in wet soil conditions. That was something to see!

Your point about design objectives is a good one. Nobody should ever design based on best case. That's a recipe for disaster.