Heat loss in a greenhouse

[tjsegan]

Hello everyone - first time post, so apologies if I break any rules or norms.

I am having a brain freeze moment with a pretty basic calculation and want a second opinion to help set me back on track.

I am trying to calculate total duty requirements to keep a greenhouse at a target temperature as an initial rough calculation for energy consumption estimates. Calculating heat losses through sides, walls and roof panels is pretty straightforward.

One piece is confusing me, and that is the floor. My engineering gut feel is that the heat loss is negligible, but double checking with conduction calculations I run into troubles with assumptions regarding the depth of concrete and soil to be used, and the temperature at the boundary. Now, soil temperatures in the region are measured by weather stations, and they show a temperature of around 20degC at the surface, rising and then dropping again as you go further down (presumably because there is some lag for the heat to transfer from solar irradiation during the day, and the earth acts a sink and ballast for heat).

My current assumption is to use a depth where the bulk is relatively steady state (~20degC), which pops out a very low heat loss.

Is the approach valid? Could I be missing anything?

Thanks in advance.

Tim

 

[chicopee]

I would treat the ground as a constant temperature heat sink, eventho, that assumption is incorrect, then do research on the coefficient of convective heat transfer for flat surface then apply the heat transfer equation for convection and for conduction thru the floor. Don't forget to estimate heat loss due to infiltration of air.

 

[IRstuff]

While the floor is a heat-sink, of sorts, the thermal conductivity isn't that great. I've seen treatments of the ground around pipelines in the literature, where the ground conductivity is modeled as 1.73 W/m-K

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[Compositepro]

The long term average heat flow though the floor is probably a small percentage of the overall heat flows, particularly if you are trying to keep the temperature constant with ventilation control. It would make sense to treat it as an insulator.

 

[tjsegan]

Thanks all.

I think what I am struggling with is what to assume the steady state temperature is at a given depth to calculate the heat loss. Here, I have used a depth where temperatures seem to settle down to a steady "bulk" temperature at about 6m, at a temperature of 19°C (see attached image).

 

[georgeverghese]

Must say I lost a fair bit of sleep, some 5-6years ago, over a similar problem with heat loss through the bottom of molten bitumen tanks (in a tank farm) operating at 170degC - the main unknown being what terminal temp to assume at the end of a composite concrete layer and a sand layer in series. EPC schedule environment meant I had to make some assumptions to get things moving - still havent figured this one yet, long after the project is closed out.
LNG tanks operating at -160degC would have the same heat transfer issues which will affect boil off gas rates during normal operation.

 

[LittleInch]

I do a fair bit of this for buried hot pipes and heated tanks.

If the temperature on the surface is 60-70 C, I would go for a 1m band of earth before hitting the heat sink. 1m works fairly well and in steady state is a little conservative as it doesn't take account of the soil heating up, which over the period of a few days will reduce heat flow. I did some measurements a long time ago on soil temperature from a 12" hot oil pipe and within 1m the temperature was within a degree or two of that at 5m.

Under something like a large tank at 170 I would take 2-3m before reaching a steady state.

LNG tanks are highly insulated, partly for the BOG issue, but mainly to stop the ground freezing and having soil heave.... Some of them raise the floor off the ground on stilts for this reason

For something like a greenhouse the floor losses might be small compared to the walls and roof of single pane glass, but should form part of the calculation

for temperature either look at the particular months you're interested in at 1m depth or choose the average.

for clay I would use something like 1.3 W/m/K

SO it should be a fairly simple calculation.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[MortenA]

the greenhouse effect is fake news.... ;-)

 

[IRstuff]

The math is relatively simple. The challenge is choosing a temperature and a depth, wisely ;-)

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[chicopee]

You could always insulate the ground before concrete ,if used, is poured, then dismiss that calculation. Nonetheless, air infiltration has to be taken into consideration.

 

[itsmoked]

The floor would be down in the noise compared to vertical or horizontal single pane, maybe the entire floor would equate to about 3% of heat loss. This would wash out in just the error of the rest of the glass.

I'm saying don't knock yourself out on the floor.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[tjsegan]

Thanks all - seems like everyone is in agreement here.