Chemical Storage Tank Heat Transfer in Sun versus Shade

[HydroAU]

I am trying to present the benefits of installing a roof structure to shield a sodium hypochlorite storage tank (FRP material) from direct sunlight and mitigate chemical degradation. I would like to try and quantify the increase in average temperature of the chemical within a sun-exposed tank in relation to a shaded tank.

I thought I could possibly use a simplified method where you have (mC_pΔT/t)_chemical + (mC_pΔT/t)_tank + UAΔT to calculate the required heat input and compare this value to solar radiation data. My biggest question is in regards to the overall heat transfer coefficient U.

Would it be reasonable to assume that, all other things being equal, the overall heat transfer coefficients would be roughly the same or at least in a similar range between the sun and shaded scenarios?

A second question is in regards to the heat used to increase the temperature of the tank material. I am not particularly interested in increasing the tank temperature, but believe it needs to be factored in since the solar radiation does not act on the chemical directly and some heat will be spent to increase the tank temperature. If I am evaluating a 1 °C temperature increase in the chemical, would it be reasonable to assume a 1 °C increase in the tank material as well?

Thanks for the help!

 

[LittleInch]

I think it's a good idea, but calculating the difference isn't straight forward for a part filled tank.

What sort of tank? vertical cylindrical or horizontal cylindrical?

Fixed roof or floating?

What even you choose, half of the tank will be in the sun and half in shade to start with.

The shade might also radiate heat from the underside, though less than the direct sun, as it heats up.

I've seen figures before that say the temperature drops by 8 to 10 degrees on the surface in the shade compared to direct sun. Then you're into ambient air temperature. 1kW/m2 is usually a good figure to sue for heat input from the sun.

If you've got a lot of fluid then this might not make much of a difference.

Most people just paint the tank white.



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

 

[HydroAU]

Thanks for the response LittleInch.

A few more details on the setup. The tank is vertical cylindrical with a flat roof. Dimensions are 14 feet in diameter and 22.5 feet high. The total storage capacity of the tank is 25,000 gallons. The tank will sit directly on a concrete slab. I thought about calculating the difference for a partially filled tank but it did occur to me that the air inside the tank would modify the heat transfer in relation to a full tank. I guess I thought that perhaps since the rate of heat transfer throughout the day would be essentially the same that I could in fact disregard the heat transfer term if I was just calculating the relative difference between a fully shaded and sun-exposed scenario.

The tank will be coated with a white gel coat paint just like you said, so that will drastically reduce any solar radiation absorbed. We also are dealing with a lot of chemical, about the size of a private swimming pull when totally full. I intuitively feel that in reality the relative temperature difference for the chemical would be minor. It is just that almost every manual I've seen for Sodium Hypochlorite storage stresses that outdoor tanks in hot climates like the Southern US should have a roof covering at a minimum. I wanted to show that, as far as temperature and chemical degradation are concerned, perhaps a roof covering does not it fact make much of a difference.

 

[IRstuff]

Depending on where you are, you can determine the sun-exposed area, which will allow you to determine the excess heat transfer into the tank from the solar load.

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

 

[LittleInch]

Or try just a bulk number.

I think when full you have about 100 tonnes of liquid.

so 100,000kg x 3.5 KJ/kg/C
so 360,000 kJ per deg c.

At 1kW/m2 let's say you have 10 m2 exposed to the sun, so 10KW
10000J x 3600 x 10 hours is err 360,000 KJ

So on a gross level that's 1C in a day ignoring any other input.

Isn't actually going to change much even for a half full tank

Now there's more to it than that especially with colour and surface temperature, but that's a lot of liquid to heat just by direct radiation.

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

 

[HydroAU]

I thought that I could use actual data from the National Solar Radiation Data Base (

https://nsrdb.nrel.gov/)

to get a solar radiation value which accounts for cloud cover. I would take the maximum Global Horizontal Irradiance (GHI) value for a respective year as an extreme case. Then I would account for albedo from the white surface (0.8?). Would that work?

I also would need to include longwave atmospheric to the overall heat input correct? I could use empirical equations, but could I conservatively assume it to be double the absorbed shortwave solar? If that doesn't make sense I could just use the 1 Kw/m2 as suggested above and then multiply by the sun-exposed area depending on time of day?

Thanks again for the help!

 

[HydroAU]

LittleInch, that was the simple answer I was looking for. Just needed to get it straight in my head. Thanks!

 

[EdStainless]

The other issue is that shade would protect your FRP from UV and lessen the issues related to that.
For large tanks with a non-volatile material in them solar heating is rarely an issue on its own.
Something volatile in a partially filled tank though can reach 'interesting' pressures fairly quickly.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[HydroAU]

EdStainless, in our discussions with the tank manufacturer, they typically coat the tank with a UV-Resistant gel coating to protect against UV. How much would sunlight alone affect the longevity of the gel-coating?

It occurred to me that the perhaps another difference in the two scenarios is that the surface temperature of the tank would be increased in the sun-exposed tank, increasing the rate of conductive heat transfer. Would that additional input still be minor compared to the heat required to increase the temperature of the chemical?

Thanks again for all the help.

 

[LittleInch]

You "might" get a very thin layer of warmer liquid on the inside skin from a sun exposed tank and if the chemical is sensitive to temperature could start to breakdown if this thin layer ( we're talking mm or microns here) gets to a high temperature before it is replaced by cooler fluid due to convection within the tank.

I don't know much about this stuff.

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

 

[HydroAU]

Thanks LittleInch!

 

[EdStainless]

The gel coat is a maintenance item, shading it will prolong its life. I have seen FRP tanks that people swore that they kept them painted and you could scrape off a 1/4" of material with your fingernail.
I have also seen FRP samples from a tank where the ones from the North side still had an ID gloss and ones from the South side had noticeable surface attack. That 5-10F difference can add up over time.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[Rogue909]

Don't assume the tank will automatically 'reset' to ambient at night. 25000 gallons is a lot to cool in southern US heat in the summer when the sun is on 14hrs a day and the cool summer air is 85F.

It would be great if you had another tank to compare to. Theory up some coefficients then test them against the tank in service. Adjust and check the numbers.

 

[HydroAU]

EdStainless, that is helpful. I can see that also being a plus for installing a roof.

Rogue, I think that it where I have been rethinking the scenario. Once the tank is heated during the day the heat will be stored at night and will cool off slower than a tank in the shade. This puts the temperatures at different starting points with each new day.

A big factor is the surface area of the tank that is exposed to the radiation. If the surface area is large then you may in fact get a lot of energy input from radiation. Even though solar radiation only acts on certain parts of the tank throughout the day, the atmospheric longwave radiation should be acting on the entire surface all day, correct? This would significantly affect the heat input and subsequent temperature difference.

Another factor I hadn't included is that the tanks have piping at the top which vents to the atmosphere. This is complicated! I wish I had time/money to do a field test but not sure if that is possible right now. Thanks all.

 

[IRstuff]

Not necessarily; the sky temperature is well below freezing, even on a warmish day, so long as the sky is clear. The sky warms up with overcast, but you lose the direct sunlight. The ground/object radiation is at ambient temperature, and that might, or might not, be a positive value.

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 sky temperature is well below freezing, even on a warmish day, so long as the sky is clear"

That certainly is not correct. The atmosphere is not transparent to all radiation.

 

[IRstuff]

It's transparent enough; if it wasn't for the scattering of sunlight by the air itself, the sky would be pretty black; nevertheless, there's a long way between "well below freezing," which is 273.15 K, and 4 K, which is the temperature of deep space. Our analytical sky model for 3-5 um sensors was around 230 K; even at 830 nm, the sky is quite dark.

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]

https://onlinelibrary.wiley.com/doi/full/10.1002/gch2.201900008

see Figure 3.

This is an actual measurement that shows sky temperature variation on a clear day varying from 294K to 286K.

Of course there are times when sky temperature can fall well below freezing. We usually see frost on the ground at those times. That is not an average occurrence. If the sky temperature is at freezing, it will not be a warmish day. That defies logic.

It does add an interesting factor as to why humid days are uncomfortably hot. Not only is evaporation of perspiration reduced, but the sky temperature also increases due to water vapor blocking heat radiation to space.

Figure 3:

https://onlinelibrary.wiley.com/cms...bb5-3a743ec9e8d5/gch2201900008-fig-0003-m.jpg

 

[EdStainless]

Night time sky is cold.
April in MI, night temp sub 50F, daytime ~70F, dig a hole 12" diameter x 18" deep.
Put a metal can with 24-32 oz of water in bottom (but sitting on some rocks to keep it off of the bottom).
Cover the top during the daylight hours and uncover at night.
If the night sky is clear in 3 or 4 nights you will have a block of slush.
If the night sky is overcast you will have a 50F can of water.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[Compositepro]

Well, I just used an infrared thermometer pointed at the sky. It is sunny with a few wispy clouds, 88F, 70% humidity. The sky temperature read from 16F to 50F. The 16F was in one small spot about 20 degrees from zenith.