Convection coefficient for wooden barrels

[sweetj]

I've got a bit of a strange application I'm working on: an accelerated aging project at a whiskey factory. I need to determine heat transfer to/from the closely-packed barrels for different proposed air distribution configurations.

I plan on approximating the barrels as a packed bed of spheres, but am stumped on what to use for a convection coefficient. The wooden barrels can be approximated as 24" dia cylinders 36" tall. They're packed 6 to a 51" x 76" pallet. Full pallets are stacked 6 high, 30 wide and 23 deep, within inches of each other.

Any thoughts or educated guesses on an coefficient I can use would be greatly appreciated.

Thanks!

 

[JStephen]

My first thought would be to model them as pipes, rather than spheres. I think you can probably find coefficents for crosswise flow through bundles of pipes- that's a heat exchanger configuration.

 

[IRstuff]

Are they stacked? That will significantly affect the HTC. Is there forced air or purely natural convection? What's the temperature delta between the barrels and the ambient air?

TTFN

 

[MintJulep]

Accelerated ageing of whiskey sounds oxymoronic, but anyway.

Since booze is big business I hope you have a reasonable budget. I would suggest setting up some test configurations to back out a coefficient. The experimentental setup could then later be used to validate the model.

 

[MintJulep]

Just another thought.

Do you know anyone in the tobacco busines? Tobacco is aged in big wooden barrels (that have a special name that I can't recall right now) stacked in huge warehouses. I's sure that RJR and P-M have fully validated CFD models for this by now. Probably closely guarded secrets, but...

 

[sweetj]

Thanks to those who have weighed in so far. Some more info on the problem:

Currently the barrels are cooled by free convection; while there is a perimeter ventilation system in the warehouse, it basically is there just to keep the perimeter barrels from freezing in the winter.

There are over 32000 barrels, 54-L each. It currently is a 6-month cycle to heat/cool the barrels.

We will be introducing some system of forced convection, bringing in cold outside air to provide the necessary cooling capacity (which is what I'm trying to evaluate). Cooling is the critical operation, as large natural gas furnace is available for heating. I've assumed "hot" barrels at +30C and cold air at -20C. Heat capacity of each barrel is approximately 500 kJ/degree C. I'm using a target for the "cold" barrel of +15C.

The barrels are stacked 6 high on pallets. I had originally considered approximating it as packed tubes in a heat exchanger, but the barrels are too close together (effectively touching at their widest diameters). The charts I found for airflow over tube banks provide a lowest spacing/diamter ratio of 1.25, while I'm looking at a ratio of about 0.08. This is why I'm leaning towards the bed of packed spheres; I have calculated a void fraction of 0.35.

But I'm still trying to figure out a convection coefficient, "h".

Any thoughts?

 

[sweetj]

Correction: spacing/diameter ratio is actually 1.08

 

[IRstuff]

so there's about a 2.5 inch spacing at the narrowest?

and there's essentially a chimney at the center of a 2x2x6 stack?

Do the pallets block the chimneys?

TTFN

 

[sweetj]

IRstuff:

Yes, there is effectively a 2-inch spacing between the barrels. There's not much of a chimney, though; the barrels are 2x3 on the pallet and the pallet does block most of the gap between the barrels.

One option we are considering for air distribution is to remove an entire column of barrels in the center of the storage areas and force air down into it (so it will then percolate through the barrels to the aisles).

Hope that helps.

 

[IRstuff]

A typical value for natural convection is something like 4 W/m²K, but that's assuming completely free access.

Wood thermal conductivity is something like 0.15 W/m*K. For a 1-inch thickness, that's equivalent to 5.9 W/m²K.

How long does it take a barrel to reach your desired temperature now?

TTFN

 

[sweetj]

Thanks for the values.

Right now, the facility works on a 6-month heating, 6-month cooling cycle. Cooling is currently accomplished by opening a few roof vents and waiting (forced ventilation is used for heating only).

We are going to try to get three complete cycles per year, i.e. 2-months per heating/cooling.