Self-Supported Cone Roof

[PEStructural]

Hi,

I have a client that is requesting to fill a 2" in 12" self-supported cone roof up to the very top (apex) of the roof with liquid. I know this raises alarms in regards to venting, etc...

What my question is - how does the liquid apply pressure on the underside of the roof? I would imagine it has a horizontal hydrostatic pressure. But does it also have a vertical uplift (buoyancy if you will) equal to the difference between the very top point of the tank and the point in question multiplied by the density of the liquid?

This is for a wine tank (designed (kind of) to API 650). According to the customer, every wine tank is filled this way to the very top of the roof (to avoid any air being trapped on top of the wine).

Thanks!

 

[JStephen]

It could be done okay. You could use the membrane stress equations from API-620. Or could calculate the pressure at the low part of the roof, then design for a uniform pressure of that amount per API-650. You may need to anchor the tank.

You'd want some type of relief valve so that thermal expansion/contraction and similar effects didn't overstress anything.

 

[msquared48]

I would think that when it is filled, there would be a vent at the top of the cone in the vat to let the air out and a small amount of liquid, and once that was achieved, it would be capped.

The only upward pressure, as it would be at atmospheric initially, would be any internal pressure generated by the fermentation process. Check with the client for this, but it could be minimized with a pressure relief valve as previously mentioned.

Bottom line is I do not think there will be much uplift on the top of the tank unless the relief valve fails.

Mike McCann
MMC Engineering
Motto: KISS
Motivation: Don't ask

 

[PEStructural]

Thanks for both your responses. So if, at the center of the roof, there is a vent of some kind, and neglecting the thermal aspects, is there a vertical pressure from the fluid itself, or just from any trapped air being "squished" (which won't happen because of the vent)?

I'm just trying to wrap my head around whether I should design the roof plates/rafters for a vertical (upward) load from the hydrostatic head.

Thanks again!

 

[msquared48]

Since this is not a government project, you do not have to worry about the low bidder syndrome for the pressure relief vent at the top.

SO, I would check with the valve supplier, or the owner, to see at what atmospheric overpressure the relief valve is supposed to trip, and design the roof structure for at least 10% more to be sure.

Mike McCann
MMC Engineering
Motto: KISS
Motivation: Don't ask

 

[Duwe6]

650 allows internal pressure w/o calc's of "weight of roof plates" thus 3/16" thick roof it is 7oz/sq. inch. [about 12-inches water column] With calcs, up to 2.5 psig [just under 6-feet w.g.]. Look at the Protectoseal website for kinds and types of pressure/vacuum relief devices. They show about every style avalable anywhere.

 

[JStephen]

The current 650 pretty much requires calcs for everything, as an internal pressure less than the weight of the roof plates still affects overturning stability, etc.

 

[MJCronin]

Wrong kind of roof.....

A cone roof is a very inefficent configuration for this kind of loading.

A domed roof or some kind of umbrella roof is a better choice...perhaps an elipsiodal configuration

If you client starts crying and complaining about not being able to use this top volume on his cone topped tank, tell him he should have considered this when the tank was specified/purchased.

Clients cant always get what they want.....

 

[MJCronin]

On futher thought....

I don't understand of what use a "fully filled" cone roof tank would be in the wine industry.

You can't fill it anymore without tank explosion, you cant empty it without extreme venting provisons.....nobody else configures tankage in this way

Oh,......and ask the client if he has ever heard of the widespread industry practice of using nitrogen blankets on wine tanks.

http://www.southtektalk.com/2010/04/nitrogen-uses-in-wine-industry.html

 

[kdgran]

This style of head is very common in the wine industry for field fabricated tanks. Normal shop fabricated tanks have truncated heads. The wine industry fills to the top to eliminate any air in the tank. The more wine is exposed to the air increases the chance for contamination. As far as uplift on the tank I normally find the average liquid head applied to the tank and use that for the calculations. Usually the diameter of these tanks is small so this approach is close.

 

[Duwe6]

If all you are qorrying about is air causing oxidation of the wine, put an Argon 'blanket'. A full 2200psi bottle of Ar costs me $24. It should take about 3/4-bottle to fill the cone of your roof. Argon displaces Oxy and Nitrogen. Wine stays good.

 

[sdz]

how does the liquid apply pressure on the underside of the roof?
Hydrostatic pressure.

 

[kdgran]

You are correct that $24 will keep wine good in a small tank. Now multiply this small tank by 200 storage tanks that Gallo, Paso Robles, etc might have at one facility and may be filled/emptied a few dozen times per year depending on the wine type. Your $24 investment in now $60k. It's much cheaper for the end user to specify that the tanks fill be filled full.

You will see uplift on the tank head due to the static head of the liquid pushing up on the head. No different than if internal pressure is pushing up on the head.

 

[Duwe6]

And API-650 tanks are not rated for significant internal pressure. This tank falls in the realm of API-620 when fully filled and having a small internal prewssure.

 

[chicopee]

A vent on the top of a wine tank??? what of waste of good alcohol vapor during fermentation.