Curd Settling Tank (Dimple Jacket) is imploded during chilled water circulation

[US Rao]

Dear Friends,

We have been used 5KL, Dimple Jacketed type Curd Settling Tank. The tank will be used for preparation of curd.

The operations in the tank is as follows:

Pasteurized milk will be taken in Settling Tank at 42 Deg. C.
Culture will be added,
Once PH comes down to 4.8, Chilled water will be circulated in the Jacket until the product inside the tank is comes down from 42 to 4 Deg.C. Once curd is formed in the tank, the product will be transferred to other settling tank for preparation of Butter Milk & Lassi.

The tank is having Air Vent at Top of the tank, NRV at chilled water return line, Safety Valve at Chilled water return line before NRV.

Despite of above protections also, product is inside the tank and during chilled water circulation inside the jacket, the tank is getting implosion. As per operator, the tank is getting imploded when the product is inside and the chilled water is under circulation in the jacket.

Photographs of damaged tank is attached attached (Tank and Air Vent)
We are unable to find out the causes for tank implosion. Kindly help us.

Regards
US Rao.

 

[Compositepro]

Obviously, you are creating sufficient vacuum in the vessel to destroy it. Steam in the vessel is condensing faster than your vent will allow air to enter the vessel. As cold air enters the vessel it will also cause additional condensation. The way your vent is designed the air enters as a jet that very rapidly mixes with the steam. This results in very rapid condensation of steam, and a sudden pressure drop that sucks in more cold air. The worst case condition is when the vessel is nearly empty.

 

[bimr]

Accidently imploding a tank that has been steamed happens frequently. However, in the operation that you describe, the fluids do not appear to be hot enough to condense.

It is more likely that your chilled water pump is pulling a vacuum. Is the chilled water system a closed system?

 

[don1980]

I'm betting that your air vent is plugged.

 

[EdStainless]

It has to be either a plugged vent or a pump pulling suction.
On a vessel that size you will not be able to cool it very quickly, at least you shouldn't be able to.
What is the initial temp of you cold water? It should not be 4C to start with.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube

 

[US Rao]

Hi Friends,
Thanks for your reply.

I want to confirm the following

1. The chilled water circulated in dimple jacket is around 2 to 2.5 Deg. C,
2. We are not using any steam/ hot water into the dimple jacket,
3. chilled water system is a closed system. The cycle starts from Chilled water storage tank (IBT) - Chilled Water Pump - Dimple Jacket Tank storage tank- Chilled Water Storage Tank (IBT).
4. 3 Tanks damaged similar to each other during chilled water circulation only. During that time, only cold product is present in tank and chilled water is under circulation in dimple jacket. And we are not drawing any product from the storage tank.
5. The dimple jacket is hydro tested for 4 bar and the maximum pressure to dimple jacket during chilled water will not be more than 4 Bar.
6. During chilled water circulation, the tank is in ideal case and we are not drawing any product from the tank.

I had small doubt. Kindly clarify. Is there any chances to create a vacuum when the tank is in ideal case (Chilled water circulated and we are not drawing any product from the tank). I hope, vacuum will create when there is no product in the pump suction line and the product is empty in storage tank and the air vent is chocked.

When product is available in tank, Is there any chances of creating any vacuum in tank ???

Regards
US Rao.

 

[gerhardl]

If all parameters are correct as given, there is in my opinion no way that a vacuum can be created by temperature and condensation issues. My suspicion is pump emptying and insufficient, by design or failure, air inlet.

Note: It is not certain that the (secondhand?) actual reported operation parameters are correct.

Insufficient air inlet compared to the rate/amount/speed of the created vacuum is anyway the cause of any vessel implosion.

The causes of insufficient air inlet can be listed as follows:

1. Damaged or stuck air inlet (Inlet should be inspected at regular intervals)

2. Too late opening of air inlet (Too high set opening pressure, too high varation range, variation from factory given accuracy etc)

3. Too small amount of air inlet. Rate of created underpressure should be estimated.

Example: Say, if the vacuum is created by pumping out alone, you have the max. pumped out volume pr time unit given by the pump capacity. Without bothering to much with temperature and pressure differences, roughly estimate the necessary opening area by the amount of air (liter) going through an opening of x (area) at an estimated air speed of 30 (conservative) to 90 m/s.

Speed of openig, set pressure, accuracy and air amount by underpressure protection designs are often underestimated.

Good luck!

 

[gerhardl]

See also thread124-444098 for comments on tank toleration of underpressure.

 

[US Rao]

Dear gerhardl

Thanks for the reply.

As you said, we cannot believe the words of operator. But during the failure, some amount of liquid ( 1200 litres) is available in tank.

I do agree that, during emptying the liquid from the tank, if the tank is getting empty and there is no product in the pump suction, vacuum will create and implosion will occur.

But in this, liquid is present in tank during implosion. Tank capacity is 5000 Litres and 1200 Litres is available in tank.

And one more point i would like to inform that, the tank is connected with two pumps in series. First pump is shear pump, which is used for breaking the curd into parts and second pump is used for transfer of product from storage tank (Dimple Tank) to packing machine. As per SOP, both the pumps should not operate.

I doubt that, during transfer of product, they might have used both the pumps.

Kindly correct me, if iam wrong.

 

[gerhardl]

The essential question is which change of operational details gives the breakdown?

If you pump our liquid from a closed tank, and there is no air inlet, this will give an underpressure regaardless of some rest amount of liqiud in the tank.

If this is not the cause, or part of the cause, something else must be changing. What?

No view of the air inlet device is given in the picture, but the air inlet device might well be malfunctioning (as suggested in the thread) or too slow or with insufficient capacity. What does an inspection of the details of the air inlet device indicate?

 

[georgeverghese]

You may have deliberately blocked the air vent from breaking the partial vacuum ( on cooldown from 42degC to 4degC) to prevent contamination of the milk with microbes in the air during curdling? One solution would be to hookup the vacuum side inlet port on the vacuum relief valve to a low pressure nitrogen tank, which in turn could be fed from high pressure N2 bottles. If you use air as the vacuum break gas, it would be have to be completely dry, oil and dust free and may even have to pass through a UV sterilisation unit - N2 bottles / LP N2 tank would be easier.

 

[bimr]

You may have pulled a vacuum on the cooling jacket, not the tank. A vacuum on the cooling jacket will also collapse the tank.

 

[US Rao]

Dear Friends,

Thank you all for your feed backs. It helped me a lot.

 

[Compositepro]

A vacuum in the cooling jacket cannot collapse the tank.

 

[bimr]

A vacuum in the cooling jacket cannot collapse the tank.

The poster has stated that the implosion was not caused by steam. The poster has also eliminated a scenario where a vacuum condition caused by a tank discharge pump.

So that leaves the cause as being external pressure from the tank jacket.

"External pressures can also be created when vessels are jacketed or when components are within multi chambered vessels. Often these conditions can be many times greater than atmospheric"

Link

 

[Compositepro]

This is a dimple jacket, so the stress created by pressure in the jacket are taken by the dimple welds. If it were a full, separate jacket, then pressure in the jacket would create an additional external pressure on the inner wall. But vacuum in the jacket would reduce the external pressure on the inner wall. Since this is a dimple jacket, vacuum and pressure have no net effect on the inner wall.

If the dimple welds fail when the jacket is pressurized, the inner wall can collapse, but not the outer wall. The pictures clearly show that the inner and outer walls collapsed inward together, which can only happen if there was a sufficient vacuum inside the vessel. The vent is fairly large with a perforated metal screen that appears clean and unobstructed. Perhaps the vent became plugged with curds and was cleaned before the photo. Since the entire tank is now spotless I guess any evidence has been removed from the photos. The vent must have been plugged.

 

[bimr]

"4. 3 Tanks damaged similar to each other during chilled water circulation only. During that time, only cold product is present in tank and chilled water is under circulation in dimple jacket. And we are not drawing any product from the storage tank"

"But in this, liquid is present in tank during implosion."

Based on the OP's comments, the scenario proposed by Compositepro is not possible.

Picture 4 shows indentations from the dimples on the interior wall, which indicates that the interior tank wall thickness is inadequate.

"A common mistake with dimple jackets is to not calculate the required thickness of the head or shell but assuming that it will be adequate because it is thicker than the dimple jacket. A shell of inadequate thickness can lead to yielding under hydrotest showing the location of the dimple welds from inside the vessel (the inside surface is no longer smooth)."

https://pveng.com/home/asme-code-design/external-pressure-methods/[/URL]]

Don't believe that the design data or code information has been supplied so far. It may be worthwhile to ask the vessel manufacturer for the design calculations on the tank.

 

[Compositepro]

bimr, it is simple physics. A vacuum on the outside of a vessel cannot implode it. The vacuum has to be inside.

 

[georgeverghese]

Agreed, if there was high pressure from the chilled water at the cooling jacket, the inner wall would have imploded and the outer wall would have exploded. In this case, the entire cooling jacket has folded inwards. Vacuum pressure within the cooling jacket is not possible for a recirculating pumped fluid in the jacket.
Looks like this is a simple free vent with no PVRV. This free vent may have otherwise plugged up with curd (if it wasnt deliberately blocked or spaded off)during one or more previous unreported overfill incidents?? Is the level indication on this tank working well?

 

[RVAmeche]

If it was a vacuum condition in the jacket and the vessel walls were too thin, I would expect the vessels walls and jacket walls to crumple towards each other, no?