vent sizing 2

[401laurent]

Hello,
I'm sizing the vent for vacuum tank in condition of cold spray after hot washing, Could any one help please giving the formulas for this calculation, the aire venting and the air inflow required, in order to size the vacuum valve.
the data given is: MAV(Maximum allowable vacuum)for the tank ( psi), Cold spray temperature (oF), cold spray flow (lbs/h).
Thanks.
Laurent.

 

[shahyar]

Can't you use API2000?

 

[Latexman]

How frequent is this?

Describe hot washing. Near 100 C with water?

Cold spray temperature? Is cold spray internal?

Good luck,
Latexman

 

[401laurent]

Hi Latexman,
The CIP Conditions : hot Washing : 85 deg.C , and internal Cold spray température is 4deg.C,the CIP frequency is 2 to 3 cleaning per day.
Thanks.
Laurent.

 

[TD2K]

I wish I had some suggestions to offer you, I've been hoping someone would post something.

I think this is a tough problem. If I understand the situation, you have a tank full of hot vapor after washing. You then start a cold water cycle which sprays cold water into this hot vapor. The vapor condenses in the cold liquid and that vapor volume has to be made up. A complicated heat transfer problem from the hot vapor into the cold liquid over a large surface area.

Have you talked to the spray nozzle vendors to see if they have any information to share with you? What about a desuperheater vendor? They have a similar issue with heat transfer from the steam being cooled using 'cold' desuperheating liquid. They might have some technical literature to share with you.

I don't think API 2000 is going to help you, this is one of those cases that the tables and formulas in API 2000 aren't set up to handle.

 

[psafety]

Google it. There are calculation methods available.

 

[mcecasf]

Hi folks,

I've been googling like mad for this, all I can find is various works for when there is cold spray on the outside of the tank surface condensing steam inside a tank... Any hints as to where I might find a solution more related to this particular problem?

Cheers
Col

 

[KvdW]

Please have a look whether the discussion in thread1203-155185
(Calculating Venting Capacity for Open Vent) helps.

 

[Latexman]

This is a very good reason to have a "vacuum tank" rated for . . . full vacuum.

Good luck,
Latexman

Need help writing a question or understanding a reply? forum1529

 

[mcecasf]

Thanks KvdW, sadly I don't think it does - the closest would be RJB32482's point but I can't see how to apply that equation to my problem (I may be daft, or over analysing it) as I think it is looking at heat losses from the tank causing the condensation, whereas I have an inflow of cold water through a sprayball. How would I define how "effective" the heat transfer from that sprayball to the steam (or more properly, the "hot humid air") in the tank?

Thanks
Col

 

[mcecasf]

Hi Latexman, were it not for the fact that I have >20 existing tanks, I might agree!

I know there are similar arrangements of hot CIP, cold rinse in many finished product storage tanks in the food industry, also e.g. road tankers for food products, but I can't find anything to define a sensible means of sizing the vent.

Col

 

[KvdW]

A good reference is the VDI-Wärmeatlas which addresses Misch- und Einspritzkondensation (translation: mix and spay condensation) on free falling flat or round surfaces, and heat transfer on individual droplets. The concept is:
- the falling liquid flow will heat up by condensation of the water vapour on the surface;
- the temperature at the end (when it reaches the liquid surface at tank bottom) can be estimated from Stanton number, outside surface and shape of the liquid flow;
- heat duty of heating up the falling liquid = heat duty of condensing vapour, which allows you to calculate the amount of vapour that needs to replaced by venting;
This will provide an estimate of the initial vent flow. As soon as vent air is introduced you will get an additional condensation due to colder air, however a decrease in heat transfer from vapour to liquid.

 

[chicopee]

Have you discussed the situation with the tank manufacturer since the rate of depressuriation and of condensation would be hard to determine.

 

[chicopee]

I have another idea, check out topics related to desuperheating, however in you case, you'll have a phase change.

 

[Latexman]

Is there any consistency in the tanks/spray nozzles, such that you could measure whats going on in one tank (is one rated for FV?), and use the results for the others?

I like TD2K's direction, call your spray nozzle company (or one of the name brand ones) and ask for their "Supreme and All Powerful Guru" to see if he/she has already cracked this nut.

Good luck,
Latexman

Need help writing a question or understanding a reply? forum1529

 

[StoneCold]

It seams to me that you could get the answer with an enthalpy balance around the tank.
If you assume perfect mixing, which is kind of a worse case scenario. Then calculate the enthalpy of the entire tank contents, ignoring the actual tank mass. Then calculate it again after one minute of cold liquid addition. If you force the pressure to be the same (imagine the tank is really a ballon) work out a new volume for the steam at the lower enthalpy value. You should have it. The tank volume - new steam volume-water will be equal to the air makeup volume. That is kind of crude and you could refine it more by breaking it down into seconds and adding air and it's enthalpy to the system. The system will also be slowed down slightly by the thermal mass of the tank. But this should get you a number in the ball park.

If you are really good you could make it into a differential equation.

Just my two cents.

Regards
StoneCold

 

[Latexman]

Yep. Perfect mixing, instantaneous heat transfer, the average or maximum mass flow rate of the cold water will limit the situation. The initial rate (t=0+) would be the worst case of this worst case.

My experiece with these "thought experiments" is they are ultra conservative and are usually unrelievable at a reasonable cost. I'd like to know how it turns out for you.

Good luck,
Latexman

Need help writing a question or understanding a reply? forum1529

 

[MortenA]

My methodology would be to:

1) Determine min pressure in your vessel (this cant be 0 barg because you need some pressure difference to drive the flow of air)
2) Use this in sizing the valve e.g. using api 2000 - remember to take losses in headers and after the valve into consideration
3) Determine air flow much as stonecold suggest - ignoring heat transfer etc assuming that the water has perfect contact with air - since this is a worst case(over design)

If the valve becomes "too big/expensive" then i might "sharpen the pensil" - if not its safe to proceede. However, tanks rated for vacuum are much better.... (i know in hindsight everything is easy)

Best regards

Morten

 

[MortenA]

or maybe not API 2000 - ubt the valve vendor will have a sizing formula or can help you with that (just remember inlet/outlet losses if appropriate

 

[mcecasf]

Hi all

Thanks for all these replies!

I've had a play about with what I can easily calculate without getting into too much complex heat transfer theory - it is a while since I had to do that kinda maths - and then come back here to see Stonecold's suggestion which is near as dammit what I came up with (I even had "assume it is a baloon" on my spreadsheet!) Didn't use differential eqn, relied on old favourite "Goal Seek" and a very small initial time increment.

I had a play with both "steam" (given it is an atmospheric tank, assumed that it was full of steam at 1 atm) and air. Given the high latent heat to condense the steam, the effect on volume from the given inlet flowrate of water was more pronounced with temperature change of air - but I'm gonna revisit now having thought about the fact that I could have a density change in the steam rather than condensing it.

for the "full of hot air" case: I'd included for incoming air at ~0 degC (outside air in winter) having additional cooling effect, and found a required flowrate; then checked choked velocity of this air and from these two determined a minimum area, stuck in a +50% fudge factor given my knowledge of compressible flow and how relevant that max rate actually is - given comments in the other linked thread above - and came up with needing a vent just over 6 inches diameter. If I drop CIP temperature to 50 degC then I get about 5 inches.

This is interesting - and kinda holds true, as the part I had not told you is
- our tanks original CIP cycle was 50 degrees with 6" vent on the tank
- at some point in the dim and distant past, before my time and my introduction of a change control system for process alterations, this was upped to 80 degC to give a better clean
- we've this week inspected a tank because it was making a "loud banging noise" during CIP / rinse and found some nice creases and bends in it (not externally visible due to lagging) suggesting it has been on a bit of a "wobble"

... so by my (and Stonecold) method, worst? case with a fudge factor, the existing 6" vents were OK for a lower temperature difference but not great for the higher temperature difference from the hotter clean.

I'll tidy up my spreadsheet and stick it up here for you all to rip holes in.......

Ref cost - remember in my case, these tanks are open to atmosphere via the vent,I don't have a valve to install, incremental cost for installing new 6" vs 12" vent is hardly worth worrying about - impact of size and design is cleanability and hygiene.

Morton - ref API 2000 - no use as specifically excludes this type of situation.
Latexman - there's a dozen different spray nozzles involved.. The tank farm has been expanded over time (again before my time) by "copy what we had but make it bigger - what sprayball did that have? - dunno just stck one in" - thankfully impact of implosion or damage is financial and a bit of mess on the floor to our contained drains, vs some form of hazardous / polluting substance showering people and environment. The remaining tanks on site still have the lower temperature clean.

Thanks again all. As a process engineering department of one on my site, it is good to be able to bounce things round a bit.