Filter Design. 1

[Chemitofreak]

I have to design some filters to remove rust particals from liquid and gas stream.
How and what values shall I specify for the following:
1. Inlet solid content
2. Inlet/outlet partical size
3. Filtration efficiency
4. Changeover time

Thanks in advance.




Dinesh S SHELATKAR.
Process Engineer

 

[TD2K]

Let's break this down

Inlet solids content. This requires testing to come up with a value or some other way of estimating it. It's not available much more than it is available. If you need to do testing, the test set-up is important to ensure you take a representative sample in order to get valid results.

Inlet particle size. This again needs testing and is also typically not available much more than it is available. There are companies that do this, google particle size testing.

Outlet particle size is up to you. What does your process require the solids size and quantity to be removed down to? Some processes require submicron, some are in the 5 to 100 micron range, some are fractions of a mm or larger. It all depends on what you need at YOUR location.

Filtration efficiency comes out of the above in terms of solids loading and size distribution and what you need to remove down to. Think of a mass balance.

Changeover time is your call. Is it acceptable to change the filters daily? Every other day? Weekly? Monthly?

Other information you want to provide are:

1. The name of the fluid, total flow rate, density and viscosity
2. Operating pressure and temperature
3. Design pressure and temperature
4. Materials of construction. CS, Cr alloy, SS (and grade), Monel is going to make a big difference.
5. Any special requirements: PWHT due to the process, do you want it stamped as a pressure vessel, etc?

 

[georgeverghese]

A particulates filter in a 2phase stream? - must say I've not come across such a service - filters are usually in single phase service (either gas OR liquid.
Also add permissible dp when fouled.

 

[LittleInch]

If the rust particles are some fraction higher density than yur liquid why not just flow through a vessel where the velocity reduces to the point where the particles simply drop out. Otherwise I would agree with all above - two phase filtration is not common.

What values you specify for the above - all up to you and your requirements - there is no one answer I'm afraid.

You might also want to consider centrifuges and cyclones.

There isn't much in your post to go on....

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

 

[Chemitofreak]

Guys,

Thanks a lot for replying to my query.

Let me clarify one thing, it is not a two phase filter. Filter for gas and liquid are separate entities.

I am told to prepare a filter data-sheet to remove rust particles, but I am unable to specify
1. Inlet solid content
2. Inlet/outlet particle size
3. Filtration efficiency
4. Changeover time

I know the inlet solid content and particle size can be specified based on laboratory experiments, but I work in an EPC Company, and I am unable to specify the same.

Hope my question is clear now.

Thanks once again.

Dinesh S SHELATKAR.
Process Engineer

 

[EmmanuelTop]

Can you design a distillation tower without knowing feed composition and purity of the products? Or design a compressor without knowing gas composition, suction and discharge conditions? Or design a heat exchanger without knowing mass flowrate and properties of hot and cold streams? Well, bad luck - the same applies for filters.

The fact that you are in an EPC company does not change the engineering facts. You need to speak up and make yourself very clear to the client that an efficient and fit-for purpose design is not possible without proper input information. If you proceed the other way, it's going to be a disaster in the field and it will cost your client a lot of money.

Dejan IVANOVIC
Process Engineer, MSChE

 

[TD2K]

You have provided no more information than contained in your first post other than clarifying you are looking at treating separate vapor and liquid streams. How do you expect to get much more of an answer based on the replies already posted?

Go back to your lead engineer and ask them for direction. I do not assign vague tasks to my younger engineers unless I want to waste hours watching them flounder and I don't have hours to waste on my projects.

 

[georgeverghese]

Tell us some more on the plant you're working on, the unit this filter is located in, and what is downstream?

The lead engineer on your project is the best source of help, but we can give you some suggestions if you find your lead engineer to be inhospitable or not available to assist.

 

[Chemitofreak]

Thanks a lot for sparing your valuable time and responding to my query.

I totally agree to you all, one cannot size a filter if one does not know the inlet particle size and concentration.

Since, this a filter used for filtering the rust particles that are accumulated in the piping over the time period, it is hard to specify the inlet characteristics and that is precisely why I posted in this forum.

Anyway, the last option is to have a word with the COMPANY and reach to a consensus.

Thanks once again.






Dinesh S SHELATKAR.
Process Engineer

 

[LittleInch]

This is not an issue which can be defined. If I read you right you are concerned about possible rust from upstream pipes.

All you need to worry about is what is the permissable size of particles downstream, your items 2 & 3.

Just specify a filter element designed to remove 95% of particles of that size. If the operators find themselves cleaning the filters every hour, that's their problem. If it gets really bad then they will fit something better. Either that or the pipes upstream will just break if there is that much rust.

To be honest the situation you describe makes no sense.

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