Design of Filter System

[KPW82]

I need help in optimizing the number of filter elements (verticle depth cartridges) used in a filtration process. We have a liquid feed with trace solid contaminant, flowing around 90gpm, continuously through 80 "5 micron" filter elements. The amount of contaminant removed averages 0.03 ppm. The initial level of contaminant is approx 0.0325 ppm.

What equations / formulas should I be using to determine the optimum number of filter elements?

Thanks in advance for any assistance!

KPW82

 

[StoneCold]

KPW82
If you already have 80 depth cartridges then you probably want to keep them. What you need to determine is how long you can run before you have to change out the cartridges. That is probably going to be based on how much pressure drop you can tollerate. If the 80 cartridges was a guess for a design then you need to go back to the cartridge vendor and talk to them about estimated pressure drop with your fluid and work from there. If the fluid was water and each cartridge was a 2.5" diameter 10" long unit then for 90 gpm you would need about 10. That being said you might have to change the cartridges every day. Only you can determine the desired run time and work with the vendor to try to match that up with the quantity of cartridges.

Regards
Stonecold

 

[KPW82]

Thanks StoneCold.

We have been alternating between 80 and 120 cartridges, but because of the huge expense and large lead time of the cartridges, we want to use as few as possible. With 80, we are having to change them once a week. With 120, we are having to change them every 16 days. We want to minimize the number of cartridges, but maximize the amount of time between change-outs. Unfortunately, the fluid we have is more viscous than water, about 1.2 times.

My preference is to run 80 cartridges, for 2 weeks or more before change-out. To my current understanding, the only way to achieve that is to adjust the inlet flowrate, but we don't have to ability to vary that too much (without spending big $ to create a flow control system).

How did you come up with the number 10 if you don't mind my asking?

Thanks again.

KPW82

 

[StoneCold]

KPW82
The idea of using 10 cartridges was based on a flow rate of about 9gpm through the cartridge. But I am obviously assuming a higher pressure drop (maybe 20 psi) across the filter. What is the pressure drop across your filters? Is differential pressure the indicator you use for changing the filters? If it is operating costs that are killing you then maybe you should look at using a filter press or a pressure leaf filter to take up most of your solids and then polishing with cartridge filters. Maybe more information would help here.

Thanks
StoneCold

 

[KPW82]

Ohhh! You typed 90gpm, instead of 9gpm (earlier). Thank you for the clarification though.

The pressure drop we see when new filter cartridges are running is around 2.8 - 3.8 psi. That is normal, although we love to see it closer to 2.8 than 3.8.

We are evaluating adding to or modifying our current filtration process, due to operating costs. We currently run 3 filters in succession, with a cooler just before the last filter, and the 3rd filter is the one that cost us so much to operate, with all the change-outs.

Thanks again.

KPW82

 

[StoneCold]

KPW82
The 9 gpm is through each filter cartridge.
Why don't you move the cooler to after the third group of filters. I would also add a pump if possible so you can get your pressure drop up to a higher level.
Just my opinion.

Regards
StoneCold

 

[shahyar]

For design:
1. Assume 5 gpm per each 2.5"x10" @2 to 20 micron filter.
By doing this calculate the number of TIE (Ten inch Equivalent): TIE= Q/5gpm
2. Then Convert the numbers to YOUR filter size by the below formula:
TIE/N2= 10/L2*(2.5/D2)^2
Where D2 and L2 is YOUR cartridge filter diameter and length.

Hope this help.

 

[koshyeng]

You may want to test 3M filter cartridges. They cost more but operating costs will be lower. I have used these filters in water filtration line. They have high dirt loading capacity than woven or pleated cartridges.

Good Luck

 

[KPW82]

Thank you all for your assistance.

Stonecold:
We have to cool before the filters for consideration of materials of construction of the 3rd stg filter housing. It will destroy the filter if we run at our normal temp, so we cool to preserve our filters.

shahyar:
Thanks for that info. This helps tremendously. Where would I find the reference for those assumptions and calculations? would they be in Perry's? Thanks again!

koshyeng:
I will look into those filters. They may or may not be a answer to our problems. Thank you.



KPW82

 

[koshyeng]

KPW82,

Chemical engineering magazine has a lot of information for young engineers. Experts in this field with a vast experience blend their articles with theory and practical experience.

Good Luck

 

[superdaigle]

Can you clarify what your process fluid is and the type of contaminant being removed?

 

[CH5OH]

have you considered to install a clarifier prior to fine filtration? its a huge investment cost, but with the rate of filter use , may be it can be justified (removal of solids through centrifugal force).

 

[RogerH]

KPW82
I have been forced to backflush cartridge filter in some places beacuse no replacements where available.

That can extend the filter "life" considerably
I usually starts as a no option but when done a few times it usually becomes a rule and there will be a reluctance to change filter at all

In situ is the most common butI have alse seen it done in the workshop

Perhaps not ideal but one can "survive" and a bad filter is better than no filter.
RH

 

[KPW82]

Thanks for all your assistance.

RogerH: We actually do back-flush several times before changing the filter. This still only provides us with a short lifespan between change-outs.

CH5OH: For this process we are unable (and more possibly unwilling) to pursue something of that large a cost for this one problem.

superdaigle: I can't specify the process fluid, but I will say it's an alkaline fluid, and a heavy metal contaminant being removed.

KPW82

 

[Compositepro]

A filter press is often used to recover heavy metal waste from water treatment facilities. It may not be suitable for your flow rates, but you don't have to pay for replacement and disposal of cartridges as hazardous waste.