Blinding Media Filters

[kap44]

We are commissioning our water treatment equipment and have found that the media filters are becoming "blinded" by a substance, probably the chemicals. The TSS of incoming water is 5-10 ppm and we are injecting both polymer (2ppm)and coagulant (3-5ppm) upstream of the filters (we do have a static mixer). Without adding chemicals we get better throughput but SDI is not <5. Any thoughts on how we can improve throughput while maintaining water quality?

 

[aspearin1]

There are a number of methods you might employ. To better advise, though, it would be best if you could describe the filters you are using. Depth type media? Membrane? bags or cartridges and the nominal retention (pore size) of your filters. Without knowing more, my best advice would be to install a pre-filter or screen to pull out the large floc. I assume your flocculant is producing some large visible debris that is getting wrapped around filter surface, and uselessly blinding off non-clogged pores. True?

ChemE, M.E. EIT
"The only constant in life is change." -Bruce Lee

 

[kap44]

The filters have anthracite (top layer where we are experiencing this problem), quartz sand, graded sand, and garnet sand. Raw water enters towards the top and filters its way down through the media layers, at least in theory. At this point, our results do not show significant TSS in the water so it appears to be a chemical dosing problem (the flocs are preventing more water from filtering through). One thought was to dose the coag first and then the polymer downstream of this point.

 

[aspearin1]

Your approach to dosing the polymer downsteam seems valid, providing it does not affect your product quality. If you don't go down this road, I might suggest a disposable bag filter prior to your fixed filter bed. It's much simpler to change out than your bed and will probably extend the life of your bed almost indefinitely compared to what you're used to, and if the proper size is chosen. Most suppliers have bags up to 200 micron. For some perspective, anything the human eye can see is at least 40 microns, so a 50-75 micron bag may pull out most of what is blinding your filter prematurely. Given the nature of your contaminant, (polymer floc), you can probably reuse a significant number of the bags you use.

ChemE, M.E. EIT
"The only constant in life is change." -Bruce Lee

 

[semo]

Have you performed any jar tests to see what concentration of your polymer/coagulant is needed for floc formation? Are both chemicals needed?

What are you trying to remove and what form of treatment is being used prior to filtration?

 

[kap44]

We have performed jar tests and concluded that 1-2 ppm polymer and 3-5 ppm coagulant is required for floc formation. These are injected upstream of the filters at the same location. I am now wondering if the coagulant should be injected first and then the polymer downstream. I believe these chemicals are required to remove excessive TSS but I'm not a chemistry expert.

 

[semo]

It depends a lot on the water chemistry; but, yes, a lot of the time, the coagulant is used during coagulation/sedimentation phase and the polymer is used just ahead of the filter as a filter aid if it is needed. Both are not necessarily needed.

The pH, water temp, etc. can play a big part on the reactiveness of the different coagulant/polymers. Different types of polymers are used for different treatments. The chemical suppliers will typically help perform the jar tests on a new plant at various pH's with different polymers to get the best floc formation/settlibility.

 

[sbush]

You do not need two coagulants. Since you are not trying to settle the floc, polymer is not needed to strengthen the floc to prevent shear. You only need to feed alum and try to feed it ahead of a 90 degree elbow (or two) for rapid mixing. The floc rippens as it passes through the interstices created within the anthracite layer until it becomes too thick and becomes trapped or adsorbed, preferably, before it reaches the fine sand layer.

In this case, since turbidity is so low, jar testing is useless. As a general rule, do not feed more than 1 grain per gallon. If you see a chalky residue on your disc filter paper, you are probably feeding too much alum.

For this direct filtration application, you should be using at least an 18 inch depth (24 inch depth or greater is better) of anthracite with an effective size of 3/16" X 3/32".

The best designs for adsorption clarification/filtration typically use two pressure vessels in series. The first contains the coarse anthracite or coarse sand; followed by the second vessel containing fine anthracite and fine sand. This approach allows you to use the deepest beds possible for longest run times.

Good luck with this challenge.

S. Bush

http://www.water-eg.com

 

[Bigginhk]

As SBUSH says you probably need to lower your coag dose and may not need poly. This is a direct filtration situation. Although I agree that normal jar testing procedures are probably useless it would be useful to dose low doses of coagulant (<1m/l as Al3+ or equivalent) and look for pinhead flocc in a litre sample. The flocc should be barely visible by holding to the light. The flocculation occurs in the bed, however, your problem is the flocc forming too soon, forming large flocc and collecting on the media surface and blinding it. You want to be looking for a steady almost linear headloss buildup on the filter indicating deep penetration into the bed rather than any sudden increases in headloss indicating blinding.

 

[cub3bead]

I would be hesitant to use alum in this case. While kap44 doesn't specifically mention RO in his post, his concern about obtaining a SDI < 5 implies it. Aluminum fouling of the RO membrane is something kap44 doesn't want to experience. I would use a ferric salt as the coagulant since it has a broader pH range than alum an d avoids the aluminum post precipitation problem.