Mixing Tank 2

[ivanhoe374]

We are including a tank and mixer in a water treatment plant prior to a clarifier to enable thorough mixing of the coagulant. For this particular application we dont believe an inline static mixer is suited.

Just after some input for how large i should make this tank.

The flow rate is 9m3/hr....i guess it is based on how long the water needs to be in the tank for..

 

[bimr]

A good answer will depend on what your process is. For example, is this process for softening, color removal, or high silt loading, etc.? What chemicals are you thinking about adding? Finally, what type of clarifier are you using?

 

[ivanhoe374]

Treating dam water for colour removal mainly.

TSS are low, Turbidity around 5NTU, colour around 15HU.

The Tender has requested Alum dosing. Im assuming other chemicals such as ferric chloride could be used.

The clarifier is a FRP vessel with a 60 degree conical bottom , overflow weir at the top(around the entire circumference of the vessel. We will probably size it at around 5000L to allow for 30 minutes retention time.

 

[ivanhoe374]

If anyone has an experience i would also be interested in your thoughts on using direct inline dosing/mixing followed by direct filtration in a media filter. (ie leaving out the clarifier)

For this particular job we are proposing clarification followed by media/carbon filtration, however on a similar job some tenderers offered this cheaper method.(ie leaving out the clarifier)

The range for the NTU goes as high as 100, colour 200 so surely the clarifier is needed?...just need to convince the client this because as you know the bottom line is what they always look at first.

 

[bimr]

I would suspect that the NTU is not going to be very high if you are taking this from a dam impoundment. The impoundment is going to function as a clarifier to reduce the NTU's.

The most widely used coagulant for color coagulation is aluminum sulfate. The coagulation should be carried out at the optimum pH, which is about 5.5 to 7 for most waters. With ferric coagulants, the effective pH range is lower from 3.5 to 5.5. Ferric will probably not be appropriate.

I agree with you that the static mixer will probably not get the job done. Static mixers operate efficiently only when operating at the design flow rate. You should consider using a mechanical backmix mixer or in-line blender. You need controlled agitation to secure good floc formation. A rapid mixing time of 2 min. will optimize the coagulation process.

Another alternative to consider is to use a reactor clarifier with an internal mixing chamber. I would not use less than 60 min retention in the clarifier unless you decide to go with a gravity plate settler. The gravity plate settler will need a longer flocculation time.

Direct filtration in a media filter is an acceptable method for color removal, as long as the amount of solids and color is not too high. High levels of solids and color will cause excessive head loss and short filter runs.

I have seen waters like this treated with a two-stage direct filtration process. High doses of polymer are added inline with a slow-rate 'working' filter, followed with higher flow rate polishing filters. This setup actually produces a water with less than 1 SDI. The working filters are backwashed daily; the polishing filters are backwashed weekly.

 

[sbush]

Discharge from rapid mixer should empty directly into clarifier. Size rapid mixer for at least 15 seconds to no more than 30 seconds of detention time. Design shape of rapid mixer to accomodate size of propellor and shaft of mixing device selected. Flow path can be either upward or downward. Add baffles to vessel walls to reduce vortex.

Static mixers work as well as mechanical rapid mixers provided that flow rate is relatively constant. I am not aware of any package water plant manufactures who still use rapid mixers.

Omitting the clarifier in favor of direct coagulation/filtration will work if the water quality is under 10 NTU. Adsorption clarifier followed by filtration (either pressure type or gravity type) will work for waters up to 40 NTU. However, in your case, 100 NTU feed water will require at least 30 minutes of flocculation detention time plus a separate settling compartment fitted with either Lamella plates or tube setters loading at under 1.5 gpm per square foot of plan area. There are relatively small package plants available that would be well suited for your purposes.

My experiences with color reduction, to date, are limited to 30 units and less (i.e., straw colored water). The source tends to be tannins and other similar yellow organic acids. I can't imagine what 200 units of color must look like! From my experience, alum and ferric chloride work equally well. Alum is cheaper and less corrosive when handling. The sludge volume is about the same for both.

S. Bush

http://www.water-eg.com

 

[ivanhoe374]

Thanks for your posts bimr and sbush.

The tender is for 7 different water treatment plants.

for the 50%ile the Turbidities range from 1.7-7
10-90%ile " " " " 1-20

50%ile for colour 5-67
10-90%ile " " 0-100


sbush can you elaporate on the "mechanical backmix mixer"

At the moment i have included for a small mixing vessel with an overhead mixer(the pH of raw water will be lowered to around 6 before dosing with Alum). This will overflow directly into the clarifier feed. Following the clarifier i have a multimedia pressure vessel and activated carbon filter (15 minute EBCT as they want removal of algal cells and toxins) then the usual pH correction/disinfection(hypo) before entering the treated water tank.

 

[sbush]

By "mechanical mixer", I'm refering to a motor-driven impeller or propellor; as opposed to "static mixer" with baffles and no moving parts.

Good luck!

S. Bush

http://www.water-eg.com

 

[ivanhoe374]

the "backmix" was the part that confused me

anyhow it sounds like thats what we have allowed for in our design.

What would you go for based on those figures i gave for NTU, HU percentiles above bush?